1655_01.pdf

SECTION 1

Cleaning Agents

© 2001 by CRC Press LLC

CHAPTER 1.1

Overview of Cleaning Agents

Barbara Kanegsberg

CONTENTS

IntroductionAqueous Cleaning Agents

Simple AdditivesCommercial Blends, AqueousWater-Soluble Organics

Solvent-Based Cleaning AgentsClassic Organic SolventsRestricted or Low Availability Ozone-Depleting ChemicalsSolvents on the HorizonOxygenated SolventsEstersHydrocarbon Blend (Mineral Spirits), Soy-Derived Cleaning Agentsd-Limonene Blends, α-Pinene BlendsCyclic Volatile Methyl SiloxanesPerfluorinated Compounds

Solvent Blends, Azeotropes, CosolventsStabilizationExtending the Solvency Range or Moderating the SolvencyCosolventsSurfactantsEmulsions: Macroemulsions, Structured Solvents, MicroemulsionsMystery Mixes

Solvency and Physical Properties; Other ParametersKauri-Butanol NumberWetting IndexOther Physical Properties; Regulatory FeaturesCosts

Overall ConsiderationsReferences


INTRODUCTION

Choosing the cleaning agent appropriate to the job at hand can be a traumatic experi-ence, even for those with some background in chemistry. While the cleaning agent sectionof this book highlights a number of different types of cleaning agents, it is by no meansexhaustive. This chapter provides an overview of cleaning agents and highlights a fewadditional cleaning agents not discussed in other chapters.

Cleaning agents are generally divided into aqueous and solvent. When most peopleinvolved in cleaning refer to the term solvent, they actually mean organic solvent. Organic sol-vents are not those processed from, say, free-range, herbicide-free lemons. Instead, organicrefers to materials that have the element carbon in them. However, when you think about it,both organic and aqueous-based cleaning agents can be thought of as solvents. If you dis-solve sugar in tea, the water is acting as the solvent; the sugar is the solute. Because manyoils are carbon based, organic solvents have been classically used for very heavy degreasingjobs. However, other mechanisms such as saponification (as described in the chapter cov-ering aqueous cleaning agents) have been successfully used to lift heavy grease off parts.Certainly, water-based cleaning is widely used; most of us have successfully removed oilsand greases from dishes using semiautomated aqueous cleaning, not a vapor degreaser.

Although a segment of the industry has traditionally been carried out by aqueouscleaning, until the last 10 to 15 years, most cleaning was conducted using classic chlori-nated solvents or with ozone-depleting compounds (ODCs). In fact, many products weredesigned specifically around the solvency properties of CFC-113 and 1,1,1-trichloroethane(TCA). The loss of ODCs has led to upheaval in the manufacturing world; and, becausethere are no true drop-in substitutes for ozone depleters, the market has fragmented.1 Thisfragmentation has continued for a number of reasons, including:

• Lack of understanding of how to use the new methods• Dissatisfaction with selected new approaches• High cleaning costs• Development of new solvent and aqueous blends• Increasingly stringent and ever-changing regulatory conditions• Differences in regulations in various geographic locations• Increasingly stringent cleaning and performance requirements• product miniaturization

To make matters even more complex, the line between cleaning agent and cleaningequipment or cleaning is often blurred. Sometimes the cleaning agent is generated in use,for example CO2 (solid, liquid, and supercritical) and plasma. Where solid or abrasivemedia2 are used, in a sense, they can also be considered as the cleaning agent.

No one product or class of products is likely to satisfy all cleaning requirements. Thecleaning agent must be matched to the soil, the substrate (the component or part to becleaned), the cleaning requirements, drying requirements, and other performance andenvironmental constraints. Inorganic soils are often referred to as hydrophilic; they dis-solve effectively in water. Organic-based soils, often referred to as hydrophobic, tend to dis-solve more effectively in organic solvents. The choice of cleaning agent should, ideally, bebased on technical considerations. Unfortunately, chemists, engineers, production people,and those involved in regulatory agencies may themselves become hydrophilic orhydrophobic. Although we all have cleaning agent prejudices, irrationally ruling out oneclass of cleaning agents can result in inadequate cleaning and can be economically and eco-logically detrimental.


Another reason to keep an open mind and to try to understand both approaches is thatthe line between aqueous and organic cleaning tends to blur, so even if you and your firmare unalterably devoted to organic solvents, it will be helpful to learn about aqueous clean-ing, and vice versa. Some inert organic cleaners are blended with small amounts of surfac-tants to improve removal of soils. Many aqueous cleaners contain significant amounts oforganic additives. Some may be basically blends of water-soluble organic compounds. Infact, certain similar organic solvent blends may be classified as semi-aqueous or cosolventonly in that very small formulation differences allow for rinsing in water (for semiaqueous)or another solvent (for cosolvent).

AQUEOUS CLEANING AGENTS

Simple Additives

Water removes some soils. With appropriate cleaning action and constant rinsing toremove soils, water alone can clean. However, additives improve performance.

Some blends are relatively simple and are accomplished by in-house blending.Such blends can be particularly desirable where residue is an issue either for the productor for disposal of the spent cleaning agent. Small amounts of peroxide (0.5%) have beenadded to water to clean and remove bacteria. Dilute hypochlorite (bleach) is often usedto prevent biological contamination. Ammonia is often used for simple cleaning whereresidue is of concern. Alcohol and acetone are sometimes added to boost cleaningpower and promote rapid drying. Sodium bicarbonate may be added. Acid washing andacid etching with Piranha, chromic acid, and other strong acids can be thought of as selec-tive cleaning. Although the solutions are simple, process control, process monitoring,employee safety, potential flammability, and environmental regulatory issues must all beconsidered.

Commercial Blends, Aqueous

Commercially available aqueous cleaning agents contain additive blends, often con-sisting of a dizzying array of organic and inorganic compounds. Some additive packagesare totally inorganic; most are a mixture. Although a few companies disclose the additivepackage, more typically, for competition-sensitive issues and other factors, the exact for-mulation is a closely held secret. A few examples of additives are provided in Table 1. Thissummary is provided for several reasons. Well-designed aqueous formulations are com-plex, sophisticated, and are specifically designed to remove certain soils. Also, be awarethat even though aqueous cleaning agents are used in dilute form, they are not formulatedfrom organic carrot juice. As with other cleaning agents, aqueous cleaning agents must beused with understanding and respect.

For general metals cleaning, aqueous formulations with relatively broad range ofacceptability for substrate and soil have been found. However, for most high-precisionapplications the aqueous cleaning agent must be specifically matched to the soil, theexpected soil loading, the substrate, and the expected end use of the product. For example,in some applications, phosphate or silicate residue is not acceptable. In addition, in clean-ing certain metals, notably aluminum, careful selection of the aqueous cleaning agent andprocess is required.


Table 1 Some Additives Used in Aqueous Formulations

Additive Function Description, Examples

Surfactants • Wettability • Single molecule with hydrophilic and• Soil displacement/dispersion hydrophobic portions• Solubilization • May have long chain organic portion,

many carbons in a row• Example: alcohol ethoxylates

Defoamers • Control excessive foam • Poorly soluble in bath at operating• Allow use in high-pressure temperature

spray applications etc. • Impart slight oil-like quality• Usually nonionic surfactants• Example: nonionic block copolymers

Solvents, assorted • Decrease surface tension • Typically soluble in water• Adjust pH • May be VOCs• Improve solubility range • Examples: butyl cellosolve,

pyrollidone,morpholine, glycol ethers,alcohols

Corrosion inhibitors • Prevent corrosion of metals • React with metal surface to reducepassivating reactivity

• Typically oxidizing agents• Examples: chromates, nitrates,

permanganates, chlorates• Some reducing agents, e.g., Na sulfite

Corrosion inhibitors, • Prevent corrosion of metals • Adsorption, formation of protectivenonpassivating films

• Examples: silicates (most common),pyrophosphates, carbonates, amines,gelatin, tannic acid, thiourea

Builders • Promote efficacy of cleaning • Typically saltsby surfactants • Chelating agents

• Sequester water hardness • Examples: sodium tripolyphosphate,• Maintain pH (acidity, sodium hexametaphosphate, sodium

alkalinity) citrate• Decrease metal, lead content • Precipitating builders (e.g.,

of waste stream carbonates)Hydrotropes • Promote solubility of organics • Important with nonorganic surfactant

in presence of high levels of packagesinorganic salts • Examples: toluene sulfonates,

short-chain alcohols, benzoate saltsOxidizers • Corrosion inhibitors • Example: hydrogen peroxide

• Adsorption, dissolution insoils, oxygen release

• Better soil removal

Source: Adapted from Cala, F.R. and A.E. Winston, Handbook of Aqueous Cleaning Technology forElectronic Assemblies, Electrochemical Publications, 1996.

Water-Soluble Organics

Some cleaning agents, nominally referred to as aqueous cleaning agents are primarilyor significantly high in organic solvent blends including long-chain nonlinear alcohols ord-limonene. They can be rinsed with water, or in some cases with either water or solvents.Providing both solvent and aqueous cleaning in the same process has advantages.However, recovery of the waste stream and carryover can become a problem.


SOLVENT-BASED CLEANING AGENTS

As in aqueous cleaning, there are mystery blends. However, it is often easier to identifythe components of solvent-based cleaning agents. With the proliferation of new organic sol-vents, a number of by-products and natural products have been or are now used in cleaningprocesses. Cleaning agents based on orange, pine, cantaloupe, and grapes have been devel-oped—an entire fruit basket of possibilities. Some solvents have been discussed in detail;others are alluded to in discussions of cleaning equipment or of specific cleaning agents.

A few additional solvents and solvent categories are worth noting. In general, one mustconsider that while many of these solvents have been used for years, long-term inhalationtoxicity data may not be available. In addition, blends and azeotropes of both new andwell-established solvents may have their own solvency, compatibility, or toxicity proper-ties. This holds true for organic and water-based cleaning agents. The best advice remainsto test the solvent or blend in the application under consideration and to be conservative—minimize employee exposure and minimize loss of cleaning agent to the environment. Inaddition, compounds with higher boiling points and fairly complex blends with certainadditives may leave undesirable residue, so rinsing is often required.

Classic Organic Solvents

Examples of classic organic solvents include toluene, hexane, heptane, benzene, andxylene. Flammability, worker exposure, air toxics, and company liability issues reduced theuse of these solvents when ODCs were available. They have remained popular for special-ized uses or as blends, particularly for specific, high-value applications. With the decreasein availability of ODCs and an increase in low-flash-point and well-contained cleaning sys-tems, these solvents have enjoyed a resurgence in popularity. They have a wide solvencyrange, but they are particularly oil-like. If they are to be used, appropriate environmentaland safety controls must be employed.

Restricted or Low-Availability Ozone-Depleting Chemicals

Chlorofluorocarbon 113 (CFC 113) and 1,1,1-trichloroethane (TCA) set the standardsfor cleaning for decades. CFC 113 has low to moderate solvency, TCA is an aggressive sol-vent. Both can be used for liquid/vapor phase degreasing, but both have high ozone deple-tion potentials. In the United States, they have been phased out of production. It is stillpossible to obtain recycled material, but at a high cost.

Hydrochlorofluorocarbon 141b (HCFC 141b) has moderate solvency and a fairly lowboiling point. It was first considered as a replacement for other ODCs, but was found tohave an ozone depletion potential similar to that of TCA. It will be phased out of produc-tion. In the United States, HCFC 141b is highly restricted and falls under a usage ban incleaning applications.3

It should be noted in general that regulation and availability of various cleaning agentsvary from country to country and often from city to city. Regulation of Class I ODCs andof HCFC 141b differs in various parts of the world.

Solvents on the Horizon

The landscape of available cleaning agents and processes varies as advances in technol-ogy, requirements of build processes, and regulatory drivers change the perceived appeal of


various options. Because the assortment of products will inevitably change, it is hoped thatthe reader will gain an appreciation not just of the specific cleaning agents, but of theunderlying, commonsense approaches to successful cleaning and contamination control.

Having said this, it is important to mention emerging solvents that are being used todevelop new cleaning agents. For example, Nippon Zeon Co. Ltd. is marketing a hydro-fluorocarbon (HFC) that may have favorable formulation qualities. Two additional HFCs,HFC-365mfc and HFC-245fa, will be discussed in greater detail. Both were initially intro-duced as foam-blowing agents, and most of the information must be extracted from litera-ture geared to the foam industry. However, both may have uses, alone and in blends, asreplacements for HCFC 141b, and for other solvents that may be heavily regulated. HCFC141b is under a federal usage ban for most cleaning applications. There are some excep-tions. For example, HCFC 141b could be used under some conditions as an aerosol wasprepellent. Suffice it to say that there is anecdotal evidence of noticeable continuing use ofHCFC 141b and of a significant number of components manufacturers who, shall we say,must be having severe problems with wasps. Those manufacturers should be aware (1) ofthe general usage restrictions on HCFC-141b as well as (2) the imminent production phase-out of HCFC-141b. These newer solvents may provide viable alternatives to HCFC-141band additional options for other applications. It is important to be aware that while bothmaterials are HFCs, they behave differently from the HFCs you may be more familiar with(described elsewhere in this book by Merchant).

One product is produced and imported to the United States by Solvay.4 It is marketedunder the trade name Solkane® 365mfc5 and it will be sold to other cleaning agent suppli-ers for formulation. DuPont recently began introduction of several blends of HFC 365 withother HFCs and with trans-1,2,-dichloroethylene. Other companies and formulators arealso developing products based on 365mfc. It should be noted that, unlike other HFCs com-monly used as solvents (Merchant), HFC 365 has a very low flash point. In nonflammableblends, and even in azeotropes, there is the potential for flammable mixtures to develop inuse. Therefore end users should work with advisors and with responsible cleaning agentsuppliers to evaluate their own production situation carefully. HFC-245fa is produced byHoneywell6, 7; it was also developed primarily as a foam-blowing agent. There is interest inuse as a solvent for aerosols, and it has potential applicability as a cleaning solvent in someapplications. With a 15°C boiling point, many end users will find HFC-245fa to be ratherevanescent; and it would not be suitable for use in classic, unmodified vapor degreasers.The material will be more suitable for use in newer vapor degreasers with subzero chillingand auxiliary cooling coils. The material is also expected to be usable in many airless andairtight systems and in flushing applications. The very low boiling point can be advanta-geous for deposition and other applications where rapid evaporation is desirable. HFC-245fa appears to have reasonable hydrolytic stability, and good materials compatibilitywith metals and with many plastics. It is miscible with some hydrocarbons, and it showspartial (over 10%) to full miscibility with a number of other solvents, including methanol,isopropyl alcohol, trans-1,2-dichloroethylene, and perfluoropolyethers. It shows good tovery good miscibility with polyolester oils, limited miscibility with fluorinated hydrocar-bon oils, and poor miscibility with mineral oil and silicone oil.

Both HFC-365mfc and HFC-245fa are very mild solvents, similar to the currently morefamiliar HFCs and hydrofluoroethers (HFEs), and costs are expected to be somewhatlower. Of the two, HFC-245fa should have better solubility for polyolester oils. With a boil-ing point of 40°C, HFC-365mfc has a significantly higher boiling point than HFC-245fa.These boiling points may be compared with the 32°C boiling point for HCFC-141b. Thedifferences may not seem significant, but remember that as a rule of thumb chemi-cal processes double in rate with every 10°C increase. For both materials, some plastics


compatibility studies have been conducted. Because synergistic behavior can occur, youwould be well advised to confirm the compatibility of any proposed blend to the specificmix of materials and in the application at hand (time of exposure, temperature, force ofcleaning action). A comparison of HFC-365mfc and HFC-245fa is presented in Table 2.

Oxygenated Solvents

Examples of oxygenated solvents include the short-chain alcohols (methyl alcohol,ethyl alcohol, isopropyl alcohol), methyl ethyl ketone (MEK), methyl isobutyl ketone(MIBK), and acetone. Compared with, for example, hexane or heptane, the addition of oxy-gen makes these compounds more polar, that is, more like water. They are more suited topolar or inorganic soils.

Despite the low flashpoint, some oxygenated compounds, such as isopropyl alcohol(IPA) can be used in appropriately designed vapor degreasing systems. However, it shouldbe noted that oxygenated solvents cannot systematically be used to replace chlorinated sol-vents. After it was delisted as a volatile organic compound (VOC), there has been increasedinterest in acetone. Although acetone has been adopted for some processes, the extremelyrapid evaporation rate, aggressiveness to some plastics, and low flash point have limitedthe extent of substitution of acetone.

Long-chain alcohols such as tetrahydrofurfuryl alcohol (THFA) are used alone and inblends. The longer the carbon chain, the more they become fatlike (able to dissolve oil).However, the alcohol portion confers some waterlike qualities. Often, these alcohols can bepart of aqueous, semiaqueous (rinse with water), or cosolvent (rinse with solvent) blends.

N-Methyl pyrrolidone (N-methyl-2-pyrrolidone, NMP) is a high boiling (295°C), high-flash-point (91°C, 196°F) solvent that is used alone and in blends. Because it is water solu-ble, it can be blended for removal of both rosin and organic acid flux, and it is used inphotoresist systems. Other pyrrolidones are being developed, notably n-octyl pyrollidone(NOP) and n-hydroxy ethyl pyrrolidone (HEP). Used alone and in blends, they may serveto extend the range of cleaning in degreasing. For example, NOP has a longer carbon chain

Table 2 Comparison of Characteristics of HFC-365mfc and HFC-245fa

Characteristic HFC-365mfc HFC-245fa

Structure CF3CH2CF2CH3 CF3CH2CF2HMolecular weight 148 134Boiling point (°C) 40.2 15Flashpoint (°C) Below �27 NoneUEL/LEL (% by volume) 3.8 to 13.3 NoneODP Zero ZeroAtmospheric lifetime 10.8 years Low to moderateVOC status Exempt ExemptSNAP status, solvents Acceptable Not yet submitted for solvent

applicationsToxicology 90-day subacute inhalation 90-day sub-acute inhalation study

study in progress complete; under evaluation byAIHA WEEL committee; suggestedinhalation levels not yet published

UEL/LEL � upper/lower explosion limit; ODP � ozone depletion potential; VOC � volatile organic compound;SNAP � Significant New Alternatives Policy.


and is therefore more oil-like, so it could be used in formulations for degreasing, paintstripping, and deinking of paper. HEP shows promise in photoresist removal.8

Relatively recently, dimethyl sulfoxide (DMSO) has also been used in some photoresistapplications as well as in other processes requiring a fairly aggressive solvent. DMSO-based products are in the process of being developed and tested.

Esters

Various monobasic and dibasic esters and notably lactate esters are used alone and inblends in semiaqueous and cosolvent applications. They have proved particularly effectivein removal of pitches, waxes, and other difficult, mixed soils. The esters have a fairly strongodor. They have high boiling points and must be rinsed in high-precision applications.Esters may also be used in coatings formulations.

Two of the esters, t-butyl acetate (VOC status pending) and methyl acetate, have rela-tively low tropospheric reactivity and may therefore be particularly useful where VOCs arean issue. In precision cleaning of electronics assemblies, t-butyl acetate has been found toprovide a good complement to acetone. t-Butyl acetate has a higher boiling point. Unlikeacetone, it does not dissolve nitrile gloves, and in one test evaluation the assemblers foundthe odor to be acceptable.9

Hydrocarbon Blend (Mineral Spirits), Soy-Derived Cleaning Agents

Hydrocarbon blends (mineral spirits or Stoddard solvent or kerosene) consisting of apetroleum cut of hydrocarbons with a range of molecular weights have been used in clean-ing applications for many years. Given the high boiling point and the potential for con-taminants in some formulations, care must be taken in use and removal, and lot-to-lotvariability may impact process control. It is possible to obtain very pure mineral spiritblends with a narrow, defined range of hydrocarbon chain lengths. Such well-definedhydrocarbon blends are better suited to high-precision applications, both for cold cleaningand, in the appropriately designed system, even for vapor-phase cleaning.

Methyl soyate is a soy-derived substitute for mineral spirits. It is being developed forcleaning applications, and may prove to a renewable-resource alternative to hydrocarbonblends.

d-Limonene Blends, �-Pinene Blends

In addition to d-limonene (citrus derived), which is discussed in more detail elsewherein this book, cleaning agents have been based on �-pinene (pine tree derived). Both havenoticeable odors; both can leave appreciable residue, depending the application, and mustbe rinsed completely. In practice, d-limonene-based cleaners have proved more useful. Aswith many of the esters and other specialty cleaning agents, long-term inhalation exposurestudies have not been conducted.

Cyclic Volatile Methyl Siloxanes

Linear volatile methylsiloxanes (VMSs) have been developed for critical cleaningprocesses and are discussed in Chapter 1.9 by Cull and Swanson. Cyclic VMSs alone and


in blends, have also been used in cold-cleaning applications where higher-boiling solventsare preferred. The cyclic VMSs do not have as favorable a toxicity profile as do linear VMSs.One supplier of cyclic VMS blends (QSOL) indicates a 10-ppm recommended inhalationlevel. Therefore, they should be used in enclosed, well-vented cleaning systems.

Perfluorinated Compounds

Perfluorinated compounds (PFCs) contain fluorine and carbon, but no chlorine orbromine. They are exceedingly mild, inert cleaners that can be used for removal of fluori-nated lubricants and as rinsing and drying agents. PFCs are very effective for particulateremoval. They are not ODCs. They are still sold alone and in various formulations.However, because of the global warming potential associated with a long atmospheric life-time, often ranging in the thousands of years, industry has been under what might betermed strong regulatory encouragement to find substitutes. HFCs and HFEs can replacePFCs in many if not most applications.

SOLVENT BLENDS, AZEOTROPES, COSOLVENTS

Aqueous additives are used for such purposes as improving wettability, improvingsolubililization and removal of soils, compensating for water quality, and preventing cor-rosion. Blends involving solvents are a much more diffuse concept. Solvents are blendedfor a number of reasons. Blends can blur the lines of demarcation among various categoriesof cleaning agents.

Stabilization

Water and acidicity are the enemy of many halogenated solvents. Stabilizer packagesare added to many chlorinated solvents (including TCA) and to n-propyl bromide (nPB) toprevent acid formation, breakdown, and reactivity with metals. Effective stabilization isimportant in degreasing (liquid/vapor cleaning). Stabilization becomes even more chal-lenging in airtight and airless systems because the solvent is reused without replenishmentover a much longer period than in traditional open-top degreasers.

Extending the Solvency Range or Moderating the Solvency

Solvent blends can provide custom-made, fine-tuned cleaning options. Sometimes,blends provide surprises.

Azeotropes are strongly preferred over blends for vapor degreasing applications. Anazeotrope is a constant-boiling mixture of two or more compounds. An extreme exampleof a nonazeotrope blend would be sugar in water. On heating, the water is boiled away, andthe sugar remains. In contrast, in specific proportions, IPA and cyclohexane form a con-stant- boiling azeotrope. This means that the vapors contain both components in a constantproportion that does not change over the life of the blend. Azeotropes have to be managedwith care. Even azeotropes can vary in composition if they are used at a temperature not inthe azeotropic range.

Blends that are not azeotropes will lose various components to evaporation at differentrates. This means that the relative concentrations in the liquid and remaining blend willvary with time. Cleaning capability, compatibility, and flash point can all change. Blends


that are not true azeotropes should be viewed cautiously, particularly in vapor degreasingapplications.

In addition, azeotropes have been known to behave synergistically (nonadditively) interms of performance and compatibility. These properties are not necessarily predicted bysolvency parameters. In other words, while two solvents may each separately show accept-able materials compatibility with a given component, the blend could produce componentdeformation. Therefore, even if one thinks the solvency and compatibility issues of eachcomponent in an azeotrope are understood, it is prudent to test the mixture.

Solvent blends are also used to modify or extend the solvency range in cold cleaningapplications. An aggressive solvent can be toned down and a mild solvent made moreaggressive. For example, HFE has been blended with NPB to tone down the aggressivenessof NPB. VMS may be blended with an alcohol to boost solvency.

Cosolvents

The terminology of cosolvents is a bit confusing. Most often, cosolvents are thought ofas two chemicals used sequentially. However, strictly speaking, any blended solvent couldbe thought of as a cosolvent system. Cosolvents can be blends that are primarily aqueousor primarily solvent. Supercritical or liquid CO2 cleaning can also be accomplished withcosolvents.

Surfactants

Surfactants are used in solvent blends to provide some qualities similar to aqueouscleaners, to change emulsifying qualities, and to allow the solvent to be readily rinsed inwater (as in cosolvent processes). Some blended cleaning agents are offered as similar for-mulations, with or without the surfactant.

Emulsions: Macroemulsions, Structured Solvents, Microemulsions

Oil and water do not mix, except in emulsions when they may coexist in a transient orrelatively permanent form. An oil-and-vinegar salad dressing is typically a transientmacroemulsion. Mayonnaise is a more permanent emulsion. Emulsifying qualities areused in aqueous blends, solvent blends, or both. In aqueous formulations, organic chemi-cals may be part of the mix only under certain conditions, such as temperature. For exam-ple, the separation of the organic phase in an aqueous cleaner at the operating temperaturemay serve to defoam the blend, allowing for spray applications.

In the same way, immiscible organic chemicals may be used as emulsions, transient orpermanent. Transient macroemulsions can be used to transfer the soil from one chemical tothe other; the part is then rinsed in more of the chemical with very low solubility for thesoil in question. Macroemulsions are typically cloudy. Sometimes one of the phases is aque-ous; in other processes both are solvent.

Microemulsions, structured solvents, liquid crystals, or continuous-phase emulsionshave been introduced as cleaning agents. Structured solvents are stable mixtures of organicsolvent, water, and coupling agents. The continuous phase may be solvent or water. Theyappear clear and may be primarily water or primarily solvent. Structured solvents can bemade to separate during the application process. Such products are useful for mixed soilswhere both solvent and waterlike characteristics are desirable.10


Mystery Mixes

Blended solvents, particularly the high-boiling blends involving hydrocarbons, esters,and nonlinear alcohols, greatly extend the specificity of cleaning that can be obtained.However, blended solvents can be particularly difficult for the components manufacturerto evaluate. As with aqueous cleaning agents, many manufacturers consider the formula-tions to be highly proprietary and competition sensitive.

Many of the comments regarding mystery mixes apply to both aqueous- and non-aqueous-based formulations. Some areas of concern in using complex mystery mixesinclude unexpected compatibility issues, regulatory constraints on one or more compo-nent, and unscheduled formulation changes. In such cases, it may be desirable to set upconfidentiality agreements so that the ingredients are understood in detail. At the veryleast, particularly where the product is used in a process requiring high levels of validationand testing, it is prudent to obtain an agreement with the vendor that the product will notbe changed.

In addition, supposed improvements in formulations can have unintended conse-quences. This author has observed many instances where a blended product was“improved” in such a manner as to impact the process adversely.

SOLVENCY AND PHYSICAL PROPERTIES; OTHER PARAMETERS

Kauri-Butanol Number

A number of solvency systems have been described in Chapter 1.2 by J. Burke. In addi-tion, other solvency systems are in use. One cloud-point test, the Kauri-butanol (KB) num-ber, is often referred to. The KB number is determined by the volume of solvent requiredto produce a defined degree of turbidity when added to standard solutions of Kauri resinin n-butyl alcohol. As a general rule, the higher the value, the stronger the solvent. The sys-tem was developed to indicate the relative solvent power of hydrocarbons, and it is notvalid for oxygenated solvents. The KB number should be considered along with the boil-ing point, because, if the solvent can be heated to higher temperatures, more entropy isintroduced into the system and better solvency occurs. Estimating solvency by mixing acleaning agent with t-butyl alcohol and tree sap is a rather unsophisticated approach.However, the KB number remains widely used, and it is somewhat predictive of solvency.11

Table 3 lists the KB number and boiling points of several representative cleaning agents.

Wetting Index

The wetting index has been used as a guideline to the ability of a cleaning agent to pen-etrate closely spaced components. The wetting index is directly proportional to the densityand inversely proportional to the surface tension and viscosity. In general, many of thevapor degreasing solvents have a higher wetting index than water or hydrocarbon blends.As with other indications, however, wetting index alone does not determine efficacy ofcleaning. The wetting index of a few common cleaning agents are provided in Table 4.12

Other Physical Properties; Regulatory Features

Physical properties of solvents constrain the range of choices. Other physical proper-ties such as boiling point, flash point, and evaporation rate must be considered in choosing


Table 3 KB Number and Boiling Point (BP), Representative Cleaning Agents

Cleaning Agent KB No. BP,°C

CFC-113 32 481,1,1-Trichloroethane 124 74HCFC 141b 56 32Methylene chloride 136 40Trichloroethylene 129 87n-Propyl bromide 125 71d-Limonene 68 150Parachlorobenzotrifluoride (PCBTF) 64 139HCFC 225 31 54HFC 43–10 9 55HFC 43–10 blend including trans-1,2,-dichloroethylene 30 37HFE 569sf2 10 76VMS OS-10 17 100

Table 4 Examples, Wetting Index

Density, Surface Tension, Viscosity, WettingCleaning Agent g/cm3 dyne/cm cp Index

Generally desirable High Low Low HighCFC 113 1.48 27.4 0.70 121TCA 1.32 25.9 0.79 65IPA 0.785 21.7 2.4 15NPB 1.33 25.9 0.49 105HCFC 225 1.40 16.8 0.61 145HFE 449 sl 1.52 14 0.6 181Hydrocarbon blend 0.84 27 2.8 11Water 0.997 72.8 1.00 14Saponifier solution, 6% 0.998 29.7 1.08 31aqueous

a solvent, and the solvent must be considered in the particular regulatory microclimatewhere the process is being carried out. Tables 5a and 5b list some physical properties ofsome commonly used solvents, and a few regulatory considerations.13

The concept of a VOC-exempt solvent refers to the U.S. federal regulatory designationsat the time of writing. Solvents that are VOC exempt are judged to have negligible reactiv-ity relative to ethane. Many solvents, including those that are VOC exempt, can be used invapor-phase cleaning applications. Those with low flash points, however, must be used inspecially designed equipment. Such equipment has a high initial capital cost. Many sol-vents do not have a flash point but do have an upper explosion level (UEL) and a lowerexplosion level (LEL); this must be considered in specialized operations and in selectingand maintaining emission control equipment.

The boiling point must be high enough to allow efficient cleaning, but not so high as todamage materials of construction or slow the build cycle. A very high boiling point maypreclude use of the solvent in a standard vapor-phase degreasing operation. The evapora-tion rate must be sufficiently rapid to allow rapid drying, but not so rapid that the solventis immediately lost. These considerations are all relative to the operation in question.


Table 5a Physical Properties, VOC, ODC Status

Comments,Evap. Rate

Boiling Point, (ref. for evapCleaning Agent °C (°F) Flash Point UEL/LEL, % rate)

1,1,1-Trichloroethane 74 (165) None 15/7.0 5 (buac � 1)a

(ODC)CFC-113 48 (118) None (TOC) NA 0.45

(buac � 1)HCFC-141b 32 (90) None 17.7/7.6 � 1 (ether � 1)Stoddard solvent, 152 (305) 40.6 (106) 6.1/1/1 Hydrocarbontypical (hydrocarbon blend, VOCblend) (VOC)n-Propyl bromide 71 (160) None 8/3 4.5 (buac � 1)(VOC)Methylene chloride, 40 (104) NA 19/12 NAVOC-exempthazardous airpollutantPerchloroethylene, 121 (250)? None (TCC) None 2.1 (buac � 1)VOC-exempthazardous airpollutantHCFC 225, 54 (130) None None 0.9 (ether � 1)VOC exemptHFE 569sf2, 76 (169) None NA —VOC exempt (TCC, TOC)HFE 449s1, 61 (142) None NA —VOC exempt (TCC, TOC)HFC 43-10mee, 55 (131) None (TOC) NA —VOC exemptWater 100 (212) None None —

Note: These data were obtained from various standard publicly available references, primarily MSDS from theCornell University Program Design Construction Web site (http://msds.pdc.cornell.edu/ISSEARCH/MSDSsrch.htm); University of Vermont Web site, with some confirmation by Lange’s Handbook of Chemistry,13th ed. McGraw-Hill, New York, and Dangerous Properties of Industrial Materials, 3rd ed., N. Irving Sax,Reinhold Book Corp). They should be used as guidelines only—the evaporation rate data in particular is proneto inconsistency among references. Boiling points rounded to nearest integer. Please confirm all information withcurrent MSDS.abuac � butyl acetate; NA = not available.

Costs

Costs are relative. Few solvents are inexpensive, particularly if total process costs14 areconsidered. The most important considerations are cleaning agent quality and productstewardship by the chemical supplier.15,16

In terms of organic solvents, pound per pound, traditional solvents such as IPA, ace-tone, and the chlorinated solvents are relatively inexpensive. nPB and the VMSs are mod-erately priced, and the engineered solvents (HCFC 225, HFEs, and HFCs) are the mostcostly.

Blended high-boiling solvents can vary markedly in price. The costs may be perceived


http://msds.pdc.cornell.edu/msdssrch.asp


Table 5b Physical Properties, VOC, ODC Status, Low-Flash-Point Solvents

Comments,Evap. Rate

Boiling Point, (ref. for evapCleaning Agent °C (°F) Flash Point UEL/LEL, % rate)

Methyl acetate, 55.8–58.2 �13 (9) 16/3.1 RecentlyVOC exempt (132–137) exempt, 5.3

(buac � 1)t-Butyl acetate, 98 (208) 15 (59) NA/1.5 Recentlyproposed, VOC proposed, VOCexempt exemptionpara-Chlorobenzo 139 (282) 43 (109) 10.5/0.9trifluoride, VOCexemptDi-siloxane VMS, 100 (212) �3 (27) TCC 18.6/1.25 Dow VMS OS-VOC exempt 10, 3.8Tri-siloxane, 152 (306) 34 (94) TCC 13.8/0.9 Dow VMS-OS-VOC exempt 20, 0.7Cyclo-tetrasiloxane, 205 (401) 76 (170) TCC Dow VMS-OS-VOC exempt 245, not calc.Acetone, 56 (134) �20 ( �4) 13/2.5 6VOC exempt

Note: These data were obtained from various standard publicly available references, primarily MSDS from theCornell University Program Design Construction Web site (http://msds.pdc.cornell.edu/ISSEARCH/MSDSsrch.htm), University of Vermont Web site, with some confirmation by Lange’s Handbook of Chemistry,13th ed., (McGraw-Hill, New York), and Dangerous Properties of Industrial Materials, 3rd ed., N. Irving Sax,Reinhold Book Corp. They should be used as guidelines only—the evaporation rate data in particular is prone toinconsistency among references. Boiling points rounded to nearest integer. Please confirm all information withcurrent MSDS.abuac � butyl acetate; NA = not available

as high in applications where soil loading is a problem and frequent solvent change-out isrequired. With heavy soil loading, it may be more effective to perform initial cleaning in arelatively inexpensive product, and then to conduct subsequent steps in the more sophis-ticated cleaning agent.

Aqueous cleaning agents must be compared against each other in the intended appli-cation. Assume that two concentrates are under consideration and that one is twice ascostly as the other. If the inexpensive cleaning concentrate must be used at a 1:4 dilutionwhile the other provides equivalent performance at a 1:20 dilution, the picture changes.15 –16

Filtration17 markedly influences bath life and therefore modifies the overall cost of thecleaning agent.

OVERALL CONSIDERATIONS

One of the problems in developing a manufacturing process is the rather daunting listof considerations and provisos. To cope with the problem, there is the tendency to think lin-early and to attempt to find the perfect cleaning agent. There is no perfect cleaning agent.However, we persist in our search for unattainable perfection.

All too often, when a cleaning process is being developed, a cleaning agent selectioncommittee is established to screen out all undesirable applicants. The safety/environmen-tal group is likely to rule out any environmentally challenged cleaning agent, even if it




Table 6 Overall Considerations, Choosing Cleaning Agents

Factor Process Consideration

Cleaning properties, • Cleaning requirements of your process (how clean is clean enough)cleaning performance • Performance under actual process conditions

• Solubility characteristics relative to soil of interest• Wetting ability• Boiling point• Evaporation rate• Soil loading capacity• Ability to be filtered• Ability to be redistilled

Materials compatibility • Compatibility under actual process conditions (temperature, time ofexposure)

• Product deformation at cleaning, rinsing, drying temperaturesResidue • Nonvolatile residue (NVR) level

• Rinsing requirement• Process time impact

Cycle time • Cleaning• Rinsing• Drying• Product cool-down• Component fixturing• Loading and unloading equipment• Product rework

Cleaning equipment • Suitability with current cleaning equipment• Ability, costs of retrofit• Costs of new cleaning equipment• Auxiliary equipment required• Maintenance, repair• Automation, component handling• Footprint (length, width, height)• Equipment weight• Component fixturing (continued )

could be used nonemissively. Company management and sometimes the customer maysubmit a series of “don’t” lists. Whole classes of cleaning agents may be ruled out as beingunacceptable on general environmental principles. The materials and process chemistsmay insist that for any cleaning agent to be considered, it should be able to be in contactwith all materials of construction at some elevated temperature for, say, 24 hours. The pur-chasing department may insist that only one or two cleaning agents be selected—period.The manufacturing engineers may insist on an extremely rapid process time, instant dry-ing, and aqueous cleaning. What’s left? Sometimes nothing, sometimes a class of cleaningagents that is totally unsuited to the cleaning application at hand.

Perfection aside, for nearly every cleaning application, there are several workable solu-tions. Some of the considerations in cleaning agent selection are indicated in Table 6. Thecleaning agent has to be considered in the context of the cleaning process and, indeed, inthe context of the overall manufacturing process. The factors indicated in Table 6 are meantto provide a starting point. It usually becomes very apparent which factors are the mostimportant in a given manufacturing situation.

It is more productive to proceed with a nonlinear approach that considers perform-ance, costs, cleaning agent, cleaning equipment, suitability to the workforce, worker safety,and the local regulatory microclimate.


Table 6 Overall Considerations, Choosing Cleaning Agents (continued )


Flash point • Choice of cleaning equipment• Process control• Control of proximal processes, activities• Choice of auxiliary equipment• Choice of emissions control

Toxicity • Acute• Long term• Anticipated exposure under process conditions (including sprays,

mists)• Employee monitoring• Inhalation• Skin adsorption

Worker acceptance • Method of application• Drying speed• Similarity to current process• Automation• Computer skills• Perceived loss of control of process• Odor

Cleaning agent • Water preparationmanagement • In-process filtration (water, organic, or aqueous cleaning agent)

• Wastewater disposal• On board redistillation

Regulatory, air • Global, national, local (ODC, VOC, HAPs, GWP)• Neighborhood concerns• Environmental justice issues• Production phaseout• Usage bans• Disposal of waste stream

Regulatory, water • Global, national, local• Disposal of waste streams

Company, customer, • Contractual requirements, restrictionsproduct performance • Company policyrequirements • Testing, acceptance qualification required

• In-house safety, environmental policy• Insurance company issues

Costs • Cleaning agent• Cleaning agent preparation and disposal• Costs as-used (dilution)• Capital equipment• Disposables• Sample handling• Total process time• Rework• Insurance• Regulatory permitting• Process qualification• Employee education, training• Process monitoring


Table 6 Overall Considerations, Choosing Cleaning Agents


Supplier stewardship, • Responsive distributor, suppliercleaning equipment • Supportive technical staffsupplier • Clear, understandable MSDS

• Provides required technical information• Provides required regulatory information• Supports process development

HAP � hazardous air pollutant; GWP � global-warming potential; MSDS � Material Safety Data Sheet.

REFERENCES

1. Kanegsberg, B., Precision cleaning without ozone depleting Chem. Ind., 20, 787–791, 1996.2. Kanegsberg E. and B. Kanegsberg, Cleaning by abrasive impact A2C2 Mag, May 2000.3. Dibble, C., EPA SNAP program update: solvent cleaning, presented at Nepcon West 2000,

February 29, 2000, Anaheim, CA.4. Information provided by K. Neugebauer, V.P. Specialty Fluorides, Solvay Fluorides, Inc.5. Zipfel, L. and P. Dournel, HFC-365mfc, the key for high performance rigid polyurethane foams,

presented at UTECH 2000, The Hague, The Netherlands, April 2000.6. Information provided by G. Knopeck, Manager, Fluorocarbon Technical Services, Honeywell

International.7. Knopeck, G., Pentafluoropropane: an HFC solvent for aerosols, presented at SATA (Southern

Aerosol Technology Association) Spring Meeting, Atlanta, GA, April 2000.8. Waldrop, M.W., BASF, communications and technical summaries, January 2000.9. Elias, W.G., Real-life applications with environmentally compliant solvents for electronics, Proc.,

Nepcon West 2000, Anaheim, CA.10. Shick, R.A., Formulating cleaners with structured solvents, in Proc., Precision Cleaning 96,

Anaheim, C.A., 1996, 285–289.11. Kenyon, W.G. and B. Kanegsberg, Accelerating the Change to Environmentally Preferred, Cost-

Effective Cleaning Processes, Tutorial with Precision Cleaning ‘95, Chicago, May 1995.12. Kenyon, W.G., Wetting Index, personal communication.13. Kanegsberg, B., Economic and environmental costs of process conversion, presented at Nepcon

West 2000, February 29, Anaheim, CA, 2000.14. Kanegsberg, B. and C. LeBlanc, The cost of process conversion, (in Report to Toxic Use Reduction

Institute, B. Kanegsberg Proc., CleanTech‘99, Rosemont, IL, May 1999.15. O’Neill, E., A. Miremadi, R. Romo, M. Shub, and B. Kanegsberg, Four steps to process conver-

sion, Parts Cleaning Mag., May 2000.16. O’Neill, E., A. Miremadi, A. Guzman, R. Romo, M. Shub, and B. Kanegsberg, Simplifying aque-

ous cleaning, the value of practical experience, Products Finishing Mag., August 2000.17. Kanegsberg, E., Liquid filtration in critical cleaning, A2C2 Mag., April 2000.


CHAPTER 1.2

Solvents and Solubility

John Burke

CONTENTS

IntroductionSolutionsMolecular AttractionsCohesive EnergySolubility ParametersThree Component ParametersThe Teas GraphVisualizing SolubilitySolvent MixturesReferences

INTRODUCTION

Solvents are ubiquitous. Not a day goes by when we don’t rely on one solvent oranother, to accomplish some essential task. Given such frequent experience with solubility,we might conclude that we all would have a pretty good idea of how solvents work. Andyet, who among us hasn’t tried in vain to remove one substance from another, guided byrules of thumb such as “like dissolves like” or vague concepts of solvent “strength.” Whilethis approach may often succeed, it also might be risky if, for example, we needed to dis-solve one material selectively while leaving other materials completely unaffected. Or, itwould be clearly inefficient if, at the same time, we were also trying to control evaporationrates, solution viscosity, material costs, or environmental and health effects.

The selection of a solvent or solvent mixture in the face of complex criteria movesbeyond trial and error, and of necessity must rely on a system that can organize and pre-dict solubility behavior. While this could be accomplished empirically, by simply testingthe effects of specific solvents on specific materials, a universal system that could encom-pass solubility behavior in general would be immensely useful. Although understandingsuch a system may seem dauntingly complex, the practical application of solubility theoryis actually quite straightforward. In fact, many solubility interactions can be predicted ona simple triangular graph. But before we can accomplish this feat, a certain amount of boththeoretical and historical background is in order.


SOLUTIONS

When two liquids are mixed together, simply stated, they either stay together or theydon’t. They may stay together because of a chemical reaction that changes the liquids intosome other compound, usually by an exchange of electrons (in which case it may be diffi-cult to retrieve the original materials). Fortunately, for the purposes of this discussion, wecan ignore solutions of this kind, since ionic and even water-based solutions are beyond thescope of simple solubility. Another reason a mixture may stay together is because the twomaterials are mutually soluble, such as gin and tonic. Simple solubility implies that theindividual materials remain essentially unchanged even while mixed, and can usually beseparated with some ease, say, by distillation. Solutions of this kind usually include sol-vents and are easier to characterize and predict.

But to say this kind of interaction is simple may be misleading. Not just any two orthree solvents may be successfully combined. When asked why oil and water do not mix,most people will reply that the oil is “lighter” than the water. Yet their mutual repulsion isnot at all due to gravity. And at the limits of solubility solutions may become clear, cloudy,or separate out again after the slightest change in concentration. How can these seeminglycomplex behaviors be accounted for? Actually, dissolving a thing is very similar to evapo-rating it—but to understand that remark we first need to know a bit of what happens onthe surfaces of molecules.

MOLECULAR ATTRACTIONS

Liquids and solids differ from gases basically because their molecules stick together,resisting the tendency to evaporate completely into space. This must mean that the mole-cules that make up liquids and solids are somehow attracted to each other, that there issome kind of intermolecular stickiness that prevents them from flying apart into a gas.

These sticky forces between molecules were first described by Johannes van der Waalsin 1873, and thus bear his name. Originally thought to be small gravitational attractions,van der Waals forces are actually caused by electromagnetic interactions between mole-cules.

The outer shell of an atom is composed entirely of a cloud of negatively charged elec-trons, completely enclosing the positively charged nucleus within. Molecules, because theyare composed of atoms, are also covered by a cloud of electrons. In both cases we can imag-ine a positively charged center surrounded by a negatively charged outer shell. These pos-itive and negative charges essentially balance out, with the result that the atom or moleculeas a whole is neutral.

It’s easy to see that the negatively charged surfaces of adjacent molecules should repeleach other like the negative poles of two magnets (and it’s a good thing too, or the universewould collapse), but we were speaking earlier of intermolecular stickiness. How can weaccount for this anomaly?

In reality, the electron cloud is never evenly distributed around a molecule. A singlemolecule, because of its structure, can exhibit a variety of electromagnetic charges on itssurface, some strong and some weak, some which cancel out, and some which reinforceeach other. The resulting sum of all the charges is what is known as the dipole moment ofthe molecule. Molecules that have permanent dipole moments are said to be polar, whilemolecules in which all the dipoles cancel out (zero dipole moment) are said to be nonpolar.

Some atomic elements attract electrons more vigorously than others, causing the elec-trons to be unequally shared between the individual atoms in a molecule. If the molecule


is symmetrical, these charges may cancel out. If, on the other hand, the electron density ispermanently imbalanced, with some atoms in the molecule harboring a greater share of thenegative charge distribution, the molecule itself will be polar. The polarity of a molecule isrelated to its atomic composition, its geometry, and its size.

A particularly strong type of polarity, for example, occurs in molecules where a hydro-gen atom is attached to an extremely electron-hungry atom such as oxygen, nitrogen, orfluorine. In these cases, the sole electron of hydrogen is drawn toward the electronegativeatom, leaving the strongly charged hydrogen nucleus exposed. In this state the exposedpositive nucleus can exert a considerable attraction on electrons in other molecules, form-ing what is called a protonic bridge that is substantially stronger than most other types ofintermolecular attractions. This special kind of polar attraction between molecules is calledhydrogen bonding, and plays a major role in solubility as we shall see.

But what about nonpolar molecules, where electron clouds are evenly distributed? Thesource of electromagnetic attractions in this case stems from the random movement of theelectron cloud. From instant to instant, random changes in electron distribution give rise topolar fluctuations that shift about the molecular surface. Since the distribution of the elec-tron cloud is uneven (maybe thicker in one place and thinner in another), small local chargeimbalances are created. The parts of the molecule with a greater electron density will bemore negatively charged, and the electron-deficient parts will be more positively charged.Although no permanent polar configuration is formed and the molecule is essentially non-polar, numerous temporary poles are created constantly, move about, and disappear.

When two molecules are in proximity, the random polarities in each molecule tend toinduce corresponding polarities in one another, causing the molecules to fluctuate together.This allows the electrons of one molecule to be temporarily attracted to the nucleus of theother, and vice versa, resulting in a play of attractions between the molecules. Theseinduced attractions are called dispersion forces.

The degree of dispersion forces that these temporary dipoles confer on a molecule isrelated to surface area: the larger the molecule, the greater the number of temporarydipoles, and the greater the intermolecular attractions. Molecules with straight chains havemore surface area, and thus greater dispersion forces, than branched-chain molecules ofthe same molecular weight.

Deviations in electron shell density, therefore, result in minute magnetic imbalances, sothat each molecule as a whole becomes a small magnet, or dipole. These electron densitydeviations depend on the physical architecture of the molecule. Certain molecular geome-tries will be strongly polar. Other molecules may be nonpolar, but are still capable of dis-persion forces. It is these electromagnetic effects that account for the intermolecularstickiness holding liquids and solids together.

Now we can explore the intriguing relationship among vaporization, van der Waalsforces, and solubility.

COHESIVE ENERGY

To bring a liquid to its boiling point, we usually add energy in the form of heat. Thisheating raises the temperature of the liquid until it finally begins to boil. Once the liquidreaches its boiling point, however, the temperature of the liquid will not continue toincrease. Any subsequent heat added is used up in continuing the boiling and separatingthe molecules of the liquid into a gas. Only when all the liquid has been completely vapor-ized will the temperature again begin to rise.

If we measure the amount of energy that we add between the moment that boiling startsand the point when all the liquid has boiled away, we will have a direct indication of the


amount of energy required to separate that amount of liquid into a gas. Interestingly, thisis also a measure of the amount of van der Waals forces that held the molecules of that liq-uid together.

It is important to note here that the temperature at which the liquid begins to boil is notimportant, but rather the amount of heat that has to be added to separate the molecules. Aliquid with a low boiling point may require considerable energy to vaporize, while a liquidwith a higher boiling point may vaporize quite readily, or vice versa. Regardless of the tem-perature at which boiling begins, the liquid that vaporizes readily has fewer intermolecu-lar attractions than the liquid that requires considerable addition of heat to vaporize. Theenergy required to vaporize the liquid is called, not surprisingly, the heat of vaporization,and reflects the attractions that exist between its molecules.

Here is the connection we have been looking for: vaporization and solubility are simi-lar because the same intermolecular attractive forces have to be overcome to vaporize a liq-uid as to dissolve it. This can be understood by thinking about what happens when twoliquids are mixed. The molecules of each liquid have to be physically separated by the mol-ecules of the other liquid. For any solution to occur, solvent molecules must be separatedfrom each other to penetrate between the molecules of the solute. At the same time, thesolute molecules must also overcome their own intermolecular stickiness to allow solventmolecules between and around them. This is similar to the molecular separations that needto occur during vaporization. The same intermolecular van der Waals forces must be over-come in both cases.

So it stands to reason that for two materials to be soluble in each other their internalenergies must be similar. The molecules of a polar liquid (such as water) with strong inter-molecular attractions just will not be separated by the molecules of a nonpolar liquid (suchas oil) that are held together only by weak dispersion forces.

SOLUBILITY PARAMETERS

If we wanted to give a number to these attractions independent of temperature (basi-cally to put everything on an even playing field), we can derive a value called the cohesiveenergy density from the heat of vaporization by the following formula:

c � ��H

V�

m

RT� (1)

wherec � cohesive energy density

H � heat of vaporizationR � gas constantT � temperature

Vm � molar volume

In 1936, Joel H. Hildebrand, in his landmark book on the solubility of nonelectrolytes, pro-posed the square root of the cohesive energy density as a numerical value indicating thesolvency behavior of a specific solvent.

δ � �c� � ��HV�

m

RT��

1/2(2)

It was not until the third edition in 1950 that the term solubility parameter was proposedfor this value, represented by the symbol δ. Subsequent authors have proposed that the


Table 1 Hildebrand Solubility Parameters

Solvent Parameters

n-Pentane (7.0)n-Hexane 7.24Freon ® TF 7.25n-Heptane (7.4)Diethyl ether 7.62Cyclohexane 8.18Amyl acetate (8.5)1,1,1-Trichloroethane 8.57Carbon tetrachloride 8.65Xylene 8.85Toluene 8.91Ethyl acetate 9.10Benzene 9.15Chloroform 9.21Trichloroethylene 9.28Tetrahydrofuran 9.52Cellosolve acetate 9.60Acetone 9.77Ethylene dichloride 9.76Methylene chloride 9.93Diacetone alcohol 10.18Butyl Cellosolve 10.24Morpholine 10.52Pyridine 10.61Cellosolve 11.88n-Butyl alcohol 11.30Ethyl alcohol 12.92Dimethyl sulfoxide 12.93n-Propyl alcohol 11.97Dimethylformamide 12.14Methyl alcohol 14.28Propylene glycol 14.80Ethylene glycol 16.30Glycerol 21.10Water 23.5

Sources: Hildebrand values from Hansen.6 Values inparenthesis from Crowley et al.3

term hildebrands be adopted for these units, in recognition of Dr. Hildebrand’s contribu-tion. Table 1 lists solvents arranged according to their solubility parameter.

In looking over a table of Hildebrand solubility parameters, it becomes apparent thatby ranking solvents according to solubility parameter a solvent spectrum is obtained, withsolvents occupying positions in proximity to other solvents of comparable “strength.” Forexample, if acetone dissolves a particular material, it may likely be soluble in neighboringsolvents, like diacetone alcohol or methyl ethyl ketone, since these solvents have similarinternal energies. It may not be possible to achieve solutions in solvents farther from ace-tone on the chart, such as ethyl alcohol or cyclohexane—liquids with very different inter-nal energies. Theoretically, there will be a contiguous group of solvents that will dissolve aparticular material, while the rest of the solvents in the spectrum will not. Some materials


will dissolve in a large range of solvents, while others might be soluble in only a few. Amaterial that cannot be dissolved at all, such as a thermosetting resin, might exhibitswelling behavior in precisely the same way.

Another interesting aspect of the solvent spectrum is that the Hildebrand value of amixture can be determined by averaging the values of the individual solvents by volume.For example, a mixture of two parts toluene and one part acetone will have a Hildebrandvalue of 18.7 (18.3 � 2/3 � 19.7 � 1/3), about the same as chloroform. Theoretically, sucha 2:1 toluene/acetone mixture should behave similarly to chloroform. Thus, for example, ifa resin was soluble in one, it would probably be soluble in the other. What’s attractive aboutthis approach is that it attempts to predict the properties of a mixture using only the prop-erties of its components. No empirical information about the mixture is required.

But, as you might expect, this is not completely accurate. Figure 1 plots the swellingbehavior of a dried linseed oil film in various solvents arranged according to Hildebrandnumber. Of the solvents listed, chloroform swells the film to the greatest degree, about sixtimes as much as ethylene dichloride, and over ten times as much as toluene. Solvents withgreater differences in Hildebrand value have less swelling effect, and the range of peakswelling occupies less than 2 Hildebrand units. Theoretically, we would expect any solventor solvent mixture with a Hildebrand value between 19 and 20 to swell a linseed oil filmseverely.

But careful examination of the graph reveals an anomaly. Two solvents withHildebrand values right in the middle of the severe swelling range, methyl ethyl ketone(19.3) and acetone (19.7), actually cause very little swelling behavior. How can this be?According to the theory, liquids with similar cohesive energy densities should have simi-lar solubility characteristics, and yet actual behavior in this instance does not bear this out.

To understand the reason for this discrepancy we need to recall our previous discus-sion about van der Waals forces. Remember the point about different molecular architec-tures giving rise to different kinds of polarity? Some materials are made of molecules that

Figure 1 The swelling behavior of linseed oil films in solvents arranged according to solubilityparameter. (Adapted from Feller et al.4)


are strongly polar, especially molecules capable of hydrogen bonding. Other materials aremade of molecules that are essentially nonpolar, with intermolecular attractions dueentirely to dispersion forces. It is precisely these differences in polarity that also must betaken into account if we want to improve our solubility predictions.

The inconsistencies in Figure 1 stem from a difference in hydrogen bonding betweenchlorinated solvents and ketones. The intermolecular forces in linseed oil are primarily dueto dispersion forces, with practically no hydrogen bonding involved. These polar configu-rations are perfectly matched by the intermolecular forces between chloroform molecules,thus encouraging interpenetration and swelling of the linseed oil polymer. Acetone andmethyl ethyl ketone, however, are more polar molecules with moderate hydrogen bondingcapabilities. Even though the total cohesive energy densities, and therefore Hildebrand sol-ubility parameters, are similar in all four solvents, the differences in polarity, primarilyhydrogen bonding, lead to differences in solubility behavior.

A scheme to overcome the inconsistencies caused by hydrogen bonding was proposedby Harry Burrell in 1955.2 His solution was to segregate the solvent spectrum into three sep-arate lists depending on hydrogen bonding capability. This is briefly summarized asfollows:

1. Weak hydrogen bonding: Hydrocarbons, including chlorinated and nitrocom-pounds;

2. Moderate hydrogen bonding: Ketones, esters, ethers, and glycol monoethers;3. Strong hydrogen bonding: Alcohols, amines, acids, amides, and aldehydes.

Accordingly, solvents with similar Hildebrand numbers and similar polarity (especiallyhydrogen bonding) should exhibit similar solubility. This system of classification doesin fact improve the prediction of solvent behavior, and is still widely used in practicalapplications.

THREE COMPONENT PARAMETERS

But it was only a matter of time before researchers needing even greater precision inquantifying solvent behavior began to look at the cumulative effects of several differentkinds of van der Waals forces. It was found that by using separate values for dispersionforces, polarity, and hydrogen bonding even greater accuracy was achievable. But the addi-tion of a third value created practical problems in portraying this information, since it couldnot be presented in a simple list. For this reason most three-component parameter systemswere represented as slices through three-dimensional solubility models. Unfortunately,this made developing formulations using solvents outside the slice impractical.

The most widely accepted three-component system was developed by Charles M.Hansen in 1966. Hansen took the existing three values farther by relating all three to thetotal Hildebrand value. This was done by first getting the dispersion parameter of a solventfrom the Hildebrand value of the nonpolar molecule most closely resembling it in size andstructure (as n-butane would be to n-butyl alcohol). Hansen then used trial-and-errorexperimentation on numerous solvents and polymers to separate the polar value into polarand hydrogen bonding component parameters best reflecting empirical evidence.

Hansen also used three-dimensional models but also found that, by doubling the dis-persion parameter axis, an approximately spherical volume of solubility could be formed.This volume, being spherical, could then be described in a simple manner: the coordinatesat the center of the solubility sphere could be located by means of the three component


parameters, along with the radius of the sphere, which he called the interaction radius (R).The mathematics involved are inconvenient, however, especially where solvent blends areconcerned.

THE TEAS GRAPH

Given the awkwardness of presenting three-component data, Jean P. Teas devised a tri-angular graph in 1968, on which polymer solubility areas could be drawn in their entirety.Because of its clarity and ease of use, the Teas graph has found increasing application inproblem solving, documentation, and analysis, and is an excellent vehicle for understand-ing complex solubility behavior.

To plot all three parameters on a single planar graph, however, a departure had to betaken from established theory. Teas constructed his graph on the fantastic hypothesis thatall materials have the same Hildebrand value. According to this assumption, solubilitybehavior is determined, not by differences in the Hildebrand value, but by the relativeamounts the three component forces (dispersion force, polar force, and hydrogen bondingforce) contribute to that value. This allows us the convenience of related percentages ratherthan unrelated sums. But because Hildebrand values are not the same for all liquids, itshould be remembered that the Teas graph is somewhat more empirical than theoretical.Fortunately, this does not prevent it from being an accurate and useful tool, and perhapsthe most convenient method by which solubility information can be illustrated.

Hansen derived his parameters from the Hildebrand value: when all three Hansenparameters for a solvent are added together, their sum will be the Hildebrand value for thatsolvent. Teas parameters, also called fractional parameters, are mathematically derivedfrom Hansen values by calculating the relative amount that each Hansen parameter con-tributes to the whole.

In other words, when all three of Teas’s fractional parameters are added together, theirsum will always be 100. Once this step has been taken, the rest is easy. Now the intersec-tion of these three values can be easily located on a triangular grid (Figure 2).

Figure 2 Any point on a triangular Teas graph is the intersection of three percentage values.


If we examine a Teas graph containing the locations of many solvents we can see thatthe alkanes, whose only intermolecular bonding is due to dispersion forces, are located inthe far lower right corner, the corner corresponding to 100% dispersion forces. Movingtoward the lower left corner are solvents with increasing hydrogen bonding contribution.Moving from the bottom of the graph upward are solvents with polarity due less to hydro-gen bonding than to an increasingly greater dipole moment of the molecule as a whole.

Overall, the solvents are grouped closer to the lower right apex than the others. This isbecause the dispersion force is present in all molecules, polar or not, and determining thedispersion component is the first calculation in assigning Hansen parameters, from whichfractional parameters are derived. Unfortunately, this greatly overemphasizes the disper-sion force relative to polar forces, especially hydrogen bonding interactions.

It can also be seen that increasing molecular weight within each class shifts the relativeposition of a solvent on the graph closer to the bottom right apex (Figure 3). This is because,as molecular weight increases, the polar part of the molecule that causes the specific char-acter identifying it with its class, called the functional group, is increasingly “diluted” byprogressively larger, nonpolar “aliphatic” molecular segments. This gives the molecule asa whole relatively more dispersion force and less of the polar character specific to its class.

This trend toward less polarity with increasing molecular weight within a class alsoaccounts for the observation that lower-molecular-weight solvent’s are often “stronger”than higher-molecular-weight solvents of the same class, although determinations of sol-vent strength must really be made in terms of the solvent’s position relative to the solubil-ity area of the solute. (Another reason for low-molecular-weight solvents seeming moreactive is that smaller molecules can disperse throughout solid materials more rapidly thantheir bulkier relatives.)

The only class in which increasing molecular weight places the solvent farther awayfrom the lower right corner is the alkanes. As previously stated, the intermolecular attrac-tions between alkanes are due entirely to dispersion forces, and, accordingly, Hansenparameter values for alkanes show zero polar contribution and zero hydrogen bondingcontribution. Since fractional parameters are derived from Hansen parameters, we wouldexpect all the alkanes to be placed together at the extreme right apex.

Observed behavior indicates, however, that different alkanes do have different solu-bility characteristics, perhaps because of the tendency of larger dispersion forces to mimic

Figure 3 Within each class, solvents with higher molecular weight tend to be closer to the lowerright axis.


slightly polar interactions. For this reason, Teas adjusted the locations of the alkanes to cor-respond to empirical evidence. Several other solvent locations were also shifted slightly toreflect observed solubility characteristics properly.

The position of water on the chart is uncertain, because of the ionic character of thewater molecule, and the placement in this paper is according to Teas.12 The presence ofwater in a solvent blend, however, can alter dramatically the accuracy of solubility predic-tions.

VISUALIZING SOLUBILITY

Now that solvent positions are located on the Teas graph, we can discuss (and visual-ize) complex solubility behavior. For example, it is easy to describe the solubility of a mate-rial by testing solvents on it and coding the results. Active solvents could have theirpositions marked with a star, marginal solvents with a filled circle, and nonsolvents withhollow circles. Once this is done, a solid area of stars will be seen, possibly with a border offilled circles. This would define the solubility window of the material (Figure 4). Not everysolvent would need to be tested in this way; only enough to circumscribe the area.

The boundaries of this solubility window could be more accurately defined by usingtwo liquids near the edge of the solubility window, one within the window and one out-side. The material could then be tested in various mixtures of these two liquids, and themixture just producing solubility noted on the graph. If this procedure is repeated in sev-eral locations around the edge of the window, its boundaries may be accurately deter-mined. Interestingly, some composite materials (such as rubber/resin pressure-sensitiveadhesives, or wax/resin mixtures) can exhibit two or more separate solubility windows,more or less overlapping, that reflect the degree of compatibility and the concentration ofthe original components.

The solubility window of a material will have a specific size, shape, and placement onthe Teas graph depending on its polarity and molecular weight, and the temperature andconcentration at which the measurements are made. Most published solubility data arederived from 10% concentrations at room temperature. Heat has the effect of increasing thesize of the solubility window, because of an increase in the disorder (entropy).

Figure 4 The solubility window of a hypothetical material.


Concentration also has an effect on solubility. Most polymer solubility windows aredetermined at 10% concentration of polymer in solvent. Because an increase in concentra-tion also causes an increase in disorder, solubility information can be considered accuratefor solutions of higher concentration as well. Solvent evaporation as the solution driesserves to increase concentration, thus ensuring that the two materials stay mixed. Solutionsof less than 10% may become immiscible, however, especially with solvent combinationsat the edge of the solubility window.

Solution viscosity of a polymer will also vary depending on where solvent is located inthe solubility window of the polymer. We might expect viscosity to be at a minimum whena solvent near the center of a polymer solubility window is used. However, this is not thecase. Solvents at the center of a polymer solubility window dissolve the polymer so effec-tively that the individual polymer molecules are free to uncoil and stretch out. In this con-dition molecular surface area is increased, with a corresponding increase in intermolecularattractions. The molecules thus tend to attract and tangle on each other, resulting in solu-tions of slightly higher than normal viscosity.

When dissolved in solvents slightly off center in the solubility window, polymer mol-ecules stay coiled and grouped together into microscopic clumps, which tend to slide overone another, resulting in solutions of lower viscosity. As solvents nearer and nearer theedge of the solubility window are used to dissolve the polymer, however, these clumpsbecome progressively larger and more connected and viscosity again increases until ulti-mately separation occurs as the region of the solubility boundary is crossed.

The position of a solvent in the solubility window of a polymer has a marked effect onthe dried film characteristics of the polymer as well. Because of the uncoiling of the poly-mer molecule, films cast from solvent solutions near the center of the solubility windowexhibit greater adhesion to compatible substrates. This is due to the increase in polymersurface area that comes in contact with the substrate. Many other properties of dried films,such as plastic crazing or gas permeability, are related to the relative position that the orig-inal solvent occupied in the solubility window of the polymer. The degree of both crazingand permeability is predictably less when solvents more central to the solubility windowhave been used.

SOLVENT MIXTURES

The Teas graph is particularly useful for creating solvent mixtures for specific applica-tions. Solvents can easily be blended to exhibit critical solubility behavior such as dissolv-ing one material but not another. The use of the Teas graph can reduce trial-and-errorexperimentation to a minimum, by allowing the solubility behavior of mixtures to be pre-dicted in advance.

This is because the solubility parameters of a mixture can be simply calculated by aver-aging its components. We can then expect the mixture to behave more or less like the sin-gle solvent with that solubility parameter. Determining the solubility parameters of amixture can be done either by calculating from the fractional parameters of the individualsolvents, or in the case of a binary mixture, by simply drawing a line between its two sol-vents and measuring their ratio.

To calculate the solubility parameters from the individual components, the fractionalparameters for each liquid are multiplied by the fraction that the liquid occupies in theblend, and the results for each parameter added together. Or, graphically, a mixture can belocated by drawing a line connecting the two solvents and determining the distance thatrepresents their ratio.


What is interesting about visualizing solvent blends on a Teas graph is the control withwhich effective solvent mixtures can be formulated. For example, two liquids that are non-solvents for a specific polymer can sometimes be blended in such a way that the mixturewill act as a true solvent (Figure 5). This is possible if the graph position of the mixture liesinside the solubility window of the polymer, and is most effective if the distance of the non-solvents from the edge of the solubility window is not too great.

This phenomenon is also valuable when selective solvent action is required, such as inselectively dissolving one material while leaving other materials unaffected, particularly ifthe solubilities of the materials involved are very similar. In this case it is helpful first to plotthe solubility windows of all the materials in question.

Once this has been done, it is easy to see the overlap of solubilities, and any areas wheresolubilities are mutually exclusive. A solvent blend can then be formulated that activelydissolves the proper material, while positioned as far away from the solubility window ofthe other material as possible (Figure 6). It is important to remember that differences inevaporation rates can shift the solubility parameter of the blend as the solvents evaporate,and this must be taken into account. Additionally, while a material may not show immedi-ate signs of solution in a solvent or solvent blend, the solvent may still adversely affect thematerial, for example, by softening it or leaching out partial components.

A further advantage of blending solvents is the ability to design mixtures with similarsolubility characteristics and lower toxicity. In such cases, it should be pointed out that thesimilarity between solvents and blends having the same numerical parameters decreasesas the distance between the components of the blend increases. Where alternate blends areeffective, however, the use of a less toxic replacement can be a sensible choice, and the Teasgraph a useful tool (Figure 7).

Figure 5 A mixture (M) of two nonsolvents (A and B) may act as a true solvent if the parameters ofthe mixture fall within the solubility window.


Figure 6 In situations where one material must be dissolved while another remains unaffected, asolvent blend that falls within the solubility window of the first and outside the window ofthe second may be effective.

Figure 7 The Teas graph (numbers indicate solvents listed in Table 2).


Table 2 Fractional Solubility Parameters

Position Solvent 100d 100p 100h

Alkanes

1 n-Pentane 100 0 01 n-Hexane 100 0 01 n-Heptane 100 0 01 n-Dodecane 100 0 02 Cyclohexane 94 2 43 VMP naphtha 94 3 34 Mineral spirits 90 4 6

Aromatic Hydrocarbons

5 Benzene 78 8 146 Toluene 80 7 137 o-Xylene 83 5 128 Naphthalene 70 8 229 Styrene 78 4 18

10 Ethylbenzene 87 3 1011 p-Diethylbenzene 97 0 3

Halogen Compounds

12 Methylene chloride 59 21 2013 Ethylene dichloride 67 19 1414 Chloroform 67 12 2115 Trichloroethylene 68 12 2016 Carbon tetrachloride 85 2 1317 1,1,1-Trichloroethane 70 19 1118 Chlorobenzene 65 17 819 Trichlorotrifluoroethane 90 10 0

Ethers

20 Diethyl ether 64 13 2321 Tetrahydrofuran 55 19 2622 Dioxane 67 7 2623 Methyl Cellosolve 39 22 3924 Cellosolve 42 20 3825 Butyl Cellosolve 46 18 3626 Methyl carbitol 44 21 3527 Carbitol 48 23 2925 Butyl carbitol 46 18 36

Ketones

28 Acetone 47 32 2129 Methyl ethyl ketone [53] [30] [17]30 Cyclohexanone 55 28 17

Diethyl ketone 56 27 17Mesityl oxide 55 24 21

31 Methyl isobutyl ketone 58 22 2032 Methyl isoamyl ketone 62 20 18


Table 2 Fractional Solubility Parameters (continued )


Isophorone 51 25 2433 Di-isobutyl ketone [67] [16] [17]

Esters

34 Methyl acetate 45 36 1935 Propylene carbonate 48 38 1436 Ethyl acetate 51 18 31

Trimethyl phosphate [39] [37] [24]Diethyl carbonate 64 12 24Diethyl sulfate 42 39 19

37 n-Butyl acetate 60 13 27Isobutyl acetate 60 15 25

38 Isobutyl isobutyrate 63 12 2539 Isoamyl acetate 60 12 2840 Cellosolve acetate 51 15 34

Ethyl lactate 44 21 35Butyl lactate 40 20 32

Nitrogen Compounds

41 Acetonitrile 39 45 1642 Butyronitrile 44 41 1543 Nitromethane 40 47 1344 Nitroethane 44 43 1345 2-Nitropropane 50 37 1346 Nitrobenzene 52 36 1247 Pyridine 56 26 1848 Morpholine 57 15 2849 Aniline 50 19 3150 N-Methyl-2-pyrrolidone 48 32 20

Diethylenetriamine 38 30 3251 Cyclohexylamine [64] [12] [24]

Formamide 28 42 3052 N,N-Dimethylformamide 41 32 27

Sulfur Compounds

53 Carbon disulfide 88 8 454 Dimethyl sulfoxide 41 36 23

Alcohols

55 Methanol 30 22 4856 Ethanol 36 18 4657 1-Propanol 40 16 4458 2-Propanol [41] [16] [43]59 1-Butanol 43 15 42

2-Butanol [44] [16] [40]Benzyl alcohol 48 16 36

60 Cyclohexanol 50 12 3861 n-amyl alcohol 46 13 41


REFERENCES

1. Burrell, H., Solubility parameters, Interchem. Rev., 14, 13–16 and 31–46, 1955.2. Burrell, H., Solubility parameters for film formers, Off. Dig. Paint Varn. Prod. Clubs, 27, 726, 1955.3. Crowley, J.D., G.S. Teague, Jr., and J.W. Lowe, Jr., A three dimensional approach to solubility,

J. Paint Technol., 38 (496), 1966.4. Feller, R.L., N. Stolow, and E.H. Jones, On Picture Varnishes and Their Solvents, The Press of Case

Western Reserve University, Cleveland, 1971.5. Gardon, J.L. and Teas, J.P., Solubility parameters, in Treatise on Coatings, Vol. 2, Characterization of

Coatings: Physical Techniques, Part II, Myers, R.R. and J.S. Long, Eds., Marcel Dekker, New York, 1976.6. Hansen, C.M., The three dimensional solubility parameter—key to paint component affinities. I.

Solvents plasticizers, polymers, and resins, J. Paint Technol., 39(505), 1967.7. Hansen, C.M., The three dimensional solubility parameter—key to paint component affinities. II.

Dyes, emulsifiers, mutual solubility and compatibility, and pigments, J. Paint Technol., 39(511), 1967.8. Hansen, C.M., The three dimensional solubility parameter—key to paint component affinities. III.

Independent calculations of the parameter components, J. Paint Technol., 39(511), 1967.9. Hansen, C.M., Solubility in the coatings industry, Skand. Tidskr. Faerg. Lack, 17, 69, 1971 (English).

10. Hildebrand, J.H., The Solubility of Non-Electrolytes, Reinhold, New York, 1936.11. Teas, J.P., Graphic analysis of resin solubilities, J. Paint Technol., 40(516), 1968.12. Teas, J.P., Predicting resin solubilities. Columbus, Ohio, Ashland Chemcial Technical Bulletin, No.,

1206, 1971.

Table 2 Fractional Solubility Parameters (continued )


62 Diacetone alcohol 45 24 312-Ethyl-1-hexanol 50 9 412-Ethyl-1-butanol 48 10 42

Polyhydric Alcohols

63 Ethylene glycol 30 18 5264 Glycerol 25 23 5265 Propylene glycol 34 16 5066 Diethylene glycol 31 29 4067 Water 18 28 54

Miscellaneous Liquids

68 Phenol 46 15 3969 Benzaldehyde 61 23 1670 Turpentine 77 18 571 Dipentene 75 20 5

Formic acid [33] [28] [39]Acetic acid [40] [22] [38]Oleic acid [62] [14] [24]Stearic acid [65] [13] [22]Linseed oil 66 17 17Cottonseed oil 67 15 18Neets foot oil 69 14 17Pine oil 70 14 16Sperm oil 75 11 14

1 Mineral oil 100 0 0

Note: Numbers in left column refer to solvent positions in Teas graph, Figure 7.

Sources: Values from Gardon and Teas.5 Values in brackets derived from Hansen’s original 1971 parametersrecalculated by the author.


CHAPTER 1.3

Aqueous Cleaning Essentials

Rick Bockhorst, Michael Beeks, and David Keller

CONTENTS

IntroductionCleaning OverviewCleaning Parameters

TemperatureAgitationConcentrationTime Required for CleaningRinsingRedepositionProtecting the SubstrateEquipmentControlling the Cleaning LineImproving Bath Life

WaterPhysical Properties of WaterImpurities

Water PretreatmentSoftening

GeneralPrinciples of Water Softening

DeionizationReverse Osmosis

GeneralRO Components

Cleaning FormulationsCleaning Chemistry

Acid CleanersAlkaline and Neutral Cleaners

IngredientsAlkaline Builders


Water ConditionersSurface-Active AgentsCorrosion InhibitorsAdditional IngredientsHydrocarbon Solvents

RinsingImportance of RinsingPart Cleanliness RequiredProduction Levels and DraginIncoming Water QualityNumber of Rinse TanksRinse Tank Design and PlacementDragin and Final Rinse Quality

DisposalConclusionAcknowledgmentReferences

INTRODUCTION

Largely as a result of environmental pressures, parts cleaners have been forced toexplore alternatives to the various solvent cleaning processes and especially vapordegreasing. These pressures have come in the form of international, federal, state and localregulations, but the most significant influence has probably been the Montreal Protocols.While aqueous cleaning is almost as old as humanity, parts cleaners have long held thatcertain soils could be cleaned adequately only by nonaqueous methods. The reality is thatmany cleaning applications that were once strictly the province of nonaqueous cleaningmethods are now being done quite successfully with aqueous processes.

CLEANING OVERVIEW *

With few exceptions there are certain principles, treated generally here, that apply toall types of cleaning. Cleaning processes combine mechanical, thermal, and chemicalenergy sources to remove a soil from a substrate. The total energy needed is the sum ofthese energy sources over a given period. Within these parameters the following generalguidelines apply:

1. Cleaning efficacy and rate improve as temperature increases.2. Agitation improves the rate and efficacy of soil removal. Agitation provides

mechanical energy to remove soils physically and assures that fresh cleaner willcontinuously contact the soil.

3. Cleaner solutions generally have a performance vs. concentration curve. A mini-mum level of cleaner is generally necessary for effective cleaning. Cleaningimproves with incremental increases of cleaner up to some point, where furtherincreases result in little or no further improvement in performance.

4. Time is the controlling factor in total energy input. If cleaning requires X mechan-ical energy for T time and the energy level is then decreased, to compensate thetime must be increased proportionally.

* Reference 16, Chapter 3.


5. Rinsing is necessary to remove any cleaner or soil residue remaining on the partsafter washing.

• Rinse type and quality are dependent on the cleanliness requirements of theapplication.

• Multiple small rinses are generally more efficient and cost-effective than onelarge rinse.

• An agitated rinse is generally more effective than a still rinse.• Final part cleanliness or, conversely, residue on the part is limited by rinse

quality.6. Soil must be prevented from redepositing on parts. The most obvious answer is

to remove the soils from contact with the substrate. This may be accomplished byvarious methods, including:• Emulsification;• Emulsification followed by demulsification and physical removal;• Flocculation;• Ultra- or microfiltration.Additionally, redeposition can be controlled by:• Choosing cleaners formulated with “antiredeposition” properties;• Using cleaning tanks of sufficient size to disperse the soil and slow the rate of

increase of contamination concentration.7. The cleaning method or solution should not harm the item (substrate) being

cleaned.8. Precleaning to remove bulk soils may be an economical and commonsense way

to increase cleaner life.9. Cleaning systems should be designed as a unit. That is, the cleaner and the clean-

ing equipment should be chosen to work together and address the particularcleaning application. Typical concerns that should be addressed include:• Cleaning temperature and its effect on the cleaner as well as the parts being

cleaned. Part of this consideration includes method of heating, insulation, andevaporation.

• Equipment design, which should include an evaluation of cleaner and partcompatibility with regard to materials of construction, economy of operation,electrochemistry, OSHA and other regulatory guidelines, and ease ofservice.

• The compatibility of the mechanical energy input, which must be addressed interms of effectiveness of removing soil from the substrate, controlling foamingtendencies of the cleaner, avoiding mechanical damage to the parts, and avoid-ing degradation of the cleaner.

Aqueous cleaners can generally be categorized as being acidic, alkaline, or pH-neutral.Alkaline cleaners are by far the predominant of the three used in all commercial/industrial cleaning applications. Thus, this discussion will mainly cover alkaline cleaning,but the principles are applicable to all types of cleaners.

Agitation techniques represent the greatest variation in cleaning methods. The mostimportant factor is that it costs money in equipment and/or labor to provide high levels ofagitation. The equipment must be designed to meet the objective of providing adequate-to-superior agitation for soil removal at the lowest cost. The major limitations for providingadequate agitation are equipment costs, equipment size (i.e., how big of a “footprint” doesthe equipment have), excessive foam generation, excessive generation of mist/spray, toxicvapors, or flammable gases.


Now that we have taken a broad look at some cleaning principles, let us look a littlemore closely at each.

CLEANING PARAMETERS

Temperature

The effect of temperature depends on the type of soil being removed and the cleaner.The first consideration is what type of soil needs to be removed. Temperature is veryimportant in speeding the removal of fats, greases, oils, and waxes. Increased temperaturereduces the viscosity of oils and greases, making them more mobile, and therefore easier todisplace from the substrate. Fats and waxes are often solids at room temperature. It is crit-ical to melt these fats and waxes to remove them by aqueous methods. If the melt range ofthe fat/wax is above the boiling point of the cleaner, aqueous cleaning may not be effectiveon this type of soil.

There is a well-established principle that the rate of a chemical reaction is doubled foreach 10°C (18°F) increase in temperature. If the cleaning process works by reaction betweena fatty acid/oil and alkali, by a paint coating undergoing chemical decomposition, or by anacid chemically removing rust and scale, then this reaction rate relationship is applicable.It is possible to remove solid fats/waxes if they can chemically react with the cleaner with-out melting them; however, the rate of removal may not be adequate.

On the other hand, excessively high temperatures could “set” proteinaceous soils ormay cause an undesirable reaction between the soil and the substrate, making the soil moredifficult to remove. Just as increasing temperature will increase the rate of cleaning, it willincrease the rate of undesired reactions. Most corrosion inhibitors work by forming a loosebarrier on the clean metal surface. Excessively high temperature can disrupt this barrierand result in chemical attack, usually seen as discoloration and etching. The majority ofindustrial cleaning is carried out at 140 to 180°F.

Agitation

As has been previously stated, agitation techniques represent the greatest variation inaqueous cleaning systems. It is usually possible to find an aqueous cleaner to remove a soilfrom a substrate. Thus, one of the biggest problems users run into is inadequate cleaningperformance because of inadequate or improper choice of agitation. The method of agita-tion should be matched to the size and shape of the part. For example, while spray wash-ing may be very effective for cleaning large relatively flat parts, it may not be suitable forparts with blind holes where direct impingement is problematic. Relatively flat objects andcomponents that do not have hidden areas can be cleaned by immersion or spray wash.Typically, parts too small or too large cannot be cleaned by spray wash; an exception forsmall parts is possible when specialty mounting racks are built. Spray wash cleaning is alsolimited on the chemical side because the formula must not be foamy. Eliminating foamrestricts the choices of raw materials a chemist can use in formulating a spray wash cleaner.Parts that can be damaged from spray impingement should only be cleaned by immersion.Virtually all parts can be cleaned by immersion. Ultrasonic cleaning is the most effectivemethod of agitation for immersion cleaning, but is restricted by cost (very expensive equip-ment) and size (not as effective on tanks above �1000-gal capacity). Spray-under immer-sion and turbulation are the next most effective methods of immersion agitation.Spray-under immersion and turbulation can sometimes create excessive foam if care is not


taken in choosing a cleaner. General agitation from pump circulation can be adequate fornoncritical cleaning applications, but is not usually suitable for precision cleaning. Airsparging can very effectively agitate an aqueous cleaner, but it has several restrictions:

1. The cleaner must be a low-foaming to nonfoaming product. Air sparging maycause foam to overflow the tank. Basically, cleaners designed for spray washingshould only be used with this form of agitation.

2. The cleaning tank should be covered to eliminate generation of mist and spray.The popping of the bubbles will generate mist and spray that will create an expo-sure problem for workers. The only cure for this is to cover the tank or place thetank within a cabinet to contain the mist/spray. It is usually impractical to put acover on most immersion tanks because of the mechanics of moving the coverand parts in and out of the tank. Adding on a cabinet increases the cost of thecleaning system and will likely increase the “footprint” taken up in the facility.

3. Air sparging can shorten the life of alkaline cleaners, especially heavy-duty caus-tic cleaners, by neutralization with carbon dioxide (CO2), a weak acid. Acids andbases will neutralize each other. Even though it makes up only a fraction of a per-cent of the atmosphere (0.035% measured at Mauna Loa Observatory, 1990, asreported in Handbook of Chemistry and Physics, pp. 14–25, 199620), the large volumeof air passed through the tank will expose the cleaner to a significant amount ofCO2. The CO2 will neutralize the alkaline builders, especially sodium or potas-sium hydroxide. Heavy-duty caustic cleaners are especially prone to this prob-lem, whereas mildly alkaline cleaners are much less sensitive. The CO2 reactswith free hydroxide ions (OH � , the cause of alkaline pH) to form bicarbonateions:

CO2 � OH� → HCO3�

The bicarbonate then goes on to react with more hydroxide ions to form a carbonateion and water:

HCO3� � OH� → CO3

2� � H2O

Thus, as the hydroxide ions (OH�) are consumed, the pH drops in the cleaner. This canadversely affect alkaline cleaners, especially heavy-duty caustic cleaners used in descalingoperations. Acidic and pH-neutral cleaners are not affected by this problem.

Finally, soaking the substrate in a stagnant tank is unacceptable, even for crude clean-ing applications.

Concentration

Concentration, also called use dilution can affect multiple attributes of the cleaningprocess. In many cases minimum or maximum cleaner concentrations can control corro-sion characteristics, chemical etching, or the deposition of protective barriers as well ascleaning efficacy. The necessary cleaner concentration will vary with the type of agitationand temperature. As an example, in the absence of foaming problems, it may be possible toobtain similar cleaning performance from the same alkaline cleaner at 5 to 10% by immer-sion, 3 to 5% by spray, 1 to 3% by steam cleaning or high-pressure hot spray, or 2 to 4% in


a high-pressure, room-temperature spray. For each of these applications, increasing thecleaner concentration will give better cleaning performance up to a certain point and thenlevel off. This leveling-off point will be dependent on the specific chemistry used in thecleaner, the soil being removed, the agitation technique, and temperature.

Time Required for Cleaning

The cleaning time is dependent on the concentration of the cleaner, the specific chem-istry used in the cleaner, the soil being removed, the agitation technique, and the tempera-ture. It is important to emphasize that cleaning is not instantaneous; some time is necessaryfor the detergent to perform its work on the soil. In a stagnant bath, cleaning may take from5 min to over an hour to occur, if at all. Most immersion cleaning situations do not exceed10 min, although numerous exceptions can be found. Spray washes typically take no morethan 5 min. One general rule of thumb does exist for ultrasonic cleaning; if it takes morethan 5 min to clean, either there is something deficient in the cleaner or the process itself isnot suited for the application. One notable exception to the “5-min rule for ultrasonics” isthe cleaning of used automobile cylinder heads and other major engine components. Manyautomobile repair facilities have adopted ultrasonic processes as an alternative to heavy-duty caustic or solvent cleaning tanks. The heavy-duty caustic tanks have fallen out of favorbecause they cannot be used on aluminum parts; nearly all major engine parts are nowmade of aluminum. Corrosion hazards and waste disposal considerations also have con-tributed to the decline of the caustic stripping tank. Solvents have been on the declinemainly because of environmental regulations that govern volatile organic content (VOC),ozone-depleting substances (ODSs), and hazardous air pollutants (HAPs). The few sol-vents (cresylic acids) that are effective on baked-on soils also pose serious health risks.Ultrasonic cleaning with highly concentrated, moderately alkaline cleaners has been foundto be effective at removing most of the baked-on soils. The drawbacks include long pro-cessing times, often 15 to 45 min, and some cavitational erosion. The cavitational erosionappears as a “star-” or “Y-shaped” pattern on the metal.

It must be stressed that the time available for cleaning is very closely related to theeconomics of the cleaning operation. The increased cost in equipment, energy, and chem-icals to reduce time must be weighed against the economic gains in increased produc-tion. Additionally, consideration must be given to effects on reject rates and customersatisfaction.

Rinsing

No matter what cleaning method has been employed, the surface of the freshly cleanedpart will contain some amount of soil and cleaner residue. In some situations this residuemay present no problem, but in many others that residue must be removed to yield accept-able parts.

This is an important issue. Just as with solvent rinses, water rinses may contain impu-rities that can dry on the parts. Water quality issues will be discussed later. The value ofpressure sprays and mechanical action in rinsing is often neglected. Experience shows thatdirect spraying is far more effective in flushing away the loosened soil than just soaking thepiece in an immersion tank.

The use of a short spray rinse followed by a soak or agitated rinse to reduce contami-nation level is most effective at reducing contamination. Static or slow-moving rinses inwhich there is improper/inadequate flow usually result in parts that need to be recleanedor discarded.


Another important consideration in rinsing is the number of rinses performed; tworinse steps are more effective than one, three are better than two, etc. Multiple rinses can beof shorter individual duration and still be more effective than single rinses because of theexponential dilution of contaminants as the parts proceed from one rinse to the next.Unfortunately, multiple rinses can lead to higher costs in equipment (i.e., number of tanksneeded, footprint taken up on the plant floor). The cost can be offset by counterflowing thewater from the last rinse back into each previous rinse. This significantly reduces waterconsumption; some of the overflow can be used as add-back into the cleaning tank toreplace evaporated water. Studies have shown that a counterflowed, triple rinse has theoptimum balance of reducing water consumption, obtaining clean parts, and capital costsin the rinse step; more than three rinses yields diminishing returns. It is possible in somesituations to equip the multistage rinse with a set of deionizing resin beds and an activatedcarbon filter. Closing the rinsing loop by deionizing the overflow water can reduce waterconsumption, replacing only the water lost to evaporation. Users with significant waste-water disposal costs should consider this kind of setup. For further reading, see Peterson13

and Spring.16

Another important consideration is the quality of water used in the rinse step. Thequality of rinsing can only be as good as the quality of water used in the last rinse.Unsoftened water obtained from a municipal source or well often contains varying levelsof hard water ions, carbonates, phosphates, and organic by-products from treatmentprocesses. The water hardness can often be extremely high, leading to hard water depositsand soap scum residue. Softening the water to remove hard water ions (calcium and mag-nesium) will eliminate those hard water salts but leaves other impurities and therefore maynot be adequate. Using deionized, distilled, or reverse-osmosis purity water gives the bestrinsing performance. As usual, as the quality of water increases, so does the cost. The levelof performance must meet the requirements of the application. The application require-ments must be evaluated for each system.

Redeposition

The design of the tank and cleaner is an important factor in reducing/eliminatingredeposition of soils. The tank must be of sufficient capacity to provide room for thesoils to move away from the parts. The size is also important to moderate the rate at whichsoil loading increases; too small of a tank can result in the cleaner becoming saturated ina matter of hours or days. Bag filters can be used to remove gross particulate matter. Thecleaner may incorporate phosphates, silicates, specialty surfactants, and synthetic poly-mers, which remove and suspend soils in solution. Use of aqueous cleaners that can “split-out” oils instead of emulsifying them in combination with an oil coalescer/skimmer willslow down, possibly even prevent, the soil from reaching a saturated condition in thecleaner. This oil splitting followed by physical removal will lead to longer tank life andprevent redeposition. Self-emulsifying oils are difficult to handle in preventing redeposi-tion. These lubricants contain their own surfactants that form stable emulsions. In general,aqueous cleaners cannot break these emulsions without being consumed themselves. Theonly viable alternative is to use microfiltration in the 0.1 to 0.5�m pore size range toremove the emulsion. The filter will not remove all of the emulsion but will remove someor most. The cleaner must also be designed to be able to pass through the filter. Micro- orultrafiltration has been successfully done but generally results in some loss of cleaner con-stituents. Thus, monitoring and adjusting the tank becomes difficult and may not be an


economically viable option. A good rinse must always follow the cleaning step to preventsoil redeposition. Waxes typically require a hot rinse to prevent resolidification on the cleansurface.

Protecting the Substrate

The cleaner must be compatible with the substrate. Unless these effects are necessary,the cleaner must not discolor, etch, cause hydrogen embrittlement, cause stress cracking, orotherwise damage the substrate. Thus, the cleaner must contain additives to protect thesubstrate from these effects. For example, aluminum cleaners may either contain silicatesand have an elevated pH or have a pH of around 7 to 9 to prevent corrosion. Stainless steelcleaners must not contain chlorides because they will cause stress cracking, especially inacidic cleaners. Rusting on steel may be avoided with strongly alkaline cleaners, especiallycleaners containing amine-based corrosion inhibitors. High-chrome steels are especiallyprone to rusting and often need to have a corrosion inhibitor added to the rinse tanks toprevent rusting during the rinse and dry cycles. Acid cleaners sometimes require inhibitorsto minimize corrosion of the base metal without reducing the efficacy on hard water ormetal oxide scales. Good rinsing, careful drying, or use of inhibitors may avoid tarnishing.Copper and copper-based alloys are notorious for tarnishing during the drying cycle.Transfer time between cleaning and rinsing tanks should be minimized to avoid drying ofcleaner residue on the substrate.

Equipment

Some of the greatest advances in the art of cleaning in the past decade relate toimprovements in equipment. Intelligent, appropriate use and care of equipment may bethe key to proper cleaning in many instances.

Controlling the Cleaning Line

It is necessary to determine when the cleaner is nearly exhausted so that fresh cleanercan be prepared or the old cleaner can be rejuvenated. This is not always easy to determine.Measuring properties such as alkalinity, conductivity, and pH are useful in determining thestate of the cleaner, but it is not uncommon for tanks to fail even when the above-men-tioned test results are within specifications. The properties that should be measured arethose that are critical to the specific process. For example, silicate-based aluminum clean-ers should be monitored mainly for pH; silicate testing should be done if affordable. Thesilicates protect aluminum from corrosion and they participate in soil anti-redeposition.The problem with silicates is that their solubility in water decreases as the pH drops in thecleaner. At some point, the silicate level will drop below the minimum level necessary toprotect aluminum. When this happens, spotting or etching may occur.

Heavy-duty caustic cleaners are best monitored by active and total alkalinity titrationmethods. Acid cleaners are best monitored by total acidity titration methods. Cleaners thatundergo microfiltration to remove emulsified oils present the greatest difficulty in moni-toring. The pH may need to be monitored if the cleaner contains silicates. Alkalinity titra-tion is useful, but does not detect many of the surfactants that are stripped out by thefiltration process. The emulsified oils may have ingredients that will severely impact thepH and skew the alkalinity titration. Monitoring how much the solution bends light, calledthe refractive index, can help in tracking loss of cleaner to stripping by the filter.


Unfortunately, this measurement requires that the solution be relatively clear; cloudy solu-tions are difficult to impossible to measure. Some tanks are so difficult to maintain that theuser must seek out a chemical maintenance firm or change some of the manufacturingprocesses that precede the cleaning step to eliminate some of these problems.

Improving Bath Life

One way of improving bath life is to have two cleaning tanks in series. Most of the soilis removed in the first tank. The part going into the second tank is relatively clean. After thefirst tank becomes heavily contaminated it can be discarded and the cleaner in the secondtank pumped over to the first tank. In this manner the cleaners can be used for long peri-ods and, if handled intelligently, it is possible to operate with the first bath very heavilycontaminated so that cleaner is seldom discarded. Contamination of the second tank canbe reduced by employing a short rinse after the first cleaning operation, the rinsing watereither going to sewer or back to the first tank. Besides better performance and economy,this procedure may be necessitated by governmental regulation of effluent that prohibitsor limits the discarding of cleaning baths.

The single biggest problem in obtaining long tank life is understanding that cleaning ispart of the manufacturing process and must be evaluated as part of the whole. The design ofthe cleaning process must consider the soils and their effects not only on processing but ondisposal. The choice of cleaning equipment and of cleaner should be made as a coordinatedeffort that takes into account soil removal from the part and from the bath. These choicesmust also consider the ultimate costs of operation, including the costs of disposal. It is inthis area that chemical and equipment manufacturers and their customers most often fail.

Great strides can be made that will allow aqueous baths with suitable replenishmentto be useful for multiple years, but cooperation of chemical manufacturers, equipmentmanufacturers, and industrial users is critical.

WATER

For most aqueous cleaners, water comprises 80 to 99% of the cleaning solution and isused in practically all rinsing steps. Although most people do not think of it in this way,water is actually a solvent in aqueous cleaners. A major key to understanding the efficacyof aqueous cleaners lies in the role played by water, its natural properties, and impurities.

Water has been vital to life and nature from the beginning of time. The basic cycle bywhich water evaporates, condenses, and flows along the surface of the Earth governs allanimal and plant life. Approximately 61.8% of the human body is water. Water coversalmost 70% of the Earth’s surface, most of it in oceans, with the balance found in lakes,rivers, the atmosphere, and absorbed in soil and rocks. Water is never absolutely pure innature and its impurities are the factors of concern in industrial applications. Humans havecontributed to impurities found in water sources. One concern of aqueous cleaning is dis-posal of spent cleaning solutions.

When an aqueous cleaner is used to remove contaminants from a surface, the water isbasically the solvent in which the cleaning takes place. The importance of its function can-not be overstated. As the solvent, water is able to dissolve and disperse the soils beingremoved. Additives such as acids, alkalis, chelants, and detergents significantly augmentthe cleaning process. These additives are not nearly as effective by themselves unless theyare dissolved in a solvent, i.e., water. The combination of these additives with water yieldsthe powerful, synergistic effects that are exploited today.


Physical Properties of Water

Pure water is colorless, odorless, and tasteless. Its chemical formula is H2O, whichshows that it is made from the two elements, hydrogen and oxygen, in a ratio of 2:1. Thesetwo specific elements combined in a 2:1 ratio yield physical properties unmatched by anyother molecule:

1. Very small size.2. Not flammable.3. Very high boiling point for its size.4. The two elements that make it up are so different that they impart a high polar-

ity to the molecule.5. The high polarity of water accounts for the high boiling point. It also accounts for:

a. The high level of thermal energy that it can absorb per degree of temperatureincrease (called the heat capacity) and to get it to boil (called the heat ofvaporization). The high heat of vaporization is what makes water so effective insteam boiler heat exchanging systems.

b. Ability to dissolve numerous substances, especially minerals and other polarsubstances.

c. Inability to dissolve nonpolar substances like fats, greases, and oils.

The very high boiling point gives aqueous cleaning the flexibility of using a wide temper-ature range. The temperature of choice can be fine-tuned to the properties of the soil andsubstrate. This property also minimizes solvent loss due to evaporation. Water loss due toevaporation may become a concern especially at temperatures above 150°F. The very highboiling point, 212°F, is beneficial in that most aqueous cleaning operations do not exceed180°F so outright boiling is not a problem. Many substrates cannot tolerate the extreme heatof boiling water without suffering from discoloration, etching, or mechanical deformation.

The high heat capacity of water makes it very effective in heating metal parts up to thecleaning temperature of the bath while having minimal impact on the bath temperature.Because metals have a low heat capacity, very little energy, relatively speaking, is expended inraising them to the bath temperature. Traditional organic solvents have low heat capacitieslike metals so they are more prone to temperature fluctuations when used as heated immer-sion cleaners.

The high polarity of water can be viewed as a double-edged sword. The high polarity makesit possible for water to dissolve many inorganic compounds, such as caustic soda, caustic potash,borates, carbonates, phosphates, and silicates. Water is also an effective solvent for many surfac-tants used in formulating aqueous cleaners. Unfortunately, this polarity results in water also beingcontaminated by numerous impurities both from the Earth’s crust and from anthropogenic pol-lution. Some of the impurities of the starting water are identical to cleaning ingredients, i.e., car-bonates and phosphates. The key is which impurities are beneficial and which are detrimental.Contaminant levels in water used to make aqueous cleaners are usually low so one must focus onwhich impurities are detrimental. Elimination/suppression of these impurities is essential in pre-venting problems including reduced cleaner performance, longevity, corrosion, contaminatedsurfaces, and water spotting.

Chemical manufacturers and industrial users spend millions of dollars annually onwater-conditioning equipment to reduce/remove the impurities as part of preventive main-tenance. Many users remain uninformed about their water quality needs. Many of these userssuffer from increased cost of recleaning and rejects as a result of the impact that poor waterquality has on the whole cleaning process. Adiscussion of water treatment options will follow.


An often forgotten property of water is its ability to dissolve oxygen gas. Oxygen gasin water can be corrosive and will attack metals. High chrome steels are exceptionallyprone to rusting in these situations. When these parts are damp and left exposed to the air,flash rusting will occur. The boiler water treatment industry knows all too well how detri-mental dissolved oxygen is in boiler systems. They have to treat these systems with whatare known as “oxygen scavengers” to remove the oxygen gas from water chemically.Oxygen scavengers are not normally used in aqueous cleaning but corrosion inhibitorsmay have to be used to combat the corrosive effects of oxygen and other ingredients dis-solved in water.

Impurities

If water were H2O and nothing else, or if all waters carried the same impurities, the use ofwater for industrial applications would be simple and straightforward. But natural waters,even rain, snow, sleet, and hail, as well as all treated municipal supplies contain someimpurities. The type and amount of contaminants in natural waters depend largely on thesource. Well and spring waters are classed as groundwaters, rivers and lakes as surfacewaters. Groundwater picks up impurities as it seeps through the rock strata, dissolvingsome part of almost everything it contacts. But the natural filtering effect of rock and sandusually keeps the water free and clear of suspended matter. Surface waters often containorganic matter, such as leaf mold, and insoluble matter, such as sand and silt. Pollutionfrom industrial waste and sewage is frequently present. Stream velocity, amount of rain-fall, and where this rain occurs on the watershed can rapidly change the character of thewater.

Below is a list of the more common and troublesome impurities:

Turbidity—Suspended insoluble matter, including coarse particles (sediment) thatsettle rapidly on standing. Amounts range from zero, in most groundwaters, tosome surface supplies of over 6% or 60,000 parts per million (ppm) in muddy andturbulent river waters.

Hardness—Water content of soluble calcium and magnesium salts equals hardnessexpressed as calcium carbonate equivalents in gpg (grains per gallon) or ppm. 1gpg � 17.1 ppm. These salts, in order of their relative average abundance inwater, are (1) bicarbonates, (2) sulfates, (3) chlorides, and (4) nitrates. Calciumsalts are about twice the concentration of magnesium salts. Hardness is undesir-able because the salts become less soluble and drop out of solution as the wateris heated and upon drying, producing a hard, stony water spot that can be diffi-cult to remove.

Iron—The most common soluble iron in groundwater is ferrous bicarbonate (blackiron). Although some water is clear and colorless when drawn, on exposure to airferrous bicarbonate can cloud up and deposit a yellowish or reddish-brown sed-iment that stains everything it contacts. Iron can also shorten the life of a watersoftener, contaminating the resin. Although the majority of iron-bearing watershave less than 5 ppm, as little as 0.3 ppm can cause trouble.

Manganese—Although rarer than iron in water, manganese occurs in similar formsand can form deposits in pipelines and tanks very rapidly with as little as 0.2ppm.

Silica—Most natural waters contain silica ranging from 1 to over 100 ppm. Whensilica spotting occurs, it can be very difficult, if not economically impossible, toremove.


Mineral acidity—Surface waters contaminated with mine drainage or trade wasteswill contain sulfuric acid, plus ferrous, aluminum, and manganous sulfates.These contaminants are corrosive and therefore waters contaminated with min-eral acidity are unfit to use without a pretreatment system.

Carbon dioxide—Free carbon dioxide is found in most natural supplies. Surfacewaters have the least, although some rivers contain up to 50 ppm. In groundwa-ters (wells), it varies from zero to concentrations so high that carbon dioxide bub-bles out when pressure is released (as in “sparkling” or seltzer water). Most wellwaters contain from 2 to 50 ppm. Carbon dioxide is also formed when bicarbon-ates are destroyed by acids, coagulants, or heating the water. This can reduce thepH of an alkaline cleaning solution. Carbon dioxide is corrosive and acceleratesthe corrosion properties of oxygen.

Oxygen—Found in surface and aerated waters. Deep wells contain very little oxy-gen. The oxygen content of water is inversely proportional to the temperature,meaning that the hotter the water, the less oxygen is present. (Note that in ele-vated temperatures, the water contains less oxygen; however, what oxygenremains is much more aggressive and corrosive.) Oxygen is very corrosive toiron, zinc, brass, and other metals. Flash rusting of metals can be a problem whenhot parts are rinsed in cold water that contains higher amounts of oxygen.

WATER PRETREATMENT

Softening

General

The impurities that cause the most trouble in aqueous cleaning processes are the inor-ganic salts. Salts are ionic compounds, which are chemicals that have a positively chargedspecies called a cation and a negatively charged species called an anion. It must be pointedout that the term salt has commonly meant sodium chloride, i.e., table salt. The chemicaldefinition is “salts are ionic compounds that contain any negative ion except the hydrox-ide ion and any positive ion except the hydrogen ion.” An ionic compound that containsthe hydrogen ion is called an acid and an ionic compound that contains the hydroxide ionis a base. Specific examples of common ionic impurities that are encountered as impuritiesin water are given in Table 1.

It must be pointed out that silicates are usually discussed/represented as silicon diox-ide, SiO2, when they are actually present in natural waters as the ions listed in Table 1.There are more complex forms of phosphate and silicate ions and numerous other traceimpurities that can be present in natural, untreated water but the above examples representthe bulk of the impurities that the cleaning industry must be concerned with.

Principles of Water Softening

The most problematic impurities are the cations. The most economical method forremoving them is by passing the water through an ion-exchange column, better known asa water softener. A standard water softener contains polystyrene beads that have beenmodified such that the surface of each bead has numerous negatively charged sites. Nature


Table 1 Dissolved Impurities in Water

Ion Type Impurity Property

Cations Ca2� HardnessMg2� HardnessFe2� and Fe3� Iron stainsMn2� and Mn4� Manganese stains and scalesNa� Too much sodium in rinse water can cause spottingK� Too much potassium in rinse water can cause spotting

Anions CO32� Alkalinity, carbonates form hard water deposits with calcium,

magnesium, iron, and manganeseHCO3

� Alkalinity, bicarbonates form hard water deposits with calcium,magnesium, iron, and manganese

PO43� Alkalinity, ortho-phosphates form hard water deposits with

calcium, magnesium, iron, and manganeseSiO4

4� and SiO32� Silicates can form the most tenacious of deposits, especially

in the presence of calcium, magnesium, iron, and manganeseCl� Chlorides promote corrosion on aluminum, iron, and steel, too

much chloride in rinse water can cause spottingNO3

2� Too much nitrate in rinse water can cause spotting

requires that charge must be balanced; which is accomplished by pairing each negativelycharged site with a sodium cation. As the impure water passes through the water softener,the hard water ions become attached to the resin beads and displace the sodium cations.The number of displaced sodium cations equals the charge of the hard water ion trappedin the softener. The hard water ions are bound more tightly to the resin because higher pos-itively charged cations bind more strongly to negatively charged surfaces. Basically, thewater softener removes highly charged cations and replaces them with sufficient sodiumcations to maintain the balance of charge. Thus, sodium contamination increases but it doesnot cause nearly as much trouble as calcium, magnesium, iron, and manganese. There areonly a finite number of resin beads in a water softener so there is a point where the softenerbecomes saturated. At this point, the softener must be recharged. This is accomplished bypassing a saturated salt solution, usually sodium chloride, through the softener. The over-whelming quantity of sodium cations slowly replaces the calcium and magnesium ions,which returns the softener back to working order. Iron and manganese are more difficult toremove from a water softener. Iron and manganese can have very high positive charges, �3and �4, respectively, which make them bind so tightly to the resin bead that the massaction of the regenerating salt solution cannot knock them off. This is typically called “ironpoisoning.” These cations can be washed out of the softener if their charge can be loweredfirst. Reducing agents can be used to lower the charge of the iron and manganese ions, typ-ical “reducing agents” are sodium sulfite, sodium hydrosulfite, and sodium thiosulfate.

Deionization

Basic water softening removes only the undesirable cations by replacing them with lessproblematic cations. It is also possible to replace anions by the same technique, only thistime the charge on the surface of the resin bead is positive and the charge is balanced bypairing up with an anion. The chloride anion is an economical choice for balancing charge


in an anion exchanger. However, excessive chloride content in water can lead to stresscracking of certain stainless steels and promote corrosion on aluminum and mild steels.When industry has to be concerned with both cation and anion contaminants, it is easier toperform a process called “deionization” than it is to replace undesirable cations withsodium and the anions with chloride. Deionization involves passing impure water througha series of cation and anion exhange resins where the cations are replaced by hydrogen ions(H� acid) and the anions are replaced by hydroxide ions (OH� base). The liberated acid andbase then neutralize each other to form water:

H� � OH� → H2O

When deionization is performed, theoretically an equal amount of acid and base is liber-ated. The neutralization reaction should then result in pH-neutral water. This is usually notthe case. Deionized water typically is slightly acidic; pH of 4.5 to 5.5. The mild acidity iscaused by the carbonate impurities initially present in the water. As the carbonate passesthrough the cation exhange column, carbonic acid is formed. The examples below willassume that the carbonates are passing through as sodium salts:

Na2CO3 � 2 H�Resin� → H2CO3 � 2 Na�Resin�

NaHCO3 � H�Resin� → H2CO3 � Na�Resin�

Carbonic acid is not stable in water so it self-destructs to form carbon dioxide, CO2, andwater:

H2CO3 → CO2 � H2O

Carbon dioxide has some solubility in water so unless the water is boiled to drive off theCO2 after deionization, the reverse reaction can occur, which liberates some acid. Thiscauses the low pH.

CO2 � H2O → H2CO3

H2CO3 → H� � HCO3�

Both water softening and deionizing systems are very effective at removing impurities,but they are not perfect. The exchange of ions is really an equilibrium process so somematerial can work its way through a column before the saturation state is reached.Imperfectly sealed control valves and incorrect flow rates can result in some impuritiesnever coming in contact with the resin so that exhange never occurs. The ion exchangecolumns will not effectively remove nonionic impurities. Incorporation of an activated car-bon filter will remove the nonionic impurities.

Reverse Osmosis

General

Reverse osmosis (RO) involves separating water from a solution of dissolved solids byforcing water through a semipermeable membrane. As pressure is applied to the solution,water and other molecules with low molecular weight pass through micropores in themembrane. The membrane retains larger molecules, such as organic dyes, cleaners, oils,


metal complexes, and other contaminants. RO membrane systems feature cross-flow fil-tration to allow the concentrate stream to sweep away retained molecules and prevent themembrane surface from clogging or fouling.

In the past, RO applications for industrial operations were mostly limited to final treat-ment of combined wastewater streams. Such applications typically involved dischargingpermeate to a publicly owned treatment works (POTW) and returning the concentrate tothe head of the wastewater treatment system. Because of the high flow rates associatedwith treating combined wastewater streams, large, costly RO units were required. Morerecent applications in cleaning involved installing RO units in specific process operations(such as wash tank or rinse water maintenance), allowing return of the concentrate to theprocess bath, and reuse of permeate as fresh rinse water.

By closing the loop, process contaminations are removed and fresh water is recycled.Furthermore, a waste stream is eliminated that would otherwise be discharged to thePOTW. RO systems have been successfully applied to a variety of industrial operations,reducing the cost of waste treatment and disposal.

RO Components

The essential components of an RO unit include strainer, pressure booster pump, car-tridge filter, and the RO membrane modules. The strainer removes large, suspended solidsfrom the feed solution to protect the pump. The booster pump increases the pressure of thefeed solution. Typical operating pressures range from 150 to 800 psi. Commercially avail-able cartridge filters are used to remove particulates from the feed solution that would oth-erwise foul the units. Cartridge filter pore sizes are typically between 1 and 5 �m.

Cleaning Formulations

Water has been used as a cleaner for centuries. The first water-soluble soaps were ablend of lye and animal fat. The chemical reaction of this mixture is a process defined assaponification. The addition of heat made the soap work better at removing the oils andgreases of the day, which were also made from animal fat.

As industry advanced and metal processing became more sophisticated, variousorganic and inorganic salts were found to enhance detergency by combining with metalions to prevent them from reacting with the soap. These salts are termed “builders” andinclude phosphates, carbonates, silicates, and gluconates, just to name a few.

The development of synthetic detergents as a substitute for soap in wartime has com-pletely changed the cleaning industry. Today, true soaps make up only a small portion ofthe surfactants used in either industrial or consumer cleaning applications.

By the mid-1970s, government regulations were starting to be felt at the job site. OHSAand Material Safety Data Sheets (MSDS) became common terms within the industrialarena. Chemicals came under increasing scrutiny for worker safety. During the followingdecades, environmental issues have played an increasingly important role in chemicalevaluation. The terms EPA, chlorofluorocarbons (CFCs), the Montreal Protocol, globalwarming, SARA Reportables, and air quality boards are all commonplace. No cleaningprocess can completely escape the concerns of health, safety, and the environment.

CLEANING CHEMISTRY 21

Aqueous cleaners are acid, neutral, or alkaline. Acid products, which have a pH of lessthan 6, are used for removal of inorganic soils and to pickle or passivate a metallic surface.Neutral and alkaline cleaners have a pH range from 6 to above 13. These products are very


effective on organic oils and greases. Additional ingredients are frequently added forincreased effectiveness on inorganic soils as well.

When defined as the removal of soil or unwanted matter from a surface to which itclings, cleaning can be accomplished by one or more of the following methods:

Wetting: Through the use of surface active agents, the cleaning penetrates andloosens the substrate–soil bond by lowering surface and interfacial tension.

Emulsification: Once wetting occurs, two mutually immiscible liquids are dis-persed. Oil droplets are coated with a thin film of surfactant, thus preventingthem from recombining and floating to the surface.

Solubilization: The process by which the solubility of a substance is increased in acertain medium. The soil is dissolved in the cleaner bath.

Saponification: The reaction between any organic oil containing reactive fatty acidswith free alkalies to form soaps.

Insoluble Fatty Acid � Alkali � Water Soluble Soap

Deflocculation: The process of breaking the soil into very fine particles and dis-persing them in the cleaning media. The soil is then maintained as a dispersionand prevented from agglomerating.

Displacement: Soil is displaced by mechanical action. Movement of the workpieceor fluid enhances the speed and efficiency of soil removal.

Sequestration: Undesirable ions such as calcium, magnesium, or heavy metals aredeactivated, thus preventing them from reacting with material that would forminsoluble products (i.e., hard water soap scum).*

Water-based cleaners are generally divided into five major pH groups as follows:

Caustic, pH 12 to 14High alkaline, pH 10 to 13Low alkaline, pH 8 to 10Neutral, pH 6 to 8Acid, pH 1 to 6

Acid Cleaners

Acid cleaners are generally not used for the removal of organic oily soils. A typicalacidic solution with a pH of 4.5 could include citric acid and nonylphenol ethoxylate. Itwould be effective for removing metal oxides or scale prior to pretreatment or painting.Systems using acid cleaners generally require constant maintenance because the aggressivechemistry attacks tank walls, pump components, and other system parts, as well as thematerials to be cleaned. Inhibitors can be used to reduce this attack. Acid cleaners often suf-fer from rapid soil loading, particularly metal loading. This loading leads to frequentdecanting and dumping of the cleaner solution. Both of these disadvantages lead to rela-tively high operating costs compared with alkaline cleaners.

Alkaline and Neutral Cleaners

Ingredients

Ingredients frequently contained in alkaline cleaners include alkaline builders, water condi-tioners, surface-active agents, corrosion inhibitors, fragrances and/or dyes, defoamers or foamstabilizers, and water. Occasionally, hydrocarbon solvents are also added to a formulation.

* See Acknowledgment at end of chapter. © 2001 by CRC Press LLC

Table 2 Alkaline Builders 22

Component Advantage Disadvantage

Caustic Cleaners (pH 12 to 14)

Hydroxides Cost-effective Corrosive

High Alkaline Cleaners (pH 10 to 13)

Amines Detergency corrosion inhibition More costlyCarbonates Detergency, soil holding, low cost ConsumableHydroxides Cost-effective CorrosivePhosphates Detergency, sequestration, corrosion inhibition Environmental restrictionsSilicates Detergency, corrosion inhibition Residues, restricted use

Low Alkaline Cleaners (pH 8 to 10)

Amines Detergency, corrosion inhibition, sequestration More costlyBorates Corrosion inhibition Limited effectSulfates Filler, carrier Restricted use

Alkaline Builders

Alkaline builders are selected based on the pH, detergency, corrosion inhibition,and/or cost limitations required or desired for a specific formulation. Environmental orprocess restrictions must also be considered. These builders may include one or more of theitems listed in Table 2.

Neutral-pH cleaners contain little or no alkalinity builder(s) or the alkalinity reserve isneutralized with an organic or mineral acid.

Water Conditioners

Sequestrants or chelators are frequently used to deactivate undesirable ions such as cal-cium, magnesium, or heavy metals. These ions or heavy metals are then no longer free toreact with bath substances that would subsequently form undesirable compounds, such ashard water soap scum. Some of the more commonly used sequestrants include:

EDTA: Ethylenediamine tetraacetic acidNTA: Nitrilotriacetic acidHEEDTA: Hydroxyethylenediamine triacetic acidSTPP: Sodium tripolyphosphateATMP: Amino tri(methylene) phosphoric acidHEDP: 1-Hydroxyethylidine-1, 1-diphosphonic acidSodium gluconateSodium glucoheptanateLow-molecular-weight polyacrylates

EDTA has maximum effectiveness in tying up calcium and magnesium, thereby softeningthe water used to dilute the cleaner bath. Sodium gluconate or glucoheptanate hasmaximum effectiveness in tying up heavy metals. Complex phosphates are extremely cost-effective, but have come under environmental pressure since the late 1960s. Low-molecular-weight polyacrylates have only found a limited market to date.


Surface-Active Agents

Surface-active agents, also known as surfactants, are used to reduce the surface orinterfacial tension on aqueous solution. Selection of the surfactant package used in cleanerformulations depends on the performance characteristics desired. Surface active ingredi-ents frequently used in water-based cleaners include fatty acid soaps or organic surfac-tants. These surfactants are classified into four basic types:

Anionic: Negatively charged ions that migrate to the anodeCationic: Positively charged ions that migrate to the cathodeNonionic: Electronically neutral ionsAmphoteric: Ions charged either negatively or positively, depending on the pH

Physical properties affected by surfactants include the cloud point, the foaming char-acteristics, and the detergency, emulsifying, or wetting mechanisms used to facilitate thecleaning process. Often a combination of ingredients is used to obtain the specific proper-ties desired.

Corrosion Inhibitors

Corrosion inhibitors are also contained in some alkaline cleaners, depending on theapplication involved. If a wide variety of substrates is involved, a combination of inhibitorsmay be used. These inhibitors are water soluble and therefore are removed with a thoroughrinse if desired.

Inhibitors frequently added to aqueous cleaner formulations include, but are not lim-ited to, aldehydes, amine benzoates, borates, carboxylates, molybdates, nitrites, thiols, tri-azoles, and urea. Phosphates and silicates could also be added to this list. The intendedcleaner application dictates the type of inhibitor package selected.

As the alloying of metals and composites becomes more complex, there is a greaterneed for sophisticated inhibitor packages, which provide protection on a broad spectrumof substrates. The synergism of chemicals allows the formulator to obtain the inhibitingproperties desired and may be limited only to the imagination of the formulator and thecost restrictions of the chemicals selected for this use.

Additional Ingredients

Aqueous cleaner formulations may also contain a broad spectrum of ingredientsdesigned to affect the appearance, odor, or physical properties of the composition. Theseinclude dyes, fragrances, thickeners, defoamers, foam stabilizers, or fillers for cost reduc-tion. Again, the intended cleaner application will dictate final composition of a formula.

Hydrocarbon Solvents

A variety of hydrocarbon solvents have been blended with surfactants to make emul-sion or semiaqueous cleaners. Glycol ethers have been added to stabilize formulations orto increase the cleaning efficiency of a composition. Environmental regulations have iden-tified these ingredients as VOCs, which are regulated by air quality boards. In addition,certain glycol ethers, including 2-butoxyethanol, have been identified as health hazards.


RINSING

Rinsing is often the most overlooked aspect of cleaning. While a process may employthe best cleaning equipment money can buy and an optimized cleaning solution, withoutadequate rinsing the overall result may be unsatisfactory.

Rinsing is no more than a reduction in contamination by dilution. Adding mechanicalor thermal energy may enhance rinsing. Although it is important to use adequate water inrinsing, the key to economy is to use no more than is necessary for acceptable parts.

Importance of Rinsing

Rinsing is a science in itself. Many factors enter into the design of a rinse system. Letus start with the end in mind. Below is a partial list of considerations:

• Part cleanliness required• Production levels required• Type of contamination to be removed (amount and type of dragin and residue)• Incoming water quality• Treatment capabilities• Number of rinse tanks, size, layout• Water usage and disposal

Each concern will be discussed further.

Part Cleanliness Required

This consideration is the controlling factor in the whole process of rinsing. If the man-ufacturer is only concerned with gross contamination, then perhaps rinsing is not evennecessary. At the other extreme, if the slightest residue leads to failure, then perhaps thefinal rinse may need a conductivity of 50 microsiemens/cm (�S/cm) or less. Obviously,final quality must be determined before other decisions can be made.

Production Levels and Dragin

Production levels and dragin will determine what measures are needed to meet theabove determined quality requirements. Dragin will depend on many factors includingpart configuration, orientation, temperature, drain time, etc. Production levels will deter-mine the amount of dragin per time. Rinsing equations must deal with these factors to pre-dict final rinse quality.

Incoming Water Quality

Incoming water quality can vary from naturally soft water with few contaminants towater that contains many hundreds of ppm hardness. Water hardness is generally measuredeither in ppm of equivalent calcium carbonate, CaCO3, or in grains of hardness.

Values greater than 120 ppm are considered to be hard water, with values greater than180 ppm considered to be very hard. Hardness can have many deleterious effects in the clean-ing process. Hardness can react with the surfactants to deactivate them, cause corrosion to


increase on steel surfaces, and leave deposits on the cleaned surfaces.These residues may cause paint adhesion problems, plating problems, or aesthetic

problems, just to name a few. Fortunately, as has been discussed, there are alternatives.Whether softened water, deionized, distilled, or RO water is chosen will depend on theapplication needs.

Number of Rinse Tanks

Generally speaking, one large rinse is less effective than multiple small rinses. As anintermediate step, a single rinse may be adequate, but, as a final rinse, that is seldom true.Experience has shown that past a certain point, generally three rinses, increasing the num-ber of rinses is of little value.

Rinse Tank Design and Placement

Rinse tank design and placement can greatly affect rinsing efficiency. Rather than anafterthought, rinsing needs should be addressed early in the planning of the cleaning line.Considerations should include tank geometry and composition, rinse flow, rinse tempera-ture, and necessary final quality of the parts. It should be noted that rinsing can only beeffective if it reaches the parts. Part orientation, loading, and rinse flow dynamics areimportant and often overlooked considerations.

The rinsing needs are quite different for the manufacturer of circuit board componentsand the rebuilder of motorcycle engines. In the first case the choice might be a heated cas-caded triple rinse with a deionized water source and, for the latter, perhaps a quick dip ina cold tap water tank is sufficient.

A whole chapter could be easily devoted just to basic rinsing concepts. Certain basicprinciples should be considered in any case.

• Multiple rinses are more efficient than single rinses with the optimum balance beingabout three rinses.

• Water usage can be minimized by cascading the final rinse overflow into the previ-ous rinse and that rinse into the one before.

• The final rinse quality is the determining factor in final residue on the part.• Rinsing will not be effective if it does not reach the parts.

For the interested reader, further reading should include Peterson’s Practical Guide toIndustrial Metal Cleaning.13

Dragin and Final Rinse Quality

Knowing and understanding how much and what kind of dragin occurs from the pre-vious step is crucial to predicting rinsing needs. Quite simply dragin over some time equalsthe quantity that must be diluted to some lower specified level. Solve for the quantity ofdiluent. In a dynamic situation at equilibrium the flow rate of overflow rinse water mustequal the dragin times the process chemical concentration divided by rinse chemical con-centration. If this equation is extended to multiple rinse tanks, the dragin is obviouslyreduced in each successive tank. If the overflow for each rinse tank is cascaded to the pre-vious tank, for all practical purposes, the equation takes the form

Overflow � Dragin (process chemical concentration/rinse chemical concentration)1/number of rinse tanks.


This cascading obviously reduces water use dramatically over either single rinsing or mul-tiple rinses that are not cascaded to achieve the same final rinse quality.

The final rinse quality should be maintained just as carefully as the processing tanks.After all, the rinse is the last liquid the parts will see. It does not make sense to go to greatlengths to clean them only to recontaminate them with a contaminated rinse.

Acceptable rinse quality will depend on the needs of the application. As such, the stepsto measure rinse quality will vary with the needs. Generally speaking, cleaning needs onlybe adequate to eliminate subsequent problems.

While rinse quality is relative, some general guidelines may be helpful. Some sourceswould classify applications into general cleaning, critical, and very critical cleaning.Although different standards may be proposed, one suggestion is to use conductivity as aguide and divide these applications as follows:

• General rinsing operations having a residual level of 1000 �S/cm• Critical rinsing with a residual of 500 �S/cm• Very critical at less than 50 �S/cm

Most plating operations call for a high-quality final rinse of less than 50 �S/cm as a mini-mum rinse quality. It can be seen that at these higher-quality rinses, higher-quality watermust be used to achieve the desired cleanliness. These guidelines are suggested as a placeto start when determining final rinse quality.

DISPOSAL

Oftentimes local regulations limit or prohibit the discharge of any process water tosewers. Thus, water conservation becomes an increasingly important issue. Bath life andwater conservation become very important issues. The result is that overall planning andcoordination includes a cradle-to-grave approach for planning and setting up a cleaningline including the rinse and its disposal. For that reason, closed cycle systems for watertreatment may actually be economical long-term alternatives. At the very least water useminimization is an important consideration.

CONCLUSION

Aqueous cleaning is an increasingly important segment of the cleaning industry. Thatimportance will probably increase with time and the development of improved cleaningsystems. The emphasis must be on cleaning systems, as the interactions between parts, soils,equipment, cleaner, and water are much more complex than they appear on the surface.Environmental, economic, and other business concerns demand that industry obtainacceptable parts with minimal impact on the environment and at the least possible cost. Thechallenge to the cleaner manufacturer and the equipment manufacturer is to develop effec-tive cleaners and equipment that meet those criteria and are compatible with each other.

The newer generation of aqueous cleaners is designed to reject contaminants ratherthan emulsify soils. This feature allows the cleaner to be filtered routinely without signifi-cant adverse effect on the cleaner chemistry. These newer formulations can be replenishedwith routine chemical additions of the cleaner concentrate, according to the maintenanceprocedures recommended by the chemical supplier.

Extension of cleaner bath life obtained with regular bath maintenance results in


reduced chemical consumption, reduced waste generation, reduced waste liability, andreduced cleaning costs. Very often the newer cleaning processes also yield cleaner parts aswell.

Aqueous cleaning is changing to meet the economic and environmental needs of thetimes. Clearly the future progress of aqueous cleaning will require close cooperationbetween the chemical and equipment industries. It is also apparent that water quality is amake-or-break issue in critical cleaning, especially as it applies to rinsing. This area is con-sistently the most neglected aspect of aqueous cleaning.

ACKNOWLEDGMENT

Portions of the text were reprinted from Metal Finishing, September 1995, J.A.Quitmeyer, The Evolution of Aqueous Cleaner Technology, pp. 34–39, copyright 1995, withpermission from Elsevier Science.

REFERENCES

1. Archer, W., Reactions and inhibition of aluminum in chlorinated solvent systems, in Corrosion1978 Conference Proceedings, 1978.

2. Betz, Handbook of Industrial Water Conditioning, 8th ed., Betz Laboratories, Inc., Trevose, 1980.3. Durkee, J.B., The Parts Cleaning Handbook, Gardner Publications, Cincinnati, 1994.4. Farrell, R. and Horner, E., Metal cleaning, Metal Finishing, 96, (1), 1998.5. Gruss, B., Cleaning and surface preparation, Metal Finishing, 96, (5A), 1998.6. Hanson, N. and Zabban, W., Plating, 46, 1959.7. Hirsch, S., Deionization for electroplating, Metal Finishing, January, 145–149, 1997.8. Kanegsberg, B.F., Aqueous cleaning for high-value processes, A2C2 Mag., 2, (8), 1999.9. Metals Eng. Q., November 1967.

10. Mohler, J.B., The rinsing ratio applied to practical problems, Part 1, Metal Finishing, May 1972.11. Nelson, W., The key to successful aqueous cleaning is water, Precision Cleaning, Flemington, April

1996.12. Permutit, Water and Waste Treatment Data Book, 18th printing, U.S. Filter/Permutit, 1993.13. Peterson, D.S., Practical Guide to Industrial Metal Cleaning, Hanser Gardner, Cincinnati, 1997.14. PPG Handbook, Vapor Degreasing, 1986.15. Schrantz, J., Rinsing: a key part of pretreatment, Ind. Finishing, June 1990.16. Spring, S., Industrial Cleaning, Prism Press, Melbourne, 1974.17. Vapor zone solvent cleaning systems, Precision Cleaning, Flemington, December 1996.18. Wolf, K. and Morris, M., Ozone depleting solvent alternatives: have you converted yet?”

Finishers’ Manage., June/July 1996.19. Zavadjancik, J., Aerospace manufacturer’s program focuses on replacing vapor degreasers,

Plating Surf. Finishing, April 1992.20. CRC Handbook of Chemistry and Physics, 77th ed., CRC Press, Boca Raton, FL, 1996.21. Quitmeyer, J., The Evolution of Aqueous Cleaner Technology, Metal Finishing, Tarrytown,

September 1995.22. Quitmeyer, J., All Mixed Up: Qualities of Aqueous Degreasers, Precision Cleaning, Fleminton,

September 1997.


CHAPTER 1.4

Review of Solvents for Precision Cleaning

John W. Agopovich

CONTENTS

Review of Solvents for Precision CleaningBackgroundGeneral Requirements of a Precision Cleaning SolventCleaning Methods for SolventsEnvironmental Issues with Solvents

Volatile Organic CompoundsGlobal WarmersFlammabilitySNAP Approval

Discussion of the Different Solvent ClassesFlammable Solvents

Hydrocarbons (Aliphatic)Hydrocarbons (Aromatic)KetonesAlcohols

Halogenated SolventsPerfluorocarbonsHydrofluoroethers

HFEsHFE Azeotropes and Blend

HydrofluorocarbonsHFCsHFC Azeotropes and Blends

Chlorinated SolventsNonozone DepletersOzone Depleters

Other SolventsReferences


REVIEW OF SOLVENTS FOR PRECISION CLEANING

Background

Since the early 1990s, organizations have been actively searching for alternatives forozone-depleting chemicals (ODCs). CFC-113 and 1,1,1-trichloroethane (also called methylchloroform) were the major ODCs utilized for precision cleaning processes. Before theMontreal Protocol mandates, these two materials were excellent for cleaning a wide rangeof contaminants from many different devices. CFC-113 was a good general cleaner for awide range of contaminants, including hydrocarbon- and halogenated-based oils and par-ticles. 1,1,1-Trichloroethane was required for degreasing of parts contaminated with heav-ier oils and residual greases, along with solder flux. Both CFC-113 and 1,1,1-trichloroethanewere available in high-purity grades required for precision cleaning operations and read-ily evaporated after use. Neither material is highly toxic and neither has a flash point. 1,1,1-Trichloroethane does have flammable limits in air, however.

This chapter discusses the several classes of solvents that have been implemented asreplacements for these two materials in precision cleaning. For purposes of definition inthis chapter, a solvent is defined as a cleaning agent that readily evaporates after use, whencleaning a component. No follow-up cleaning is required after the use of this class of mate-rial. These solvents will evaporate even after cold-cleaning. This will generally mean thatthe solvent has a vapor pressure of greater than 25 torr at ambient temperature. Anotherbetter quantifier of volatility is the evaporation rate compared to n-butyl acetate. n-Butylacetate is set at unity (ether and trichloroethylene are also used as reference materials). Themajority of solvents discussed in this chapter all have an evaporation rate � n-butyl acetateor 1.

Another parameter related to volatility is the heat of vaporization, the heat requiredwhen a solvent transforms to the vapor phase from the liquid phase. When possible, it isbest to choose solvents that have a low heat of vaporization, as these materials will evapo-rate without absorbing heat (the part cools). Solvents with high heats of vaporization coola part as more heat is absorbed from the environment. Often water condenses on a devicethat is cleaned, if the environment is humid. This is not desirable in precision cleaning.

Volatility and ease of vaporization have drawbacks. These include issues of contain-ment, flammability, toxicity, local regulations on emissions, and cost. These issues are dis-cussed later in the chapter. Cleaners, such as any aqueous based terpenes and hydrocarbonsin the combustible range (flash point �100°F), do not fall under the definition of a solventand will not be discussed in this chapter.

Critical precision cleaning areas that use solvents under this definition include but arenot limited to medical devices, directional devices (gyroscopes, accelerometers, and com-ponents within), computer components (disk drives), precision ball bearings, oxygentransport systems, and circuit boards.

General Requirements of a Precision Cleaning Solvent

In addition to the need for volatility discussed previously, a precision cleaning solventmust meet other critical requirements. Obviously, the contaminant requiring removal musthave a finite solubility in the cleaning solvent. This definition varies as factors such as time,temperature, and agitation can be altered. Generally speaking, increasing the temperatureof a solvent will improve cleaning effectiveness, as will increasing the exposure time. As anexample, 3M defines “soluble” when a material dissolves in another in the range of 5 to 25g/100 g of solvent at room temperature.1 In another article, solubility has been defined as


50 g/100 ml of solvent.2 There are many other ranges of definitions. The solubility andcleaning effectiveness required will vary depending on how contaminated the part is, aswell as the user’s final requirements.

Other solubility parameters include a Kauri-butanol (KB) number. The KB numbers forCFC-113 and 1,1,1-trichloroethane are 31 and 124, respectively. Generally speaking, thehighly chlorinated compounds have the higher reported KB values. The KB numberreflects the ability of a solvent to dissolve heavy hydrocarbon greases. Specifically, it is ameasure of ability to dissolve a solution of butanol and Kauri resin. ASTM D 1133-90describes the standard test method for determining the KB value of hydrocarbon solvents.This procedure has been extended to evaluate ODC replacements discussed in this chap-ter. It is not applicable to oxygen-containing solvents.

However, the author believes that as a first approximation, the “like dissolves like”concept is very useful. This means that polar solvents dissolve polar contamination andnonpolar solvents dissolve nonpolar contaminants. Hydrocarbon solvents will dissolvehydrocarbon oils and fluorocarbon-based solvents dissolve fluorocarbon oils and greases.To clean solder flux residues from printed circuit boards, polar oxygen-containing solventslike alcohols or chlorinated solvents are required.

In precision cleaning, it is beneficial to have a solvent with a low viscosity and low sur-face tension. This property will allow solvents to enter very narrow gaps in a complicateddevice, to clean a contaminant. Particle removal is often a critical part of precision cleaningoperations. A solvent with a low viscosity and low surface tension also facilitates particleremoval. A high-density solvent provides additional momentum to remove particles fromsurfaces.

Another critical requirement is chemical stability during use and also a solvent havinga long or infinite shelf life. This was been a problem with the use of 1,1,1-trichloroethanewithout stabilizers. Some azeotropes and mixtures discussed later require stabilizers.

Of even more importance is compatibility with the component one is cleaning. Therecannot be a chemical reaction or a physical change such as irreversible swelling or extrac-tion of the materials of construction of the component being cleaned. CFC-113 and 1,1,1-trichloroethane were compatible with most materials. ODC alternative solventmanufacturers are very cognizant of the concerns of customers about solvent compatibil-ity. Extensive solvent/materials compatibility tests are performed on a wide range of mate-rials when a new solvent is introduced to the public. If one still has a question of thecompatibility of material and solvent, it is best to have the solvent user perform the com-patibility test in one’s particular application.

Another critical solvent property is the nonvolatile residue (NVR). It is critical that,when a solvent evaporates from a surface after cleaning, no residue is left behind. Solventmanufacturers typically have NVR specifications in the range of 1 to 10 parts per million(ppm) for precision cleaning. Very often the reported NVR of a given batch of solvent iswell under the company-set specification. The issue of NVR is also important when expen-sive solvents such as the perfluorocarbons (PFCs), hydrofluoroethers (HFEs), and hydro-fluorocarbons (HFCs) are reclaimed and recycled. Any recycling process (such asdistillation) must produce a product with the NVR meeting the original manufacturer(OEM) specifications.

Recycling is desirable for cost savings when using expensive solvents. The halo-genated-containing solvents (PFCs, HFCs, and HCFCs) are particularly expensive.Recycling of used solvents is possible when solvents are used in cleaning and subsequentlycontaminated with particles or high-boiling oils and greases. A simple strip or low theo-retical plate distillation can be performed to purify the reclaimed solvent, to obtain NVRlevels equal or to better than the OEM specifications.


Table 1 Physical Properties of CFC-113 and 1,1,1-Trichloroethane

Vapor SurfaceBoiling Pressure, Density Viscosity TensionPoint mm Hg Flash TWAa (g/cc), (cP), (dyn/cm),

Solvent (°C) at 25°C Point (ppm)b 25°C 25°C 25°C

CFC 113 48 334 None 1000 1.56 0.68 17.31,1,1-Trichloroethane 74 121 None 350 1.32 0.80 25.0

Note: Physical properties compiled from published information.aTime-weighted average.bAmerican Conference of Governmental Industrial Hygienists (ACGIH) 8-h TWA (1998).

Later in this chapter, the topic of azeotropes is discussed. The use of azeotropes is ben-eficial if the user requires expanding the capabilities of an existing single-component clean-ing solvent. In the days when CFC-113 was used, many azeotropes of this material wereavailable and popular for a wide range of applications. In many cases, an azeotrope willallow the ability of a precision cleaning agent to remove contaminants not possible whenusing the pure component. An example is using a PFC/hydrocarbon azeotrope that canclean light hydrocarbon oils. Many existing PFC alternatives form azeotropes with alco-hols, chlorinated solvents (nonozone depleters), and hydrocarbons. The alcohol azeotropesare useful for removing ionic contamination from solder flux residuals.

Listed in Table 1 are typical physical properties of CFC-113 and 1,1,1-trichloroethane.This table is presented for comparison purposes to the upcoming discussion of the recentgeneration of ODC replacement solvents.

Cleaning Methods for Solvents

These solvents can be used in vapor degreasers (both single component and as a cosol-vent), ultrasonic cleaners, power spray (low- and high-pressure), dip, and wipe cleaning.The issues with the flammable solvents and toxicity in equipment must be addressed inlight of local regulation requirements and standards set by each manufacturer.

Environmental Issues with Solvents

Unfortunately, even though most of the solvents discussed in this chapter do notdeplete the ozone layer, there are other environmental issues that must be dealt with. SomeODC alternatives are classified as volatile organic compounds (VOCs), hazardous air pol-lutants, (HAPs), global warmers, or fall under National Emission Standards for HazardousAir Pollutant (NESHAP) regulations.

Volatile Organic Compounds

Any compound that is released in the troposphere can undergo chemical reactions viahydroxyl radical extraction. According to the Environmental Protection Agency (EPA), allsolvents are under the definition of VOC unless specifically exempted. VOCs can lead tothe formation of ground-level, tropospheric ozone. The formation of ozone can in turn leadto the formation of smog via a series of complex free radical and photochemical reactions.Many compounds are exempted from the VOC restrictions, when it is determined that therelease of this material in the troposphere will not lead to ozone formation.


Global Warmers

There is an environmental concern about materials such as PFCs that have no mode ofbreakdown in the atmosphere after release. These materials do not react with hydroxyl rad-icals or ultraviolet (UV) radiation and break down. They simply accumulate and haveatmospheric lifetimes on the order of hundreds or thousands of years. These materials cantrap heat (infrared radiation) from escaping the troposphere. Alternatives to PFCs havebeen developed and have much shorter lifetimes. Another measurement of atmosphericlifetime is global warming potential (GWP). Typical materials are compared with carbondioxide. The carbon dioxide GWP is set at 1.0 for a 100 year integration time horizon (ITH),per the Intergovernmental Panel on Climate Change (IPCC).3

Flammability

CFC-113 and 1,1,1-trichloroethane had the advantage of not being flammable materi-als. Many of the alternatives discussed in this chapter possess flash points. Others do notflash, but have flammable limits in air, like 1,1,1-trichloroethane. Some azeotropes andazeotrope-like mixtures that will be discussed contain a high percentage of a nonflamma-ble solvent (inerting agent).

SNAP Approval

The EPA has required that the Significant New Alternatives Policy (SNAP) programmust approve all alternatives to ODCs. This program was formalized in 1994. Unlessspecifically mentioned in this chapter, all solvents discussed are SNAP approved. Somematerials are SNAP approved with conditions. An example is the PFCs. The PFC materialscan only be used if no other solvents can be used for technical or safety concerns. Othermaterials are conditionally SNAP approved for safety reasons, such as requiring that per-sonal exposure limits be met. This is especially the case with chlorinated solvents.

The safety issues with solvents, such as flammability and exposure limits, should beverified by a user by reviewing the Material Safety Data Sheets (MSDS) and product infor-mation literature from the solvent manufacturer.

DISCUSSION OF THE DIFFERENT SOLVENT CLASSES

Flammable Solvents

Hydrocarbons (Aliphatic)

Aliphatic hydrocarbons such as n-hexane, n-heptane, and isooctane have utility incleaning a wide range of hydrocarbon-based contaminants. These include hydrocarbon-based oils and greases. These materials can be used as wipe solvents to remove residualhydrocarbons and fingerprints. A list of physical properties is shown in Table 2. All theseare exceedingly flammable with flash points well below room temperature. Hydrocarbonsolvents are SNAP approved. These materials have KB values in the range of 30, from com-parison to mineral spirits. Cyclohexane is added to this group as a saturated cyclic hydro-carbon.


Table 3 Physical Properties of Toluene

Vapor Flash SurfaceBoiling Pressure Point Density Viscosity TensionPoint (torr) at (TCC), TWA (g/cc), (cP), (dyn/cm),

Solvent (°C) 20°C °F (ppm)a 20°C 20°C 20°C

Toluene 111 29 35 50 0.87 0.59 28.5

Note: Physical properties compiled from published information. TCC = tag closed cup.aACGIH 8-h TWA (1998).

Table 2 Physical Properties of Aliphatic Hydrocarbons



n-Hexane 69 124 �15 50 0.66 0.31 18 (25°C)n-Heptane 98 36 25 400 0.68 0.41 20.3Isooctane 99 41 10 300b 0.69 0.50 18.8Cyclohexane 81 78 �17 300c 0.78 1 25

Note: Physical properties compiled from published information. TCC � tag closed cup.aACGIH 8-h TWA (1998).bTWA for gasoline.cPlanned to be reduced to 200 ppm per the ACGIH, 1998 TLVs ® and BEIs ® ThresholdLimit Values for Chemical Substances and Physical Agents, “Notice of intended changes for1998,” p. 74.

Hydrocarbon solvents are not terribly aggressive cleaning solvents. However, they arecompatible with a wide range of materials. Flammable hydrocarbons must be used perVOC and HAP regulations, as required. All should be high-performance liquid chro-matography (HPLC)-grade solvents or equivalent.

Hydrocarbons (Aromatic)

Aromatic hydrocarbons such as toluene often display increased cleaning effectivenessas compared with aliphatic hydrocarbons. Toluene is used instead of benzene for the obvi-ous toxicity reasons. The physical properties of toluene are shown in Table 3. Aromatichydrocarbons have KB numbers greater than the aliphatic hydrocarbons and mineral spir-its. The KB number of toluene is 105.

Aromatic solvents are not as popular as the aliphatic materials because of toxicity rea-sons. Toluene is classified as a VOC and HAP. Toluene is SNAP approved. The HPLC gradeor equivalent should also be used.

Ketones

Acetone and methyl ethyl ketone (MEK) are popular cleaning solvents where aggressivecleaning is required. These materials are used for removal of polar contaminants and alsosolder flux and conformal coatings. These materials are sometimes too aggressive for gen-eral cleaning operations so care must be taken if these are used. Compatibility verificationwith organic polymers (plastics, elastomers, etc.) is required for each application. Physical


Table 4 Physical Properties of Acetone and MEK



Acetone 56 185 0 500 0.79 0.36 23.3MEK 80 74 30 200 0.80 0.43 24 (25°C)

Note: Physical properties compiled from published information.aACGIH 8-h TWA (1998).

Table 5 Physical Properties of Short-Chain Alcohols

Vapor Flash SurfaceBoiling Pressure Point TLV/ Density Viscosity TensionPoint (torr) at (TCC), TWA (g/cc), (cP), (dyn/cm),


Methyl 65 97 54 200 0.79 0.55 22.6Ethyl 78 45 58 1000 0.79 1.10 22 (25°C)(200 proof)

Isopropyl 82 32 53 400 0.78 2.40 21.8(15°C)

Note: Physical properties compiled from published information. TCC � tag closed cup.aACGIH 8-h TWA (1998).

properties of acetone and MEK are shown in Table 4. Acetone and MEK are SNAPapproved. HPLC grades are adequate.

Acetone has recently been exempted as a VOC and is not a HAP. KB values are notapplicable for acetone and MEK as they are oxygen-containing solvents.

Alcohols

Short-chain alcohols such as methanol (MeOH), ethanol (EtOH), and isopropyl alcohol(IPA) are also more polar than hydrocarbons and can therefore clean a wider range of con-tamination. Many water-soluble materials are miscible in alcohols. Also, alcohols can beused to follow up an aqueous cleaning process to remove traces of water left behind.Alcohols are popular in mixtures and azeotropes for cleaning solder flux and ionicresidues. These mixtures are discussed later. KB values are not applicable for alcohols.Physical properties are shown in Table 5.

Alcohols are SNAP approved and must be used per VOC and HAP regulations, asrequired.

Halogenated Solvents

This section discusses a popular class of solvents that are heavily halogenated con-taining fluorine and/or chlorine. These solvents are all expensive cleaning materials andrange from $10 to $20/lb.


Perfluorocarbons (PFCs)

3M has had a range of PFCs under the name Fluorinert™ for several years. As thereplacements of ODCs became an issue, 3M introduced a line of solvents referred to asPerformance Fluids (PFs). Some of these were from the Fluorinert line (FC) of fluids thatdid not meet the tight requirements for these FC applications, but were acceptable forcleaning requirements. Examples are FC-84 and PF-5070. The predominant material in bothis perfluoroheptane, but the PF-5070 is acceptable for cleaning. PFCs have a high liquiddensity, a low viscosity, and low surface tension. These physical properties make thesematerials excellent replacements for ODCs in critical particle removal applications.Because these are fluorinated, they are not miscible with hydrocarbon oils. Therefore, theyare not candidates for general cleaning operations (fingerprints, oils, etc.). Properties ofPFCs have been discussed elsewhere.4

Other popular PFCs for cleaning are PF-5060 (predominant component is perfluoro-hexane) and perfluoro N-methyl morpholine (PF-5052). The structure of PF.-5052 is shownin Figure 1. The ring carbons are completely fluorinated.

PFCs are excellent solvents for PFPEs (perfluoropolyethers such as Krytox™ fluids),which are used as lubricants in computer disk drives and precision ball bearings. PFCs areused as carriers to deposit PFPEs on a computer disk drive and also clean them from sur-faces. PFCs are also excellent solvents for cleaning the Halocarbon Corporation productline of chlorotrifluoroethylene (CTFE)-based fluids, as these materials are heavily fluori-nated. Additional details of the use of PFCs in precision cleaning have been discussed pre-viously.4

When PFCs are used for critical particle removal applications, the solvent may have tobe filtered before use. This will depend on the application.

PFCs are relatively inert, nontoxic, and compatible with essentially all materials. Theyare not classified as VOCs. PFCs have a KB number of 0 as they have no miscibility withthe Kauri resin. PFCs are SNAP approved with conditions when they are the only materi-als available for technical and safety reasons.

The physical properties of PFCs are shown in Table 6.

Hydrofluoroethers

HFEsThe HFEs are the 3M second generation of replacements for ODCs. HFEs are ethers

with an n-butyl/isobutyl fluorocarbon group and a shorter hydrogen-containing alkyl

Figure 1 Perfluoro-N-methyl morpholine.


Table 6 Physical Properties of PFCs

Vapor SurfaceBoiling Pressure, Density Viscosity TensionPoint mmHg at Flash TWA (g/cc), (cS), (dyn/cm),

Solvent (°C) 25°C Point (ppm) 25°C 25°C 20°C

PF-5060 56 232 None N/A 1.68 0.4 12.0PF-5070 80 79 None N/A 1.73 0.6 13.0PF-5052 50 274 None N/A 1.70 0.4 13.0

Note: Physical properties taken from recent 3M product information. N/A — not applicable.

group. The HFE-7100 (3M™ Novec™ HFE-7100) has a methyl group and HFE-7200 (3M™Novec™ HFE-7200) has an ethyl group. The structures are shown below.

HFE-7100 CH3–O–C4F9

HFE-7200 CH3CH2–O–C4F9

The atmospheric lifetime, although still significant, is reduced drastically from that of thePFCs because of the hydrogen-containing moiety (carbon–hydrogen bonds). This becomesa site in the molecule for hydroxyl radical attack and breakdown in the troposphere. Thesematerials are SNAP approved by the EPA without conditions. HFEs are therefore the logi-cal replacements for PFCs.

The HFEs have cleaning characteristics similar to PFCs as they are solvents for PFPEsand CTFE fluids. The hydrocarbon side chain allows for increased cleaning effectivenessfor light hydrocarbon oils, especially in the case of HFE-7200, and at elevated temperatures.Because of these solubilities, these materials have KB numbers greater than zero but stillvery low, on the order of 10. Physical properties of the HFE-7100 and HFE-7200 are sum-marized in Table 7. These materials still have the low viscosity, high density, and low sur-face tension (like PFCs) to assist particle removal from surfaces. For some critical particleremoval applications, the HFEs may require filtering before use. 3M has recently intro-duced a high-purity HFE-7100DL (disk lubricant) material, which has controls on particles,ions, metals, etc.

HFEs are compatible with most materials2 and are not classified as VOCs. Both mate-rials have no flash points. HFE-7200 has flammable limits in air. Other details about HFEsare also discussed in Reference 2, 3M product literature, and in Chapter 1.5 by Owens.

Table 7 Physical Properties of HFE-7100 and HFE-7200

Vapor SurfaceBoiling Pressure, Density Viscosity TensionPoint mmHg at Flash TWA (g/cc), (cP), (dyn/cm),

Solvent (°C) 25°C Point (ppm)a 25°C 25°C 20°C

HFE-7100 61 202 None 750 1.52 0.61 13.6HFE-7200 76 109 None 200 1.43 0.61 13.6

Note: Physical properties taken from recent 3M product information.aHFE-7100 exposure guideline from the American Industrial Hygiene Association. 3Mexposure guideline for HFE-7200.


Table 8 Physical Properties of HFE Azeotropes and Blend

Vapor SurfaceBoiling Pressure, TLV/ Density Viscosity TensionPoint mmHg at Flash TWA (g/cc), (cP), (dyn/cm),

Solvent (°C) 25°C Pointa (ppm)b 25°C 25°C 25°C

HFE-71DE 41 383 None See below 1.37 0.45 16.6HFE-71DA 40 381 None See below 1.33 0.45 16.4HFE-71IPA 55 207 None See below 1.48 nlc 14.5

Note: Physical properties taken from recent 3M product information.aThe HFE-71IPA material has no closed-cup or open-cup flash point.See 3M literature about details of the flash points of these materials.b3M reports separately the level of HFE-7100 (750 ppm) and the other components(components were mentioned earlier in chapter).cnl — not listed in literature reviewed.

HFE Azeotropes and Blend

Azeotropes are defined as mixtures that maintain an essentially constant relative pro-portion in the vapor and liquid phase as the mixture components boil. 3M has two com-mercially available azeotropes and one azeotrope-like blend to improve cleaningeffectiveness of hydrocarbon-based materials. These include HFE-71DE, HFE-71IPA, andHFE-71DA, whose physical properties are summarized in Table 8.

HFE-71DE is a 50/50 mixture by weight of HFE-7100 and trans-1,2-dichloroethylene.This material can be used to clean a wider range of hydrocarbon-based oils. The KB valueof this azeotrope is 27, which approaches the KB number of mineral spirits. In general,HFE-71DE is compatible with most materials. However, the high percentage of trans-1,2-dichloroethylene warrants users to verify the compatibility of this azeotrope-like mixturewith certain organic materials.

Another product is a ternary azeotrope, HFE-71DA. This is similar to HFE-71DE, buthas a low percentage of ethyl alcohol to remove ionic materials in flux removal applica-tions. HFE-71IPA is an azeotrope-like mixture of HFE-7100 and approximately 5% IPA. Thesmall percentage of IPA increases solubility of light oils and hydrocarbon oils.

The HFE azeotropes have no flash points. The HFE-71DA and HFE-71IPA have a flam-mability range in air. The HFE-71DE does not have a flammability range in air.

Hydrofluorocarbons

HFCs

HFCs are PFCs where selected fluorines are replaced in the molecule by hydrogen. Inactuality, it is usually not chemically possible to replace fluorines in a PFC molecule. TheHFC molecule is synthesized via a different route than a fluorocarbon, allowing a selectedfew hydrogens in the molecule. As with the HFEs, the HFC have a weak area in the mole-cule where hydroxyl radical attack can occur to break these molecules down in the tropo-sphere. The atmospheric lifetime (global warming potential) of this molecule is much lessthan a PFC and they are SNAP approved without conditions.

DuPont introduced an HFC material with the trade name Vertrel® XF. This solvent isperfluoropentane, with fluorine replaced by hydrogen at the second and third carbons:


CF3CF2CHFCHFCF3

This material is still very similar to a PFC for cleaning effectiveness. PFPEs and CTFE flu-ids are exceedingly soluble in this material. However, XF has limited utility for dissolvinghydrocarbon materials, as noted by a KB value of 9. This material can be used for particleremoval also, as it still has a relatively high density, low viscosity, and low surface tension.

HFC Azeotropes and Blends

There are several azeotropes of XF available, expanding the cleaning capability of theseproducts. These include XM, XE, SMT, and MCA. MCA Plus and XMS Plus are blends. TheXM and XE azeotropes have a small amount of methyl alcohol and ethyl alcohol, respec-tively. These azeotropes can be used for light oil cleaning and particle removal.

The remaining four cleaners contain trans-1,2-dichloroethylene, analogous in conceptto the HFE azeotropes. MCA is a binary azeotrope of XF and trans-1,2-dichloroethylene.The SMT is a ternary azeotrope containing a small amount of methyl alcohol in addition tothe trans-1,2-dichloroethylene, making it useful for cleaning ionic and solder flux contam-ination from circuit boards. MCA Plus is a ternary blend useful for heavy oil and greasecleaning. It contains trans-1,2-dichloroethylene and cyclopentane. XMS Plus is a quater-nary blend and is similar to MCA Plus, but contains a small amount of methyl alcohol. Ittherefore has potential cleaning effectiveness of both the SMT and MCA Plus.

DuPont has the details of the percentages of each component of the above-mentionedmixtures. Physical property and toxicity information is shown in Table 9. All details of XFformulations (compatibility, environmental issues) can be found in the latest DuPontproduct specification literature.

Discussions of some DuPont HFC materials are also presented in Reference 2 and inChapter 1.6 by Merchant.

Another HFC material, benzotrifluoride (BTF), is toluene except the methyl group iscompletely fluorinated. The fluorinated methyl group results in this material having alower KB number than toluene (49 compared with 105). However, it evaporates more

Table 9 Physical Properties of HFC Azeotropes and Blends

Vapor Flash SurfaceBoiling Pressure, Pointa AEL Density Viscosity TensionPoint mmHg at PMCC, Limit, (g/cc), (cP), (dyn/cm),

Solvent (°C) 25°C °F ppm 25°C 25°C 25°C

XF 55 226 None 200b 1.58 0.67 14.1XM 48 298 None 200c 1.49 0.63 14.1XE 52 250 None 235d 1.52 0.73 14.1MCA 39 464 None 200c 1.41 0.49 15.2SMT 37 471 None 192d,e 1.37 0.47 15.5MCA plus 38 461 None 214d 1.33 0.49 16.1XMS plus 38 470 None 197d,e 1.34 0.46 14.9

Note: Physical properties taken from recent DuPont product information.aPensky-Martens closed cup except for XE and XM, which are tag-open-cup tested.bDupont AEL/8- and 12-h TWA.ctrans-1,2-Dichloroethylene and methyl alcohol have a 200-ppm TLV/ 8-h TWA per ACGIH.dACGIH calculation for TLV of mixtures.eSmall amount of stabilizer required.


Table 10 Physical Properties of BTFa

Vapor Flash SurfaceBoiling Pressure Point Density Viscosity TensionPoint (mmHg) @ TCC, CEL (g/cc), (cP), (dyn/cm),

Solvent (°C) 20°C °F (ppm) 20°C 25°C 20°C

BTF 102 30 54 100 1.19 nl 23

Note: Physical properties taken from OxyChem product literature. TCC � tag closed cup, nl � not listed inliterature reviewed.aBTF is SNAP approved per the condition of the 100-ppm exposure limit.

quickly after use, is less toxic, and has a higher density and lower surface tension thantoluene. This KB value is higher than any other pure, single-component HFC, likelybecause of the presence of aromatic hydrogens. This material forms azeotropes with a widerange of materials and has a flash point in the flammable range6. Other details can be foundby review of literature from OxyChem and in Chapter 1.10 on benzotrifluorides by Skelly.Properties are shown in Table 10. BTF is also known as Oxsol® 2000.

Chlorinated Solvents

Nonozone Depleters

Chlorinated solvents are popular because of their excellent solvency for cleaning awide range of contaminants. These materials do not have flash points but all have flam-mable limits in air, similar to 1,1,1-trichloroethane. When 1,1,1-trichloroethane was avail-able for use, it was the preferred degreasing chlorinated solvent. All of these have high KBnumbers, ranging from 90 for perchloroethylene to 136 for methylene chloride.

Methylene chloride CH2Cl2

Trichloroethylene CCl2�CHClPerchloroethylene (PCE) CCl2�CCl2

trans-1,2-Dichloroethylene (trans) CHCl�CHCl

In a recent issue of Parts Cleaning, an article details the toxicity and carcinogenicity of thesethree solvents: trichloroethylene, methylene chloride, and perchloroethylene.5 All thesematerials are suspected of being carcinogenic.

Even though these materials contain carbon–chlorine bonds, none is an ozone depleterbecause of their short atmospheric lifetimes. However, chlorinated ethylenes are VOCs.trans is not a HAP. These materials are SNAP approved with the condition of meeting per-sonal exposure limits.

The chlorinated ethylene materials are not completely stable, need to be protected frommoisture, and some require the presence of stabilizers. A listing of physical properties isshown in Table 11. They are also discussed in Chapter 1.8 by Risotto.

Ozone Depleters

AK-225 is a product offered by AGA chemicals. Of all the solvents discussed in this chap-ter, this material is most similar to CFC-113 in physical properties and therefore cleaningeffectiveness. Some of these details are discussed in Reference 2. They have identical KBnumbers of 31. AK-225 is a solvent for hydrocarbon oils but because it contains fluorine andchlorine, it is also a solvent for heavily fluorinated materials such as the PFPEs and CTFE


Table 11 Physical Properties of Non-Ozone Depleting Chorinated Solvents

Vapor Flash SurfaceBoiling Pressure Point TLV/ Density Viscosity TensionPoint (torr) at TCC, TWA (g/cc), (cP), (dyn/cm),


Methylene 40 350 None 50 1.33 0.44 28.1chloride

Trichloroethylene 87 47b None 50 1.46 0.57 29.5Tetrachloroethylene 121 18b None 25 1.62 nl nltrans-1,2- 48 324b 36d 200 1.26 nl nl

Dichloroethylenec

Note: Physical properties compiled from published information; nl � not listed in literature reviewed.aACGIH TWA (1998).bAt 25°C.cThis material is generally used as a part of a mixture or azeotrope with HFEs and HFCs.dClosed-cup flash point from 7/6/94 Dupont MSDS.

fluids. This material is a mixture of the ca and cb isomers. The chemical structures of thetwo isomers are as follows:

HCFC-225ca isomer CF3CF2CHCl2

HCFC-225cb isomer CF2ClCF2CFHCl

This material is a Class II ozone depleter, with an ODP of 0.03 (CFC-11 is 1.0 on this scale).It is SNAP approved and is not classified as a VOC. These materials are not scheduled forstart of phaseout of production until the year 2015, with complete phaseout by 2030.Interestingly, the ca isomer is more toxic than the cb isomer, necessitating the conditionsplaced upon the SNAP approval. The SNAP conditions that a company-set exposure limitof 25 ppm for the ca isomer is required (the level for the cb isomer is 250 ppm). The expo-sure limit for the mixture was set at 50 ppm by the manufacturer. Physical properties of AK-225 are shown in Table 12. Note at the writing of this chapter that AGA chemicals hasmentioned that a material designated as AK-225 G will possibly be made available for crit-ical government applications. This material is the purified cb isomer, which is less toxic.

Several azeotropes or mixtures with HCFC-255ca/cb exist and are discussed in prod-uct literature. These can be used for heavier oil and grease removal or solder flux removal.Some are discussed in Reference 2.

Table 12 Physical Properties of AK-225

Vapor SurfaceBoiling Pressure, TLV/ Density Viscosity TensionPoint, kg/cm2 Flash TWA (g/cc), (cP), (dyn/cm),

Solvent °C at 25°C Point (ppm) 25°C 25°C 25°C

AK-225(HCFC-255ca/cb) 54 283 None 50 1.55 0.59 16.2

Note: These physical properties are taken from recent AK-225 product information and are further discussed inChapter 1.11 by Miki et al.


Other Class II ozone depleters include HCFC-141b and HCFC-123. The SNAP programdeemed HCFC-141b unacceptable for precision cleaning because of its high ODP, which issimilar to 1,1,1-trichloroethane. HCFC-123 has a similar ODP to AK-225, but has a boilingpoint near room temperature, making it difficult to handle. It can have utility in properlydesigned vapor degreasing equipment. It is SNAP approved with the condition of meetingthe 30 ppm exposure limit.

Other Solvents

This group includes:

n-Propyl bromide (nPB) CH3CH2CH2BrVolatile methyl siloxane (VMS) (CH3)3–Si–O–Si–(CH3)3

para-Chlorobenzotrifluoride (PCBTF)

There are three other solvent-type precision cleaners that merit discussion. n-Propylbromide is a replacement for 1,1,1-trichloroethane in degreasing operations. As of this writ-ing, this material has not yet been SNAP approved but is listed as SNAP pending andtherefore can be legally sold and used. The EPA has petitioned inputs from users of thismaterial to assist in the SNAP-approval process. This material has a low ODP. The exactvalue is being reexamined because of uncertainties in the model when the material has ashort atmospheric lifetime, such as n-propyl bromide. It likely will range from 0.006 to0.027. The EPA also has questions about the toxicity of this material. It will likely be SNAPapproved with conditions. n-Propyl bromide is discussed in Chapter 1.7 by Shubkin.

The n-propyl bromide material has a KB value of 129, which is similar to 1,1,1-trichloroethane. It is therefore an aggressive solvent. Physical properties of this material areshown in Table 13.

Dow Corning has a series of VMSs (volatile methyl siloxanes) under the trade nameOzone Safe (OS) fluids. The OS-10 material has a vapor pressure in the solvent range. OS-10 is SNAP approved, compatible with many materials, and not classified as a VOC. It isflammable as noted in Table 13. The KB values of these materials are low, with OS-10 at16.6. It is suitable for precision cleaning of materials with light to medium hydrocarbon andsilicone oil contamination. VMS is further discussed in Chapter 1.9 by Cull and Swanson.

Table 13 Physical Properties of Other Solvents

Vapor Flash SurfaceBoiling Pressure, Point Density TensionPoint, mmHg at CC, TWA (g/cc), Viscosity (dyn/cm),

Solvent °C 25°C °F (ppm) 25°C 25°C 25°C

n-Propyl (nPB) 70 110 None TBDa 1.33 0.49 cP 26 (20°C)bromideOS-10 100 42 27 200b 0.76 0.65cS 15.2para-Chloro-benzotrifluoride(PCTBF) 139 7.9 109 25c 1.34 0.79cP 25

Note: Physical properties compiled from published information. nPB data from EnSolv; OS-10 properties fromDow Corning; PCTBF data from OxyChem.aTo be determined, EPA estimates 50 to 100 ppmbDow Corning limit.cThe OxyChem Corporate exposure limit for 8-h TWA is also 25 ppm (May 5, 1998 MSDS).


OxyChem produced para-chlorobenzotrifluoride (PCBTF), which is a replacement for1,1,1-trichloroethane. (Oxsol® 100) has a KB value of 64, intermediate for a wide range ofcleaning operations. It is compatible with a wide range of organic polymers. It was SNAPapproved with the condition of a 25-ppm exposure limit. It is exempt from VOC restrictions,is not an ODC, and is not an HAP. The vapor pressure of PCTBF is lower than solvents dis-cussed previously and has a flash point in the combustible range, but does have utility as acleaner if used safely to comply with the low exposure limit. Other details can be found byreview of OxyChem literature or in Chapter 1.10 on benzotrifluorides by Skelly.

[Editor’s note: As of press time, OxyChem is exiting the market and is ceasing production ofbenzotrifluorides. Therefore, the future availability of PCBTF is uncertain. — B.K.]

REFERENCES

1. 3M Fluorinert™ Liquids Product Manual, 1991.2. Agopovich, J.W., PFC alternatives analyses, Precision Cleaning, March 1997 and references cited

therein.3. IPCC, Radiative Forcing of Climate Change, The 1994 Report of the Scientific Assessment

Working Group of the IPCC.4. Agopovich, J.W., Fluorocarbons and supercritical carbon dioxide serve niche needs, Precision

Cleaning, February 1995, and references cited therein.5. Reynolds, R., TCE and cancer fact and fiction, Parts Cleaning, March 1999.6. Ostrowski, P., Benzotrifluoride: a new HFC for cleaning, Precision Cleaning, April 1997.


CHAPTER 1.5

Hydrofluoroethers

John G. Owens

CONTENTS

IntroductionProperties of Segregated HFEs and Their Impact on Cleaning Processes

General PropertiesPhysical PropertiesSolvency and MixturesSafety ConsiderationsEnvironmental ConsiderationsMaterials Compatibility

Cleaning Systems and EquipmentNeat Cleaning SystemAzeotrope Cleaning SystemCosolvent Cleaning SystemParts Cleaning in HFE Cleaning SystemsDrying/Water Removal ProcessesOperation Practices to Maximize Solvent Containment

References

INTRODUCTION

The identification of suitable long-term alternatives to ozone-depleting substances(ODSs) is challenging because of the complex combination of performance, safety, andenvironmental properties required. Researchers investigating alternatives have evaluatedhundreds if not thousands of compounds. This effort has significantly increased the under-standing of the structure–property relationship for a number of compound classes.

Much of the initial research has focused on various halogenated alkanes (i.e., organicmolecules with a carbon–carbon backbone substituted with fluorine, chlorine, or bromine).However, some researchers believe that the requirement that a long-term alternative beboth nonflammable and nonozone depleting effectively eliminates compounds containingchlorine or bromine.1 Recently, a number of researchers have investigated the new class of


hydrofluoroethers (HFEs).2 –7 This class of compounds is nonozone depleting since thecompounds contain neither chlorine nor bromine.8

A variety of HFE structures have been investigated including hydrofluoropolyethers.9

The insertion of an ether oxygen atom into the backbone of the molecule was often done tomodify the thermophysical properties of a compound for specific end uses. However, oneof the principal advantages of the HFE class is that certain HFE structures lead to signifi-cantly improved environmental properties.

Results on numerous segregated HFE compounds demonstrated that they could havesignificantly shorter atmospheric lifetimes and, as a result, lower global warming poten-tials (GWPs) when compared with alkanes such as hydrofluorocarbons (HFCs) and per-fluorocarbons (PFCs).10 Segregated HFEs are those in which all of the hydrogen atomsreside on carbons with no fluorine substitution and are separated from the fluorinated car-bons by the ether oxygen, i.e., CxFyOCmHn. This segregated structure maximizes the effectof the ether oxygen in reducing the atmospheric lifetime of the compound.

The research also showed that segregated HFEs combined these environmental attrib-utes with the performance and safety properties required in a suitable alternative solvent.7

The HFE solvents commercialized to date have all been of the segregated HFE class. Thesematerials have been shown to be low in toxicity.7 They also have physical properties simi-lar to the solvents they replace. Although their inherent solvent power is relatively low,HFEs are readily mixed with other components to produce nonflammable, high-strengthsolvents. As a result, HFEs have been identified as a class of compounds capable of replac-ing ODSs, high-GWP solvents, and chlorinated solvents for a number of new as well asconventional industrial cleaning processes.

PROPERTIES OF SEGREGATED HFEs AND THEIR IMPACTON CLEANING PROCESSES

General Properties

The novel structure of the segregated HFEs results in a unique combination of proper-ties. HFEs are clear, colorless liquids that have very little odor. This class of compoundsoffers a wide range of boiling points from 34°C to well over 100°C. The materials have verylow freezing points, often well below �100°C. The liquid HFEs are low in toxicity, non-flammable, noncorrosive, thermally stable, and electrically nonconductive. Like the sol-vents they replace, these fluids have high densities, but low viscosity and surface tension.

To date, two HFEs compounds have been commercialized. Each consists of two insep-arable isomers with essentially identical properties. The structures and several identifyingnames and numbers are listed in Table 1.

Physical Properties

The physical properties that determine the performance of solvents in typical end-usesare listed in Table 2 in comparison with CFC-113. The higher boiling points of the HFEsmean that they have lower vapor pressures, resulting in easier containment of the fluid incleaning systems. These fluids have extremely low freezing points, which allows the use oflow-temperature cooling coils to enhance fluid containment further. The wide liquid rangeof the HFEs also enables their use in other applications such as heat transfer fluids. Likemany of the materials they are intended to replace, HFEs are nonflammable, having noopen or closed cup flash point.


Table 1 Structures and Names of Commercially Available HFEs

Chemical Structures C4F9OCH3 C4F9OC2H5

Common name Methyl perfluorobutylether Ethyl perfluorobutyletherCAS numbers 163702-07-6 163702-05-4

163702-08-7 163702-06-5CAS names 2-(Difluoromethoxymethyl)- 1-Ethoxy-1,1,2,2,3,3,4,4,4-

1,1,1,2,3,3,3-heptafluoropropane nonfluorobutane1,1,1,2,2,3,3,4,4-Nonfluoro-4- 2-(Ethoxydifluoromethyl)-

methoxybutane 1,1,1,2,3,3,3-heptafluoropropane

Halocarbon numbers HFE-449sccc1 HFE-569sfccc2HFE-449scym1 HFE-569sfcym2

Simplified halocarbon HFE-449s1 HFE-569sf2numbers

Commercial names 3M™ Novec™ HFE-7100a 3M™ Novec™ HFE-7200aRegistered trademark of 3M Company, St. Paul, Minnesota, USA.

Table 2 Physical Properties of HFEs Compared with CFC-113

CFC-113 HFE-449s1 HFE-569sf2

Structure CCl2FCClF2 C4F9OCH3 C4F9OC2H5

Boiling point, °C 48 61 76Freezing point, °C �31 �135 �138Flash point, °C, open or None None None

closed cupSolubility for water, ppmw 110 95 92Solubility in water, ppmw 170 12 3Density, ρ, g/ml 1.56 1.52 1.43Viscosity, µ, cp 0.7 0.6 0.6Surface tension, γ, dyn/cm 17 14 14Heat of vaporization, cal/g 35 30 30Wetting index (1000* �/� * �) 131 181 170

The solubility of water in the HFEs is very low, which limits their ability to absorbhumidity from the air. And their solubility in water is extremely low, minimizing the poten-tial to contaminate contacting water streams. The high density of the HFEs and their verylow viscosity and surface tension are important properties in cleaning applications, partic-ularly for penetrating and cleaning components having complex geometry. Someresearchers in the cleaning industry have suggested that a useful parameter for assessingthe potential performance of a precision cleaning solvent is the “wetting index,” which isdefined as the ratio of the density of the solvent to its viscosity and surface tension.11 Thehigher the wetting index, the better the ability of a solvent to wet component surfaces andpenetrate into tight spaces, especially for the removal of particulate contamination. TheHFEs’ solvent properties indicate that they are well suited for precision cleaning applica-tions with wetting indices exceeding that of CFC-113 (see Table 2).

Another advantage of superior wetting capability is that it provides good drainage ofthe solvent from components at the end of the cleaning cycle. This property, combined with


the low heat of vaporization, allows the HFEs to dry rapidly from part surfaces and mini-mizes fluid dragout from the cleaning process.

Measurements of solvent dragout have confirmed the effectiveness of a solvent with ahigh wetting index. If a part is subjected to a typical cleaning cycle with a dwell in the vaporzone, it will carry residual solvent when it enters the freeboard zone (the area between thetop of the vapor zone and the lip of the machine). Thus, it is usually necessary to hold partsin the freeboard for a period of time prior to removal from the cleaning system to minimizesolvent dragout. The amount of liquid remaining on a representative, high-surface-areapart is shown in Figure 1* for several solvents as a function of residence time in the free-board. Solvents with high wetting indices, such as the HFEs, drain more completely fromthe part surface and dry faster. HFEs exhibit lower dragout compared with the ozone-depleting solvents they are intended to replace. The difference is even greater when com-pared with conventional chlorinated solvents such as trichloroethylene (TCE) andmethylene chloride (MeCl).

Solvency and Mixtures

The segregation of the hydrogen and fluorine atoms on the HFE molecule leads tohigher solvency than if the hydrogen atoms had been placed on a simple fluorochemicalbackbone. However, the pure HFEs are relatively mild solvents as indicated by the solventparameters listed in Table 3. They do not have the hydrocarbon solvency to match the ODSsolvents, but since they can be readily mixed with other materials to form nonflammableazeotropes or cosolvent systems, they can be just as effective in cleaning.

Azeotropes are mixtures that are inseparable by boiling. As a result, azeotropes havefound significant use in cleaning processes. For example, the binary azeotropic mixture ofC4F9OCH3 with trans-1,2-dichloroethylene (t CClH � CClH) as well as the ternaryazeotrope of C4F9OCH3 with trans-1,2-dichloroethylene and ethanol are nonflammable andhave Kauri-butanol values similar to CFC-113. trans-1,2-Dichloroethylene is used in a num-ber of azeotropic solvent mixtures since it has a wide margin of safety with an exposure

Table 3 Solvency Properties of HFEs Compared with CFC-113

CFC-113 HFE-449s1 HFE-569sf2 Azeotrope 1 Azeotrope 2

Structure CCl2FCClF2 C4F9OCH3 C4F9OC2H5a b

Hildebrand solubility 7.3 6.5 6.3 7.7 7.8parameter

Kauri-butanol (KB) 31 10 10 27 33value

Solubility for mineral Mc � 1 � 1 20 20oil (weight %)

Solubility for silicone M � 1 � 1 M Moil (weight %)

Solubility for M M M M Mfluorinated oil(weight %)

aAzeotropic mixture of 50% by weight of trans-1,2-dichoroethylene in C4F9OCH3,commercially available as HFE-71DE from 3M.bAzeotropic mixture of 44.6% by weight of trans-1,2-dichoroethylene and 2.7% by weight ofethanol in C4F9OCH3, commercially available as HFE-71DA from 3M.cM � miscible in all proportions.

* Chapter 1.5 Color Figure 1 follows page 104.


guideline of 200 ppm, is nonozone depleting, and its flammability is easily inerted by mix-ing with a nonflammable solvent such as an HFE.12 The HFEs are miscible with a widerange of organic solvents and form homogeneous, azeotropic mixtures with numerouscompounds.13

Since they are highly fluorinated, HFEs have very high solubility for fluorocarbons andother halogenated compounds. Fluorinated oils and greases are typically completely mis-cible in an HFE solvent.

Safety Considerations

A wide range of safety considerations is necessary for proper design of cleaning sol-vents. As indicated in Table 2, the commercially available HFE compounds are nonflam-mable. These products as well as the commercially available azeotropes have beenassigned NFPA (National Fire Protection Association) flammability indices14 of zero to onesimilar to the ODS solvents. These products do not become flammable under any condi-tions of normal use.12

Beyond flammability, the next most important safety consideration is the toxicity of thesolvent. Under normal conditions, workers can be exposed to a small amount of a solventduring its use. The risk of greater exposure also exists in the event of a solvent spill, leak,or equipment failure. Extensive toxicological tests should be completed to determine if asolvent is safe in its intended applications. A number of the key findings for the HFEs15 arelisted in Table 4 in comparison to CFC-113.

The HFE solvents have very high acute lethal concentrations. That is, workers can beexposed to the solvent in very high doses for short periods of time without adverse effects.The HFEs were found to be nonmutagenic, are not cardiac sensitizers, and are dermallyand ocularly nonirritating. The high exposure guidelines and lack of exposure ceiling indi-cate that the HFEs have a wide margin of safety.

Another important safety consideration is the thermal stability of the solvent. TheHFEs are capable of being continuously refluxed without degradation. Even in the pres-ence of air and metals there has been no evidence of peroxide formation, which is commonto many hydrocarbon ethers.

All solvents can degrade if severely overheated. This degradation can produce by-products that are more hazardous than the original solvent. For example, it was known that

Table 4 Toxicological Properties of HFEs15 Compared with CFC-113



Acute lethal conc., 55,000 � 100,000 � 92,0004 h LC50, ppmv

Mutagenicity Negative Negative NegativeCardiac sensitization 10,000 � 100,000 � 18,900

threshold, ppmvOcular irritant No No NoDermal irritant No No NoExposure guideline 1000 750 200

8-h TWA, ppmvExposure ceiling, ppmv None None None


CFC-113 could produce decomposition products such as HCl and HF if overheated.16

Similarly, the HFEs can generate hazardous decomposition products such as HF if severelyoverheated (e.g., exposure to temperatures 150°C or higher).17 Conventional cleaningequipment has safety interlocks incorporated into its design to prevent overheating thesolvents. In addition, the high temperatures required to decompose an HFE solvent (90°Cor more above its boiling point) provide a wide margin for safe use. Solvent manufacturer’srecommendations should be followed when recycling and recovering solvent for reuse.

The HFEs also have a high degree of chemical stability. The materials are hydrolyticallyand oxidatively stable under normal use conditions. The pure HFEs are stable whenrefluxed in the presence of water or strong aqueous base. Contact with many other rela-tively strong acids and bases produces little, if any, reaction. Exceptions are reactions withamines such as piperidine. As with all halogenated solvents, the HFEs should not be con-tacted with finely divided active metals, alkali, or alkaline earth metals (i.e., groups IA andIIA of the periodic table).

Environmental Considerations

An increasing number of environmental properties need to be considered when select-ing cleaning solvents, such as those listed in Table 5. These properties affect both local envi-ronmental issues, such as smog formation, and global issues, such as ozone depletion andglobal warming. The atmospheric lifetimes of the HFE solvents are in a very desirablerange. They are long enough not to contribute to formation of photochemical smog (i.e.,volatile organic compound, VOC, exempt18) yet short enough to preclude concerns of accu-mulation in the atmosphere.1,19 These short atmospheric lifetimes, compared with PFCs,lead to lower global warming potentials.1,20 Since the HFEs contain no chlorine or bromine,they have no ozone depletion potential. The HFEs are not considered hazardous air pollu-tants since they are low in toxicity.

The stability of the HFEs allows their recovery and recycling for repeated use in clean-ing processes. Used HFE solvents are not classified as hazardous waste; however, the soilspresent in them may change that classification (e.g., metals from a defluxing operationwould typically be considered hazardous waste). For this reason, manufacturers haveestablished return programs to assist users in disposal of used solvents.

Materials Compatibility

The pure HFE solvents are compatible with essentially all common metals, most plas-tics and a number of elastomers. Table 6 lists a number of the specific materials that havebeen tested with the HFEs. Test coupons of the materials listed in Table 6 were exposed to

Table 5 Environmental Properties of HFEs Compared with CFC-113



Atmospheric lifetime, years1 85 4.1 (19) 0.77Ozone depletion potential,1 0.8 0 0

[CFC-11 � 1]Global warming potential,1 6000 320 (20) 55

[CO2 � 1]Photochemical smog precursor18 No No No


the HFE solvents for 7 days at the boiling point of the solvent and subsequently examinedfor weight, volume, and appearance changes. The HFE azeotropes containing trans-1,2-dichloroethylene exhibit compatibility similar to the pure HFEs with all metals but due totheir higher solvency have limited compatibility with most polymeric materials.Compatibility of the components to be cleaned as well as materials of construction for thecleaning equipment should be evaluated with the various cleaning fluids during processselection.

Table 6 Materials Compatibility with HFE Solvents

HFE-449s1 HFE-569sf2 Azeotrope 1

C4F9OCH3 C4F9OC2H5 C4F9OCH3t-CClH � CClH

Aluminum 1 1 1Copper 1 1 1Carbon steel 1 1 1302 stainless steel 1 1 1Brass 1 1 1Zinc 1 1 1Molybdenum 1 1 1Tantalum 1 1 1Titanium 1 1 1Tungsten 1 1 1Acrylic 1 1 3Polyethylene 1 1 2Polypropylene 1 1 2Polycarbonate 1 1 3Polyester 1 1 2Nylon 1 1 1Epoxy 1 1 2PVC 1 1 3PET 1 1 3ABS 1 1 3PTFE 1 1 1Butyl rubber 1 1 2Natural rubber 1 1 3Nitrile rubber 2 2 3EPDM 2 2 2Fluoroelastomer 2 2 2

Polychloroprene 2 2 2

1. Compatible over extended exposures to the solvent with less than 5% changes in weight or volume.

2. Compatible with limited exposure to the solvent (i.e., short time exposures such as a cleaning cycle).

3. Typically incompatible.


CLEANING SYSTEMS AND EQUIPMENT

Several different cleaning processes exist that employ HFE solvents. These include:

1. Neat cleaning systems using pure HFEs2. Azeotrope cleaning systems using azeotropic mixtures3. Cosolvent cleaning systems using zeotropic mixtures

The appropriate cleaning system is selected based upon the soil to be removed, the mate-rials of construction of the part to be cleaned, and a number of other factors as indicated inTable 7. All of the HFE cleaning processes can be conducted in conventional cleaningequipment such as a vapor degreaser (Figure 2) as well as in-line cleaning systems.

Neat Cleaning System

The neat cleaning process, employing pure HFE solvents, is used when a single-com-ponent, mild solvency cleaning fluid is required. This process is conducted in a conven-tional vapor degreaser (Figure 2) and can effectively clean light hydrocarbon and siliconeoils, particulate contamination, and halogenated lubricants, oils, and greases from parts.The HFE solvent can be used in a single or multiple sump system. Parts cleaning with thisprocess is conducted in a manner similar to that used with conventional vapor degreasingsolvents. Both vapor-phase and liquid-phase cleaning is possible depending upon the partsand soils to be removed. The mechanism of cleaning with neat HFE systems can be by dis-solving or displacement of the soil.

Table 7 HFE Cleaning Process Selection Guidelines

Neat Cleaning Azeotrope Cleaning Cosolvent CleaningProcess Process Process

KB value of 10 27–33 20 to � 150cleaning solvent

Soils Light hydrocarbon Medium oils, Heavy oils, greases,and silicone oils, lubricants, release buffing compounds,halogenated oils, agents, some heavy fluxparticulate waxes and fluxes

Substrates

Plastic partsa OK Not appropriate OKMetal parts OK OK OKCircuit boards Not appropriate OK OKCoils OK OK OKBall or roller bearings OK OK OKCathetersa OK May be applicable Not appropriateElastomer partsa OK May be applicable OKGlass parts OK OK OKaCareful evaluation of the compatibility of the parts with the various cleaning fluids shouldbe considered during process selection.


Figure 2 HFE cleaning equipment.

CleaningAgent

CleaningAgent

Cleaning Agent Vapor

FreeboardFreeboard

Secondary Coils

Cleaning Agent Vapor

Cleaning Agent

Primary Coils

Water Separator

Single Sump Multiple Sump

Azeotrope Cleaning System

The azeotrope cleaning process uses an HFE azeotropic mixture, such as those listed inTable 3, as the cleaning fluid. This process is used in applications requiring a stronger sol-vent mixture. Since the solvents used in this process are true azeotropes, the compositionsof the cleaning and rinse sumps remain essentially constant throughout use. The mecha-nism of cleaning with azeotrope systems is almost exclusively via dissolving the soil.The equipment required is similar to the conventional degreasers used with CFCs andtheir azeotropes. The HFE azeotropic mixtures can be used in single or multiple sumpequipment in the same manner as the neat cleaning system described above. This processeffectively cleans many oils, waxes, greases, and fluxes depending upon the azeotropicsolvent used.

Cosolvent Cleaning System

A cosolvent process combines two different fluids to conduct the cleaning process. Oneis a low-volatility, high-solvency organic solvent that is used to dissolve the soil from apart’s surface. The second component, the HFE, functions as a rinsing agent since it is usedto rinse the solvating agent from the component. This process typically uses solvatingagents that are miscible with the HFE. These mixtures are not azeotropes (i.e., they arezeotropes) since the components separate when boiled, resulting in very different compo-sitions in the cleaning sump and rinse sump. The process operates analogously to a two-sump vapor degreaser, with a mixture of the solvating agent and rinsing agent in the boilsump and pure HFE in the rinse sump (see Figure 3). Because of the large difference in boil-ing points between the two components, very little of the solvating agent is distilled intothe rinse sump. The rinse sump contains essentially pure HFE throughout the process. Themechanism of cleaning in this process is most often by dissolving the soil into the mixtureof solvating and rinsing agents in the first sump.

A wide variety of high-boiling, combustible solvents can be used as solvating agents inthis process, provided that their flash points are well above the operating temperature ofthe system. Use of the HFE rinse agent renders the solvating agent nonflammable during


use. Higher-solvency mixtures are created by increasing the ratio of solvating agent to rins-ing agent in the boil sump. The process is controlled by monitoring the boiling temperatureof this mixture since the operating temperature is determined by composition in the boilsump as shown in Figure 4. Temperature and composition control can be accomplishedmanually through periodic measurements of the boiling temperature followed by solventadditions when necessary or via an automated system.

Since this process requires immersion into the solvent mixture in the cleaning sump,vapor cleaning is not possible. The HFE cosolvent process is capable of effectively cleaninga very wide variety of soils including heavy oils, greases, fingerprints, waxes, and flux.Cosolvent processes offer a great deal of flexibility by selecting a solvating agent and rins-ing agent combination that best meets the needs of a particular cleaning application. Theprocess can provide sufficient solvency for a given soil while maintaining compatibilitywith the part’s materials of construction. The higher solvency mixtures of the solvatingand rinsing agents can accommodate higher soil loading levels than the neat or azeotropesystems.

Figure 3 HFE cosolvent cleaning equipment.

SolvatingAgent +

Rinsing

Agent Mix

Rinsing Agent Vapor

Figure 4 Operating temperature of HFE cosolvent cleaning process. Example using C4F9OCH3 andmethyl decanoate as a solvating agent.

RinsingAgent

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .


Parts Cleaning in HFE Cleaning Systems

HFE solvents can be used with a variety of cleaning equipment designs such as con-ventional vapor degreasers, in-line systems, enclosed systems, and vacuum systems. Themost common equipment employed with HFE solvents is a vapor degreaser. Both singleand multiple sump systems are available (see Figure 2). Single sump systems are generallylimited to vapor-phase cleaning and may require some form of spray rinsing. Multiplesump systems typically require immersion in a cleaning sump followed by rinsing in oneor more rinse sumps.

Efficient fluid containment requires holding the cleaned parts in the vapor zone fol-lowed by a dwell in the freeboard zone (the space between the vapor zone and the top ofthe machine) prior to exiting the equipment. This allows the maximum amount of solventto drain and evaporate from the part while it is in the equipment, minimizing fluiddragout. The dwell times during the various steps of the process (cleaning sump, rinsesump, vapor zone, freeboard zone) are typically only a few minutes each, resulting in a rel-atively short cleaning cycle.

Several basic steps are followed in most cleaning processes using HFE solvents toensure the parts are thoroughly cleaned, rinsed, and dried in the specified cycle time. Anexample of a liquid immersion cleaning process would include:

1. Maintain cleaning system at operating temperature (cleaning sump, fluids boil-ing, full vapor zone established, condensate flow into the rinse sump, ultrasonicgenerators operating, etc.).

2. Arrange parts to maximize the flow of solvent around them. Avoid shadowingthe parts from direct spray. Prevent the nesting of parts and the cupping of flu-ids.

3. Slowly lower the basket of parts into the cleaning sump. The parts must be com-pletely covered by the cleaning fluid. The parts must remain covered for theentire immersion time. Ultrasonics or recirculating pumps may enhance cleaningin some applications.

4. Raise the basket of parts above the boil sump and allow excess cleaning solventto drain back into the sump.

5. Move the basket over the rinse sump and completely immerse it into the rinsingfluid.

6. Let parts dwell in the rinse sump for a specified time.7. Slowly raise the basket into the vapor zone, dwelling in the vapor until fluid

drainage has stopped. Rapid ascent of the basket can result in increased vaporlosses.

8. Slowly raise the basket into the freeboard zone. Dwell in the freeboard shouldlast until the solvent has visibly evaporated from basket and parts.

9. Ensure that soil loading in the cleaning solvent is maintained below the leveldetermined for process efficacy. Soil loading is typically monitored via the boil-ing temperature in the cleaning sump (the boiling temperature increasing withsoil loading).

Additional factors to consider when selecting the appropriate combination of cleaning sol-vent and equipment are listed in Table 8.


Drying/Water Removal Processes

HFE solvents are sometimes used as a drying fluid following processes such as aqueouscleaning or metal plating. The drying processes can be conducted in equipment similar to vapordegreasers used for cleaning applications. Water can be absorbed from the surface of a compo-nent using HFE/alcohol solutions such as the azeotrope that forms between C4F9OCH3 and iso-propanol (6.7% isopropanol by weight). These mixtures are typically useful for applicationswith relatively low water removal needs since the solutions can become saturated with water,reducing their efficacy. The solutions can be distilled to remove the water but care must be takento avoid extracting the alcohol into the water and removing it from the drying solution.

Applications demanding larger water removal capability typically use HFE/surfactantmixtures to displace the water from the component’s surface. Parts to be dried are immersedinto a dilute solution of a silicone- or fluorochemical-based surfactant in the HFE solvent. Thesurfactant allows the solvent to wet the surface preferentially. This wetting action causes thewater to bead up and displace from the surface, floating to the top of the solution. The part isthen immersed in one or more sumps of pure HFE solvent to rinse the surfactant from its sur-face. This process is capable of efficiently removing water from complex components in shortcycle times.

Table 8 Equipment Comparisons for HFE Cleaning Processes

HFE Neat HFE Azeotrope HFE CosolventCleaning Process Cleaning Process Cleaning Process

Vapor-phase Possiblea Possiblea Not possiblecleaning

Single sump Possibleb Possibleb Not possibleBoil-up ratea 1–3 times rinse 1–3 times rinse 1–3 times rinse

sump volume/h sump volume/h sump volume/hUltrasonics Suggested for use Suggested for use Suggested for use

where possible where possible where possiblec

Distillation equipment Recommended Recommended Recommended with(external) (external) control features

(internal)Water separator Decanter Desiccant DecanterPump seals PTFE or seal-less PTFE or seal-less PTFE or seal-lessHeaters Sized to field Sized to field Sized to field voltage

voltage with 15% voltage with 15% with 15% safetysafety margin safety margin margin

Filtrationd Recommended Recommended Recommended withwith immersion with immersion all sumpssumps sumps

System piping Welded or Welded or Welded orcompression compression compressionfittings fittings fittingsrecommended recommended recommended

aDepends on part geometry, soil, soil loading, throughput, and part type.bSingle sump systems should be limited to vapor-phase cleaning and may require sprayrinsing.cMay be necessary in both clean and rinse sumps.dAs required by part cleaning specifications.


Operation Practices to Maximize Solvent Containment

In nearly all applications, the HFE solvent emissions are significantly lower than thoseof the ODS it replaces.21 However, to ensure economical operation of the cleaning systemas well as reduce emissions of solvent to the environment, the following practices are rec-ommended to maximize containment of the solvent:

1. Eliminate drafts near the cleaning equipment. Drafts can increase vapor lossesby causing disturbances in the vapor–air interface.

2. A sliding cover is recommended to reduce losses while the system is idling (i.e.,the system is operating but parts are not being processed) and during down-time. Hinged or plug-type lids are typically not as effective since their operationcan cause disturbances in the vapor–air interface.

3. Cleaning equipment should incorporate extended freeboard (120%) and lowtemperature (��20°C) secondary cooling coils to minimize diffusive losses ofsolvent.

4. Do not begin cleaning operation until the equipment is at the operating tem-perature. This will ensure the parts are adequately heated and that the solventwill dry in the freeboard zone prior to exiting the equipment.

5. The cycle times established for specific parts should be strictly followed.Inadequate dwell times in any of the zones (cleaning sump, rinse sump, vaporzone, or freeboard zone) can have a deleterious effect on process performance.Suitable residence times should be established to ensure that the parts are com-pletely cleaned, rinsed, and dried.

6. Use of spray wands should be limited to only those parts where absolutely nec-essary and then only below the cooling coils since misdirected spray canincrease losses.

7. Parts should be arranged so that solvent can efficiently drain and trapping orcupping of fluid is avoided. Tumbling of parts before removing them from thevapor zone can help remove excess solvent and reduce dragout.

8. Parts should not be raised above the vapor zone until the end of the cleaningcycle.

9. Use automated hoists and transfer equipment where possible.10. Perform periodic checks for fluid leaks and follow equipment maintenance pro-

cedures.11. Soil-loaded solvent should be recovered and recycled to every extent possible.

Solvents used in the neat, azeotrope, or cosolvent cleaning processes can bedistilled to be efficiently recovered for reuse following the manufacturer’s direc-tions. Often this distillation can take place within the cleaning equipment.

REFERENCES

1. WMO (World Meteorological Organization), Scientific Assessment of Ozone Depletion: 1998, GlobalOzone Research and Monitoring Project, Rep. 44, WMO, Geneva, 1999.

2. Cooper, D. L., Cunningham, T. P., Allan, N. L., and McCulloch, A., Tropospheric lifetimes ofpotential CFC replacements: rate coefficients for reaction with hydroxyl radical, Atmos. Environ.,26A, 1331, 1992.

3. Cooper, D. L., Cunningham, T. P., Allan, N. L. and McCulloch, A., Potential CFC replacements:tropospheric lifetimes of C3 hydrofluorocarbons and hydrofluoroethers, Atmos. Environ., 27A,117, 1993.


4. Bartolotti, L. J. and Edney, E. O., Investigation of the correlation between the energy of the high-est occupied molecular orbital (HOMO) and the logarithm of the OH rate constant of hydro-fluorocarbons and hydrofluoroethers, Int. J. Chem. Kinetics, 26, 913, 1994.

5. Zhang, Z., Saini, R. D., Kurylo, M. J., and Huie, R. E., Rate constants for the reactions of hydroxylradical with several partially fluorinated ethers, J. Phys. Chem., 96, 9301, 1992.

6. Owens, J. G. and Minday, R. M., Update on hydrofluoroethers alternatives to ozone-depletingsubstances, presented at the International CFC and Halon Alternatives Conference, Washington,D.C., October, 1995.

7. Koenig, T. A. and Owens, J. G., The role of hydrofluoroethers in stratospheric ozone protection,presented at the International Conference on Ozone Protection Technologies, Washington, D.C.,October, 1996.

8.Wallington, T. J., Schneider, W. F., Sehested, J., Bilde, M., Platz, J., Nielsen, O. J., Christensen, L. K.,Molina, M. J., Molina, L. T., and Wooldridge, P. W., Atmospheric chemistry of HFE-7100(C4F9OCH3): reaction with OH radicals, UV spectra and kinetic data for C4F9OCH2 andC4F9OCH2O2 radicals, and the atmospheric fate of C4F9OCH2O radicals, J. Phys. Chem. A, 101,8264, 1997.

9. Marchionni, G., Silvani, R., Fontana, G., Malinverno, G., and Visca, M., Hydrofluoropolyethers,J. Fluorine Chem., 95, 41, 1999.

10. Owens, J. G., Segregated hydrofluoroethers: low GWP alternatives to HFCs and PFCs, presentedat Joint IPCC/TEAP Expert Meeting on Options for the Limitation of Emissions of HFCs andPFCs, Petten, the Netherlands, May 26–28, 1999.

11. Kenyon, W. G., New ways to select and use defluxing solvents, presented at Nepcon West,Anaheim, 1979.

12. Owens, J. G., Hydrofluoroethers: a growing family of alternatives to ozone-depleting com-pounds, presented at the International Conference on Ozone Protection Technologies, Baltimore,November, 1997.

13. Flynn, R. M., Milbrath, D. S., Owens, J. G., Vitcak, D. R., and Yanome, H., U.S. patent 5,827,812,Minnesota Mining and Manufacturing Company, 1998.

14. NFPA (National Fire Protection Association) 49, Hazardous Chemicals Data, 1994 ed., NFPA,Quincy, MA.

15. Product Toxicity Summary Sheets: HFE-7100 and HFE-7200, 3M Company, 1998.16. Ellis, B. N., Cleaning and Contamination of Electronics Components and Assemblies, Electrochemical

Publications Limited, Ayr, Scotland, 1986, 175.17. Material Safety Data Sheet: HFE-7100, 3M Company, 1998.18. U.S. Fed. Regist., 62, August 25, 44900, 1997.19. Wallington et al. (1997) reported that the n- and i-isomers of HFE-7100 were expected to have sim-

ilar reactivity with OH based upon the observation that there was no difference in isomer reac-tivity toward Cl and F radicals. The n-C4F9OCH3 was reported in Wallington et al. (1997) as � 5years and was requoted in WMO Report No. 44 as 5.0 years. The value calculated to two signifi-cant figures is 4.7 years. Later measurements by Molina et al. at MIT on the pure i-C4F9OCH3

determined the lifetime to be 3.7 years. The commercial HFE-7100 is an approximate 60/40 mix-ture by weight of the iso and normal isomers, which results in an average lifetime of 4.1 years.

20. WMO Report 44 reports the GWP of HFE-7100 as 390 over a 100-year integration time horizonusing the lifetime of 5.0 years. Calculation using the 4.1-year lifetime of the commercial productyields a GWP of 320 (100 year ITH).

21. Warren, K. J., Use of hydrofluoroethers in electronics cleaning applications, presented at theInternational Conference on Ozone Protection Technologies, Baltimore, November, 1997.


CHAPTER 1.6

Hydrofluorocarbons

Abid Merchant

CONTENTS

IntroductionStructure and Properties of HFC-43-10mee

Chemical Structure and NomenclaturePropertiesElectrical Properties

Replacement of Ozone-Depleting Substance (ODS) and High Global Warming Gases—Opportunities and Alternatives

Ozone-Depleting SubstancesPFC AlternativesAlternatives to ODSToxicityEnvironmental Regulatory Considerations

Compatibility of Materials of ConstructionPlastics and ElastomersMetals CompatibilityChemical and Thermal StabilitySelective Solvent PowerApplications of HFC-43-10mee, Neat

Carrier FluidParticulate RemovalRinsing AgentFingerprint SolventDisplacement Drying Application

HFC-43-10mee FormulationsBinary Alcohol AzeotropesBinary Heptane AzeotropeBinary 1,2-trans-Dichloroethylene AzeotropesMulticomponent AzeotropeAzeotrope-Like FormulationsNonazeotropic Blends


Equipment Design Considerations for Low EmissionEmission Measurement Data

SummaryReferences

INTRODUCTION

Hydrofluorocarbons (HFCs) are a family of compounds containing carbon, fluorine,and hydrogen. These compounds were developed and commercialized as a result of theMontreal Protocol to replace ozone-depleting substances (ODS), such as chlorofluorocar-bons (CFCs) for refrigeration, foams, propellants, and solvent applications. The absence ofchlorine in the HFC molecules makes them nonozone-depleting substances. The presence ofhydrogen atom(s) reduces atmospheric life, and therefore these compounds have signifi-cantly lower GWP than the similar fully fluorinated or CFC compounds. The presence of alarge number of fluorine atoms tends to make these compounds nonflammable, low in tox-icity, stable to heat, low in reactivity, and compatible with most materials of construction.

In mid-1990s, an HFC containing 5 carbon, 2 hydrogen, and 10 fluorine atoms,CF3CFHCFHCF2CF3 was introduced to replace CFC-113, perfluorocarbons (PFC) (C6F14,C7F16), and hydrochlorofluorocarbons (HCFC-141b and HCFC-225) in some solvent appli-cations. This new HFC, called HFC-43-10mee in ASHRE (American Society of Heating,Refrigeration and Air Conditioning/Engineers) nomenclature, has physical propertiessimilar or better than CFC-113 and C6F14. Compared with CFC-113, this HFC has very lowsurface tension, higher boiling point, and lower heat of vaporization so that it dries rapidlywithout leaving any residue. These properties combined with nonflammability, chemicaland thermal stability, low toxicity, and ease of recovery by distillation make this HFC idealfor a broad range of applications. Solvency lies between CFC-113 and PFC and can beenhanced significantly by use of appropriate azeotropes and blends with alcohol, hydro-carbons, esters, and hydrochlorocarbons.

STRUCTURE AND PROPERTIES OF HFC-43-10MEE

Chemical Structure and Nomenclature

HFC-43-10mee is a straight-chain HFC. Its molecular structure and that of CFC-113 areshown in Figure 1. The absence of chlorine in the HFC molecule requires a large number ofcarbon (5), fluorine (10), and some hydrogen (2) atoms to achieve the desired boiling point,good environmental properties, and nonflammability similar to CFC-113. In the IUPAC(International Union of Pure and Applied Chemistry) nomenclature this compound is1,1,1,2,3,4,4,5,5,5-decafluorpentane or 2,3-dihdrodecafluorpentane. It is marketed under atrade name of Vertrel XF. Because of the position of hydrogens on the second and thirdcarbons, it consists of an identical pair of diesteriomers with very similar properties.

Figure 1 Structures of HFC-43-10mee and CFC-113.

HFC-43-10mee

F

F

F

C C C C C

F F F

F

F

F

F

HH

F

F

F

FC C

C1

C1

CFC-113


Properties

HFC-43-10mee is a clear, dense, colorless liquid with a faint ethereal solvent odor likeCFC-113. Table 1 gives a list of physical, transport, and thermodynamic properties of theHFC along with those of CFC-113. The HFC has a boiling point of 55°C, which is higherthan that of CFC-113. The freezing point is �80°C which is lower than the freezing point ofCFC-113. Thus, it can be used as a liquid over a broader temperature range than CFC-113.The higher boiling point also reduces emissions in the existing degreasing equipment,making it environmentally more friendly. The high liquid density helps to displace soilsand particulate from the surfaces of parts being cleaned and to float these soils to the sur-faces of the solvent. One of the clear advantages of CFC-113 was its low surface tensioncompared with other solvents such as chlorocarbons, hydrocarbons, alcohols, and water.The surface tension of the HFC is lower than that of CFC-113, which makes it easier to wetsurfaces and thus assist in removal of soils through small crevices and openings found in

Table 1 Physical Properties

Propertya HFC-43-10mee CFC-113

Molecular weight 252 187Boiling point, °C (°F) 55 (130) 47.6 (117.6)Vapor pressure, mmHg (psia) 226 (4.4) 334 (6.46)Freezing point, °C (°F) �80 (�112) �35 (�31)Liquid density, g/cc (lb/gal) 1.58 (13.2) 1.56 (13.1)Surface tension, dyn/cm 14.1 17.3Viscosity, cPs 0.67 0.68Solubility in water, ppm 140 170

Solubility of water, ppm 490 110Critical temperature, °C (°F) 181 (357) 214 (417)Critical pressure, psia (atm) 331.9 (22.6) 495 (33.7)Critical volume, cc/mol 433 325Heat of vaporization 31.0 (55.7) 35.1 (63.1)

(at boiling point),cal/g (Btu/lb)

Specific heat at 20°C (68°F), 0.27 (0.27) 0.21 (0.21)cal/g°C (Btu/lb·°F)

Diffusivity, cm2/s 0.066 0.068Thermal conductivity Btu/h ft °F

Vapor 0.0057 0.0043Liquid 0.036 0.043

Refractive index 1.24 1.35Flash point

Closed cupb None NoneOpen cupc None None

Flammable range in air None NoneAutoignition point in air Noned NoneaAt 25°C (77°F) except where indicated.bPensky–Martens closed cup tester (ASTM D 93).cTag open cup tester (ASTM D 1310).dNone detected up to 540°C.


surface mount printed wiring boards and close tolerance precision inertial guidance com-ponents. The energy consumption of a recirculating solvent system in a degreaser is a directfunction of the heat of vaporization of the solvent. The heat of vaporization is lower thanthat of CFC-113, and much lower than that of alcohol, hydrocarbons, chlorocarbons, andwater, making it more energy efficient in use. It has no flash point by both open and closecup methods and no flammable limits in air. Also, it does not have autoignition tempera-ture in air measured up to 540°C.

Table 2 gives both vapor pressure and specific gravity data as a function of temperature.From Antoine’s equation for vapor pressure of HFC-43-10mee,

Log10 P � a � �(c �b

T)�

wherea � 7.03668b � 1093.094c � 208.3936P � millimeter of mercuryT � °C

Electrical Properties

Electrical properties are given in Table 3. The dielectric constant is slightly higher thanthat of CFC-113 and the breakdown voltage is lower than that of CFC-113. Thus, the dielec-tric strength of the HFC is lower but still considered acceptable. The volume resistivity isin the most desirable range to minimize electrostatic discharge (ESD). ESD has becomevery important in the computer and electronic industry. The high-density disk drives usenew technology consisting of giant magneto resistive (GMR) heads, which are extremelysensitive to small changes in current flow through the devices. ESD can result in a momen-

Table 2 HFC-43-10mee Density and Vapor Pressure Change with Temperature

Temperature, °C (°F) Density, g/cc (lb/g) Vapor Pressure, mmHg (psia)

�20 (�4) 1.70 (14.2) 16 (0.3)�10 (14) 1.68 (14.0) 36 (0.7)

0 (32) 1.66 (13.8) 62 (1.2)10 (50) 1.62 (13.5) 109 (2.1)20 (68) 1.60 (13.3) 176 (3.4)30 (86) 1.57 (13.1) 284 (5.5)40 (104) 1.55 (12.9) 434 (8.4)50 (122) 1.51 (12.6) 641 (12.4)60 (140) 1.49 (12.4) 921 (17.8)70 (158) 1.46 (12.2) 1288 (24.9)80 (176) 1.43 (11.9) 1753 (33.9)90 (194) 1.40 (11.7) 2343 (45.3)

100 (212) 1.38 (11.5) 3072 (59.4)110 (230) 1.34 (11.2) 3961 (76.6)120 (248) 1.32 (11.0) 5032 (97.3)130 (266) 1.30 (10.8) 6309 (122.0)


tary current surge, which could as a minimum cause severe damage to the magnetic prop-erties or, at most, a complete failure by the fusion of the GMR head. Therefore, it is desir-able for all materials used in the manufacturing process to have low propensity to generateESD. The ESD properties of a solvent can be measured by measuring its volume resistivity.The resistivity can be classified per EIA-541 and other specifications as:

Conductive Less than 1 � 105 ohm/sq.Dissipative Between 1 � 105 and 1 x 1010 ohm/sq.Insulative Above 1 � 1012 ohm/sq.

Lower value of the resistivity means lower propensity to generate ESD. As can be seen, theHFC has a resistivity number in the range called the “dissipative,” which is consideredmost desirable to minimize ESD generation and at the same time have adequate insulativeproperties.

REPLACEMENT OF OZONE-DEPLETING SUBSTANCE (ODS) AND HIGHGLOBAL WARMING GASES—OPPORTUNITIES AND ALTERNATIVES

Ozone-Depleting Substances

HFC-43-10mee was primarily developed to replace ODS’s CFC-113 and methyl chloro-form in applications where not-in-kind alternatives and other alternative solvents were notacceptable because of safety, process, or material incompatibility problems. To aid under-standing of solvent markets and applications in 1989, before ODSs were restricted, solventuses of CFC-113 and methyl chloroform are provided in Figures 2 and 3, respectively.

With the implementation of the Montreal Protocol, the manufacture of both of theseODSs for solvent applications ceased in January 1996 (except in a few Article 5 countries).It is estimated that the current production of these ODS substances in these countries is lessthan 5% of the peak production of 1989.

PFC Alternatives

PFCs such as C5F12, C6F14, C7F16 and C8F18 were introduced as substitutes for the ODSCFC-113 in the late-1980s to early 1990s, and their uses grew rapidly as the phaseout of CFC-113 began. Prior to this, PFCs were used in small quantities in niche applications. PFCs wereheavily promoted as a substitute for CFC-113 in such applications as carrier fluid for fluori-nated lubricants for the computer hard disk drives; flush fluid for particulate removal inprecision cleaning; drying fluid in displacement drying application; coolants in other elec-tronic components; and rinsing agents in a cosolvent process for cleaning printed circuitboards and mechanical components containing oil, grease, and other soils.

The PFC uses for 1998 are estimated to be around 2500 MT of C6F14 (PFC-5060) and 500to 1000 MT for other PFCs.1

Table 3 Electrical Properties of HFC-43-10mee

Resistivity 2.9 � 109 ohms-cm Dielectric constant 7Dissipation factor 0.14Breakdown voltage 17.8 kV


Alternatives to ODS

The use of CFC-113 and methyl chloroform in electronic applications, especiallydefluxing, has been primarily replaced by three not-in-kind technologies: aqueous, semi-aqueous, and “no-clean.” However, there are a few instances where these not-in-kind alter-natives have not performed satisfactorily or were rejected because of compatibility, safety,flexibility, or reliability reasons and other solvent alternatives have been considered.

HFC solvent (neat and in formulations) is used in very small niche specialty applica-tions. It is estimated that total HFC-43-10mee use will be less than 1000 to 2000 MT and isequal to less than 1% of the CFC-113 market in 1989. The vast majority of the CFC-113 useshave been replaced by not-in-kind technology such as aqueous, no clean, and hydrocarbon.

Figure 3 Methyl chloroform uses in 1989, 706,000 MT. (From ECSA, 1996.)

Aerosols13%

Coatings10%

CPI7%

Electronics4%

Adhesives13%

Vapor Degreasing/Cold Cleaning

53%

Figure 2 CFC-113 uses in 1989, 251,000 metric tons (MT). (Breakdown by DuPont internal data,AFEAS Report, 1998.)

Medical Uses3% Miscellaneous

9%Dielectric/Coolant

2%

Carrier4%

ParticulateRemoval

4%

Drying8%

Metal Cleaning30%

Electronic40%


The ODSs, CFC-113 and methyl chloroform have been replaced primarily by non-HFCalternatives such as no-clean flux, aqueous, and semi-aqueous cleaning processes. It isestimated that the total organic solvent market is only 10% of the pre-phaseout solvent usesand consists of hydrocarbons, chlorocarbons, bromocarbons, HCFC, PFC, HFC, HFE, andHFC.2 Although several HFC solvents have been developed, only HFC-43-10mee has beencommercially available since mid-1995. The volume of HFCs as a solvent substitute isextremely small. Figure 4 provides an estimate of the fluorinated and HFC solvent market,1999 to 2004. It is estimated that the HFC uses will be less than 1% of the CFC-113 uses inpeak year 1989, before the phaseout of CFC-113 began. Also, HFC-43-10mee is an ideal sub-stitute for highly global warming PFC (C6F14, C7F18) compounds and is successfully replac-ing them in many applications. Although HFEs have slightly lower global warmingpotential (GWP), they may not have all the desired performance attributes to be substitutedfor the HFCs in many applications.

For many cleaning applications, HFC-43-10mee blends and azeotropes are preferred.The most common azeotrope or azeotrope-like blends are of HFC with hydrochlorocar-bons, alcohols, and hydrocarbons. These will be discussed in greater detail in other sectionsof this chapter.

Toxicity

HFC-43-10mee has a low order of toxicity on both an acute and chronic basis. DuPonthas established an allowable exposure limit (AEL) for HFC-43-10mee of 200 ppm (v/v: 8-h time weighted average). The AEL is the atmospheric concentration of an airborne chem-ical to which nearly all workers can be exposed repeatedly, day after day, during an 8-h day,or 40-h week without any adverse effect. It has low acute inhalation toxicity as estimatedby the lowest concentration that causes mortality in experimental animals, the approxi-mate lethal concentration (ALC). The 4-h ALC is 10,000 ppm in rats.

In a 90-day inhalation study (6 h/day, 5 days/week) in rats exposed to 0, 500, 2000, and3500 ppm, no effects on body weight, hematology, organ weight, or histopathology wereobserved. The 500 ppm was a no-observed-adverse-effect level (NOAEL). Since there wasno exposure level between 500 and 2000, a subsequent 2-week repeated inhalation study(6 h/day, 5 days/week) was conducted at 1000 ppm. No adverse clinical signs of toxicitywere evident at this 1000 ppm level throughout this study.

Figure 4 Fluorinated solvent market size from 1989 to 2004. (From DuPont estimate.)

Fluorinated Solvents in 2004(Estimated 5% of 1989)

HFC projected use in 2004 (<1%)

Total CFC-113 usein 1989

(251 M Metric Tons)


Several genetic toxicity studies have also been completed and included an in vitroAmes assay, a chromosomal aberration study with human lymphocytes and an in vivomouse micronucleus. No mutagenic changes were observed in any of these studies.Therefore, the evidence from these studies suggests that HFC-43-10mee is not genotoxic.

Inhalation of certain other compounds followed by intravenous injection of epineph-rine, to stimulate human stress reaction, can result in a cardiac sensitization response inexperimental screening studies with dogs. For the HFC, no cardiac sensitization responsewas observed at 1000 and 5000 ppm. At 10,000 ppm seizurelike behavior was observed inthe dogs, thereby preventing detection of a cardiac sensitization response.

The results from a developmental toxicity study show that this material does not haveembryotoxic or teratogenic effects in rats. The results from skin and eye exposure studiesshow that it is a slight skin and eye irritant.

In summary, HFC-43-10mee has low toxicity, an AEL of 200 ppm, and a ceiling limit of400 ppm.

Environmental Regulatory Considerations

The environmental properties of HFC-43-10mee and CFC-113 are given in Table 4.Because HFC has no chlorine, it has zero ozone depletion potential. The presence of hydro-gen in the molecule reduces the atmospheric lifetime to about 17 years, low compared to100 years for CFC-113 and 3200 years for PFC-5060. The reactivity with hydroxyl ions in thelower atmosphere is very low; therefore, emissions to the atmosphere do not contribute toformation of the smog or air pollution. Therefore, the U.S. Environmental ProtectionAgency (EPA) has exempted HFC-43-10mee from the volatile organic carbon (VOC)regulations.

The HFC-43-10mee molecule is recently developed and hence required TSCA listingfrom the EPA for commercial applications. This approval was obtained and the substancewas added to the TSCA inventory listing in 1995. Since 1995, similar approvals have beenobtained in other industrialized countries of the world and the HFC has been added to thechemical inventories of Canada, Japan, European Common (EC) Market countries, Korea,Taiwan, Singapore, and Australia. Countries that are not listed above either do not requireformal application and/or accept approval of the United States and the EC as sufficient forpermitting its use in their countries.

The EPA, under the Significant New Alternatives Program (SNAP), also accepted HFC-43-10mee as a substitute for ODS like CFC-113 for various application categories. It is alsonot considered as a hazardous waste (RCRA) nor a hazardous air pollutant (HAP) andtherefore not subject to NESHAP regulations or SARA-III reporting requirements. Table 5summarizes regulatory status for HFC-43-10mee.

The Montreal and Kyoto Protocols are interconnected because HFCs, developed as sig-nificant substitutes for some important uses of ODSs, are included in the basket of green-

Table 4 Environmental Properties

Property HFC-43-10mee CFC-113

Formula C5H2F10 C2Cl3F3

Atmospheric lifetime, years 17.1 100Ozone depletion potential (ODP) 0.0 0.8Global warming potential (CO2 � 1) 1700 4800VOC Exempt Exempt


house gases (HFC, PFC, SF6, N2O, CO2, and CH4) to be controlled by the Kyoto Protocol.HFCs are effective substitutes in that they are safe in use, they are very energy efficient, andthey have unique properties that help meet key needs of society—such as refrigeration, airconditioning, some areas of health care, and precision cleaning of high tech equipment.

When considered total “greenhouse gases” the fluorocarbon gases contribute less than2% of total carbon equivalent greenhouse gas emission.2 The HFC-PFC task force spon-sored by the Montreal Protocol to resolve differences in Montreal and Kyoto protocol con-cluded that some HFC uses as ODS substitutes are critical and must be continued. The taskforce also recommended that the HFC uses must be responsible and continued where theyprovide the greatest overall value in terms of energy efficiency, environmental protection,performance, and safety.

COMPATIBILITY OF MATERIALS OF CONSTRUCTION

Two types of compatibility testing are conducted, short term and long term. Short-termcompatibility tests, generally a 15-min exposure under a given set of conditions, are usefulfor evaluating compatibility with materials of construction in articles being cleaned.Longer-term compatibility tests, generally a 1- to 2-week period, at or above the boilingpoint, are useful for evaluating compatibility of the materials of construction of cleaningequipment. HFC-43-10mee is a milder solvent than CFC-113 and its compatibility withmost plastics, elastomers, and metals is as good as or better than that of CFC-113.

Plastics and Elastomers

The effect of solvents on plastics or on elastomers (such as swelling, shrinkage,extractables, and hardness) depends on the nature of polymer, the compounding method,the curing or vulcanizing conditions, the presence of plasticizers or extenders, and otherfactors. For these reasons, it is difficult to make generalizations on the effects of any sol-vents on these materials. Testing in the proposed application is particularly important.

Table 6 summarizes the effects of the HFC on a large variety of plastics in theunstressed state exposed at 50°C in a sealed tube tests for a period of 2 weeks. The plasticspecimens were examined before and after the exposure tests for physical change as wellas weight gain, and were assigned empirical ratings from 0 to 2; i.e., 0 means no effect, 1means borderline, 2 means incompatible. These tests show that the HFC has minimaleffects on most commonly used plastics except acrylic. Fluoropolymer plastics show highweight gain; however, the weight returns to normal after air drying. The 2-week test datashould be used for selecting the vapor degreaser construction material and not for assess-ing the compatibility of the parts to be cleaned. The tests also suggest that the HFC may

Table 5 HFC-43-10mee Regulatory Status

Regulations Status

SARA III NoHAP NoRCRA NoHalogenated NESHAP NoSNAP ApprovedVOC ExemptODP Zero


dissolve and extract plasticizer from flexible, highly plasticized PVC tubing and contami-nate the solvent. Therefore, flexible PVC tubing use should be minimized if plasticizers areof concern in the cleaning cycle.

Table 7 gives the effects of the HFC on a large variety of elastomers exposed at 50°C ina sealed tube test for a period of 1 to 2 weeks. The elastomer specimens were examinedbefore and after the tests for linear swell, hardness change, and physical appearance. Thesamples were assigned 0 to 2 empirical ratings similar to the plastics as discussed above.The fluoroelastomers as a group, and two others (“Vamac” and “Alcryn”), were consideredincompatible because of excessive linear swells and hardness change. Fluoroelastomersswelling and shrinking will, in most cases, revert to within a few percent of the original sizeafter air-drying.

Table 6 Plastic Compatibility with HFC-43-10mee SealedTubes (2 weeks at 50°C, 122°F)

Plastic Common Brand Name Rating Weight Gain, %

HDPE Alathon 0 0.3PP Tenite 0 0.5PS Styron 0 0.3PVC 0 0.1CPVC 0 0.1PTFE Teflon® 1a 3.5ETFE Tefzel® 1 1.4PVDF Kynar 0 0.4Ionomer Surlyn® 0 0.5Acrylic Lucite® 2 —b

ABS Kralastic 0 0.0Phenolic 0 0.0Cellulosic Ethocel 1c 4.7Epoxy 0 0.0Acetal Delrin® 0 0.2PPO Noryl 0 0.2PEK Ultrapek 0 �0.1PEEK Victrex 0 �0.1PET Rynite® 0 0.2PBT Valox 0 0.0Polyarylate Arylon® 0 0.0LCP 0 0.1Polyimide

A Vespel® 0 0.0PB Ultem 0 0.1PAI Torlon 0 0.0

PPS Rython 1 2.7Polysulfone Udel 0 �0.1Polyaryl sulfone Rydel 0 �0.1

Rating: 0 � compatible; 1 � borderline; 2 � incompatible.

Physical change: amore flexible; bsample dissolved; csome extraction.


Metals Compatibility

Table 8 summarizes the effect of the HFC on most commonly used metals exposed at100°C in a sealed tube test for a period of 2 weeks. The tests were conducted under two con-ditions: dry solvents and solvents fully saturated with water. The metal coupons wereexamined for change in appearance and weight, and solvent was tested for any reactionproducts. All metals were judged to be compatible, although a slight discoloration of zinc,brass, and copper was seen with the solvent saturated with water.

Chemical and Thermal Stability

HFC-43-10mee is stable to most soils and chemicals commonly encountered in solventuses. It does not hydrolyze with water or moist air. Therefore, no stabilizer is required.

Like all hydrogen-containing halocarbons, it does react with strong bases and amines,particularly in the presence of alcohol. Therefore, contact with such materials should beavoided.

Table 7 Elastomer Compatibility with HFC-43-10mee, Sealed Tubes (2 weeks at 50°C, 122°F)

Elastomer Rating Linear Swell, % Units HardnessChange

Natural rubber 0 �0.6 �1Butyl rubber 0 1.0 �1Nordel® EDPM 0 �1.0 �2Neoprene CR 0 0.2 1Buna S 0 0.7 0Nitrile Rubber

Buna-N 0 �0.6 2NHBR 0 3.9 �8

Vamac® EA 2a 13.9 �12Hypalon® CSM 0 1.3 0Fluoroelastomer

Viton® A 2 17.3 �14Viton® B 2 22.8 �34Zalak® 2a 13.7 �13Kalrez® 2 21.6 �20Fluorinated silicone 2 14.1 �11

Silicone 0 0.5 �4Epichlorohydrin

Homopolymer 0 �0.5 1Copolymer 0 0.0 2

Adiprene U 1a 2.7 �2FA polysulfide 0 1.5 0Thermoplastic

Alcryn® 2a �1.2 13Santoprene 0 0.1 0Geoplast 1a �0.5 �3

Hytrel® Polyester 0 0.3 0

Rating: 0 � Compatible; 1 � Borderline; 2 � Incompatible.aNoticeable extraction affecting rating.


HFC-43-10mee has excellent thermal stability. In laboratory tests using an acceleratingrate calorimeter, no decomposition could be detected at temperature reaching 300°C.

Selective Solvent Power

HFC-43-10mee is a relatively mild solvent, with selective solvency for soils, oils, andgreases. Some commonly used tests to measure the solvency are Hansen and Hildebrandsolvency parameters and Kauri-butanol number (KB). Hansen and Hildebrand solubilityparameters3 for HFC-43-10mee as well as those for C6F14 and CFC-113 are given in Table 9.Based on these data, its is apparent that the HFC has overall solvency between C6F14 andCFC-113. Unlike C6F14, the HFC is completely miscible with most esters, ketones, ethers,ether-alcohols, and the lower alcohols such as methanol, ethanol, and isopropanol. Thelower hydrocarbons, such as pentane, hexane, and heptane, have good solubility in HFC-43-10mee as well as most chlorocarbons and fluorocarbons, including high-molecular-weight fluorocarbon lubricant such as Krytox® and Fomblin®. Thus, HFC-43-10mee is usedas an application carrier fluid or to remove these kinds of compounds.

Unlike CFC-113, the neat HFC-43-10mee has limited solvency for many higher-molec-ular-weight materials such as hydrocarbon oils, silicon oils, fluxes, waxes, and hydrocar-bon greases. However, many HFC formulations containing chlorocarbons, hydrocarbons,esters, and alcohols have enhanced solubility and cleaning efficiency to remove these soils.

Applications of HFC-43-10mee, Neat

Carrier Fluid

HFC-43-10mee has excellent solvency for high-molecular weight fluorocarbon lubricants.Thus, it is ideally suited to replace PFC C6F14, (PFC®-5060), which is currently used inapplying fluorolubricant to computer hard disks. The lube application process consists of

Table 8 Metal Compatibility with HFC-43-10mee(2 weeks at 100°C in sealed tube test)

Rating

Metal Dry Wet

Zinc1 0 0a

Stainless steel 0 0Brass 0 0a

Aluminum 0 0Copper 0 0a

aSlight discoloration.

Table 9 Solubility Parameters

HydrogenCompound Dispersive Polar Bonding Hildebrand

C6F14 5.6 0 0 5.6CFC-113 7.2 0.8 0 7.2HFC-43-10mee 6.3 2.2 2.6 7.8


dipping the disks in a solvent bath containing a small amount of the high-molecular-weight, high-boiling fluorocarbon lubricant. The lubricated disks are either removed fromthe bath or the bath is pumped out of the chamber. The solvent from the disk surface evap-orates, leaving a desirable thin film of the lubricant on the surface. A similar lubricationprocess is also used to apply fluorolubricant to bearings. Also, a dilute lubricant mixture inHFC-43-10mee is being used to apply the lubricant to a specific joint, shutter, or movingpart in precision equipment such as cameras, videos, tape decks, etc.

Particulate Removal

The physical properties such as low surface tension, low viscosity, and high densitymake the HFC very efficient for particulate removal. One common measure of particulateremoval efficiency is wetting index, which is defined as a ratio of density over viscosity andsurface tension. Table 10 gives wetting indexes for selected solvents. A high wetting indexmeans the solvent readily wets the soil and easily rinses away contaminant such as partic-ulate. The wetting index for HFC-43-10mee is higher than many common solvents.

HFC-43-10mee is commercially used in several particulate removal applications. Onesuch application is removal of particulates from camera-original motion picture negativesprior to making a large number of prints for use in theaters. Another such applicationinvolves particulate removal (size range from 0.5 to 60 µm) from miniature ion pumps usedin the Shuttle Hardware Program.4 Wipes saturated with HFC-43-10mee are used on pre-viously cleaned parts of the shuttle hardware to remove particulates before final assembly,and by electricians to clean cable surfaces prior to splicing the wires.

Rinsing Agent

As discussed earlier, HFC-43-10mee has limited solvency for high-molecular-weighthydrocarbon oils and greases. However, HFC-43-10mee is fully miscible with manyaggressive organic solvents such as hydrocarbons, esters, ester-ethers, and ether-alcohols.Unfortunately, these solvents (also called solvating agents) are flammable, combustible, orhave high boiling points. A mixture of HFC-43-10mee with one of these solvents can pro-vide increased solvency, suppression of flammability, and rapid drying. This process iscalled a cosolvent process and is carried out in a conventional two-sump degreaser with aminimum modification. Typically, the first sump of the degreaser is charged with a mixtureof HFC-43-10mee and a solvating agent, and the second sump is filled with pure HFC as arinsing agent. The first sump primarily removes the soil and the second sump provides a

Table 10 Wetting Indexes of Solvents

Surface WettingCleaning Agents Density Tension Viscosity Index

Freon® TF (CFC-113) 1.48 17.4 0.7 122

1,1,1-Trichloroethane (TCA) 1.32 25.9 0.79 65

Isopropyl alcohol (IPA) 0.79 21.7 24 15

DI water 0.997 72.8 1.00 14

Saponifier solution

(6% water) 0.998 29.7 1.08 31

HFC-43-10mee 1.58 14.1 0.67 167


clean rinse of fluorocarbon solvent to remove traces of high-boiling solvating agent left onthe components to be cleaned. Table 11 gives a list of potential solvating agents.

Fingerprint Solvent

Ninhydrin dissolved in CFC-113 is commonly used by law enforcement agencies todevelop latent fingerprints from porous surfaces such as papers. This CFC-113-based for-mulation is now being modified to use HFC-43-10mee by many law enforcement authori-ties to comply with the Montreal Protocol. The HFC-based new formulation works as wellor better than the CFC-113 formulations.

Displacement Drying Application

CFC-113 containing a surfactant was commonly used for spot-free drying after aque-ous cleaning. HFC-43-10mee, with physical properties similar to those of CFC-113, makesan attractive replacement in this application. HFC drying fluid with a proprietary surfac-tant is now commercially available and effectively used in many drying applications.

The solvent drying process is carried out by dipping the wet parts in a boiling solventcontaining a small amount of surfactant. The surfactant and the dense solvent lift the waterfrom the surface of the parts. Water floats to the top of the solvent and is pushed out into acompartment by the circulating solvent, where it is decanted. The parts are subsequentlyrinsed in one or two baths of the pure solvent to remove residual surfactant. The wholeprocess typically takes 3 to 10 min. This solvent drying process is more energy efficient thanconventional oven or hot air drying processes.

Two recent case studies5 indicate effectiveness of HFC-based displacement drying. Onestudy involves drying of optical glass products (changing from a conventional oven dryerto the solvent displacement dryer). The HFC-based process reduced the reject rate(depending on the parts size) from 23 to 65%, and reduced the drying cycle time from 45 to5 min. The second case study involved drying of microprocessor lids where the HFC-43-10-based formulation replaced PFC-5060 (a high global warming fluid).

HFC-43-10MEE FORMULATIONS

Since the chemistry of soils varies greatly, the chemistry of a cleaning agent must be sim-ilarly adjusted to remove them. HFC-43-10mee can be effectively blended with a variety ofother solvents to form specialized cleaning agents for widely divergent cleaning applications.The resulting formulations are used in vapor degreasing, aerosol, or cold cleaning processes.

A thorough search was conducted to select materials for HFC-43-10mee formulations.The materials qualified for blending must have properties that would enhance the perform-ance of the neat HFC-43-10mee for removal of certain soils and must also have acceptable

Table 11 HFC-43-10mee Solvating Agents

Dibasic esters (DBE) Aliphatic hydrocarbonsDiisobutyl DBE Aliphatic alcoholsMethyl decanoate Dipropylene glycol butyl etherIsopropyl myristate Propylene glycol N-propyl etherN-Methyl-2-pyrrolidone (NMP) Dipropylene glycol monomethyletherTetrahydrofurfuryl alcohol (THFA)


environmental and toxicity profiles. The candidates were further screened to pick those thatwould either form azeotropes or behave like azeotropes. Based on these criteria the candi-dates that met these requirements were alcohols (methanol, ethanol, and isopropanol),hydrocarbons (cyclopentane and heptane) and 1,2-trans-dichlotoethylene. Butyl Cellusolve,surfactants, and hexamethyldisiloxane were chosen for some specialty blends.

There are three basic types of HFC-43-10mee formulations: azeotropes, azeotrope-likeblends, and specialty mixtures. Table 12 gives their trade names, components, compositions,and key applications, Table 13 gives their physical, thermodynamic, and environment prop-erties, and Tables 14 and 15 give their compatibility with plastics and elastomers, respectively.

Table 12 HFC-43-10 Formulations

Vertrel Components Composition, wt% Applications

Azeotropes

XM 43-10/Methanol 94/6 Ionic and particulateremoval

Pneumatic systemcleaning

XE 43 10/Ethanol 96/4 Ionic and particulateremoval

XP 43-10/IPA 97/3 Ionic and particulateremoval

KCD-9566 (XH) 43 10/Heptane 93/7 Light oil and particulateremoval

MCA 43 10/t-DCE 62/38 Oxygen systemPrecision cleaningCleanliness verification

Near Azeotropes

SMT 43-10/t-DCE/Methanol 53/43/4 DefluxingPrecision cleaning

MCA Plus 43–10/t-DCE- 50/45/5 Precision cleaningCyclopentane Oil and grease removal

XMS Plus 43–10/t-DCE/ 50.9/43/4/2 DefluxingMeOH/Cyclopentane Precision cleaning

Nonazeotropes

Xsi 43–10/OS-10 57/43 Silicone deposition andremoval

Swelling of siliconetubing

X-B3 43–10/Butyl Cellusolve 97/3 Cutting/drilling fluidX-P10 43–10/IPA 90/10 Ionic and particulate

removalAbsorption drying

X-DA 43–10/Surfactant/ 99.4/0.1/0.5 Dryingantistatic


Binary Alcohol Azeotropes

Three binary azeotropes of HFC-43-10mee with alcohol have been commercialized:XM for methanol azeotrope, XE for ethanol azeotrope, and XP for isopropanol (IPA)azeotrope.

The presence of alcohol makes these azeotropes superior in removing particulates andionic contamination. The methanol-based azeotrope is used for cleaning laser disks andflushing pneumatic lines. The ethanol azeotrope is used for precison cleaning as well asremoval of trace moisture from thin tubes. The isopropanol azeotrope is effectively used forparticulate removal from the large satellite mirrors and disk drive components. Also, theisopropanol azeotrope meets the California South Coast Air Management District VOCexemption guidelines of 50 g/l.

Binary Heptane Azeotrope

HFC-43-10mee forms an azeotrope with n-heptane. The azeotrope has improved sol-vency compared to neat HFC-43-10mee for light oils. The compatibility of this azeotrope isthe same as neat HFC-43-10mee, which makes this material suitable for cleaning polycar-bonates and polyurethane polymers. This azeotrope is marketed as a developmental prod-uct as KCD-9566.

Binary 1,2-trans-Dichloroethylene Azeotropes

Although 1,2-trans-dicholoethylene (trans-DCE) contains chlorine, the presence ofhydrogen and the double bond makes the molecule reactive in the lower stratosphere.Therefore, it has near zero ozone depletion potential. trans-DCE has excellent solvency forhigher-weight-molecular hydrocarbon oils and greases and for many fluxes. The permissi-ble exposure limit of 1,2-trans-DCE is 200 ppm, the same as that of HFC-43-10mee. 1,2trans-DCE is flammable, but the presence of the HFC in the azeotrope makes the azeotropicmixture nonflammable. The azeotrope is marketed under the trade name MCA.

This azeotrope is effective for precision cleaning of mechanical components, bearings,and compressor parts. It is also used as a verification fluid in the NASA shuttle hardwareprogram and other aerospace applications.6 Because 1,2-trans-DCE is somewhat aggressiveto plastics and elastomers it is important that the compatibility of the component materialsto be cleaned should be tested.

Multicomponent Azeotrope

HFC-43-10mee, 1,2-trans-DCE, and methanol form a ternary azeotrope. A small con-centration of nitromethane is added to prevent free radical reaction of alcohol with activemetal (marketed as SMT).

This azeotrope is effective in removing residual flux and ionic impurities in defluxingprinted circuit boards and precision cleaning electronic components. Processes using thisazeotrope have met or exceeded the cleaning requirement for ionic impurities for the mil-itary and the surface insulation resistance.7


Table 13 HFC-43–10 Formulations: Physical,Thermodynamics, and Environmental Properties

Blend

Property XM XE XP XH MCA SMT MCA Plus XMS Plus XSi X-B3 X-P10 X-DA

Physical

Boiling point (°C) 48 52 52 53 39 37 38 38 56.6 61 54 55Liquid density (g/ml) 1.49 1.52 1.53 1.47 1.41 1.37 1.33 1.34 1.0473 1.54 1.42 1.58Liquid density

(lb/gal) 12.43 12.68 12.76 12.26 11.76 11.43 11.12 11.15 8.73 12.84 11.84 13.18KB 9.5 9.4 9.4 9.7 20 38 29 32 17 10.5Vapor pressure

(psia at 25°C) 5.8 4.8 3.9 9.0 9.1 8.90 9.10 2.6 4.4 4.6 4.4Surface tension

(dyn/cm) 14.1 14.1 15.1 14.4 15.2 15.5 16.1 14.9 14 14.1 14.1Freezing point (°C) � �80 � �80 � �80 � �80 � �50 � �50 � �50 � �50 � �50 � �80 �80Heat of vaporization

at bp (cal/gm) 43 35 tbd 35 43 53 51 54 38 31Heat capacity at

25°C (cal/g C) 0.27 0.27 tbd 0.29 0.27 0.28 0.28 0.29 0.27 0.27Viscosity (cps) 0.63 0.73 0.7 0.64 0.49 0.47 0.49 0.46 0.6 0.75 0.67Molecular wt. 178.5 212 228 230.9 155.5 128 136.2 125 203.4 190.8

Environmental

AEL (ppm) 200 235 213 217 200 192 214 197 200 92 238 186ODP 0 0 0 0 0 0 0 0 0 0 0 0GWP (100 year ITH) 1222 1248 1258 1209 806 688 650 662 741 1261 1170 1292HGWP (vs. CFC-11) 0.235 0.240 0.240 0.233 0.155 0.132 0.130 0.130 0.143 0.243 0.225 0.249VOC (g/l) 89 61 50 103 536 645 665 658 Exempt 46 142 8Flash point

(closed cup) None None None None None None None None None None None NoneLEL (vol%) 9 None None None None tbd 6 4 5 None 5 NoneUE (vol %) 11 None None None None tbd 11 14 a None 11 None

Negligible contribution from other compounds.aCondensation—could not determine.

tbd � to be determined.


Table 14 HFC-43-10 Formulations: Plastic Compatibility, 15 Min. @ Room Temperature (22°C)

Compatibility

Tradename Generic XF X-P XE XM XSi MCA MCA Plus SMT XMS Plus

Alathon HDPE 0 0 0 0 0 0 0 0 0Tenite Polypropylene 0 0 0 0 0 0 0 0 0Styron Polystyrene 0 0 0 0 0 2 2 2 2

PVC 0 0 0 0 0 0 0 1 0CPVC 0 0 0 0 0 0 0 1 0

Teflon TFE 1 0 0 0 0 0 0 0 0Tefzel ETFE 0 0 0 0 0 0 0 0 0Surlyn Ionomer 0 0 0 0 0 0 0 0 0Plexiglas Acrylic 2 0 1 1 0 2 2 2 2

ABS 0 0 0 0 0 2 2 2 2Phenolic 0 0 0 0 0 0 0 0 0

Ethocel Ethyl cellulose 1 0 0 2 0 2 0 0 2Epoxy 0 0 0 0 0 2 2 2 0

Delrin Acetal 0 0 0 0 0 2 0 0 0Noryl PPO 0 0 0 0 0 0 0 0 2Ultrapek PEK 0 0 0 0 0 2 2 2 0Victrex PEEK 0 0 0 0 0 0 0 0 0Rynite PET 0 0 0 0 0 0 0 0 0Valox PBT 0 0 0 0 0 0 0 0 0Arylon 0 0 0 0 0 0 0 0 0Zenite LCP 0 0 0 0 0 0 0 0 0Vespel PI 0 0 0 0 0 0 0 0 0Ultem PEI 0 0 0 0 0 0 0 0 0Torlon PAI 0 0 0 0 0 0 0 0 0Ryton PPS 1 0 0 0 0 0 0 0 0Udel PSO 0 0 0 0 0 1 1 2 2Radel PES 0 0 0 0 0 0 0 0 0

0 � Compatible

1 � Borderline

2 � Incompatible


Table 15 HFC-43-10 Formulations: Elastomer Compatibility, 15 Min. @ Room Temperature (22°C)

Compatibility

Tradename Generic X-P XE XM X-DA XSI MCA MCA Plus SMT XMS Plus

Adiprene Polyurethane 0 0 0 0 0 2 2 2 2Krynac Nitrile 0 0 0 0 0 2 2 2 2Plioflex Styrene/ 0 0 0 0 0 0 0 0 0

ButadieneEthylene

Nordel Ethylene/ 0 0 0 0 1 0 0 1 1Propylene

Hypalon Chlorosulfonated 0 0 0 0 0 2 2 2 2Polyethylene

Natural Isoprene 0 0 0 0 0 1 1 1 1Rubber

Neoprene Chloroprene 0 0 0 0 0 0 1 1 1Butyl Isobutyl isoprene 0 0 0 0 0 2 2 1 2Silopren Silicone 0 0 0 0 0 1 1 1 1Thiokol FA Polysulfide 0 0 0 0 0 2 1 1 1Viton Fluoroelastomer 0 1 1 1 1 2 2 2 2

0 � Compatible

1 � Borderline

2 � Incompatible


Azeotrope-Like Formulations

Two azeotrope-like formulations of HFC-43-10mee are available for precision cleaningand defluxing, respectively. The first azeotrope-like formulation is a ternary consisting ofHFC-43-10mee, 1,2-trans-DCE, and cyclopentane and is marketed as MCA Plus. This for-mulation has higher solvency for hydrocarbon oils and greases than MCA and is used forheavier oils and greases. This formulation is used for precision cleaning components andmachined parts with close tolerances.

The second formulation is a four-component azeotrope-like mixture consisting ofHFC-43-10mee, 1,2-trans-DCE, cyclopentane, and methanol, and is marketed as XMS Plus.A small amount of nitromethane is added to protect against free radical reaction of alcoholwith active metals. This formulation has a slightly better solvency for the fluxes andimproved compatibility with PWB components. Therefore, it is used in defluxing PWB andprecision cleaning of metal components.

Nonazeotropic Blends

There are four nonazeotropic blends of HFC-43-10mee commercially available for spe-cialty applications.

1. A proprietary blend of HFC-43-10mee with hexamethyldisiloxane is marketed asXsi. It has excellent solvency for the high-molecular-weight silicones and, thesame time, good compatibility with polycarbonates and polyurethane. The sol-vent is used as a carrier fluid to apply or remove silicone lubricant from the sur-faces of medical devices used for insertion into the human body. It is also used asa swelling media for silicone rubber tubings. Both components of this formula-tion are VOC exempt and therefore the blend is a non-VOC. This blend is flam-mable and in use can separate into flammable compositions; hence, it should beused in flammable-rated equipment.

2. A proprietary blend of HFC-43-10mee with 2-butoxyethanol (butyl Cellosolve) ismarketed as X-B3. This blend replaces a CFC-113 blend with 2-butoxyethanolthat was used to serve a lubricant and heat removal function in cutting anddrilling thick aluminum sheets or plates. The HFC blend was developed to servea similar purpose.

3. A proprietary blend of HFC-43-10mee with 10% isopropanol is marketed as X-P10. The higher alcohol helps in the removal of water and ionic impuritiesfrom nonporous surfaces and for absorption drying. Although the blend is non-flammable as formulated, because it is nonazeotropic, its composition may shiftin use and it could become flammable.

4. A proprietary blend of HFC-43-10mee with a small amount of fluorosurfactantadditive (� 0.1 % by wt), an antistatic additive (0.5 %), is commercially availableas X-DA. This formulation works extremely well as a displacement drying fluid.As discussed earlier, this blend offers one-step, low-energy, spot-free drying thatis efficient, safe to use, and environmentally responsible.


EQUIPMENT DESIGN CONSIDERATIONS FOR LOW EMISSION

With increased environmental awareness and increased cost of solvents, it is impera-tive that the equipment designed to operate with the HFC solvents have the state-of-the-art emission reduction technology. The primary enhancement features8 recommended areas follows:

• An extended deep freeboard: freeboard to width ratio of 1.2 to 2.0• A secondary condenser for vapor diffusion control operating at�20 to�30°F• Piping systems containing welded or soldered joints to minimize joint leaks• Hoods and/or sliding doors on the top-entry machines

Some additional enhancements, although costly, may further reduce emission:

• Automated work transport facilities• Facilities for superheated vapor drying

The emission measurement tests9 run with an unmodified degreaser and with the pri-mary enhancement features showed emission rate reduction from 0.075 to 0.0155 lb/hr ft2,or about 80% reduction in emission loss. A rough calculation indicates the cost of retro-fitting an existing degreaser can be recovered in reduced solvent consumption in as little as6 months of operation.

Emission Measurement Data

Several vapor-in-air concentration measurements have been made for HFC-43-10meesolvents in the working space immediately adjacent to the degreasers with and withoutenhancement features.9 In all cases, the emission has been well below the allowable expo-sure limit of 200 ppm. Table 16 gives the actual emission measurement data under variousoperating conditions around an enhanced Ultronix-modified Barons–Blakeslee 120 modeldegreaser (higher freeboard and a secondary low-temperature coil). As one can see, theaverage emission is less than 10 ppm.

Table 16 HFC-43-10mee Vapor Concentration in Air Measured Adjacent to a ModifiedBBI-MSR-120 Vapor Degreaser

Vapor-in-Air Conc, ppm HFC-43-10mee

Sample Location Operating Conditions Range of Values Average Values

Front and behind Degreaser idling with lid open 2.3–18 7Front Degreaser idling with lid open 2.3–12 624 in, behind Degreaser idling with lid open 4.9–18 9.6

degreaser Cleaned six basket loads of 1.6–31 8.4parts/lid open at all times

Cleaned four basket loads of 0.6–38 8.5parts/lid closed duringidling period


SUMMARY

HFC-43-10mee is an option that is critical but only in a very small segment of the pre-vious CFC-113 markets. The replacement for the ODS solvents and/or high-GWP PFCsshould be selected based on a holistic assessment of performance, environmental protec-tion, safety, and health impacts. Responsible use of any alternatives including HFC-43-10mee is strongly encouraged. Where possible, equipment should be upgraded withemission-reduction technologies.

REFERENCES

1. Harnisch J. et al., Primary aluminum production: climate policy, emission and cost, presented atJoint IPCC/TEAP Meeting on Options for the Limitation of Emission of the HFC and PFCs,Petten, the Netherlands, 1999.

2. Report of the HFC and PFC Task Force of Technology and Economic Assessment Panel (TEAP)of the Montreal Protocol, October 1999.

3. Fritz, H. L., DuPont KSS report.4. Jones, D. W., Evaluations of ODS-free particulate removal from miniature ion pumps, presented

at Int. CFC and Halon Conf., October 1995.5. Westbrook, G. A., et al., Solvent-based drying system offer fast spot-free results, Precision

Cleaning, June 1999.6. Wittman, C. L., et al., The search for a replacement for CFC-113 in the precision cleaning and ver-

ification of shuttle hardware, presented at Int. CFC and Halon Conf., October 1995.7. Hanson, M. R. B., et al., Performance testing of HFC azeotropes for precision cleaning, presented

at Int. CFC and Halon Conf., October 1995.8. Ramsey, R. B., et al., Considerations for the selection of equipment for employment with HFC-43-

10mee, presented at Int. CFC and Halon Conf., October 1995.9. Ramsey, R. B., et al., DuPont KSS report.


CHAPTER 1.7

normal-Propyl Bromide

Ronald L. Shubkin

CONTENTS

IntroductionHistorical Developmentnormal-Propyl BromidePhysical PropertiesCleaning PowerDryingCompatibility

Plastics and ElastomersMetalsDrums and Drum Linings

Thermal StabilityHydrolytic StabilityComparison of nPB Properties with Other New Cleaning SolventsSpecial Formulations—Electronics

Compatibility with MetalsCompatibility with Plastics and ElastomersRemoval of Ionic Residues from Integrated Circuit BoardsPrevention of Silver Tarnish

Health, Safety, Environmental, and Regulatory IssuesToxicology—Acute

Mammalian Genetic ToxicityMammalian MetabolismAquatic Acute Toxicology

Recommended Exposure LimitEnvironmental and Health Regulatory StatusEnvironmental and Health-Related Parameters Compared with Other

New SolventsSummaryCase Histories

1. Electronic Equipment2. Electric Motor Stators and Refrigeration Coils


3. Implantable Body Parts4. Aluminum Parts for Optical Applications5. High-Performance Inertial Navigation Systems

References

INTRODUCTION

The cleaning of complex parts to exacting specifications has become a major consider-ation in the manufacture of a wide variety of machines, appliances, and instruments. Theintroduction of chlorinated solvents provided manufacturers and fabricators a convenientand economical way to clean a host of difficult soils contaminating strategic parts. Efficientcleaning, rapid drying, low flammability, residue-free parts, and relatively low solventcosts all contributed to the popularity of chlorocarbon and hydrochlorocarbon fluids.However, many chlorine-containing solvents have now been banned or restricted becauseof environmental and/or health considerations. With the mandated elimination of themost popular cleaning solvents, many manufacturers switched to aqueous or semiaqueouscleaning systems. Although these proved to be viable solutions in many applications, theywere not suitable for all situations. In the search to find more appropriate alternatives, awide variety of new solvents were developed.1 Many of the new solvent cleaners do excel-lent jobs, but still suffer from one or more deficiencies relative to the overall cost/perform-ance of the chlorinated materials they replaced. Some of the newer solvents, such as somehydrochlorofluorocarbons (HCFCs), have been shown to have environmental problemsand either have been banned from use in cleaning applications or are scheduled for phase-out. In other cases, the newly introduced solvents meet the environmental and toxicologi-cal requirements, but do not meet the performance standards. The need in specializedapplications for a high-performance cleaning agent that could be used in a safe and effi-cient manner led to the development of cleaning systems based on the solvent normal-propyl bromide (nPB).

HISTORICAL DEVELOPMENT

In 1991, the total U.S. market for 1,1,1-trichloroethane (TCA) as a cleaning solvent was700 million lb. Another 200 million lb were sold as emissive solvents. TCA was, and stillcould be, a very effective and popular cleaning solvent, but it suffers from two importantenvironmental drawbacks. It has a relatively high ozone depletion potential (ODP) and arelatively high global warming potential (GWP). The manufacture of TCA was banned bythe Montreal Protocol effective January 1, 1996.

A host of new cleaning systems have been introduced in recent years to meet the chal-lenges presented by today’s industrial cleaning requirements. All of the new systems offeradvantages in specified niche applications, but none has met all the requirements of themarketplace. Some of the important cleaning options are as follows:

• Alternative chlorocarbon solvents. Most chlorinated solvents are effective clean-ing agents. However, most have been banned from manufacture, or restricted tospecified applications, or scheduled for phaseout, or have regulations restrictingemissions and requiring extensive reporting.


• Hydrocarbons and oxygenated hydrocarbons. These solvents have the obviousadvantage of a historical low cost. However, most are readily flammable andpresent serious hazards when used in cleaning operations. Many also have toxi-cological problems.

• Hydrochlorofluorocarbons (HCFCs). As with many chlorocarbon solvents, themost popular of these (HCFC-141b) has been restricted to noncleaning applica-tions. HCFC-225 is still being used in cleaning, but it is scheduled for phaseout.Other types of HCFC molecules are too volatile, have only moderate cleaningability, or are not considered cost-effective in specific application areas.

• Fluorocarbons. Fluorocarbons are nontoxic, nonflammable, and very safe to use.However, they tend to be expensive and have poor solvency for most soils. Thenotable exception is their ability to solubilize highly fluorinated oils and greases.

• Hydrofluorocarbons (HFCs). HFCs have low or moderate solvency for most soilsof interest and historically have been expensive. They are excellent for nicheapplications, and they are often blended with more aggressive cleaning solvents.The environmental acceptability of the blend tends to be dependent on the sec-ond component.

• Hydrofluoroethers (HFEs). HFEs are similar to HFCs in solvency, cost, and overallperformance. Again, blends are sometimes used to boost cleaning performance.

• Volatile methyl siloxanes (VMSs). VMSs, such as hexamethyldisiloxane, are lowin toxicity and contain no halogen atoms. They are chemically very stable. On theother hand, they have flash points and only moderate solvency.

• Semiaqueous systems. The problem of proper disposal of semiaqueous systemsis often overlooked. Because of the high organic content, it is not appropriate (orlegal in most cases) to dispose of these systems down the drain. Separation andrecycle of the organic phase is usually difficult and is not cost-efficient. Slow dry-ing and potential corrosion problems may also come into play.

• Aqueous systems. Aqueous systems are very inexpensive in terms of detergentcost, but they are not suitable for all applications. High capital investment, mul-tistep processing, a large equipment footprint, and high energy costs are oftenreported. Residuals on the clean parts and difficult drying are also problems.Corrosion of metal parts may become a factor. Finally, electrical and electronicapplications often cannot tolerate the presence of remaining traces of water.

• No-clean systems. Some manufacturers have eliminated the need to clean at var-ious stages of manufacture. Sometimes this requires a change in the manufactur-ing process or the order of assembly.

normal-PROPYL BROMIDE

In the mid-1990s, a new solvent/cleaner based on nPB was developed to meet theneeds of those who require the cleaning efficiency of the chlorinated solvents but who mustmeet the strict environmental standards for a replacement solvent.2 The newsolvent/cleaner does not suffer from many of the shortcomings of other alternative sol-vents that have been offered to the market. nPB is an effective cleaning agent. It is safe touse under the proper conditions, has a low ODP and a low GWP, and it is not regulatedunder most environmental and worker safety rules. It is compatible with metals, has a lowtendency to cause corrosion, and may be used in most current vapor degreasing equip-ment. It is easily recycled and is moderately priced.


Table 1 Physical Property Comparisons

1,1,1-Tri-n-Propyl chloro- Trichloro-Bromide ethane ethylene HCFC-141b HCFC-225

Boiling point, °C 71 74 87 32 54Specific gravity, 25°C 1.35 1.32 1.46 1.24 1.55Viscosity, 25°C, cP 0.49 0.79 0.54 0.43 0.59Vapor pressure, 20°C, 110.8 100 57.8 593 285

torrSpecific heat, 25°C 0.27 0.25 0.22 — 0.25Latent heat of vapor, 58.8 57.5 57.2 52.3 33

cal/gSolubility in water, 0.24 0.07 0.11 0.18 0.033

g/100 g waterSolubility of water, 0.05 0.05 0.03 0.042 0.03

g/100 g solventSurface tension, 25°C, 25.9 25.6 26.4 19.3 16.2

dyn/cmFlash point, TCC, °C None None None None NoneFlammability limits, 4–7.8 7–13 8–10.5 7.6–17.7 None

vol %

PHYSICAL PROPERTIES

Table 1 compares the physical properties of nPB to two hydrochlorocarbons and twoHCFCs. HCFC-141b is CH3–CCl2F, and HCFC-225 is a mixture of the two isomersCF3–CF2–CHCl2 and CF2Cl–CF2–CHClF. The physical properties of the nPB are similar toall four of these solvents, but particularly so to TCA.

CLEANING POWER

Indicators that relate to the cleaning ability of a solvent are the solubility parameters.These are the Hildebrand parameter, the Kauri-butanol number (KB) and the Hansenparameters. As indicated by the data in Table 2, the values for nPB compare quite well withthe common chlorocarbons.

Although solubility parameters provide a guide to cleaning power, they do not take theplace of experimental results. To compare the cold cleaning ability of neat nPB to neat chlo-rinated solvents, a simple test procedure was devised.3 Solutions of typical soil contami-nants (30 wt%) were prepared in solvent. Steel wool wedges were weighed and thensoaked in the contaminated solvent, drained, and dried at 100°C for 30 min. The wedgeswere reweighed and the weight of retained soil recorded. The impregnated wedges werethen placed in short glass tubes and washed with 3 ml of the test solvent. The wedges weredrained, dried, and weighed as before. The grams of soil lost per milliliter of test solventgive a measure of cleaning power.

Figure 1, the cleaning power of chlorinated solvents relative to that of neat nPB, showsthat nPB-based cleaners have fractionally lower cleaning ability than TCA when used toclean mineral oil, equivalent performance on grease and silicone oil, and superior per-formance in removing polyol esters. These experimental results are consistent with the


Figure 1 Relative cleaning ability, at ambient temperature, of nPB and chlorinated solvents.

Table 3 Comparisons of Solvency Characteristics

Solvent KB

nPB 125HFC-43-10 mee 9HFE (methyl ether) 10HCFC-225 31VMS (hexamethyldisiloxane) 17

Hansen parameters, which show that nPB has a slightly lower value for nonpolar materi-als and higher values for polar and hydrogen-bonding compounds.

The cleaning power of nPB-based cleaners is clearly equivalent to the popular chlori-nated solvents that have been banned or restricted. Comparisons to the new alternative sol-vents that have been introduced to replace the chlorinated materials, however, are morerelevant to today’s needs. Again, it is instructive first to compare the relative solvencypower of nPB with some of the new solvents that are being offered in the cleaning market.

Table 3 compares the KB of nPB with that of several alternative solvents that have beenintroduced in recent years. By this one measure, superior performance is expected from the

Table 2 Solubility Parameters

n-Propyl 1,1,1- Trichloro- Methylene Perchloro-Bromide TCAa ethylenea chloridea ethylenea

Hildebrand parameterb 18.2 17.4 18.8 19.8 19.0KB 125 124 129 136 90

Hansen parameters

Nonpolarb 16.0 17.0 18.0 18.2 19.0Polarb 6.5 4.3 3.1 6.3 6.5Hydrogen bondingb 4.7 2.1 5.3 6.1 2.9a From CRC Handbook of Solubility Parameters and Other Cohesion Parameters, 2nd ed., Allan F. M. Barton,Ed., CRC Press, Boca Raton, 1991.b Data presented as ��(MPa)1/2.


cleaning formulations based on nPB. The abbreviations used here and throughout the restof this chapter are as follows:

• nPB: normal-propyl bromide• HCFC-225: dichloropentafluoropropane (two isomers)• HFC-43–10mee: decafluoropentane• HFE: perfluorobutyl alkyl ethers (especially the methyl ether)• VMS: volatile methyl siloxanes (especially the dimer, hexamethyldisiloxane)

Of course, the KB is only an indication of cleaning performance. Figure 2 shows theresults of experiments that were conducted in the same manner as those carried out to com-pare nPB with the chlorinated solvents. For these experiments, performance was comparedfor the removal of mineral oil, silicone oil, and a complex mixture of common soils that sim-ulates the ASTM 448 Standard Soil (some minor substitutions were made). These experi-mental results indicate clearly superior cleaning performance for nPB-based cleaners. It isfor these reasons that the other solvents are often sold as blends with more aggressive sol-vents that for one reason or another cannot be sold by themselves as cleaning solvents.

Cold cleaning is suitable for a variety of cleaning needs, but the very difficult soils usu-ally require more severe conditions. For this reason, difficult soils are often removed usingvapor degreasing equipment. In this equipment, the parts may be cleaned in the hot vaporsof the solvent. For the most difficult soils, especially those that have been “baked on,” thepart to be cleaned is often immersed in the boiling solvent. Because cold cleaning per-formance does not always reflect performance at higher temperatures, a new cleaning testwas designed.

Steel coupons (C1010) were weighed and coated with the selected soil (�0.1 g). Thecoupons were placed in an oven at 250°C for 1 h, cooled, and reweighed. The coupons werethen immersed in boiling solvent for 5 min, dried, and reweighed. Each trial was con-ducted in triplicate. The weight percent of soil removed from each coupon was recordedand averaged for the three trials.

Because formulation plays a part in total cleaning efficiency, commercial cleaning sol-vents were used for this comparison. Five formulated solvents were chosen. The solventswere an nPB-based cleaner (Albemarle ABZOL™ VG Cleaner), HFC-43–10 mee (DuPontVertrel® XF), HFE (3M HFE-7100), HCFC-225 (Asahi Glass AK-225), and VMS (Dow

Figure 2 Relative cleaning ability, at ambient temperature, of nPB and four newer solvents.


Figure 3 Cleaning of tough, baked-on soils. Coupon immersed in boiling solvent for 5 min.

Figure 4 Relative evaporation rates of nPB and four common chlorinated solvents.

Corning OS-10). The soils chosen were mineral oil, a polyol ester, and a rosin-based solderflux. The results that are shown in Figure 3 indicate that the nPB cleaner removes thebaked-on mineral oil much more efficiently than the other cleaning solvents. The baked-onpolyol ester and the rosin-based solder flux remain nearly untouched by the other solvents,but are mostly removed after 5 min in boiling nPB cleaner.

DRYING

From the standpoint of raw material costs, the most economical class of substitutes forchlorinated solvents would appear to be aqueous-based systems. However, aqueous clean-ing procedures suffer from a number of difficulties, one of which is the slow drying. Dryingrates can be improved by the installation of specialized drying equipment, but this is costlyand usually contributes significantly to the process time. Cleaning systems based on nPBhave drying rates that are comparable to the chlorinated solvents. To compare evaporationrates, loss of weight from 2 ml of solvent at room temperature (24°C) was measured after 5min. Figure 4 compares the relative rate of evaporation of nPB with four common chlori-nated solvents.


An alternative, but popular, benchmark is the evaporation rate relative to butyl acetateat room temperature. For 1,1,1-TCA, the rate is 5.8 times that of butyl acetate. For nPB, therelative evaporation rate is 6.2—marginally faster than 1,1,1-TCA.

COMPATIBILITY

Because nPB is a very aggressive cleaner base, care must be taken to ensure that it iscompatible in the short term and at the appropriate temperature with all of the componentsof the parts being cleaned. Similarly, care must be taken that the nPB-based cleaner is com-patible for the long term with the materials of construction for the cleaning equipment andthe storage and handling facilities.

Plastics and Elastomers

Plastics and elastomers that pass short-term compatibility tests with nPB (immersionin boiling solvent for 15 min) include the following:

Plastics Elastomers

Acculam™ epoxy glass Adiprene™ polyurethaneAlathon™ HDPE Aflas™ PTFEDelrin™ acetala Buna-N™ rubberKynar™ polyvinyl fluoridea Kalrez™ fluorelastomera

Nylon™ (6 and 6.6) Neoprene™ polychloroprenePhenolicsa Viton-A™ fluoroelastomerb

Polyester (filled and unfilled) Viton-B™ fluoroelastomerb

PolypropyleneTeflon™ PTFEa

Tefzel™ ethylene/PTFEa

XLPE™ cross-linked PEa These materials are also compatible for long-term (2 months)immersion at elevated temperature (65°C).b The Viton fluorelastomers are marginal for long-term (2 months)immersion at elevated temperature (65°C).

Plastics and elastomers that were found to be unsuitable (U) or marginal (M) for con-tact with nPB at elevated temperature for short periods (15 min) include:

Plastics Elastomers

Low-density polyethylene (M) Butyl rubber (M) EPDM-60 (U)Ultem™ polyether imide (M) NBR nitrile rubber (M) Silicone (U)

Metals

nPB must be formulated with passivating agents to be used with many metals. Becausethe type and concentration of the passivators determines the performance, the followinginformation applies only to ABZOL® VG Cleaner (VG). It was tested for compatibility with


metals according to Mil-T-81533A 4.4.9. This is a metal corrosion test that was originallydesigned to test the suitability of TCA for military applications. The metal coupon is heldhalf-submerged in the refluxing cleaning fluid for 24 h. It is then examined for signs of cor-rosion. All of the following metals passed this test:

Nickel Inconel TitaniumBrass Copper ZincMonel Aluminum TantalumStainless steel 316L Carbon steel 1010 MagnesiumStainless steel 304L

Fresh aluminum surfaces react immediately with TCA at room temperature. nPB ismuch less reactive toward aluminum. If an aluminum coupon is scratched beneath the sur-face of TCA, which contains no metal passivators, there is an immediate formation of adark, brownish red color. If the test is repeated using neat nPB, no color formation isobserved. At reflux, some small dark spots are observed on the edges of the aluminumcoupons after 3 to 4 hours. Similar tests were done with magnesium. Fully formulated nPBis completely safe for use with aluminum, magnesium, and other active metals.

Drums and Drum Linings

In addition to the corrosion test, 2-month immersion studies were carried out at 130°Fwith carbon steel 1010, stainless steel 316, and high-baked phenolic linings. All of thesematerials were shown to be suitable for long-term storage of cleaning fluids containingnPB. Most perfluorinated plastics are also suitable for storage.

THERMAL STABILITY

Knowledge of the thermal stability and thermal degradation products of a new solventis important for safety considerations during use. Buildup of contaminants on heating ele-ments, for example, can cause localized hot spots that may degrade the solvent. It is impor-tant to know at what temperature this will occur and to ensure that the products of thedegradation are not dangerous or highly toxic. Two different approaches were taken todetermine thermal stability.

Thermal degradation studies were conducted by Columbia Scientific using the accel-erating rate calorimetry (ARC) method. This method identifies the temperature for theonset of degradation by detecting the accompanying exotherm. For a fully formulatedcommercial cleaner based on nPB (VG), a significant exotherm occurred at 226.5°C. The testwas terminated at 347°C. For nPB containing no additives, no exotherm was reported upto 395°C. However, the final pressure of the bomb after it was cooled was considerablyhigher than for the VG cleaner. Scrutiny of the temperature and pressure data shows thatthe temperature curve for the pure nPB flattened briefly at 226.5°C and that the rate ofincrease in pressure increased simultaneously. A possible interpretation of the data is thatnPB thermally degrades at 226.5°C, but the event is either endothermic or it is slightlyexothermic and is not detected by the ARC instrument. In the case of the VG cleaner, whichcontains stabilizers, the products of the degradation react exothermically with one or moreof the stabilizers present.


The degradation products formed in the experiments described above were trapped ina stainless steel bomb and analyzed by gas chromatography/mass spectrometry (GC/MS).The products are essentially the same from both the stabilized and the unstabilized nPB,although the ratios of the products are somewhat different. No free bromine or HBr isdetected. Although trace amounts of methyl bromide and benzene are found, no productsof a highly toxic nature form in significant quantities. Unlike chlorinated solvents, it ischemically impossible to produce an extremely toxic compound such as phosgene, whichcontains two chlorine atoms.

The second approach to determining thermal stability was to simulate a real-worldfailure of a heating element in a vapor degreaser.4 A coiled nichrome wire was immersed inVG cleaner in the bottom of a 250-ml flask. The flask was connected to a dry ice/acetonetrap that was vented to a laboratory hood. Electric current was passed through thenichrome wire until the exposed part glowed red hot. The cleaning solution boiled vigor-ously. The vaporized products were collected in the trap and analyzed by GC/MS. Unlikethe ARC experiment that was conducted in the absence of air, some of the products formedin this experiment contain oxygen. The decomposition products detected after the twoexperiments were as follows:

ARC Method Submerged Nichrome Wire

Propane PropeneIsobutane Methyl bromideButane Ethyl bromideMethyl bromide Benzene2-Methyl butane ToluenePentane Dipropyl etherEthyl bromide 1,3,5-TrioxacycloheptaneBranched C6H14 isomers 4-Bromo-2-butanolIsopropyl bromide 4-Bromo-1-butanol

Excessively high temperature “hot spots” on the interior walls of a vapor degreaser canoccur if the heating elements short-circuit. The thermal degradation studies indicate thatthe use of nPB-containing cleaners creates no risks that are not normally encountered in theevent of catastrophic equipment failure.

HYDROLYTIC STABILITY

Laboratory tests show that nPB is subject to a small degree of hydrolysis when con-tacted with water for extended times, particularly at elevated temperatures. In laboratorytests, an nPB cleaning solvent (VG) was compared with TCA cleaning solvent. After reflux-ing with water for 164 h, the layers were separated and analyzed. The nPB formulationshowed two to three times as much hydrolysis as the TCA. Other tests were then per-formed to determine the relative corrosivity of HBr and HCl. In these tests, dilute HBr wassix to seven times less corrosive at 50°C than was an equimolar concentration of HCl.2

Hydrolysis of nPB is less likely to cause corrosion than hydrolysis of TCA.


Table 4 Comparison of Properties for New Cleaning Solvents

nPB HCFC-225 HFC-43-10 HFE VMS

Boiling point, °C 71 54 55 60 100Flash point, °C None None None None �2.77Flammability limits, vol% 3–8 None None None 1.2–18.6Density, g/m 1.35 1.55 1.58 1.43 0.76

Cleaning

Wide range of soils Ga Gb/F Gb/P Gb/P G/FHigh soil loading in liquid G Gb/F Gb/P Gb/P G/FFluorinated oils and greases Gb/P G G G PSilicone oils and greases G Gb/P Gb/P Gb/P GFast drying G G G G F

Compatibility

Metals Gc Gc G G GPlastics and elastomers G/F/Pd G/F/P G/Fe G/Fe G/F/PLiquid oxygen—direct contact P P G G Pa G � good, F � fair, P � poor.b Applies to blend or azeotrope, not pure solvent.c Must be properly formulated for metals compatibility.d G/F/P � compatibility varies significantly with different plastics and elastomers.e Fluorinated solvents may have poor compatibility with fluoroplastics and elastomers.

COMPARISON OF NPB PROPERTIES WITH OTHERNEW CLEANING SOLVENTS

Table 4 is a brief summary of the physical performance characteristics of the relativelynew entries in the cleaning solvent application area.5 Data are for the pure solvents, not thecommercial blends.

SPECIAL FORMULATIONS—ELECTRONICS

The electronics industry faces some cleaning challenges not found in other types ofcleaning applications. In particular, it is of utmost importance that ionic residues remain-ing on integrated circuit boards after soldering operations be removed to very low levels.In addition, conventional formulations for nPB-based solvents have a tendency to tarnishsilver and silver-plated leads. To address these specific challenges, a formulation has beenintroduced that is designed specifically for the electronics industry. The new formulation,6

designated EG (electronics grade), is exceptionally efficient at the removal of ionic residues.

Compatibility with Metals

The EG of nPB-based cleaner was tested for compatibility with metals using the Mil-T-81553A protocol. Metal coupons were polished with a fine emery cloth and then exposedto boiling solvent (liquid and vapor) for 24 h. The coupons were examined for signs of cor-rosion. The metals that passed this test were as follows:


Zinc Monel Titanium-2Copper Magnesium Tin plateBrass Inconel 6000 Nickel 2000C1010 steel Galvanized steel 316L steelAl 2024 304 Steel

The only metal that failed this test was C12L14 leaded carbon steel.

Compatibility with Plastics and Elastomers

Plastics and elastomers that pass short-term compatibility tests (immersion in boilingsolvent for 15 min) with EG cleaner are shown below. Those that were (M)arginal or(U)nsatisfactory after 2 weeks at 65°C are marked accordingly. The marked materials maybe cleaned but should not be incorporated in equipment where they may contact the hotcleaning solvent for extended periods.

Plastics Elastomers

Acculam™ epoxy glass (M) Teflon™ PTFE Adiprene™ polyurethane (U)Alathon™ HDPE Teflon™ PTFE Aflas™ PTFEDelrin™ acetal Tefzel™ ethylene/PTFE Lucite™ (U)Kynar™ polyvinyl fluoride Ultem™ polyether imide Natural rubber (U)Low-density PE XLPE™ cross-linked PE NBR polyether imidePhenolics Neoprene™ polychloroprene (U)Polypropylene Viton-A™ fluoroelastomer (M)

Removal of Ionic Residues from Integrated Circuit Boards

The testing of EG cleaner was carried out in conjunction with Contamination StudiesLaboratories, Inc. (CSL). CSL prepared IPC B-24 boards fluxed with Higrade RMA liquidflux and processed through a standard wave solder system. The boards were then shippedto the Albemarle Technical Center where they were cleaned using a Branson 5-gal labora-tory vapor degreaser. The cleaning cycle was as follows:

1. Vapor zone—2 min2. Boil sump—3 min3. Rinse sump—1 min4. Vapor zone—2 min

Table 5 Removal of Ionic Residues with EG nPB Cleaner

Omega Meter, Ion Chromatography,Board No. �g/in.2 chloride ion, �g/in.2

1 4.3 2.372 5.2 2.493 4.1 2.08Average 4.5 2.31


The cleaned boards were returned to CSL for evaluation. Three techniques were used:

1. Resistivity of solvent extract (ROSE) using an omega meter;2. Ion chromatography (IC) after extraction;3. Surface insulation resistance.

The results of the ROSE (omega meter) tests and the IC tests are given in Table 5. Notethat the IC values are for chloride ions only. Bromide ions were also found, but control testsshowed that these were most likely extracted from the brominated flame retardantsincorporated in the board material. The IC test employs a very vigorous extraction tech-nique that involves heating the boards at 80°C for 1 h in a 3/1 IPA/water solution.

The CSL report concludes that “The ionic residue testing of ROSE (Resistivity ofSolvent Extract—Omega Meta) and Ion Chromatography show that the samples have lowresidues.” They also report, “The SIR (Surface Insulation Resistance) data all pass the SIRtesting protocol for good product performance.”

At a presentation for the IPC/SMTA Electronics Assembly Expo in October of 1998,Howard Feldmesser of The Johns Hopkins University Applied Physics Laboratoryreported that both the VG and EG formulations were superior to AK-225 AES in theremoval of ionic residues from circuit boards. In addition, the EG formulation was supe-rior to the three other nPB-based formulations that he included in his evaluation.7

Prevention of Silver Tarnish

Reports from the field indicated that nPB-based cleaning formulations were tarnishing sil-ver and silver-plated contacts. A study of the additive technology resulted in a solution thatwas then incorporated into the nPB EG formulation.8,9 Silver-plated steel coupons were placedvertically into beakers of several commercial nPB-based solvents such that the solvent cameapproximately one third of the way up the coupon. The beaker was then heated on a hot plateto boiling and held at the boiling temperature for 10 min. The heat-up time was approximately5 min. The coupons were removed from the boiling solvent, dried, and examined. All of thecommercial grades badly tarnished the coupons, but the EG cleaner showed no tarnish.

A second experiment involved the cleaning of lead frames that are made of copper andare tipped with silver. One frame was cleaned with a typical nPB-based formulation and theother frame with EG cleaner. The cleaning cycle included 3 min in an ultrasonic sump, 3 minin the boil sump, and 4 min in the vapor zone. The difference is dramatic and obvious. A sol-vent cleaner based on nPB and formulated in the usual way for vapor degreasing severely tar-nishes silver and silver plate. The EG formulation, on the other hand, does not tarnish silverwhen used in a normal cleaning cycle (including ultrasonics and/or submersion in the boil-ing solvent).

HEALTH, SAFETY, ENVIRONMENTAL, AND REGULATORY ISSUES10

Toxicology—Acute

nPB is somewhat toxic, and it should be handled with appropriate precautions. Belowis a compilation of the current acute toxicology data for nPB. Current Material Safety DataSheets (MSDSs) should be obtained from the manufacturer for additional and/or the latestinformation.


Mammalian Genetic Toxicity

nPB is negative for dominant lethal activity in rats at 400 mg/kg/day given for 5 daysto male rats prior to breeding once weekly for 8 successive weeks. No difference in matingperformance was noted in treated males. Their frequency of fertile matings, mean numbersof corpora lutea, number of implants per female, number of live embryos per female, andthe dominant lethal index were comparable to the negative control group at weeks 1, 2, 3,4, 5, 6, 7, and 8 after treatment. The frequency of dead implants was higher at week 8 oftreatment compared to the control group, but no increase was observed in the dominantlethal index at that or any other time. The frequency of dead implants in the treated groupwas comparable with the control group at weeks 1, 2, 3, 4, 5, 6, and 7.

Mammalian Metabolism

The half-life of nPB in the rat is very short (approximately 2 h). The majority of theadministered dose is eliminated rapidly in expired air as the unchanged parent compound.The remainder is metabolized and excreted in the urine (predominant route) or in theexpired air as CO2 (minor route).

Following a single intraperitoneal dose (200 mg/kg), the initial rate of excretion ofunchanged 14C-labeled parent compound in the expired air of the rat was rapid; 2 h afteradministration, 56% of the administered dose was exhaled as the parent compound. After4 hours, 60% had been exhaled; only trace amounts were detected in expired air after thistime. An earlier study also reported the elimination of the unchanged parent compound inexpired air. Oxidation to CO2 occurred only to a minor extent. Only 1.4% of the total dose(or 3.5% of the metabolized dose) was exhaled as CO2 over 48 hours. Approximately 40%of the total intraperitoneally-administered dose was available for metabolism in the rat andexcretion in the urine.

Aquatic Acute Toxicology

The solubility of nPB in is approximately 0.25 g/100 ml water at 20°C. The 96-h LC50 inflathead minnows is 67,300 mg/l.

Recommended Exposure Limit

Based on 90-day inhalation studies and preliminary two-generation inhalation repro-ductive tests on rats, the Albemarle Workplace Exposure Guideline (AWEG) was set forworkers using nPB on an 8-h shift, five shifts per week. The AWEG has been set at 25 ppm.At this level, solvent/cleaners based on nPB can be used safely with modern vapordegreasers and in other applications with proper handling. As of this writing, the U.S. EPAhas not made a final decision on nPB.

[Editor’s note: At the time of publication, cooperative studies by the Bromine SolventsConsortium (BSOC) to support determination of an appropriate inhalation level for nPBare in progress. BSOC is working closely with the U.S. EPA. Studies will also be conductedunder the U.S. National Toxicology Program. It should be noted that producers of nPB dif-fer somewhat in their interim recommendations. For example, another major producer,which, like Albemarle Corporation, is a member of BSOC, recommends that exposure lev-els be controlled to below 50 ppm pending evaluation of the BSOC studies and resolutionof areas of uncertainty. — B.K.]


Table 6 Environmental and Health Regulatory Status

Regulation n-Propyl Bromide 1,1,1-Trichloroethane Trichloroethylene

SARAa No Yes YesHAPb No Yes YesNESHAPc No Yes YesRCRAd No Yes YesHGWPe 0.0001 0.023 Almost zero?ODPf 0.0019, 0.02711 0.1 Almost zero?Atmospheric 15 days10 5.4 years NALifetimeg

PELh 25 ppm (additional 350 ppm 50 ppm (25 ppm in CA)data under review)

VOCi Yes No YesSNAPj Pending Unacceptable AcceptableaSARA—Superfund Amendments and Re-authorization Act. This act requires reporting of inventories andemissions of listed chemicals and groups. nPB is not regulated. Most common chlorinated solvents are regulated.bHAP—Hazardous Air Pollutant. A listing of chemicals that the EPA has declared hazardous. nPB is not on thelist, but chlorinated solvents are.cNESHAP—National Emission Standard for HAP. Sets standards for use of HAPs. Since nPB is not a HAP,these standards do not apply.dRCRA—Resource Conservation Recovery Act. Defines hazardous wastes and how to manage them. nPB isnot regulated under this act. Chlorinated solvents are regulated.eGWP and HGWP—Global Warming Potential, Atmospheric Lifetime, and Ozone Depletion Potential calculationswere carried out by Atmospheric and Environmental Research, Inc. GWP is calculated relative to CO2, whereasHGWP (Halocarbon GWP) is calculated relative to CFC-11. GWP calculations were done using different integrationtime horizons. By the HGWP method, CFC-11 is 10,000 times more detrimental as a global warming agent thannPB. By the GWP method, it is 14,000 times worse than nPB, and nPB is only one tenth as bad as CO2.

GWP

Compound HGWP 20 years 100 years 500 years

CFC-11 1.0 4500 3400 1400nPB 0.0001 1.01 0.31 0.1

fAtmospheric Lifetime—Ozone Depletion Potentials (as well as GWP) depend on the atmospheric lifetime ofthe substance in question. Ozone depletion takes place in the stratosphere. In order for a substance to have ahigh ODP, it must be able to work its way to the stratosphere. The atmospheric lifetime of nPB is only 15 days. Incomparison, TCA has a lifetime of 5.4 years.gODP—Ozone Depletion Potential. Models for calculating ODPs have generally been based on the assumptionthat the chemicals are relatively long-lived in the atmosphere. Because of the short atmospheric lifetime of nPB,certain assumptions had to be made relative to the rate of transport of the molecules and the free radicals thatare formed when they dissociate. Two different models were used, resulting in ODPs of 0.0019 and 0.027(Atmospheric and Environmental Research, Inc.). For comparison, TCA has an ODP of 0.1. Refinement of thismodeling for short-lived chemicals is ongoing.hPEL—Permissible Exposure Limit. These are exposure guidelines for workers using the given chemical. PELsmay be set by EPA or OSHA. For nPB, neither agency has set a PEL at the time of this writing but studies are inprogress. The value given has been assigned by Albemarle Corporation and is referred to as the AlbemarleWorkplace Exposure Guideline (AWEG). Based on inhalation studies, Albemarle set the exposure levelguideline at 25 ppm for a time-weighted 8-h exposure.iVOC—Volatile Organic Compound. All VOCs are classified as VOCs until there is experimental evidence thatthey do not contribute to the formation of smog. Therefore, nPB is currently classified as a VOC and must beused in accordance with local regulations regarding VOCs. Studies have been conducted to determine thedegree of photoreactivity of nPB and the types of photochemical products formed. The data has been suppliedto the EPA. At the time of this writing, the classification is still under review.jSNAP—Significant New Alternatives Policy. This is the policy under which EPA gives approval for the marketingof a replacement for an ozone depleting chemical. EPA may approve, disapprove, or give restricted approvalwithin 90 days of the application. Alternatively, EPA may choose to delay any ruling on an application. If the EPAdoes not rule within 90 days, as is the case with nPB, the replacement chemical may be commercialized,pending final determination. On February 18, 1999, the EPA published an “Advance Notice of ProposedRulemaking (ANPR)” on nPB, requesting further input. No final action has been taken at the time of this writing.


Table 7 Comparison of Environmental Parameters for New Cleaning Solvents

nPB HCFC-225 HFC-43-10 HFE VMS

Atmospheric 15 days 2.7–7.9 years 17.1 years 4.1 years � 30 dayslifetime

VOC Yesa No No No NoODP 0.002, 0.027 0.03 None None NoneGWPb 0.31 170/690c 1700 320 LowHalocarbon 0.0001 0.04/0.06c 0.33 0.06 Low

GWPd

Exposure 25e 50 200 750 200guideline,TWA, ppm

a Application for delisting as VOC has been filed with EPA.b GWP (CO2) � 1, 100-year ITH).c Different values for the two isomers.d HGWP (CFC 11 � 1.0).e Please see Editor’s Note in the section “Recommended Exposure Limit.”

Environmental and Health Regulatory Status

Worker safety, public safety, and environmental protection are paramount in the devel-opment of any new product. The status at the of nPB and solvent/cleaner systems basedupon it is given in Table 6. The table includes information on 1,1,1-TCA and TCE for com-parison. Explanations of the regulations follow the table.

Environmental and Health-Related Parameters Compared with OtherNew Solvents

It is also useful to compare the environmental impact and safe usage factors of the var-ious new solvents. As one can see by studying the data in Table 7, there remain difficultconflicts in deciding which solvent has the least impact on both environmental and healthand safety issues. The ODP and the GWP and HGWP are often in conflict. The atmos-pheric lifetime is an important consideration, although there remains debate about how tocalculate ODP of molecules with short half-lives. One must also consider whether the com-pound is a VOC, i.e., a compound that causes smog formation.

SUMMARY

Solvent/cleaner systems based on nPB have been introduced as replacements for chlo-rinated solvents in cleaning applications. nPB is an aggressive, fast-drying solvent that issuitable for a variety of difficult cleaning and degreasing applications. The use of nPB-based solvents is not regulated under SARA, HAP, NESHAP, or RCRA, and they areapproved for sale under SNAP (further review by EPA is under way). nPB has low poten-tials for ozone depletion and for global warming, but it is currently classified as a VOC. TheAlbemarle Workplace Exposure Guideline is 25 ppm, which indicates the use of properprecautions for worker safety. In comparative performance testing, nPB-based formula-tions have been demonstrated to be as effective as chlorinated solvents and more effectivethan HCFCs, HFCS, HCES, and VMSs.


The newest developments in nPB-based cleaning solvent evolution include the exten-sion of the formulation technology to solve specific requirements in specialty niche appli-cations. One such example is for the electronics industry. In addition to the expectedadvantages for a nPB-based cleaning solvent, the EG formulation has the advantages ofproviding enhanced efficiency in the removal of ionic residues while not tarnishing silver.

CASE HISTORIES

1. Electronic Equipment

A company that produces electronic components for clinical equipment used in bio-medical applications had used CFC-113 blends in both liquid and vapor phase defluxing.12

Exacting performance standards, rate of throughput, space limitations, limited capitalequipment budget, and lack of an adequate industrial water system were major constraintson a changeover to a new cleaning system.

The company attempted to switch to a system based on d-limonene. Residue from thecleaning agent, buildup of rosin flux, and reactivity to produce assorted oxidation productsmade the electronic components unsuitable for use. An unacceptable green residueremained on the assemblies and there was an increase in product failures.

After switching to VG cleaner, the company found that the new cleaner did a better jobof removing flux than the CFC-113 blend. There were no residue problems as there werewith d-limonene. Some of the plastic components did show some discoloration, but thisproblem was solved by shortening the exposure time—an added benefit that increasedproduction throughput.

2. Electric Motor Stators and Refrigeration Coils

The Galley Products Division of B/E Aerospace has a need to clean burned oils fromused electric motor stators. They sent two such stators for test cleaning. One was clean andthe other was covered in oil. Both were cleaned using VG cleaner. They were first loweredinto the vapor zone of a vapor degreaser and held there until condensation of the vapor onthe parts ceased. This required about 5 min. They were then lowered into the ultrasonicbath for 10 min and back into the vapor zone for 5 min. The parts were returned to B/E forexamination.

B/E Aerospace reported that the parts were “perfect.” No residual oils or other con-taminants were found, and there was no damage to the electrical wiring or casings. Thisfinding led B/E Aerospace to examine other applications for VG cleaner. It has sincedevised a flushing station for cleaning the long coils of copper tubing used in refrigerationunits. B/E presented a paper at the CleanTech’98 conference outlining the extensive selec-tion process that led it to choose VG for its tough cleaning problems.13,14

3. Implantable Body Parts

A company manufactures artificial body parts, such as hip joints, for implantation. Theparts consist of a titanium bone replacement and an ultrahigh-molecular-weight polyethyl-ene (UHMWPE) cartilage replacement. Standards for cleanliness are, of course, very high.In addition, the company expressed concern about retention of solvent in the UHMWPEparts. A final criterion is that the cleaning solvent must kill at least 50% of the bacterialspores on an artificially inoculated UHMWPE part. There were two cleaning solvents that


the company wished to replace in its process. The first was based on HCFC-141b. The sec-ond was based on trichloroethylene.

The initial set of experiments was for part cleanliness, and these were performed atBaron Blakeslee, Inc. (Long Beach, CA). UHMWPE parts were exposed to VG vapors for 90s. Both cleaned and uncleaned parts were returned to the company, which found the clean-ing to be satisfactory. Acetabular components and hip stems were contaminated with buff-ing compound. These were cleaned using the Baron Blakeslee AutoBatch vapor degreaser.The cycle consisted of 30 s in the vapor, 5 min immersion in the ultrasonic sump at 130°F,2 min TopHat drying, and 2 min freeboard dwell. The parts were reported to be completelydry with no solvent dragout. Baron Blakeslee returned the parts to the company, whichagain determined that the cleaning was satisfactory. In a third experiment, a femoral partwith fingerprints was exposed to vapor for 90 s. Evidence of fingerprints remained afterthis test.

The second company concern was retention of the solvent by the UHMWPE parts. nPB(VG) was compared directly with commercial cleaning grades of HCFC-141b and TCE. Thetest procedure was to place five parts in the boiling solvent for 3 min. The parts wereremoved and placed in the same solvent at ambient temperature for 2 h. The parts werethen removed and placed in an open dish. The dish was left in a fume hood with the fangoing until it was time to take the measurements. To obtain the quantity of headspacevapors, the parts were placed in a sealed glass chamber and allowed to equilibrate for 1 h.The vapors were then analyzed using GC/MS and quantified against a standard.Measurements were taken at 24 and 96 h for the nPB. For the other two solvents, the meas-urements were at 24 and 106 h. The vapor in the headspace is reported in ppm by volume.

Concentration of Solvent in Headspace, ppm

24 h 96 or 106 h

nPB (VG) 30 3.7HCFC-141b 169 4.4Trichloroethylene 469 27

These experiments indicate that the nPB is retained in the UHMWPE parts to a lesser extentthan either the HCFC-141b or the TCE.

The third criterion was that the solvent/cleaner reduce bacterial spore counts on theUHMWPE parts by at least 50%. The company supplied two sets of parts (six parts each)contaminated with Spordex® Bacillus subtilis (globigii) spores. Each set was divided into twosets of three parts each—one set to be cleaned and one set as a control. The cleaning pro-cedure was as follows:

1. Test parts were placed in the basket of a laboratory vapor degreaser and coveredwith a metal screen to prevent the parts from floating to the surface. Thedegreaser contained nPB (VG).

2. The basket was lowered into the vapor zone and remained there until condensa-tion stopped.

3. The basket was then lowered into the boil-up sump (71°C) for 3 min (Set A) or 1.5min (Set B).

4. The parts were placed in the rinse sump for 1 min, followed by the vapor zonefor 1 min. The basket was allowed to hang in the freeboard zone for an additionalminute to assure that the parts were dry.


The two sets of cleaned parts along with the accompanying control sets were returnedto Smith & Nephew for analysis. The bioburden validation was done at Axios, Inc.(Kennesaw, GA). The results were as follows:

Sample Spore Count % Reduction

Sample AControl 5.5 � 106

Cleaned 1.5 � 106 73%Sample B

Control 8.8 � 106

Cleaned 2.8 � 106 68%

4. Aluminum Parts for Optical Applications

A manufacturer of optical equipment that uses anodized aluminum componentsrequested these tests. There were two areas of concern. First, aluminum is a very activemetal. It is especially active toward halogenated materials. The manufacturer was con-cerned about possible interactions of the aluminum with nPB-based cleaners. The secondconcern involved the lettering and other markings that the parts have on them. The cus-tomer requested an evaluation to make sure that the markings would not be damaged inthe normal cleaning process. It sent four sets of components, each set containing six differ-ent parts. Two sets were cleaned in nPB (VG) by immersing in the boil-up sump for 10 minfollowed by 1 min in the rinse sump. The other two sets were immersed for only 3 min inthe boil-up sump and then 1 min in the rinse sump. All 24 parts were returned to the cus-tomer for examination. No damage to the parts was observed.

5. High-Performance Inertial Navigation Systems

The Guidance & Control Systems Division of Litton Industries builds high-perform-ance navigation systems. Systems include gyroscope instruments and associated electron-ics assemblies. Producing inertial navigation systems involves exacting, multistep cleaningof complex subassemblies. Each subcomponent of a system may require various cleaningsteps. Fluxes, oils, and flotation fluids must be removed from an array of materials of con-struction including a wide assortment of metals, plastics, and epoxies. For this application,soil residues, cleaning agent residues, and water are not acceptable. When faced with theprospect of having to replace 1,1,1-TCA and CFC-113, Litton undertook an ambitious pro-gram to evaluate cosolvent systems, fluorinated solvents, and nPB-based cleaners.15

Litton listed a number of requirements for new cleaning systems. These included:

• Replacement of ozone-depleting chemicals,• Process improvement,• Maintenance of superior performance,• Minimization of contamination,• Cost-effective operation,• Efficient processing,• Compliance with national and local regulatory requirements,• Assurance of worker safety,• Production to meet exacting customer standards.


A variety of systems were evaluated. Based on feedback from the assemblers in theplant, the cleaning sequences were refined and implemented. The approach initiallyadopted was a cosolvent system consisting of initial cleaning with a hydrocarbon blendcontaining various alcohols followed by 2 to 3 rinses with IPA. This new process allowedelimination of TCA cleaning. Some perfluorinated material continues to be used as a finalrinse to assure thorough removal of fluorolube. Overall, the number of process steps wasreduced. In some cases, 18 steps were reduced to 4 to 6 steps. There were still problems,however. Because IPA was found to react with beryllium periodically, intermittent residueswere found. Eventually, the IPA was replaced with VMS.

The cosolvent system developed to replace TCA was still far from optimal. The sub-assemblies are very complex, with close tolerances and blind holes. While the cleaning agentcan be removed with careful process control, extreme and constant care is required to assurethat no cleaning agent residue is left. In addition, the hydrocarbon blends were costly, someof the operators found the odor to be disagreeable, and the blends were flammable.

Cleaners based on nPB had been introduced to the cleaning market by this time, andLitton undertook an extensive evaluation. One consideration was that the nPB is a veryaggressive solvent that could be expected to have cleaning performance similar to TCA. Inthe end, VG cleaner was chosen as the most suitable for the Litton requirements. In addi-tion to the cleaning capability, the reliability and the environmental and safety acceptabil-ity of the VG, Litton found improved processing time and lower cleaning agent usage.Litton has reported that the processing time has been reduced by over 40%. The cleaningagent usage has been reduced to one third of the previous amount.

REFERENCES

1. Kanegsberg, B., Precision cleaning without ozone depleting chemicals, Chem. Ind., #20, 787, 1996.2. Shubkin, R. L., A new and effective solvent/cleaner with low ozone depletion potential, in 1996

International Conference on Ozone Protection Technologies, Proc., Presentation, Washington, D.C.,October 21–23, 1996.

3. Unless otherwise noted, experimental procedures and results referred to in this chapter were per-formed at the Albemarle Technical Center, Baton Rouge, LA, or by contract laboratories under thedirection of Albemarle technical personnel.

4. Shubkin, R. L. and Liimatta, E. W., A new cleaning solvent based on n-propyl bromide, NEPCONWest ‘97 Conference, Anaheim, CA, February 23–27, 1997.

5. Shubkin, R. L., Solvent cleaning into the next century and beyond, presented at CleanTech99,Rosemont, IL, May 19, 1999.

6. Shubkin, R. L. and Liimatta, E. W., n-Propyl bromide based cleaning solvent and ionic residueremoval process, U.S. patent 5,792,277, to Albemarle Corp. August 11, 1998.

7. Feldmesser, H. S., Loyd, K. M., Clausen, M., and Karvar, P., Examining the compatibility of elec-tronic assembly materials with cleaning solvents, presented at IPC/SMTA Electronics AssemblyExpo, Providence, RI, October 25–29, 1998, and at the Ninth Annual Solvent SubstitutionWorkshop, Scottsdale, AZ, December 1–4, 1998.

8. Shubkin, R. L., Method for inhibiting tarnish formation when cleaning silver with ether stabi-lized, n-propyl bromide based solvent systems, U.S. patent 5,990,071, to Albemarle Corp.,November 23, 1999.

9. Shubkin, R. L., Method for inhibiting tarnish formation during the cleaning of silver surfaceswith ether stabilized, n-propyl bromide based solvent systems, U.S. patent allowed to AlbemarleCorp.

10. Shubkin, R. L. and Smith, R. L., normal-Propyl bromide: formulation technology and productstewardship for the electronics industry, presented at NEPCON West ’99 Conference, Anaheim,CA, February 23–25, 1999.


11. Nelson, D. D., Jr., Wormhoudt, J. C., Zahniser, M. S., Kolb, C. E., Ko, M. K. W., and Weisenstein,D. K., OH reaction kinetics and atmospheric impact of 1-bromopropane, J. Phys. Chem. A, 101, 27,4987–4990.

12. Kanegsberg, B., Cleaning high value components for biomedical and other applications, in 1996International Conference on Ozone Protection Technologies, Proc., Washington, D.C., October 21–23,1996.

13. Petrulio, R. and Kanegsberg, B. F., Practical solutions to cleaning and flushing problems, pre-sented at CleanTech ’98, Rosemont, IL, May 19–21, 1998.

14. Petrulio, R. Kanegsberg, B. F., and Chang, S.-C., A practical search solves aerospace cleaningquandary, Precision Cleaning Mag., 6(8), August, 1998.

15. Carter, M., Anderson, E., Chang, S.-C., Sanders, P. J., and Kanegsberg, B. F., Cleaning high preci-sion inertial navigation systems, a case study and panel discussion, presented at CleanTech ’99,Rosemont, IL, May 18–20, 1999.


CHAPTER 1.8

Vapor Degreasing with TraditionalChlorinated Solvents

Stephen P. Risotto

CONTENTS

OverviewPhysical and Chemical InformationEnvironment and Worker Health Considerations

EnvironmentWorker Health

Regulatory OverviewGeneralNESHAP Requirements

Life Cycle AssessmentObjectives and MethodologyInterpretation of the Results

Cleaning ProcessesVapor DegreasingCold Cleaning

Summary

OVERVIEW

Today’s manufacturing engineers and plant managers can face a difficult challenge whenchoosing among the many available surface cleaning options. Aqueous, semiaqueous, flam-mable solvents, and new fluorinated and brominated solvents are just a few of the possibili-ties. Among these many choices, however, one process stands out for its ability to produce aclean, dry part at a reasonable price—vapor degreasing with the chlorinated solvents.

Trichloroethylene (TCE, TRI), perchloroethylene (PCE, PERC), and methylene chloride(MC, METH), have been the standard for cleaning performance in precision parts cleaningfor more than 50 years. Today, the development of new equipment and processes that min-imize emissions and maximize solvent recovery makes TCE, PCE, and MC more effectivethan ever.


Despite their superior performance attributes, however, some companies havereplaced TCE, PCE, or MC with other solvents or processes. Their decision often was basedon misperceptions about the regulatory status, continued availability, and safety in use ofthe chlorinated solvents. The facts are as follows:

• TCE, PCE, and MC have not been banned. Among the commonly used chlorinatedsolvents, only 1,1,1-trichloroethane (methyl chloroform) was phased out of produc-tion, because of its ozone-depletion potential. Meanwhile, the U.S. EnvironmentalProtection Agency (EPA) issued a 1994 decision under its Significant NewAlternatives Policy (SNAP) program that the other three chlorinated solvents areviewed as acceptable substitutes for ozone-depleting solvents.

• Chlorinated solvents will continue to be available. TCE and PCE demand hasremained steady or increased in recent years as a result of their use as raw materialsin the production of refrigerant alternatives to chlorofluorocarbons (CFCs). MC con-tinues to be used in a wide variety of applications. The producers of these solventsremain committed to serving their markets for many years to come.

• TCE, PCE, and MC can be used safely. From the point of view of health and theenvironment, the chlorinated solvents are among the most thoroughly studiedindustrial chemicals. Animal tests and epidemiological studies indicate that whenthe solvents are handled, used, and disposed of in accordance with recommendedand mandated practices, they do not cause adverse health or environmental effects.

• The potential impacts of the solvents can be minimized. Environmental, health,and safety regulations governing the chlorinated solvents are strict, but manage-able. In complying with these regulations companies can get help from severalsources—EPA, the Occupational Safety and Health Administration (OSHA), stateand local agencies, producers and distributors of solvents and degreasing equip-ment, and organizations like the Halogenated Solvents Industry Alliance (HSIA).

PHYSICAL AND CHEMICAL INFORMATION

TCE, PCE, and MC are clear, heavy liquids with excellent solvency. All are virtuallynonflammable, since they have no flash point as determined by standard test methods.Each has its own advantages for specific applications, based on its physical profile (seeTable 1). These solvents work well on the oils, greases, waxes, tars, lubricants, and coolantsgenerally found in the metal-processing industries. They are widely used in the vapordegreasing process.

TCE has been long recognized for its cleaning power. TCE is a heavy substance(12.11 lb/gal) with a high vapor density (4.53 times that of air) that allows for relativelyeasy recovery from vapor degreasing systems. The ability of the solvent to provide con-stant pH and to protect against sludge formation has helped make it the standard by whichother degreasing solvents are compared. Its high solvency dissolves soils faster, providinghigh output.

TCE is used extensively for degreasing zinc, brass, bronze, and steel parts during fab-rication and assembly. It is especially suited for degreasing aluminum without staining orpitting the work, because its stabilizer system protects the solvent against decomposition.For cleaning sheet and strip steel prior to galvanizing, TCE degreases more thoroughly andseveral times faster than alkaline cleaning, and it requires smaller equipment that con-sumes less energy.

PCE has the highest boiling point, weight (13.47 lb/gal), and vapor density (5.76times that of air) of the chlorinated solvents. The high boiling point of PCE gives it a clear


Table 1 Typical Properties of the Chlorinated Solvents

MethyleneChloride Perchloroethylene Trichloroethylene

Chemical formula CH2Cl2 C2Cl4 C2HCl2Molecular weight 84.9 165.8 131.4Boiling point °F (°C)

at 760 mmHg 104 (40) 250 (121) 189 (87)Freezing point °F (°C) �139 (�95) �9 (�23) �124 (�87)Specific gravity at

68°F (g/cm3) 1.33 1.62 1.46Pounds per gallon at 77°F 10.99 13.47 12.11Vapor density (air � 1.00) 2.93 5.76 4.53Vapor pressure at 77°F

(mmHg) 436 18.2 74.3Evaporation rate at 77°F

Ether � 100 71 12 30n-Butyl acetate � 1 14.5 2.1 4.5

Specific heat at 68°F(BTU/lb/°F or cal/g/°C) 0.28 0.205 0.225

Heat of vaporization(cal/g) at boiling point 78.9 50.1 56.4

Viscosity (*cps) at 77°F 0.41 0.75 0.54Solubility (g/100 g)

Water in solvent 0.17 1.01 0.04Solvent in water 1.70 0.015 0.10

Surface tension at 68°F 28.2 32.3 29.5Kauri-butanol (KB) value 136 90 129Flash point

Tag open cup None None NoneTag closed cup None None None

Flammable limits(% solvent in air)

Lower limit 13 None 8Upper limit 23 None 9.2

advantage in removing waxes and resins that must be melted to be solubilized. The highertemperature also means that more vapors will be condensed on the work than with othersolvents, thus washing the work with a larger volume of solvent.

PCE is effective in cleaning lightweight and light-gauge parts that would reach theoperating temperature of lower-boiling solvents before cleaning is complete. When clean-ing parts with fine orifices or spot-welded seams—especially if there is entrapped mois-ture—the high boiling point of PCE is essential for obtaining good penetration.

Inherently more stable than other chlorinated (and brominated) solvents, PCE alsoincorporates a multicomponent stabilizer system that provides the greatest resistance tosolvent decomposition available in the industry. While it can be used to degrease all com-mon metals, PCE is especially applicable to cleaning those that stain or corrode easily,including aluminum, magnesium, zinc, brass, and their alloys.

MC has the lowest boiling point of the chlorinated solvents, as well as the lightestvapor density (2.93 times that of air) and weight (10.98 lb/gal). MC is uniquely suited for


use as a vapor degreasing solvent in applications where low vapor temperatures and supe-rior solvency are desirable. The low boiling point of MC makes it a popular choice forcleaning temperature-sensitive parts such as thermal switches or thermometers.

Vapor degreasing with MC allows more rapid processing and handling, particularlywhen cleaning large, heavy parts. The more aggressive nature of MC is especially useful whendegreasing parts soiled with resins, paints, or other contaminants that are difficult to remove.

ENVIRONMENT AND WORKER HEALTH CONSIDERATIONS

The potential health effects of TCE, PCE, and MC have been very well studied. Eachcan cause acute health effects at elevated exposure levels, but these effects have been foundto be reversible. The primary concern with these solvents has been their potential to causecancer, based on the results of laboratory animal tests showing an increase in certaintumors following lifetime exposure to the solvents. Scientific questions have been raisedregarding the relevance of these animal tumors to human health, however, and epidemiol-ogy studies of workers exposed to the chlorinated solvents over extended periods of timehave failed to produce a consistent pattern of increased cancer incidence. The animal data,nevertheless, have traditionally been viewed by regulatory agencies as indicating a poten-tial for risk in humans.

Environment

Chlorinated solvents have been found as contaminants in soil and groundwater as aresult of past handling and disposal practices. Releases of the chlorinated solvents to landand water are minor, however, in comparison with atmospheric emissions. The residencetimes of TCE, PCE, and MC in the atmosphere are very short and, despite many years ofuse, concentrations in the ambient air are very low. TCE and PCE are oxidized to carbondioxide, water, and hydrogen chloride in the lower atmosphere by reaction with either oxy-gen (ozone) or hydroxyl (OH) radical. Methylene chloride is oxidized by OH only, formingthe same naturally occurring organic breakdown products. Because of their relatively shortlifetimes in the atmosphere, TCE, PCE, and MC are not considered to contribute to deple-tion of the stratospheric ozone layer. Similarly, the chlorinated solvents have negligibleglobal warming potential.

TCE is photochemical reactive and is believed to contribute to the formation of ozone(smog) in the lower atmosphere under certain conditions. The decomposition of PCE andMC contributes only negligibly to the formation of ozone.

In reviewing the acceptability of TCE, PCE, and MC in its SNAP review, the EPA notedthat these compounds are regulated under several other environmental laws and regula-tions, including the occupational limits and national emission standards described else-where in this chapter. The agency concluded that compliance with these regulations willsignificantly reduce the potential for environmental releases and worker exposure fromdegreasing operations. As a result, the SNAP program did not impose further use restric-tions on the three solvents in degreasing.

Worker Health

As with all industrial chemicals, occupational exposure to the chlorinated solventsshould be kept as low as practical. OSHA has set permissible exposure limits (PELs) for


chlorinated solvents. The PEL for PCE and TCE is 100 parts per million (ppm) for an 8-htime weighted average (TWA). The limits for MC are 25 ppm for an 8-h TWA and 125 ppmfor a 15-min short-term exposure limit, or STEL. In addition to the TWA and STEL, theOSHA standard for MC imposes several additional requirements. The AmericanConference of Governmental Industrial Hygienists (ACGIH) also recommends exposurelimits for the chlorinated solvents. The solvent producers recommend maintaining work-place exposure levels within the OSHA limits or the ACGIH levels, whichever is lower (seeTable 2).

The OSHA Hazard Communication (HAZCOM) standard specifies a minimum ele-ment of training for people working with hazardous materials, including the chlorinatedsolvents. This includes how to detect the presence or release of a solvent, the hazards of thesolvent, and what protective measures should be used when handling it. The OSHA HAZ-COM standard also requires labeling of all hazardous chemicals and preparation of aMaterial Safety Data Sheet (MSDS). Labels must contain a hazard warning, the identity ofthe chemical, and the name and address of the responsible party.

REGULATORY OVERVIEW

U.S. federal regulations affecting the use, handling, transportation, and disposal ofchlorinated solvents can be found under the Clean Air Act, the Clean Water Act, theResource Conservation and Recovery Act (RCRA), and the Comprehensive EnvironmentalResponse Compensation and Liability Act (CERCLA, or Superfund). State and local regu-lations also exist for the purpose of controlling emissions. Although numerous, these reg-ulations are manageable and companies can obtain compliance assistance from numeroussources. The major federal regulations pertaining to the chlorinated solvents are summa-rized below.

Table 2 Workplace Limits for the Chlorinated Solvents (ppm)

Methylene Perchloro- Trichloro-Chloride ethylene ethylene

OSHA Permissible Exposure Limitsa

8-h TWA 25 100 10015-min STEL 125 — —Ceiling — 200 200Peak — 300 300

ACGIH Threshold Limit Valuesb

8-h TWA 50 25 5015-min STEL — 100 100aAn 8-h TWA is an employee’s permissible average exposure in any 8-h workshift of a 40-h week. The short-term exposure limit (STEL) is a 15-min TWAexposure that should not be exceeded at any time during the day. Theacceptable ceiling concentration is the maximum concentration to which aworker may be exposed during a shift, except that brief excursions to theAcceptable Maximum Peak are permissible.bThreshold Limit Values (TLVs) are established by the American Conferenceof Governmental Industrial Hygienists (ACGIH).


General

Volatile organic compound (VOC) regulations under the Clean Air Act apply to TCEand limit its emissions to reduce smog formation, particularly in ozone nonattainmentareas. Exact requirements vary by state, but generally include obtaining a permit allowinga specific amount of VOC emissions from all sources within a facility. PCE and MC, how-ever, are exempt from VOC regulations in most states.

The Clean Air Act also calls for the three chlorinated solvents to be regulated as haz-ardous air pollutants (HAPs). The EPA has issued National Emission Standards forHazardous Air Pollutants (NESHAP) for solvent cleaning with halogenated solvents,which are discussed below. Other NESHAPs govern dry cleaning with PCE and the use ofMC in aerospace manufacture and rework, wood furniture manufacture, and polyurethanefoam manufacture.

The Clean Water Act defines chlorinated solvents as toxic pollutants and regulatestheir discharge into waterways. Under RCRA, wastes containing chlorinated solvents fromsolvent cleaning operations are considered hazardous. Generators, transporters, and dis-posers of such hazardous waste must obtain an EPA identification number and must com-ply with “cradle-to-grave” management requirements for these wastes.

The Superfund law requires that if a reportable quantity of a chlorinated solvent orother hazardous chemical is released into the environment in any 24-h period, the federal,state, and local authorities must be notified immediately. Reportable quantities are 1000 lbfor MC and 100 lb for PCE and TCE.

NESHAP Requirements

The EPA NESHAPs for new and existing halogenated solvent cleaning operations gov-ern emission standards for chlorinated solvent degreasing operations. These standardscover both vapor degreasing and cold cleaning with TCE, PCE, and MC.

In developing the standards, the EPA focused on equipment and work practicerequirements that permit a level of control between 50 and 70%. Companies operatingbatch or in-line degreasers are given three options for compliance: (1) Installing one of sev-eral combinations of emission control equipment and implementing automated parts han-dling and specified work practices; (2) meeting an idling-mode emission limit, inconjunction with parts handling and work practice requirements; or (3) meeting a limit ontotal emissions.

The multiple compliance options in the NESHAP recognize the vast number of differ-ent industries and operating schedules associated with the use of halogenated solventcleaners. The EPA standard allows companies considerable flexibility in complying withthe control requirements. The alternative idling and total emissions limits allow the use ofnew and innovative technologies to achieve a level of control equivalent to the availableequipment combinations.

When a company chooses the first option, it may choose from a series of combinations oftwo or three procedures, which include freeboard ratio of 1.0, freeboard refrigeration device,reduced room draft, working-mode cover, dwell, and superheated vapor. In addition to theseoptions, solvent cleaning processes must include an automated hoist or conveyor that carriesparts at a controlled speed of 11 ft/min or less through the complete cleaning cycle.

Compliance with one of the control options for batch or in-line vapor equipment isdemonstrated by periodic monitoring of each of the control systems chosen. Work practicesare also required as part of the new EPA standards. Rather than require direct monitoring of


work practice compliance, however, EPA has developed a qualification test, included as anappendix to the standard. The test is to be completed by the operator during inspection, ifrequested.

A company choosing to comply with the second option, the idling-emission limit(0.045 lb/ft 2/h for batch vapor equipment, 0.021 lb/ft 2/h for in-line equipment), isrequired to demonstrate initial compliance by using the EPA idling reference test method307. Data from the equipment manufacturer may be used, provided the unit tested is thesame as the one for which the report has been submitted. Compliance with the idling-emission limit also requires installation of an automated parts-handling system and com-pliance with work-practice requirements. In addition, the company must show that the fre-quency and types of parameters monitored on the solvent cleaning machine are sufficientto demonstrate continued compliance with the idling standard.

Complying with the third option, the limit on total emissions, requires the company tomaintain monthly records of solvent addition and removal. Using mass-balance calcula-tions, the company determines the total emissions from the cleaning machine, based on a3-month rolling average, to ensure they are equal to or less than the established limit forthe cleaner (30.7 lb/ft2/month for small-batch vapor machines, 31.4 lb/ft2/month for largebatch vapor machines, 20.37 lb/ft 2/month for in-line machines). For new machine designswithout a solvent/air interface, the EPA has established an emission limit based on clean-ing capacity (� 330 � (vol)0.6).

Companies meeting the total emission limit requirements do not need to conduct mon-itoring of equipment parameters, but must maintain records of their solvent usage andremoval of waste solvent. According to the EPA, this compliance option provides an incen-tive for innovative emission control strategies to limit solvent use. For some cleaningmachines, the EPA calculates that the alternative total emission limit could be more strin-gent than the equipment specifications. In particular, the EPA expects that this alternativestandard will be more difficult to meet for larger machines, for machines operating morethan one shift, and for machines cleaning parts with difficult configurations.

LIFE CYCLE ASSESSMENT

Substitution of chlorinated solvents for cleaning metal parts is frequently proposed byregulators and others. Alternatives, such as aqueous cleaning with detergents, are oftenperceived as having less environmental impact than cleaning with chlorinated solventsused in a vapor cleaning process.

Objectives and Methodology

A life cycle assessment (LCA) was conducted in 1996 by the European ChlorinatedSolvents Association (ECSA) to provide robust data relating to the environment impact ofmetal parts cleaning in TCE vapor degreasing and aqueous processes. Five major environ-mental indicators were considered: nonrenewable resource depletion, greenhouse effect,air acidification, eutrophication (water impacts), and solid waste.

For a viable comparison, the cleaning processes were studied assuming the same per-formance (the “functional unit”) and a clear definition of what is and is not included in thesystem (the “system boundary”). As requested by standard LCA methodologies, the studyalso incorporated the environmental impacts caused by the manufacturing of the cleaningagents, upstream of their use in cleaning installations.


The functional unit adopted in this study was the complete removal of 46.8 g (1.6 oz)of grease from 1 m2 (10.7 ft 2) of metal parts. Thus, a single level of contamination isassumed in all the cases studied, as well as the removal of the whole quantity of grease bythe cleaning step.

To allow for viable comparisons, the same boundaries were used for all the cleaningprocesses assessed, as required by the standard methodology for LCAs. In particular, bothfor solvent cleaning and for aqueous cleaning, the following were included in the life cycleassessment.

• The environment impacts incurred by the manufacturing of the cleaning agents• The environmental impacts of the cleaning steps themselves• The environmental burdens associated with the treatment of the cleaning residues

or effluents

Seven cleaning scenarios were chosen, three for solvent cleaning and four for aqueouscleaning. Data on the environmental impacts of metal parts cleaning were collected at fivecleaning sites in Europe. The sites were selected to be representative of various technolo-gies across different countries.

Interpretation of the Results

Each cleaning technology was found to have potentially significant environmentalimpact (see Figures 1 through 6) The primary disadvantage of TCE, air pollution (i.e., airacidification), can be minimized with emission controls. The water pollution disadvan-tages of aqueous cleaning, however, remain significant even after significant physiochem-ical � biological treatment of the cleaning residues. With aqueous cleaning, impact onwater was between 200 and 2000 times higher than with TCE degreasing, depending on thesite under consideration.

For cleaning and drying metal parts, solvent technology has a lower overall environ-mental impact than aqueous technology. This is true even without the use of carbon recov-ery of solvents in the vapor phase, provided that equipment meets the NESHAPrequirements and is operated to best practice. However, air acidification can be higher as aresult of solvent releases. In the ECSA investigation, the air pollution impact of solventcleaning varied from 8.5 times greater to 5 times less than that for aqueous cleaning with adrying step.

Generally, a given solvent machine can treat a wider range of metal parts than a givendetergent installation. This occurs because a detergent is often specific to a kind ofcontamination and a shape of metal part. A solvent technology without carbon recovery iscompetitive with an aqueous technology from an environmental viewpoint, provided thatsolvent emissions are limited through other means.


Figure 1 Nonrenewable resource depletion (kg/year) results of LCA (per square meter of metal partcleaned) Scenarios: VDG1—open-top degreaser without NESHAP-compliant controls;VDG2—NESHAP-compliant degreaser with on-site distillation; VDG3—NESHAP-compli-ant degreaser with on-site distillation and carbon adsorption; AQ1—aqueous cleaningequipment with primary wastewater treatment; AQ2—aqueous cleaning equipment withprimary and secondary wastewater treatment and drying. (Two additional aqueous sce-narios are not included in the graphs.) (From CA Comparison of Metallic Parts Degreasingwith Trichloroethylene and Aqueous Solutions, prepared for the European ChlorinatedSolvents Association by Ecobilan, December 1996).

Figure 2 Total energy use (MJ) results of LCA (per square meter of metal part cleaned).


Figure 3 Greenhouse effect (gram equivalent CO2) results of LCA (per square meter of metal partcleaned).

Figure 4 Solid waste (kg) results of LCA (per square meter of metal part cleaned).


Figure 5 Air pollution (gram equivalents H�) results of LCA (per square meter of metal partcleaned).

Figure 6 Water pollution (gram equivalents PO43 � ) results of LCA (per square meter of metal part

cleaned).


CLEANING PROCESSES

The chlorinated solvents have been used traditionally in both vapor degreasing andcold cleaning applications. Recent advances in vapor degreasing help to ensure thatworker and environmental regulations and concerns can be effectively addressed.

Vapor Degreasing

The vapor degreasing process is the ideal technology for high-quality cleaning of parts.It is able to remove the most stubborn soils. It reaches into small crevices in parts with con-voluted shapes. Parts degreased in chlorinated solvent vapors come out of the process dry,with no need for an additional drying stage.

Vapor degreasing is particularly effective with parts that contain recesses, blind holes,perforations, crevices, and welded seams. Chlorinated solvent vapors readily penetratecomplicated assemblies as well. Solid particles such as buffing compounds, metal dust,chips, or inorganic salts contained in the soils are effectively removed by the washingaction of the solvent vapor.

Vapor degreasing can be carried out in either a batch or an in-line degreaser. The tra-ditional batch degreaser is a covered tank, with cooling coils at the top, into which the dirtyparts are lowered. Solvent in the bottom of the tank is heated to produce vapor. On contact-ing the cooler work, the vapor condenses into pure liquid solvent. The condensation of sol-vent dissolves the grease and carries off the soil as it drains from the parts into the solventreservoir below. This process continues until the parts reach the temperature of the vapor, atwhich point condensation ceases and the parts are lifted out of the vapor, clean and dry(Table 3).

Many degreasers contain one or several immersion tanks below the vapor zone, so thatparts can be lowered into liquid solvent—often in a tumbling basket—before being raisedinto the vapor for final rinsing. Ultrasonic cleaning can be added to remove heavy oildeposits and solid soils by installing transducers in the degreaser. When ultrasonic energyis transmitted to a solution, it imparts a scrubbing action to the surface of soiled partsthrough cavitation—the rapid buildup and collapse of thousands of tiny bubbles.

Several types of conveyorized equipment provide in-line vapor degreasing. Theselarge, automatic units, which can handle a volume of work and are enclosed to provideminimal solvent loss, include the monorail and the cross-rod degreasers. They are particu-larly valuable when production rates are high.

Table 3 Operating Parameters for the Chlorinated Solvents

Methylene Perchloro- Trichloro-Chloride ethylene ethylene

Vapor thermostat setting, °F (°C) 95 (35) 180 (82) 160 (71)Boil Sump Thermostat Setting,a °F (°C) 110 (43) 260 (127) 195 (91)Steam pressure (psi) 1–3 40–60 5–15Solvent condensate temperature,b °F (°C) 100 (38) 190 (88) 155 (68)Cooling coil outlet temperature range (°F) 75–85 100–120 100–120aMaximum boiling temperature, based on 25% contamination with oil.bTo facilitate effective separation of the solvent from the water.


Although conveyorized degreasers are enclosed, there is still some solvent lossthrough the openings where work enters and leaves the equipment. Consequently, somecompanies have found it cost-effective to install one of the advanced types of degreasersthat have no air/vapor interface (see Chapter 2.11 by Gray and Durkee). These sealed unitswere first introduced in Europe, but have become available in the United States in recentyears.

Typically these degreasers perform the cleaning operation in a sealed chamber intowhich solvent is introduced after the chamber is closed. Solvent vapor then performs thefinal drying stage, and all vapors are exhausted after each cycle and passed into a solventrecovery system. With the sealed chamber, control of solvent loss exceeds 90%. Operationis programmed and automated, permitting a variety of cleaning programs, including hotsolvent spray.

Although these sealed units can be costly and may not be effective for some cleaningjobs, a few U.S. plants have installed them to ensure compliance with safety and environ-mental regulations.

Cold Cleaning

The manufacturers of the TCE, PCE, and MC generally do not recommend the use ofmethylene chloride, trichloroethylene, and perchloroethylene in hand wipe and other cold(room-temperature) cleaning applications. In circumstances where workplace exposure,NESHAP, and other requirements can be met, these solvents may provide a viable optionfor companies searching for an effective cold cleaning solvent.

SUMMARY

Among the surface cleaning options available, one process stands out for its ability toproduce a clean, dry part at a reasonable price—vapor degreasing with the chlorinatedsolvents. TCE, PCE, and MC are clear, heavy liquids with excellent solvency. All are virtu-ally nonflammable, and can effectively remove the oils, greases, waxes, tars, lubricants, andcoolants generally found in the metal processing industries.

The potential health effects of TCE, PCE, and MC have been very well studied.Environmental, health, and safety regulations governing these chlorinated solvents arestrict, but manageable. While cleaning with these solvents is often perceived as havinggreater environmental impacts, life cycle assessment suggests that TCE, PCE, and MC mayhave a lower overall environmental impact in many situations.

The chlorinated solvents have been used traditionally in both vapor degreasing andcold cleaning applications. Recent advances in vapor degreasing help to ensure thatworker and environmental regulations and concerns can be effectively addressed.


CHAPTER 1.9

Volatile Methylsiloxanes:Unexpected New Solvent Technology

Ray A. Cull and Stephen P. Swanson

CONTENTS

OverviewChemistry and PropertiesEnvironmental/Regulatory ProfileHealth and Safety TestingPerformance AssessmentCompatibilitySilicone Conformal Coating RemovalProcess ConsiderationsSummaryReferences

OVERVIEW

A new class of fluid chemistry has been introduced to the precision and industrialcleaning markets, based on linear and cyclic volatile methylsiloxane (VMS). The use of asilicone-based cleaner may seem “counterintuitive” to some people, since the low surfacetension of silicone contamination has historically made it very difficult to remove.However, in a cleaning solvent the low surface tension becomes a definite asset, since ithelps wet out and undercut soils. Dragout is also reduced, because of the low viscosity ofthe liquid. One of the keys to success with what appears at first to be an unlikely technol-ogy is the ability to manufacture VMS that dries with ultralow nonvolatile residue (NVR)so that it will evaporate completely, leaving behind a clean surface.

While VMS materials are new as cleaning solvents, they have been commercially avail-able since the 1950s, primarily used as the building blocks for higher-molecular-weight,nonvolatile silicone fluids and polymers. In addition, VMS fluids are widely used in thepersonal care industry, including many antiperspirant, hair care, and skin care products.The majority of the personal care applications employ VMS materials with a cyclic struc-ture, designated “cyclomethicones” by the Cosmetic Toiletry and Fragrance Association. In


Figure 1 Chemical structure of linear VMS.

Figure 2 Cyclic VMS structure.

contrast, industrial and precision cleaning fluids are made primarily with linear VMS flu-ids, which have a faster rate of evaporation and higher recommended exposure levels, buthigher cost than the cyclics.

As shown in Figures 1 and 2, both the linear and cyclic materials are relatively simple,and contain little opportunity for reaction or breakdown into harmful compounds.Common traits for nearly all “neat” VMS materials include low toxicity and practically noodor. They are compatible with a variety of surfaces, and can be used on metals, glass, poly-carbonate, acrylic, and other plastics. With their Kauri-butanol (KB) value, pure VMS flu-ids are not very aggressive cleaners, however, making them primarily effective onnonpolar contaminants like silicones, oils, and light greases. VMS fluids can be blendedwith other, more polar solvents to increase cleaning performance. [Editor’s note: Cyclic VMSfluids are being introduced in cold cleaning applications (e.g., automotive) where VOCsare an issue. – B.K. (See Chapter 1.1 by Kanegsberg.)] Patented azeotropes have been devel-oped that improve cleaning effectiveness on more difficult soils such as rosin solder flux.

The primary limitation of the linear VMS products is their flash points. The materialsare available in a range of drying rates, and each has a corresponding flash point. The dry-ing rates of the di-, tri-, and tetrasiloxane materials are comparable to acetone, isopropyl


alcohol (IPA), and mineral spirits, respectively. The cyclics typically dry more slowly thanmineral spirits, but have a higher flash point.

Regardless of molecular weight, all VMS fluids have the advantage of rapid, naturaldegradation in the environment. Their atmospheric life is just 10 to 30 days,1,2 and they donot contribute to smog. The U.S. EPA has declared VMS materials as a class to be exemptfrom the federal VOC regulations, and to date they have also been granted VOC-exemptstatus in 48 states (including California). Several commercial versions of linear VMS havealso earned Clean Air Solvent certification by the California South Coast Air QualityManagement District.

Based upon their low-odor, good-toxicity profile, VOC-exempt status, and flash pointlimitations, most of the applications for VMS cleaning technology have been in benchtopwork and other hand cleaning operations. Most VMS fluids can also be used in flammable-rated cleaning equipment, with some minor changes to the fire detection system. In gen-eral, any operations involving the removal of greases and oils or the cleaning of siliconeswould be a good fit for VMS fluids, because of their inherent ability to dissolve many sili-cone oils and residues, as well as to soften cured sealants and coatings.

CHEMISTRY AND PROPERTIES

VMS materials are siloxanes with relatively low molecular weight (�600) and highvapor pressure. When distilled and handled to ensure ultrahigh purity, these liquidsremove light oil, silicone residue, and other nonpolar contaminants. For more aggressivecleaning, a recently patented VMS azeotrope helps remove polar soils, such as rosin fluxand liquid crystal residue. Overall, VMS chemistry has demonstrated broad compatibilitywith metals, plastics, elastomers, glass, and many other materials. As an example, a 1995Battelle Memorial Institute study found that VMS fluids were compatible with gyroscopematerials, including adhesives and sensitive metal components.3

Individual VMS compounds have different evaporation rates and flammability. (SeeTable 1 for typical properties.) Drying of disiloxane resembles acetone, for example, and ithas a �3°C (24°F) flash point vs. the more flammable acetone flash point of �18°C (1°F).Trisiloxane dries similarly to IPA, with a flash point of 34°C (94°F), while the tetrasiloxanehas a 57°C (135°F) flash point and dries like mineral spirits. Because the commercialazeotrope is based on disiloxane, their flash points are about the same. The di- and trisilox-ane materials are considered flammable, and tetrasiloxane is classified as combustible.

The key requirement when using a VMS-based material for precision cleaning opera-tions is low NVR, usually less than 5 ppm in a typical lot. Although VMS fluids have beenused as commodity industrial products for decades, it has been only recently that com-mercial grades have become available that are tested and certified at the low NVR levelsrequired for precision cleaning operations. This is especially critical in applications wheresecondary bonding or coating is required, since even 50 ppm residue may cause problemsin adhesive and coating operations.

ENVIRONMENTAL/REGULATORY PROFILE

VMS compounds offer a very favorable environmental profile. They were declaredexempt from VOC regulation at the federal level in 1994,4 and so far 48 states (includingCalifornia) have followed suit. These SNAP-approved materials are not considered haz-ardous air pollutants (HAPs) or ozone-depleting compounds (ODCs), and are not regu-lated under NESHAP or Clean Air Act National Ambient Air Quality standards.


Table 1 Typical Properties of VMS

Cyclotetra- Cyclopenta- Di- Tri- Tetra-siloxane siloxane siloxane siloxane siloxane Azeotrope(244 fluid) (245 fluid) OS-10 OS-20 OS-30 OS-120

Flash point, 55 76 �3 34 57 �4closed cup (°C)

Freezing point (°C) �68 �82 �68 ��68Boiling point (°C) 172 205 100 152 194 98Evaporation rate 0.2 3.8 0.7 0.15 3.5

(ASTM D 1901)Viscosity, cSt 2.5 4.2 0.65 1.0 1.5 0.65

at 25°CSpecific gravity 0.953 0.956 0.76 0.82 0.85 0.77

at 25°CSurface tension, 17.8 18.0 15.2 16.5 17.3 16.3

dyn/cm at 25°CHeat of vaporization, 42 39 46a 44a 36a 61

cal/gm at 25°CKB value 14.5 16.6 15.1 13.4 18.5VOC content, 0 0 0 0 0 11

weight %aEstimated.

Further, VMS fluids are not controlled under U.S. EPA Title III, Air Toxics, nor are theyclassified as EPA Criteria Pollutants. In 1996, the California South Coast Air QualityManagement District (responsible for air quality in Los Angeles) ranked linear VMS as thenumber 1 preferred cleaner for substrate and precision cleaning, an option second only tono-clean technology for electronic applications.5 In 1998, several commercial versions ofVMS technology (Dow Corning® OS Fluids) were awarded Clean Air Solvent Certificatesby the California South Coast Air Quality Management District.

Because of the volatility of VMS fluids, the most likely route for environmental expo-sure is through evaporation. VMS compounds degrade quickly via natural oxidation in thepresence of ultraviolet light; they have an atmospheric life span of less than 30 days. Theultimate degradation products are water, dissolved silica, and carbon dioxide, all of whichare benign compounds that occur abundantly in nature. Because VMS fluids are effectivelyoxidized long before reaching the stratosphere, they are believed to have no impact on theEarth’s protective ozone layer and negligible contribution to global warming (see Figure 3).

VMS emissions from industrial sources are very limited, because the compounds aremanufactured and used as intermediates in enclosed systems. Although VMS fluidvolatilizes into the atmosphere from consumer applications, the amount released per useis extremely small, and environmental emissions are therefore very diffuse. Small amountsof VMS can also end up in wastewater treatment plants (WWTPs), primarily incidentalquantities from industrial wastewater.

If these essentially insoluble compounds should end up in surface waters with treatedeffluent, they have been shown to rise to the surface as a result of their low specific gravity(less than 0.9) and evaporate to the atmosphere. With its short lifespan, the atmosphericcontent of VMS is expected to remain far below the No Observable Effect Level (NOEL).

Because they contain no chlorine, fluorine, or bromine atoms and degrade quickly inthe troposphere,1,2 VMS materials are considered to have no potential to deplete (or even


Figure 3 OS fluids life cycle.

reach) the stratospheric ozone layer. They do not contribute significantly to global warm-ing in comparison with organic compounds, which have much higher carbon levels thatare converted to CO2.

Recent studies demonstrate that VMS compounds have negligible potential to impacturban air quality adversely. Specifically, these studies show that VMS materials do not con-tribute to the formation of ozone in the urban atmosphere.6 Similar conclusions weredrawn when atmospheric reactivity data were used to calculate the Photochemical OzoneCreation Potential (POCP) of VMS materials using the Harwell Photochemical TrajectoryModel.7

VMS compounds are not persistent in the atmosphere. No significant contribution hasbeen observed from VMS materials to ground-level ozone generation6 or aerosol forma-tion.8 When partial oxidates of VMS are deposited or scrubbed out of the atmosphere, theyare not expected to present any threat to aquatic or terrestrial biota because of theirdecreased lipophilicity.

Used VMS fluids from precision cleaning applications can typically be purified toextend service life and increase cost-effectiveness. Techniques include distillation, filtra-tion, and desiccant water removal. At least one supplier has established a no-charge fluidreturn program for customers, under which returned material is fuel blended for its BTUvalue and silica content in making Portland cement. The technique produces a necessaryingredient in cement manufacturing, captures energy from the used VMS fluid, andreleases no hazardous by-products.

The cyclic tetramer [(CH3)2SiO]4 (octamethylcyclo-tetrasiloxane, or OMCTS) has beenextensively tested for ecotoxicological properties as part of an industry testing programcarried out under the U.S. EPA Toxic Substances Control Act.9 A recent regulatory-driveneffort evaluated the possible transport of OMCTS from consumer products to aquaticecosystems and the potential effects on aquatic organisms. The testing defined the NOELfor this material, and the program (under a consent order negotiated with the EPA)


included physicochemical and environmental fate studies, as well as extensive aquatic tox-icity testing. Additional voluntary work sponsored by the industry through the SiliconesEnvironmental, Health, and Safety Council (SEHSC) included broad-scale environmentalfate modeling, WWTP monitoring, and the development of a comprehensive aquatic riskassessment.

Although the studies indicated that toxic aquatic effects can be observed in laboratorytests,10 a subsequent exposure assessment and exposure monitoring at WWTPs11 concludedthat levels in the aquatic environment can be conservatively estimated at 64 to 444 timesbelow the NOEL.12 For benthic organisms, 157- to 1080-fold margins of safety13 have beenestimated. Rapid volatilization and additional dilution in most aquatic environments willincrease this ratio even further.

Because of their physical properties (low water solubility, low density, rapid volatiliza-tion) and short lifetimes, OMCTS and other VMS compounds are not expected to reach eco-logically significant levels in the aquatic environment. Concentrations are expected to below and transient in water and sediments. No adverse effects are expected in the aquaticenvironment from the use of VMS in consumer products.

HEALTH AND SAFETY TESTING

Linear VMS fluids have favorable toxicological characteristics as demonstrated in oral,inhalation, and dermal exposure testing. In acute studies, no adverse effects have beenobserved, even at the maximum achievable doses in laboratory animals. However, thematerials may result in mild, transient discomfort upon direct eye contact.14

Subchronic oral and inhalation studies have revealed no toxicological responses sig-nificant to human health. Tissue culture biocompatibility studies have assessed whetherthese fluids can damage or destroy cells, and no adverse effects have been observed fromany of the materials tested. The linear VMS liquids tested have been found nongenotoxicin short-term evaluations for DNA damage or mutation.14

Threshold limit values (TLVs) have not been established for VMS fluids by theAmerican Conference of Governmental Industrial Hygienists (ACGIH). However, basedon a 90-day study, an Industrial Hygiene Guideline (IHG) has been set by Dow Corning at200 ppm for linear VMS, defined as the maximum average worker exposure level over an8-h workday.

In studies that included worker exposure monitoring, hexamethyldisiloxane (themajor ingredient of the azeotrope formulation) has been measured at two typical electron-ics assembly facilities (see Table 2). The results of these studies demonstrate the long-termsafety of manual cleaning from pressure-dispensed containers. Since the disiloxane meas-ured is the most volatile of the linear VMS fluids, this represents a worst-case scenario.Other linear VMS fluids would have even lower exposure levels as a result of lower volatil-ity. For the 14 different operator stations tested, vapor exposures remained at less than 10%of the 200 ppm guideline at all times. Unlike more traditional cleaning solvents such asalcohol, acetone, and terpenes, VMS fluids have little or no odor, a characteristic especiallywelcomed in benchtop operations.

PERFORMANCE ASSESSMENT

A variety of surface analysis techniques have been used to evaluate the cleaning per-formance and residue levels of commercially available, ultrahigh-purity VMS fluids.Techniques have included electron spectroscopy chemical analysis (ESCA) and contact


Table 2 Operator Exposure to Hexamethyldisiloxane (measured as an 8-h,time-weighted average)

Operator, Hexamethyl- Operator, Hexamethyl-Facility A disiloxane, ppm Facility B disiloxane, ppm

1 9.3 1 7.52 0.4 2 1.63 1.7 3 2.54 7.5 4 1.25 2.2 5 16.06 10.0 6 0.77 1.2 — —8 3.1 — —

Note: 200 ppm is maximum recommended level.

angle testing. Standard coatings tests (including appearance and crosshatch adhesion testssuch as ASTM D 3359 78B) have been conducted on steel and aluminum panels. Test coat-ings have been observed to adhere well to the ultrahigh-purity VMS-cleaned panels.

In another evaluation, clean metal panels were dipped in commercial linear ultrahigh-purity VMS fluids and allowed to air-dry. Lap shear specimens were then made from thepanels using an epoxy, a urethane, and a quick-set adhesive. As shown in Table 3, panelsexposed to the ultrahigh-purity VMS had as good or better adhesion than the clean controlpanels, which were wiped with IPA.

Independent testing of eight cleaning fluids on threaded stainless steel fasteners fur-ther demonstrated the cleaning ability of VMS. By measuring particle removal in ultrasoniccleaning equipment, researchers found that the high-purity, linear VMS fluids were moreeffective than CFC-113 or IPA, and were outperformed only by a cleaning agent/surfactantmixture.15

High-purity VMS materials offer desirable evaporation rate, typically with less than 5ppm NVR. Their low surface tension allows penetration into small spaces, and promotescleaning, rinsing and drying. With their ability to undercut and displace soils, VMS fluidsare most effective on nonpolar contaminants, including greases, oils, dirt, and uncured sil-icone residue (see Table 4).16

Even though VMS fluids have a low KB value, they are miscible in most solvents. Theycan be used in conjunction with more polar fluids, providing a solution with greater ver-satility than either type alone. Often the combination results in a formulation that cleans abroad range of contaminants, yet reduces the amount of the more aggressive solvents used.

Table 3 Lap Shear Adhesion Testing with Ultrahigh-Purity VMS Fluids (in psi)

Sample Epoxy Urethane Quick-Set Adhesive

Control 1213 344 475 �50Disiloxane (OS-10) 1233 �20 494 �20 581Trisiloxane (OS-20) 1440 �50 383 640Tetrasiloxane (OS-30) 1478 �100 395 � 507

Note: Lap shears were based on soaking the substrate panels in VMS fluids, then allowingthem to air-dry. Control samples were cleaned with IPA and wiped.


Table 4 Cleaning Performance of OS Fluids (% of soil removed from steel substrate)

Disiloxane Trisiloxane Tetrasiloxane

Motor oil 99 100 99Quenching oil 100 100 100Cutting oil 100 100 100Gyroscope fill fluid (Newark AFB) 100 100 100200® Fluid 350 cst 100 100 100200® Fluid 100,000 cst 94 88 89Grease HD #2 84 75 76Dow Corning 550 Fluid 99 99 99Dow Corning 710 Fluid 97 96 100Uncured silicone conformal coating 100 99 99

COMPATIBILITY

The broad compatibility of VMS has been demonstrated in studies with a number ofmaterials, using an approach similar to the one described by the National Center ofManufacturing Sciences. For example, results of immersion testing with 16 different plas-tics in linear VMS showed negligible changes in weight, volume, and appearance.

Even acrylic and polycarbonate, which typically experience clarity loss (clouding)from solvent exposure, displayed no change in optical properties from VMS immersion, asdetermined by colorimeter.17 Tests on 1 � 2 in plastic substrates immersed in VMS fluid for3 h at 50°C (122°F) have revealed negligible changes in specimen weight, volume, orappearance. Colorimeter assessment of acrylic and polycarbonate materials (often suscep-tible to crazing from solvent exposure) have shown no effect on optical properties, evenafter exposure at 100°C (212°F). Similar testing of various elastomers revealed little or nochange in most cases. However, exposure to VMS fluids has produced swelling in somelow-consistency silicone sealants, natural rubber, and neoprene.

Because of their compatibility, VMS fluids are suited to a variety of cleaning applica-tions, including sensitive optical devices, gyroscopes, fiber optics, and electronic compo-nents. They can be used on most plastics and sensitive metals, and make effective agentsfor precleaning circuit boards or other surfaces before coating, sealing, or bonding opera-tions. A patented VMS azeotrope is especially useful for benchtop cleaning of electronicassemblies and removing rosin flux and liquid crystal residue. The hexamethyldisiloxane-based formulation includes propylene glycol ether for enhanced cleaning capability.

SILICONE CONFORMAL COATING REMOVAL

VMS-based fluids are also effective at swelling and softening cured silicones so theycan be lifted from substrate materials. This allows electronics manufacturers to remove sil-icone conformal coatings from circuit boards and components to rework failed units.16 Theboards can then be recoated after repair, and very low NVR helps ensure a good bond withadhesives and coatings, whether organic or synthetic based. The ultrahigh-purity VMS flu-ids are well suited for applications requiring secondary bonding or coating, and will notcause adhesion problems in subsequent operations, even on surfaces previously coatedwith silicone.


In the past, the electronics industry has typically used nonpolar solvents such astoluene and xylene to remove silicone conformal coatings from circuit boards when repairsare necessary. Although they are very effective removal agents, these compounds are alsohazardous air pollutants.

Diphenyl oxide disulfonate formulations are also popular for coating removal. Thechemical structure of these compounds actually depolymerizes the silicone coating, caus-ing a partial reversion toward a gel state. In fact, diphenyl oxide disulfonates provide effec-tive silicone removal from both primed and unprimed surfaces. However, the causticnature of the formulations also deteriorates FR4 board and other plastics used in manufac-turing printed circuits. Any process involving conformal coating removal with these com-pounds requires close monitoring of exposure times to prevent damage to boards andcomponents. Further, the strong odor of diphenyl oxide disulfonates can be extremelyunpleasant in the workplace. In contrast, the plastics compatibility of VMS allows circuitboards to soak in the fluid for an unspecified period of time without damage, and the lowodor helps avoid worker discomfort.

PROCESS CONSIDERATIONS

When used in applications other than benchtop cleaning, VMS fluids require equip-ment and systems designed to accommodate their flash points. One such process employsa pneumatically controlled machine that requires no electrical components. Air-drivenpumps are used to circulate the cleaning fluid, with the compressed gas supply directed tothe final stage to speed drying. Multitank systems are also in successful operation withmethylsiloxane fluids.

SUMMARY

When manufactured to ultralow NVR levels, VMS fluids and VMS-based azeotropeshave been proved to be effective cleaning agents for precision and industrial applications,and for removal of cured silicone coatings. Although the low KB value of straight VMS lim-its its cleaning performance to nonpolar contaminants such as greases, oils, and silicones,the inherent low surface tension helps undercut and lift many soils. Since VMS fluids aremiscible in most solvents, they can be used in conjunction with more polar fluids, provid-ing a solution with greater versatility than either type alone. Often the combination resultsin a formulation that cleans a broad range of contaminants, yet reduces the content of themore aggressive solvents.

These ozone-safe materials demonstrate a favorable health and environmental profile,and their use is not likely to be restricted by foreseeable legislation. VMS fluids can usuallybe reprocessed using established technology, depending on the specific contaminants.Spent fluid is currently being fuel blended in cement manufacturing, used for its silica con-tent and BTU value.

Equipment designed to handle flammable liquids safely can typically be used forcleaning with VMS. Developed as an alternative to conventional solvent technology, thesefluids help contribute to greater worker comfort and safety. While no drop-in replacementhas yet been found to equal the cleaning performance and cost of ODCs, VMS fluids offeran exceptional balance of performance and ecological properties.


REFERENCES

1. Atkinson, R., Kinetics of the gas-phase reactions of a series of organosilicon compounds with OHand NO3 radicals and O3 at 297� 2 K, Environ. Sci. Technol., 25, 863, 1991.

2. Sommerlade, R., Parlar, H., Wrobel, D., and Kochs, P., Product analysis and kinetics of the gas-phase reactions of selected organosilicon compounds with OH radicals using a smog chamber-mass spectrometer system, Environ. Sci. Technol., 27, 2435, 1993.

3. Battelle Lab Report to Newark Air Force Base, Experimental Evaluation of the AdhesiveDegradation and Corrosion Potential of Silicone Fluids, Contract F09603-90-D-2217-Q805,January, 1995.

4. U.S. EPA, Fed. Regist., 192, 50693–50696, 1994.5. Iwata, T., Motavassel, F. and Perryman P., Ozone Depleting Compounds Replacement

Guidelines, California South Coast Air Quality Management District, Office of Stationary SourceCompliance, January, 1996.

6. Carter, W.P.L., Pierce, J.A., Malkina, L.L., and Duo, D., Investigation of the Ozone FormationPotential of Selected Volatile Silicone Compounds, final report from the University of Californiato Dow Corning Corporation, November 20, 1992.

7. Jenkin, M.E., and Johnson, C.E., Photochemical Ozone Creation Potentials of Volatile Siloxanes,AEA Technology Consulting Services (AEA/CS/16411030/001 Issue 1), August, 1993.

8. Carter, W.P.L., Luo, D., Malkina, I., and Venkataraman, C., Screening Experiments to Evaluate theAerosol Forming Potential of Selected Volatile Silicone Compounds, final report from theUniversity of California, Riverside to Dow Corning Corporation, June 16, 1994.

9. U.S. EPA, Testing Consent Order for Octamethylcyclotetrasiloxane, Fed Regist., 54, 818–821, 1989.10. Sousa, J.V., McNamara, P.C., Putt, A.E., Machado, M.W., Surprenant, D.C., Hamelink, J., Kent,

D.J., Silberhorn, E.R., and Hobson, J.F., Effects of octamethylcyclotetrasiloxane (OMCTS) onfreshwater and marine organisms, Environ. Toxicol. Chem., 14, 1639, 1995.

11. Mueller, J.A., DiTorio, D.M., and Maiello, J.A., Fate of octamethylcyclotetrasiloxane (OMCTS) inthe atmosphere and in sewage treatment plants as an estimation of aquatic exposure, Environ.Toxicol. Chem., 14, 1657, 1995.

12. Hobson, J.F., and Silberhorn, E.M., Octamethylcyclotetrasiloxane (OMCTS), a case study: sum-mary and aquatic risk assessment, Environ. Toxicol. Chem., 14, 1667, 1995.

13. Kent, D.J., McNamara, P.C., Putt, A.E., Hobson, J.F., and Silberhorn, E.M., Octamethylcyclote-trasiloxane in aquatic sediment toxicity and risk assessment, Ecotoxicol. Environ. Saf., 29, 372,1994.

14. Dow Corning OS Fluids Product Stewardship Summary, Dow Corning form 10-678-96.15. Ambati, R.R. and Kaiser, R., Silicone Particle Removal Study for Gyro Components, Entropic

Systems, report prepared for Defense Construction Supply Center, June, 1994.16. Witucki, B.A. and Cull, R.A., Evaluation of volatile methylsiloxane fluids as potential replace-

ments for ozone depleting solvents, presented at the 213th ACS National Meeting, San Francisco,paper TECH-06, April 13–17, 1997.

17. Swanson, S., Cull, R., Bryant, D., and Moore, J., Cleaning Performance and New TechnologiesBased on Volatile Methyl Siloxanes, Dow Corning Corporation paper, presented at SAMPESeattle, November, 1996.


CHAPTER 1.10

Benzotrifluorides

P. Daniel Skelly

CONTENTS

OverviewProperties of the BenzotrifluoridesSolvent ToxicityWorker ProtectionCleaning Systems for Benzotrifluorides

General ConsiderationsSize and Shape of PartsMaterials of ConstructionVolume of Parts to Be Cleaned and Amount of Soil They ContainSoil LoadingCold CleaningMechanical AgitationHeated CleaningSolvent Blends

CompatibilityCompatibility with MetalsCompatibility with Polymers and Elastomers

Approved Military and Aerospace ApplicationsReclamation and DisposalReferences

[Editor’s note: Many factors impact the cleaning options available to components manufac-turers. The situation with the benzotrifluorides illustrates the impact of overall businessplans of chemical producers. Recently, as this book was going to press, OccidentalChemical, the U.S. producer of benzotrifluorides, announced its intention to exit the mar-ket, ceasing production of parachlorobenzotrifluoride (PCBTF) and offering the businessfor sale. At this point, the future of PCBTF and related compounds as cleaning agents is notknown, although some imported PCBTF may be available.]


*OXSOL and OxyChem are registered trademarks of Occidental Chemical Corporation.

Despite the uncertainty, this chapter has been included for two reasons. For one thing,PCBTF has been found to be a valid option for some components manufacturers. In addi-tion, Mr. Skelly has written a thoughtful approach to cleaning options. PCBTF has beenintroduced as a substitute for mineral spirits, and it is particularly valuable in areas of poorair quality. In Southern California, some local regulations explicitly depend on PCBTF forcold cleaning. In addition, because it is neither a VOC or a HAP, PCBTF has been usedalong with acetone in the reformulation of coatings to meet regulatory constraints. This isperhaps an object lesson to all of us, including end users, formulators, equipment manu-facturers, and regulators that depending on a single or primary chemical to resolve majorproblems may not be prudent. PCBTF is produced throughout the world, but it is prima-rily a chemical intermediate, not a cleaning agent. In the interest of assuring options, it ishoped that the material will continue to be made available with the appropriate technicalsupport and product stewardship.—B.K.]

OVERVIEW

Three commercially available benzotrifluorides, benzotrifluoride (BTF), parachloroben-zotrifluoride (PCBTF) and 3,4-dichlorobenzotrifluoride (DCBTF), have potential as replace-ments for ozone-depleting compounds and other organic solvents. Because it is exempt asa volatile organic compound (VOC), PCBTF is used in cleaning applications, particularly inareas with high regulatory constraints.

“Likes dissolves like.” This is why, over the years, people have used organic solvents toclean oils, greases, and other organic contaminants off their parts and assemblies. However,most organic solvents used in cleaning operations are VOCs, which contribute to ground-level ozone or smog formation. As a result, restrictions have become burdensome and per-mits difficult to obtain, especially in metropolitan areas where air pollution is a significantproblem. Although the three benzotrifluorides have been commercially produced in theUnited States since the 1960s, until the mid-1990s, their use was limited to chemical inter-mediates for the agricultural and pharmaceutical industries. Interest in their use as clean-ing agents developed when it was discovered that the atmospheric lifetimes of thebenzotrifluorides are short enough to avoid ozone depletion. The compounds are not listedas hazardous air pollutants (HAPs). The low tropospheric reactivity of PCBTF led to VOCexemption by the U.S. EPA.

PROPERTIES OF THE BENZOTRIFLUORIDES

The structures of BTF, PCBTF, and DCBTF are indicated in Figure 1. OccidentalChemical Corporation (OxyChem®) is currently the only U.S. producer of these com-pounds, and they are currently sold under the OXSOL® 2000, OXSOL 100 and OXSOL 1000trademarks, respectively.* Some physicochemical properties are summarized in Table 1;their regulatory status is indicated in Table 2.

In addition, properties of the benzotrifluorides obviate many of the issues associatedwith aqueous cleaning. Water pretreatment and wastewater disposal are not issues, nor areproblems of flash rusting. Because the benzotrifluorides are organic, they have an affinityfor a wide range of organic contaminants. The fluorine content provides a low surface ten-sion, which allows penetration into blind holes and crevices. Although the efficiency ofmost processes tends to increase with use of multiple tanks, multiple cleaning and rinsingsteps are not required. While parts will still come out wet after cold cleaning, drying is rel-atively fast. Cycle times are comparable with many traditional organic solvent systems.


Figure 1 Structures of the benzotrifluorides.

SOLVENT TOXICITY

Because no official exposure levels have been established by either OSHA or theACGIH, Occidental Chemical Corporation has set a CEL (corporate exposure limit), 8-hTWA (time-weighted average) for each of the three benzotrifluoride molecules. These val-ues, which appear in Table 2, are as follows: BTF � 100 ppm, PBCTF � 25 ppm, andDCBTF � 5 ppm. These values were established as a safe level of exposure for an average8-h workday. Additional studies have been completed and are available.

WORKER PROTECTION

Benzotrifluorides can be used with appropriate consideration given to worker protec-tion. The cleaning system must be designed to minimize hazards to the worker and theenvironment. This may include mechanical controls such as tank covers and auxiliary cool-ing coils to condense solvent vapors, or fans and exhaust hoods to remove solvent vaporsfrom the workstation. The flash point must also be considered in appropriate equipmentdesign.

CLEANING SYSTEMS FOR BENZOTRIFLUORIDES

General Considerations1

No single cleaning system can be used universally, and what works for a neighbor ora sister plant may be ill-fitted for your requirements. Educate yourself on the options, thenchoose the solvent(s) and equipment that best meet your needs. The system should becapable not only of removing your soils efficiently and economically, but also of meetingthe regulatory and safety requirements of today and the future. A critical first step in select-ing the cleaning system is to address, “How clean is clean?” Only you, the owner of theparts that need cleaning, can properly evaluate if a given system will meet the majority ofyour needs and expectations.

Size and Shape of Parts

Where there are blind holes, lap joints, or other small crevices, the solvent must notonly have a good solubility for the soil, but also a low viscosity and low surface tension.Unlike water, the benzotrifluorides have a naturally low surface tension allowing them topenetrate and clean these hard-to-reach areas.


Table 1 Physical Properties of the Benzotrifluorides

Physical Properties BTF PCBTF DCBTF

CAS number 98-08-8 98-56-6 328-84-7Chemical formula C7H5F3 C7H4F3Cl C7H3F3Cl2Molecular weight 146.11 180.56 215.00Boiling point (BP) at 760 mmHg (°C) 102 139 173.5Density, liquid at 25°C (lb/gal) 9.88 11.16 12.3Density vapor (air � 1) 5.0 6.2 7.4Dielectric breakdown (kv) — 49 —Dielectric constant at 25°C 11.5 — —Electrical resistivity M � (M �-cm) �2.5 (330) �20 (�2640)a �20 (�2640)a

Evaporation rate at 25°C (n-BuAc � 1) 2.8 0.9 0.2Flammability

Flash point, Tag Closed Cup, °C (°F) 12 (54) 43 (109) 77 (170)Fire point, Tag Open Cup, °C (°F) 23 (74) 97 (207) �114 (�238)Autoignition temperature (°C) — �500a �500a

Upper flammability limit (% vol) — 10.5 7.8Lower flammability limit (% vol) — 0.9 2.9Limited oxygen index (LOI) (% vol) 18.2 26.2 �29

Freeze point (°C) �29 �36 �12.4Hansen solubility parameter (MPa)1/2 16.1 17.7 18.2

Nonpolar 13.0 13.9 14.1Polar 9.5 9.9 10.0Hydrogen bonding 0 4.7 5.7

Kauri-butanol value (KB) 49 64 69Latent heat of vaporization at BP (cal/g) 53.9 49.6 46.6Refractive index at 25°C 1.4131 1.4444 1.4736Solubility

Solvent in water at 20°C (ppm) 250 29 12Water in solvent at 25°C (ppm) 290 240 153

Specific gravity at 25/15.5°C 1.185 1.3380 1.478Specific gravity correction factor (per °C) �0.0013 �0.00146 �0.00146Specific heat, liquid at 20°C (cal/g/°C) 0.31 0.27 —Surface tension in air at 25°C (dynes/cm) 23 25 29Vapor pressure at 20°C (mmHg) 30 5.3 1.6Viscosity, liquid at 25°C (cps) 0.53 0.79 1.52Volatiles (% wt) 100 100 100a Limitation of test equipment.

Materials of Construction

The selection of a strong solvent may be good at removing the soil, but you must alsobe sure that it does not damage the substrate or your equipment. Based on the Kauri-butanol (KB) scale, all three benzotrifluorides have a moderately high solvency power.They have about twice the cleaning power of mineral spirits and half that of toluene. Thismakes them effective for a wide range of organic soils, but not so strong that they damagemost plastics or elastomers.


Table 2 Regulatory Summary of the Benzotrifluorides

Regulatory Issue BTF PCBTF DCBTF

Regulated as an ozone depleter No No NoRegulated as a VOC Yesa No YesRegulated as an HAP No No NoSARA Title III, Sect. 313 Reportable No No NoGlobal warming potential Low Low LowSuspected carcinogen No No NoCorporate exposure limit, 8-h TWA (ppm) 100 25 5DOT hazard class Flammable Not regulated Not regulatedOSHA hazard class Flammable Combustible CombustibleRCRA hazardous waste number D001 D001 Not regulatedaPetitioned the Federal EPA for VOC exemption on 3/11/97. Still pending at the time of this writing.

If a heated process is planned, the substrate must be able to withstand the prescribedtemperature. For vapor degreasing, keep in mind that the boiling points of PCBTF andDCBTF are 139 and 173.5°C, respectively. These relatively high temperatures could dam-age some parts.

Volume of Parts to Be Cleaned and Amount of Soil They Contain

A process involving occasional cleaning of a few small parts is very different from acritical process in a high-volume assembly line operation. Do not shortchange yourself.You do not want to pay for excess equipment capacity, but make sure to account for futuregrowth potential. Fabricating additional sheet metal tanks does not generally significantlyincrease capital costs, and additional and/or slightly larger tanks can make a big differencein the capability and capacity of the equipment.

Soil Loading

You cannot clean parts with dirty solvent. If only a small amount of soil needs to beremoved, your solvent will remain relatively clean for a long time. However, if there is alarge amount of contaminant, you might want to consider multiple dips in progressivelycleaner solvent, and/or on-site distillation. With solvent recovery, a stable, single-compo-nent cleaner (such as PCBTF) is more easily maintained and recycled than is a complex pro-prietary blend.

In most applications, once the soil loading exceeds 30%, it is recommended that it bereclaimed or disposed. Soil loading may often be approximated by measuring its specificgravity at a given temperature. In the case of PCBTF, a simple graph was developed at threedifferent temperatures and various levels of mineral oil. Once such a chart is developed foryour “typical” contaminant, you can then approximate the soil loading with a quick com-parison to a chart similar to Figure 2. It will be up to end users to determine what soil load-ing level is tolerable in their system for adequate cleaning performance.

Cold Cleaning

Recent regulations have focused on the VOC content of mineral spirit solvents tradi-tionally used in these systems. PCBTF, alone or in blends, may be an appropriate replace-ment in recirculating overspray applications where aqueous cleaning is ineffective.


Figure 2 Specific gravity vs. oil loading in PCBTF2 (From OXSOL 100—Used as a Solvent Flush inRefrigeration Conversions, pending bulletin BCG-OX-A/C, 8/95, Occidental ChemicalCorp., Niagara Falls, NY. With permission.)

In some cases, single or multiple tank immersion units can be adapted for use withPCBTF and DCBTF with only minor modification, with the proviso that all equipmentmust be designed for use with low flash point solvents. However, it is important to assurethat top enclosures and side workload entry be included to minimize worker exposure andto eradicate possible solvent odors. Because of its relatively high vapor pressure and theresulting solvent losses, BTF would not generally be used in immersion dip tanks.

Mechanical Agitation

The cleaning operation is generally enhanced with some level of mechanical agitationto help carry the soil away from the surface of the part. Where ultrasonics are used with thebenzotrifluorides, the possible elevation of the solvent bath temperature must be taken intoconsideration.

Heated Cleaning

With waxes, buffing compounds, and similar soils, heating (from slightly above ambi-ent temperature to the boiling point) may be required. Unless the equipment is properlydesigned for elevated temperatures, solvent losses will increase. Further, given the flashpoints, cleaning systems designed for flammable solvents must be used.

Because of its volatility and flammability, BTF is ill-suited for this cleaning method,and both DCBTF and PCBTF use is limited in heated dip tanks. Benzotrifluorides must notbe used in a standard, unmodified vapor degreaser. PCBTF and DCBTF could be used inapproved, specifically designed low flash point systems with sufficient containment andautomation to maintain employee exposure well below the recommended limit (seeChapter 2.12 by Bartell). Initial capital expenditure will be significant for such systems.


Automation is desirable to distance the worker from the cleaning operation, therebyminimizing unnecessary solvent exposure.

Solvent Blends

Although a single-solvent system is generally cheaper and easier to maintain, in someapplications a blend of the benzotrifluorides with other compounds is desirable. A systemcontaining a mixture of solvents can have advantages in areas such as solubility parame-ters, flash point suppression, odor reduction, and evaporation rates. However, dependingon the additive selected, you may add to the regulatory requirements, complexity of sol-vent recovery or disposal, and total system cost.

COMPATIBILITY

Compatibility with Metals

In one study run for the U.S. Navy, PCBTF passed the ASTM F 483 total immersion cor-rosion test with no visible corrosion on any of the following alloys: AMS 4037, 4041, and4049 aluminum, AMS 5046 grade 1020 steel, cadmium-plated 1020 steel, AMS 4911 tita-nium, and AMS 4377 magnesium.

Compatibility with Polymers and Elastomers

The solvency power of the benzotrifluorides ranges between that of mineral spirits andperchloroethylene (dry cleaning solvent). As a result, these moderately strong solventshave the ability to dissolve a wide range of uncured resins (of interest to the paint andadhesive markets) and selectively remove many organic soils from the workload. At thesame time, they are generally mild enough to preserve the integrity of the articles beingcleaned. A representative sample of a study of commercially available polymers and elas-tomers exposed to PCBTF is summarized in Table 3. Similar data are available for the otherbenzotrifluorides. In the study, test strips were weighed before and after submersion inPCBTF at room temperature for 24 h. They were removed, blotted dry, and reweighed. Thedifferences in the weights are reported in Table 3 as “�wt% Initial” (% gain or loss). Thesesame test strips were then allowed to air-dry at room temperature for 4 days and wereagain reweighed and the results reported as “�wt% Overall” (% gain or loss).

APPROVED MILITARY AND AEROSPACE APPLICATIONS

In a study contracted by Occidental Chemical Corporation for the U.S. Navy, PCBTFpassed or conformed to the following corrosion/compatibility tests:

• ASTM F 483—Total Immersion Corrosion (see section on Compatibility)• ASTM F 945—Stress Corrosion of Titanium Alloys by Aircraft Maintenance

Materials–Method A• ASTM F 519—Test Method for Mechanical Hydrogen Embrittlement Testing of

Plating Processes and Aircraft Maintenance Chemicals

As a result of these favorable test results, both the U.S. Navy and Air Force have approvedthe use of PCBTF in several protective coatings for military aircraft applications.


Table 3 Examples of PCBTF Compatibility with Polymers and Elastomers

Brand orPCBTF (� wt%)

Material Tested Common Name Initiala Overallb

Acrylic Lucite® �16.5 �16.6Butyl rubber Bucar 83.1 9.0Carboxylic rubber NBR 120.4 �0.5Chloroprene rubber Neoprene W 73.6 �1.8Chlorosulfonyl PE Hypalon® 40 74.5 3.1EPDM Nordel® 6962 69.7 �6.6EPD terpolymer Nordel (Elastomer) 61.2 �5.1Fluorocarbons

PTFE Teflon® 0.1 0.0PVDF Kynar® �0.1 �0.5VF-HFP copolymer Viton® A 30.6 6.5

Ionomer Surlyn® 1.8 0.6Nitrile butadiene rubber Buna N 79.9 �10.4Polyamide Nylon 6/6 0.0 0.0Polycarbonate PC c 12.4Polyester

PBT Valox® 0.1 0.0PET Rynite® 0.1 �0.1

Polyethylene, high density HDPE 1.8 0.6Polyphenylene sulfide Ryton® 0.0 0.0Polypropylene PP 17.5 0.2Polysulfone Polysulfone 0.1 �0.1Polyurethane Adiprene® L 29.1 6.3Polyvinyl chloride

CPVC — 0.0 0.0PVC — 0.1 0.0

Silicone Silicone 77.3 27.2Styrene — d d

Styrene butadiene rubber Buna S 124.3 �3.7a Initial � initial weight change after 24-h solvent soak.b Overall � overall weight change after a 4-day drying period.c Attacked and crackedd Test strips dissolved.

Adiprene® L is a registered trademark of Uniroyal Chemical Corporation; Delrin®, Hypalon®, Lucite®,Nordel®, Rynite®, Surlyn®, Teflon®, Tefzel®, and Viton® are registered trademarks of I.E. du Pont deNemours and Company; Kynar® is a registered trademark of Pennwalt Corporation; Noryl®, Ultem®, andValox® are registered trademarks of General Electric Company; Panacea® is a registered trademark ofPrince Rubber & Plastics Co., Inc; Ryton® is a registered trademark of Phillips Petroleum Company;Thiokol® FA is a registered trademark of Morton International, Inc.

Source: Modified from OXSOL Solvents, Compatibility with Polymers and Elastomers, BCG-OX-205197, Occidental Chemical Corp., Dallas, TX, 1997, 2–4. With permission.

Parachlorobenzotrifluoride has also been approved by the U.S. Air Force under T.O. 1-1-8for use as a cleaning solvent on aircraft, as a wipe solvent for primer reactivation and as apaint thinner. After an extensive study to replace 1,1,1-trichloroethane (TCA), AlliantTechSystems (formerly Hercules Aerospace) has approved the use of PCBTF as a wipecleaning solvent on the Navy’s Trident II D/5 first- and second-stage rocket motors.


RECLAMATION AND DISPOSAL

Recovery of the benzotrifluoride containing waste stream4 may be done by a compe-tent and properly permitted contractor, or by investment in on-site explosion-proof distil-lation equipment. Most stills available today have efficiencies of 90% or greater. Recoveryincreases to 95% or more with a high-efficiency, thin-film evaporation still. To increase theamount of reclaimed solvent further, and reduce the volume of waste that must be dis-posed, stills can be equipped to employ steam stripping or vacuum. Because additive sta-bilizer packages are not required, the distillate can easily be returned into the cleaning tankfor reuse; recycling equipment may have a short payback period.

The still bottoms from a typical on-site distillation operation contain between 1 and10% solvent and must be disposed of according to proper Resource Conservation andRecovery Act (RCRA) hazard classifications. Although DCBTF is not regulated underRCRA, pure BTF and PCBTF have flash points of less than 140°F, which qualifies them asD001 “Characteristic Hazardous Wastes.” Because they do not contain any listed concen-trations of compounds recognized by RCRA as hazardous wastes, many state and local reg-ulations allow these still bottoms to be added to other combustible products andincinerated as fuel oils. This can minimize costly hazardous waste disposal fees. Fuelsblending programs for cement kilns are preferred as the flames are generally hotter and thealkalinity of the cement will neutralize the acid gases that will be generated.

An alternative disposal method is hazardous waste incineration in licensed equipmentcapable of handling HCl and HF. The heat of combustion values for BTF, PCBTF, andDCBTF are 8060, 7700, and 4830 BTU/lb, respectively. Carefully analyze the characteristicsof your particular waste prior to selecting any waste disposal option.

Recycling, disposal. Is the spent cleaning solution classified as a hazardous waste?This needs to be determined on an individual basis, taking into consideration the compo-nents and characteristics of the waste stream.

REFERENCES

Information was taken from the following technical bulletins produced by OccidentalChemical Corporation:1. OXSOLs for Metal Cleaning, BCG-OX-19 1/95.2. OXSOL 100—Used as a Solvent Flush in HFC Refrigeration Conversions, pending bulletin BCG-

OX-A/C 8/95.3. OXSOL Solvents—Compatibility with Polymers and Elastomers, BCG-OX-20 5/97.4. Disposal of OXSOL Degreasing Wastes, BCG-OX-11 6/97.


CHAPTER 1.11

HCFC-225: Alternative Precision andElectronics Cleaning Technology

Toshio Miki, Masaaki Tsuzaki, and Kenroh Kitamura

CONTENTS

IntroductionFundamental PropertiesToxicology SummaryApplications

Precision CleaningOther Sequential Cleaning ApplicationsDefluxing of Printed Circuit Boards for AutomobilesSpecial Applications

RegulatoryReferences

INTRODUCTION

CFC-113 has been widely used as a cleaning agent in various industrial applications suchas electronics cleaning, precision cleaning, metal cleaning, and dry cleaning, due to suchfactors as excellent chemical inertness, thermal stability, nonflammability, and low toxicity.Various alternatives to CFC-113 such as hydrocarbons, alcohols, aqueous and semiaqueouscleaning have been commercialized. These alternatives can easily replace some of the CFC-113 applications. However, their properties are different from CFC-113.

Asahi Glass has developed and commercialized HCFC-225 (a mixture of two isomers,HCFC-225ca and HCFC-225cb, sold as Asahiklin™ AK-225) as an alternative to CFC-113.1,2

HCFC-225 has most of the characteristics of CFC-113 as a cleaning agent, such as physicaland chemical properties and nonflammability, and can be applied to the range of CFC-113applications with few changes in the process or cleaning equipment. HCFC-225 has beenapplied as a replacement of CFC-113 in applications where other alternatives cannot beused. HCFC-225 has a very low ozone depletion potential (ODP) and a very low globalwarming potential (GWP). It has been widely used for more than 10 years. In this chapter,


Table 1 Physical Properties of HCFC-225 Products

HCFC-225 HCFC- CFC-113 CFC-113AESa

Boiling point (°C) 54 52 47.6 46.5Freezing point (°C) �131 �138 �35 �41.8Density (g/cm3)b 1.55 1.49 1.57 1.51c

Viscosity (cP)b 0.59 0.61 0.65 0.66Surface tension (dyne/cm)b 16.2 16.8 17.3 18.5c

Latent heat of vaporization (cal/g, b.p.) 34.6 40.6 36.1 42.9Relative evaporation rate (ether=100) 90 81 123 120Specific heat (cal/g °C)b 0.24 0.27 0.229 0.272c

Solubility of water (wt%)b 0.031 0.33 0.109 0.25Solubility in water (wt%)b 0.033 0.053 0.017 —Flash point (°C) None None None NoneKB value 31 41 31 39ODP (CFC-11 � 1.0) 0.03 0.03 0.8 0.8GWP (CO2 � 1.0, 100 years) 370 310 5000 4800aAzeotrope of CFC-113 and ethanol.bAt 25°C.cAt 20°C.

the fundamental properties, range of applications, and typical cleaning processes of HCFC-225 and its blends are discussed. The emphasis is on applications used by manufacturersin Japan.

FUNDAMENTAL PROPERTIES

The physical properties of HCFC-225, HCFC-225AES, CFC-113, and CFC-113AES aresummarized in Table 1. The boiling point, surface tension, and Kauri-butanol (KB) valueare the key properties of a cleaning agent. A boiling point in the range from 40 to 60°C isimportant to clean parts without raising their temperature to the point where the temper-ature-sensitive elements or materials are affected. Low surface tension enables a cleaningagent to penetrate into small gaps, and a KB value of approximately 30 to 35 allowsremoval of soils without damage to metals, plastics, or elastomers. Latent heat of vapor-ization is also a key property in vapor cleaning. A low latent heat of vaporization results inless energy consumption in vaporization and rapid drying. As shown in Table 1, the phys-ical properties of HCFC-225 and its blends are very similar to those of CFC-113 and itsblends. HCFC-225 and its blends can be used with a few or no changes of the existing clean-ing equipment and procedure for CFC-113 and its blends.

Compatibility of HCFC-225 with representative materials of construction is summa-rized in Table 2. In general, HCFC-225 is equivalent to CFC-113; in some cases it is some-what more aggressive. Some plastics such as stressed polycarbonate and ABS resin maycrack, craze, or swell.

TOXICOLOGY SUMMARY

HCFC-225ca and HCFC-225cb were individually tested by PAFT (Programme forAlternative Fluorocarbon Toxicity Testing). Testing was conducted on the separate isomersto allow characterization of the toxicity of each isomer and of the mixed isomer products.


Table 2 Material Compatibilities of HCFC-225

HCFC-225 HCFC-225AES

Plastica Similar to CFC-113; Similar to CFC-113/EtOH;exception: acrylic exception: acrylic

Elastomerb Similar to CFC-113 Similar to CFC-113/AESMetalc Similar to CFC-113 Similar to CFC-113/AESa21 kinds of plastics tested.b10 kinds of elastomers tested.c17 kinds of metals tested.

Data from acute toxicity studies demonstrate that HCFC-225ca and HCFC-225cb havevery low acute toxicity. Neither isomer causes eye irritation or dermal toxicity in stan-dardized tests; skin application of both isomers at high doses (2000 mg/kg body weight)produces no adverse effects. Oral administration of either isomer at high doses (5000mg/kg body weight) does not cause any mortality. Therefore the oral LD50s are greater than5000 mg/kg body weight. Both isomers also have very low acute inhalation toxicity asmeasured by the concentration that causes 50% mortality in experimental animals, theLC50. The 4-h exposure LC50s for both isomers are approximately 37,000 ppm in rats.Anesthetic-like effects are observed in rats at high inhalation concentration (greater than5000 ppm).

As with many other halocarbons and hydrocarbons, inhalation of HCFC-225ca andHCFC-225cb followed by intravenous injection of epinephrine, which simulates humanstress reactions, results in a cardiac sensitization response in experimental screening stud-ies with beagle dogs. This cardiac sensitization response is observed at approximately15,000 ppm for the mixture of HCFC-225 at ca/225cb (45/55 weight percent) and 20,000ppm for the isomer 225cb, which are levels well above expected exposures. By comparison,a cardiac sensitization response is observed with CFC-113 at 5000 ppm.

In 28-day inhalation studies with rats, the activity and responsiveness of the animalswere reduced at exposures of 5000 ppm or greater for each isomer. Toxicity was otherwiseconfined to the liver. To investigate the biological relevance of the liver toxicity to humans,comparative repeated inhalation studies have been conducted with rats, hamsters, guineapigs, and marmosets. The results indicated that although liver effects were observed in therats, the effects did not transcend to higher primates.

In conclusion, HCFC-225 is not a carcinogen, a teratogen, does not cause birth defects,and is not harmful to unborn children. From the results of the toxicological findings, an8-h/day/40-h workweek exposure was set at 50 ppm (AEL).

Emergency exposure limit guidelines have been set at 1000 ppm for a period of 15 min,and 2000 ppm limit for a 1-min time period. This information is considered supplementaland applies to the material without additives. Please review the Material Safety Data Sheetfor any chemical before using.

APPLICATIONS

Of the HCFC-225 produced, 70% is used in Japan. About 70% is used neat (withoutadditives) in precision cleaning, for rinsing and drying in cosolvent systems, as a carriersolvent, and as a solvent for chemical reaction.


Blends are used when the solvency of HCFC-225 is insufficient for the cleaning appli-cation. In some cases, including precision cleaning applications, cleaning agents are usedsequentially, with HCFC-225 typically used as the rinsing and/or drying agent; an advan-tage is that rinsing and drying times typically are reduced.

Precision Cleaning

HCFC-225 has been used as an effective cleaning agent for removal of oils, greases,dusts, and so on. Especially, it has been applied for degreasing and dust removing of pre-cision components such as bearings, miniature motors, coils, relays, and connectors. Thesecomponents usually have to avoid contact with water, and they can hardly be dried withalternatives other than HCFC-225 because of their minute and complicated shapes.

Cleaning procedures of HCFC-225 are quite similar to those of CFC-113 as shown inFigure 1. The procedure consists of immersing a work load into the warm solvent, rinsingwith cold solvent, and drying in solvent vapor. An existing vapor degreaser for CFC-113can be used for HCFC-225 as described above. Agitation and ultrasonic cleaning are oftenused together in the immersing step. Total cleaning time may vary from less than 1 min to30 min, depending on the required cleanliness level. Typical cleaning time is a few minutesin total. In certain cases, rust preventative is added into HCFC-225, and then parts to becleaned are immersed into HCFC-225 to clean and prevent them from corroding at thesame time.

Other Sequential Cleaning Applications

In some processes, parts are cleaned with hydrocarbon, and then rinsed with HCFC-225 vapor. In other typical semiaqueous/cosolvent processes, HCFC-225 is added to theend of the process to improve drying and spotting. Parts are cleaned with an aqueous orsemiaqueous cleaning agent and then rinsed with water. Residual water is removed withalcohol. Finally, alcohol is removed with HCFC-225.

In both cases the content of hydrocarbon or alcohol should be monitored by determin-ing the specific gravity or boiling point of the mixture. The isopropyl alcohol mixture will

Figure 1 Schematic diagram of six-sump type cleaning equipment used by bearing manufacturer.


Table 3 Fluxes and Solder Pastes for HCFC-225AES

Manufacturer Flux Solder Paste

Alphametals SM4592C-10T, PRM615-15, SR-9100-39R5186

Asahi Chemical Res. AGF-200FZ —Harima Chemicals — F6, F16, F27 typesKoki JS-64-ND-3 SE4-A228Nihon Handa — RX-463-AK-3, RX-462-AK-3Nihon Superior NS-829 SM63RA-FMQ7SSenju PO-4003-K3, PO-F-210-K4, OZ 63-606F-AA-10.5,

SR-210-K2 OZ 2062-606F-AA-10.5,OZ 63-633F-42-10,OZ 2062-633F-42-10

Tamura CRF-225-20 SQ-1040AK-7, SQ-1030AK-7,SQ-2030AK-7

become flammable at greater than 20% by weight alcohol. It is necessary to maintain thealcohol content at or below 10% by weight by monitoring the boiling point. The content canalso be monitored by specific gravity of the mixture.

Defluxing of Printed Circuit Boards for Automobiles

High reliability and durability are required for printed circuit boards (PCBs) installedin automobiles. The manufacturing process of the PCBs consists of soldering electronicdevices, cleaning, and coating. Given the high production volume, rapid cleaning isrequired. Many manufacturers use HCFC-225AES for defluxing. Most use fluxes and sol-der pastes specifically suited to the solvent to maximize process efficiency (Table 3).

In one application, a three-sump process is used (Figure 2). The procedure is the sameas that using HCFC-225 as described above. The tact (labor) time is 15 s, total process timeis 1 min.

In a second application, a vertical two-sump is used (Figure 3). PCBs are cleaned withwarm solvent (40 to 45°C), then rinsed by spraying with cold solvent and dried in solventvapor. The tact time may be 10 s or more; total process time is 1 min.

It is important to note that in both cases, less solvent is used than in the original CFC-113 process. Manufacturers no longer clean all PCBs. Instead, they evaluate the necessity ofcleaning, and they only clean for essential applications. They also use recovery and/or dis-tillation equipment to reduce solvent usage.

Special Applications

HCFC-225 is used in special cleaning and noncleaning applications. Such special clean-ing includes refrigeration cycle cleaning to replace CFC-11, liquid oxygen cleaning foraerospace, and iron manufacturing to replace CFC-113 or carbon tetrachloride. It is used asa carrier solvent for silicones to coat syringe needles and for lubricant to coat hard disks.Furthermore, it is used as a solvent in chemical reactions.

HCFC-225 is gradually replacing CFC-113 for some dry cleaning processes. Specificallydesigned dry cleaning equipment is available from several equipment manufacturers.Surfactant additives for HCFC-225 are also available in Japan.


Figure 2 Schematic diagram of three-sump-type cleaning equipment. (A) Warm sump (15 s), (B)cold sump (15 s), (C) vapor (15 s).

Figure 3 Schematic diagram of vertical two-sump-type cleaning equipment. (1) Warm sump immer-sion (15 s), (2) rinse with spray (10 s), (3) vapor (20 s).

REGULATORY

HCFC-225 is considered a transitional alternative that will be phased out by the year2020 based on the Montreal Protocol. However, as described in the report of the 15th meet-ing of the open-ended working group of the parties to the Montreal Protocol,4 for certainhigh-reliability applications, the use of this chemical has remained important in view of thelack of appropriate alternatives with good technical or environmental performance.


REFERENCES

1. K. Kitamura et al., in Proceedings of the 1994 International CFC & Halon Alternatives Conference,October, Washington, D.C., 1994, 585.

2. K. Kitamura et al., in Proceedings of the 1995 International CFC & Halon Alternatives Conference,October, Washington, D.C., 1995, 779.

3. K. Kitamura et al., in Proceedings of the 1996 International Conference on Ozone ProtectionTechnologies, October, Washington, D.C., 1996, 669.

4. UNEP, Report of the 15th meeting of the open-ended working group of the parties to theMontreal Protocol, INEP/Ozl.Pro./WG.1/15/5, 12 June, 1997.


CHAPTER 1.12

d-Limonene: A Safe and VersatileNaturally Occurring Alternative Solvent

Ross Gustafson

CONTENTS

IntroductionTypical PropertiesSafety and Environmental ConcernsEffectivenessStand-Alone Parts WashersAutomatic Parts WashersConversion of Vapor DegreasersCircuit Board CleaningAerosol ApplicationsSample FormulationsCase StudiesOther UsesOther ConcernsConclusionReferences

INTRODUCTION

With the production phaseout of chlorofluorocarbons (CFCs) and other ozone-depleting chemicals and increased awareness of workplace safety, many different cleaningsolvents have been introduced. d-Limonene, a naturally occurring substance extractedfrom citrus rind during the juicing process, has shown great effectiveness in the cleaningmarket and is experiencing a growing acceptance as the solvent of choice in a number ofdifferent applications. d-Limonene is a non-water-soluble solvent. It can be used straight orblended with an emulsification system to produce a water dilutable/rinsable product. It iscapable of effectively removing organic dirt loads ranging from light cutting oils and lubri-cants to heavy greases, such as cosmoline.


Figure 1 Structure of d-limonene.

+

d-limonene

d-Limonene is in the chemical family of terpenes. These are products, produced by allplant life, based on isoprene. All terpenes are combinations of two or more isoprenemolecules. The d-limonene structure is shown in Figure 1 as the addition of two isoprenemolecules. Some common terpenes besides d-limonene are pinene, menthol, camphene,and b-carotene.

d-Limonene is extracted at two different points in the citrus-juicing process. During thepressing of the fruit to remove the juice, a significant quantity of the peel oil is also pressedout. This floats on top of the juice, and is separated by decanting. This fraction is calledcold-pressed oil, and contains many flavor and fragrance compounds along with thed-limonene. The cold-pressed oil can be distilled to separate the d-limonene fraction fromthe flavor and fragrance compounds. The separated d-limonene is termed food-graded-limonene in the industry. After the fruit has been juiced, the peels are sent to a steamextraction step where more d-limonene is recovered. During this process, the peels areexposed to steam, which carries the d-limonene to a condenser. Here the water andd-limonene separate into aqueous and oil phase, and the d-limonene is recovered. Thisproduct is termed technical-grade d-limonene.

Historically, orange oil and d-limonene have been used extensively as flavor and fra-grance ingredients in a wide variety of products, including perfumes, soaps, and bever-ages. d-Limonene has also been used to make paint solids through a polymerizationprocess. The first uses of d-limonene as a cleaning solvent began in the 1970s, but with theincreased environmental awareness in the 1980s and 1990s, use of d-limonene as a cleanerhas shown a tremendous increase.

TYPICAL PROPERTIES

d-Limonene is a thin, relatively colorless liquid. The typical physical and chemicalproperties for technical grade are:

Color Slight yellow to water whiteOdor Orange aromaSpecific gravity (25°C) 0.83800.843Refractive index (20°C) 1.4710 to 1.4740Optical rotation (25°C) �96° to �104°Flash point 115°FBoiling point 178°C (310°F)Freezing point �96°C (�140°F)Evaporation rate 0.05 vs. butyl acetateWater solubility InsolubleVapor pressure (20°C) 1.4 mmHg


SAFETY AND ENVIRONMENTAL CONCERNS

From a personal-safety standpoint, d-limonene is a much safer product for use thanmost other solvents. The oral LD50 of d-limonene is greater than 5000 mg/kg body weight.The product also is classified as a food additive and has been granted the Food and DrugAdministration GRAS (Generally Recognized as Safe) status. For comparison, the typicalmineral spirit LD50 is around 2000 mg/kg body weight. A formal threshold limit value(TLV) or permissible exposure limit (PEL) has never been established for d-limonene.

d-Limonene is also noncaustic and nonreactive to metal surfaces. It has been classifiedas a slight skin irritant, because it can remove the naturally occurring oils from skin, buthas not been shown to cause lasting damage. It is not carcinogenic or mutagenic, and is cur-rently being evaluated for its chemopreventative and chemotherapeutic properties.

d-Limonene is not itself and does not contain any ozone-depleting chemicals. It is cur-rently regulated as a volatile organic compound (VOC). The evaporation rate of d-limoneneis relatively low, so the actual VOC emissions are small. d-Limonene is not considered anair toxic or hazardous air pollutant (HAP), and is not regulated under the Clean Air Act orSARA Title III. It is an approved solvent substitute under SNAP (Significant NewAlternatives Program).

The issue of global warming as it pertains to the recovery and use of d-limonene is dif-ficult, and no reliable estimate has been completed. When plants create d-limonene, or anyterpene, they use carbon dioxide and water. When a terpene is destroyed or degrades, car-bon dioxide and water are produced. So the creation and destruction of d-limonene wouldresult in a net zero global-warming effect. Generally, d-limonene solutions are not heatedwhen used. In those applications there would be essentially no global-warming effect. Atsome point in a cleaning process, a small amount of energy may be used for heating dry-ing air or rinse tanks, or at least circulation of fluids. This will have a global-warmingimpact, but in general is not significant. The real difficulty lies in estimating the impact ofproducing the fruit and extracting the product from the rind. Since d-limonene is truly a by-product of the juicing industry, very little or none of the impact from growing and pro-cessing the fruit should be attributed to d-limonene production. Citrus growers do notgrow fruit for the express purpose of producing d-limonene. They are in the business ofproducing fruit and juice. So the global-warming impacts in the form of fertilizers, irriga-tion, transportation, and other energy uses associated with agriculture are attributable tofruit production, but not to d-limonene production. The major source of global-warmingpotential attributable to d-limonene production is the energy required to perform theextraction. Although this has not been formally calculated or estimated, one would notexpect to see a significant amount of energy used and one could therefore consider pro-duction of d-limonene to have a very small global-warming potential.

The closed cup flash point for d-limonene has been established at 117° F. This makesd-limonene a Class III combustible under Department of Transportation (DOT) regulations.For transportation purposes this means that when shipping d-limonene by ground, thevehicle must be placarded as hazardous if a single container contains over 110 gal ofd-limonene. For storage purposes, sprinkler systems in the storage area are required. It isalso recommended to use explosion-proof pumps and wiring.

When d-limonene/surfactant/water systems are made, the closed cup flash point willgenerally rise to about 130°F. These solutions will have a very high open cup flash point,and will not support a flame at any temperature below boiling.

Under RCRA regulations, d-limonene is classified as a characteristic ignitable hazard,and must be disposed of as a hazardous waste. However, water/d-limonene emulsions will


Table 1 KB Values, ComparativeStrength (Solvency) of IndustrialSolvents (Higher Values � HigherDissolving Power)

Solvent KB Value

Methylene Chloride 136Trichlorethylene 129Benzene 107Toluene 105Xylene 98Perchloroethylene 92d-limonene 67Mineral Spirits 37Naphtha 34Kerosene 34Stoddard Solvent 33MEK N/AAcetone N/A

have no flash point at less than 1.5% d-limonene concentration. So when using a water/d-limonene system and rinsing the cleaned materials, the rinse water can generally be dis-posed of as a nonhazardous material if the concentration is below this amount.

EFFECTIVENESS

In many cleaning applications, the soils to be removed are organic oils and greases.Because of the inherent chemical differences between organic and inorganic materials, suchas polarity and ionic effects, organic solvents tend to perform much better for cleaningthese types of soils than water-based solutions. To compare organic solvent strengths, theKauri-butanol (KB) value, an ASTM method (D1133-97), has been established. The moretoxic chlorinated solvents and benzene and its related compounds are all extremely effec-tive cleaning solvents and have high KB values. The KB value of d-limonene is a bit lower,but higher than that of petroleum-derived products (Table 1). It is not possible to performthe KB test on oxygenated compounds, so there is no listed value for methyl ethyl ketone(MEK) or acetone.

STAND-ALONE PARTS WASHERS

The typical tub-and-basin parts washers found in various mechanical maintenanceshops can be adapted to use d-limonene as the washing solvent, and in most cases cleaningeffectiveness will be increased and the time necessary for cleaning will be reduced.Typically the parts washer holds 18 to 20 gal of solvent and the solvent is pumped througha nozzle or brush into the tub for cleaning the parts. As the solvent is used and soil loadingincreases, the effectiveness of the solvent diminishes. The solvent is replaced at some pre-determined point. When using d-limonene as the solvent, an in-line cartridge filter can beinstalled. A wound cotton filter preferentially absorbs the oils and greases, allowing thed-limonene to maintain its cleaning effectiveness for extended periods of time. The filtersare changed out every 1 to 3 weeks, depending on workload, and approximately 1 gal offresh d-limonene is added to the parts washer monthly to make up for dragout, which is


solution lost by adhesion to the parts being cleaned when they are removed from the bath.While a petroleum-based product may need replacement every 1 to 2 weeks, parts wash-ers using d-limonene as the solvent have stayed in service for up to 3 years without a com-plete replacement of the solvent. The used filters are incinerated by a waste disposalcompany. Because these filters are the sole waste to dispose of, a company’s waste statusmay change to a small-quantity generator.

AUTOMATIC PARTS WASHERS

In much the same manner, d-limonene can be used in batch-loaded spray and immer-sion systems. In many of these systems, a water-soluble detergent is used as the cleaningagent. Typically, these detergents have an elevated pH and are used at temperatures wellabove 120°F. This combination of temperature and pH can have adverse effects on metalparts, such as corrosion. Depending on the temperature and cleaning system, appropriatecontrols may be required in consideration of the flash point. In addition, a hot caustic solu-tion can potentially cause injuries to personnel. Water usage is also a concern. d-Limonenecan be used in automatic parts washers in much the same manner as in stand-alone partswashers. By inserting a filter in-line on the system, the solvent can be used for extendedperiods of time. It can also be used effectively at room temperature, reducing costs associ-ated with heating the solution. If water rinsing of the parts is necessary to remove residueof soil or cleaning agent, an emulsifier can be added to the d-limonene, which will allowthe parts to be water-rinsed.

CONVERSION OF VAPOR DEGREASERS

The easiest conversion is the one most likely to be used. In some applications, d-limonene can be substituted directly into the line as a replacement solvent. One issue is thatd-limonene will not work as a vapor-phase cleaner. The solution must be brought intodirect contact with the parts being cleaned. Such conversions are suited to applicationswhere rinsing to remove d-limonene is not required.

The approach to conversion depends partly on the design of the original cleaning sys-tem. In systems where a conveyor carries parts into the vapor cleaning zone, it can be mod-ified so that parts are lowered into the liquid instead of into the vapor. The dwell time forthe parts in contact with the cleaner will be about the same, or a little less. The parts can beair-dried just as they are with a vapor cleaning solvent. The existing equipment can prob-ably be used with only minor modifications, as long as the parts conveyor system can befitted to allow direct contact with the solvent and the pumps and piping are checked formaterial compatibility with the solvent.

Another method of conversion for a straight solvent substitution is to build a spray sys-tem into the cleaning line. This requires installing an enclosed booth with spray headersdirected at the parts. The parts can then be air-dried. Some new equipment and line mod-ifications are needed (the spraying time is determined by testing), but it is not an extensivechange.

In both these cases, the d-limonene is used neat, or undiluted. It is not water soluble; itis not rinsed; it will not cause rusting or oxidation of any materials. As thed-limonene is used, soil loading of oils and greases from the parts being cleaned willincrease. Also, due to dragout, d-limonene will need to be periodically replaced to main-tain the initial volume.


Once the d-limonene has reached the limits of its effective lifetime, it must be disposedof properly. Ideally, waste d-limonene is incinerated. The d-limonene has a fuel value of18,000 BTU/pound, and is valuable as a secondary fuel. Alternatively, the d-limonene canbe distilled for reuse.

d-Limonene blended with surfactants can be used in aqueous emulsion. In diluting thed-limonene with water, the cost of cleaning agent is reduced. Usually these solutions in useare 10 to 25% d-limonene/surfactant and the remainder is water. This requires more exten-sive equipment changes when converting from a vapor-type cleaning system. Because ofthe nature of the mixture of d-limonene, surfactants and water, the parts must be water-rinsed after cleaning to remove any residue that may remain on the parts. In general, mostvapor cleaning equipment can be converted to a wash tank, as suggested above, for usingthe water emulsion. Again, the line must be altered to allow direct contact of the parts withthe cleaning solution, but no additional equipment is necessary at the wash step. A newpiece of equipment needs to be inserted after the wash tank to rinse the parts, usually awater-spray system. After rinsing, the parts can then be dried or go on for further process-ing. Periodic addition of cleaning agent is needed to replace the dragout, and filtration canbe used to keep the solution usable. When this solution can no longer be used, again it isbest to have it incinerated. A number of incinerators are equipped to deal with the presenceof water in the solution. Because of the mixture created by the surfactants and water, thismixture cannot be distilled and regenerated.

CIRCUIT BOARD CLEANING

d-Limonene has been used for flux removal on circuit boards. The d-limonene can besubstituted directly into the washing equipment and used either in the spray or flushingmethod. d-Limonene does not dry as quickly as the CFCs that have historically been usedfor this process. But, if a cleaning system designed for use with low-flash point solvents isbeing used, by following the d-limonene washing stage with a rinse of either acetone or iso-propyl alcohol (IPA), drying times and residue levels are comparable. In such applications,low flash point cleaning systems must be adopted. High-purity/low-residue grades of d-limonene are being introduced for printed circuit board applications with some success,although cost of this material may be twice that of regular d-limonene.

AEROSOL APPLICATIONS

d-Limonene has been used as an aerosol for various cleaning applications, includingelectronics and parts cleaning. It can be sprayed on motor windings and allowed to dry byevaporation and sprayed on contacts, switches, and connections and wiped or let air-dry.It can also be sprayed into tight areas or onto bolts to facilitate removal. In many cases theresidue left after evaporation does not interfere with the process. d-Limonene can also becombined with 10% IPA or acetone to promote quicker drying times and lower residue.When packaging d-limonene in an aerosol container, carbon dioxide at 15 to 20% can beused as the propellant.

A water-rinsable product can be made by combining d-limonene in a 90/10 ratio ofd-limonene to emulsifier. Various types of emulsifiers can be used for this, includingethoxylated alcohols, alkylamine dodecylbenzene sulfonates, coconut diethanolamides,and sodium xylene sulfonates. The emulsified product can then be sprayed on the surfaceto be cleaned and rinsed off with water. Again, carbon dioxide is the propellant of choice.


It is also possible to make aqueous dilutions of the d-limonene/emulsifier blends or foam-ing products for aerosol packaging. These types of products require propane/isobutanepropellant.

d-Limonene can be combined in aerosol formulations to impart a pleasant citrus odor.Care must be taken to evaluate proper packaging materials. d-Limonene may causeswelling of gaskets and valves of some conventional dispensers. Viton and neoprene maybe some of the best choices for aerosol stem gaskets (better than butyl or buna). Valves andcans should have an epoxy coating. Aerosol packagers and gasket suppliers should be con-sulted as to materials recommended for d-limonene.

SAMPLE FORMULATIONS

A few examples of possible formulations for various applications for critical cleaningare presented here. The emulsifier in the formulations can be a wide variety of products, ora blend of several different products, depending on the properties desired.

A general-purpose concentrate for water dilution:

85 to 90% d-limonene10 to 15% emulsifier

A water dilutable/rinsable printing press cleaner:

90% d-limonene7% emulsifier3% tripropylene methyl glycol ether

This product can be diluted to 70 to 75% water to clean soy- and water-based inks.The following guidelines can be used to formulate products for various applications.

In this case the d-limonene and emulsifier or surfactant blend should be mixed in equalproportions.

Application % d-Limonene/Emulsifier % Water

Engine degreaser 25 75Tar/asphalt removal 50 50Adhesive removal 15 85Marine vessel cleaner 10 90Industrial metal cleaner 20 80

These ratios are meant as guidelines.

CASE STUDIES

The following case studies show a variety of uses and benefits that have been foundfrom switching to d-limonene-based cleaning systems.

Martin Marietta Astronautics has replaced 1,1,1-trichloroethane (TCA) and MEK witha terpene cleaner for hand-wiping operations.1 The terpene cleaner was selected after 16months of extensive testing of citrus- and alkaline-based compounds. Workers prefer the


citrus-based cleaner because it is more efficient. The terpene cleaner leaves less residueresulting in higher coating bond strength. Martin Marietta estimates the change hasreduced toxic emissions by thousands of pounds per year. Research costs were $350,000 tofind a suitable replacement for MEK and TCA. Estimated savings are $250,000/year.

In a joint research effort, the U.S. EPA and APS Materials, Inc., have investigatedthe use of a limonene cleaner to replace TCA and methanol.2 APS Materials, Inc. is ametal-finishing company that plasma-coats parts for use in hostile environments. In thebiomedical parts division, cobalt/molybdenum and titanium parts are coated with aporous titanium layer for use as orthopedic implants. APS Materials has converted to theterpene cleaner as a result of the investigation. Cleaning efficacy is excellent with a slightincrease in bonding strength for the limonene-cleaned parts. Changing to the aqueous sys-tem required the addition of rinse and dry stations. The new system cost $1800 to installwith annual operating expenses of $850. Net savings are $4800/year.

GE Medical Systems of Waukesha, Wisconsin, is a manufacturer of medical diagnosticequipment. Spray cleaning (degreasing) of parts using TCA resulted in fugitive air emis-sions. GE Medical Systems eliminated fugitive TCA emissions by converting to a terpenecleaner.3 With TCA, 800 gal of solvent were purchased annually, all of which was lost toatmosphere. Because terpene cleaner is much less volatile, only 30 gal are purchased peryear. In addition, terpene cleaner is recycled. No capital expenditure was required.

Northern Precision Casting of Lake Geneva, Wisconsin, switched to a citrus-based sol-vent for cleaning the wax patterns used in making molds.4 Previously, they used TCA. TCAfugitive emissions amounted to 18,000 lb in 1988. The terpene solvent is water soluble andis discharged to a publicly owned treatment works. No capital costs were incurred for thechange. Maintenance and operating costs are equivalent.

The Marine Corps Air Station Naval Aviation Depot, Cherry Point, North Carolina, isresponsible for the complete maintenance/rebuilding of naval aircraft. In 1990 the depotused 8000 gal of CFC-113 and 15,600 gal of 1,1,1-TCA. By the end of 1992, CFC-113 usagehad been reduced to 500 gal annually and TCA usage had been cut to about 4800 gal annu-ally. Terpene cleaners were used as one of a number of approaches.5 Approaches includedsoap bubbles for leak checks; aqueous power washers for electronics, motor, and engineshop use; terpene cleaners for hand wiping; steam cleaning or wet sodium bicarbonateblasting for soil and carbon removal; and plastic media blasting for paint removal.

AT&T has reduced usage of CFC-113 by converting to a semiaqueous chemistry forcleaning surface mount assemblies.6 Parts are carried by conveyor into a power washerconsisting of wash and rinse/dry modules. Low- and high-pressure sprays of a terpenecleaner are followed by nitrogen knives to reduce cleaning solution dragout and blanketthe washer with an inert atmosphere to prevent fire. In the second module, the parts arerinsed with low-pressure, then high-pressure water sprays to remove the terpene cleaner.Rinsing is followed by water removal by air knives within the same module. Care must betaken in selecting materials of construction in the surface-mount components because theterpene cleaner swells some plastics and elastomers. AT&T has found that the new clean-ing method is more economical than the previous CFC-113 method.

In 1988, the Motorola Corporation had 29 flux-removal cleaning systems using 250,000lb of CFC-113 annually. By August 1991, Motorola had eliminated CFC-113 usage. Many ofthe printed circuit board assemblies are now assembled using a no-clean flux. Assembliesthat require cleaning now use terpenes and water.7 Benefits reported include cleanerassemblies, lower production downtime, and decreased cleaning cost. Cleaning costs arenow about $8/h using the terpene/water vs. $38/h for CFC-113.

Crown Equipment Corporation, New Bremen, Ohio, manufactures electric lift trucksand television antenna rotors. Parts cleaning involves mild steel, aluminum, cast iron, and


copper. In 1988 Crown used 208,000 lb of TCA in cold-cleaning (immersion) and vapordegreasing operations. Hand dipping now uses a water-based cleaner with rust inhibitoradded for corrosion resistance, and 100% d-limonene spray cleaner has replaced TCA forhand-wiped parts.8 An alkaline aqueous immersion cleaner has replaced one degreaser(with inhibitor added for ferrous parts). The other degreaser was replaced with an aqueouspower washer that uses heat, agitation, and forced air drying to produce clean parts. Thepayback period for capital expenses was 10 months. In 1989, Crown saved $100,000 inchemical costs. Crown Equipment has switched to water-based cleaning with no decreasein production. Employees prefer the water-based cleaner for hand dipping.

The Bureau of Engraving, Industrial Division, manufactures printed circuit boards. In1990, it decided to eliminate the use of methylene chloride and TCA, which were beingused at the rate of 681,000 lb/year. Several changes in the manufacturing process were nec-essary to accomplish this goal, including the use of water-based and terpene-based clean-ers. The Bureau of Engraving, Industrial Division, saves $250,000 annually in purchase costand $20,000 in maintenance, energy, and disposal costs.9

OTHER USES

Several other cleaning applications exist for d-limonene. d-Limonene is also usedextensively in the asphalt industry for clean up and aggregate analysis, as an ink cleaner inprinting operations, in the oil and gas fields for maintenance and as a well recovery solvent.It is used in hand cleaners and a wide variety of other general cleaning and householdapplications. In general, if an organic soil is to be removed, d-limonene may perform aswell or better than other solvents.

In addition to cleaning applications, d-limonene has been found to have thermody-namic properties that make it a very good heat transfer fluid, especially for cryogenic appli-cations, below �100°C. It has also been used as a carrier solvent in paints and similarcoatings, as well as adhesives. Studies have also shown d-limonene to be an effective pes-ticide and bactericide, although these are not yet approved uses. These are in addition tothe best-known uses of d-limonene and other citrus derivatives as flavor and fragrancecomponents.

OTHER CONCERNS

A “perfect” solvent probably does not exist, and d-limonene is no exception.d-Limonene can cause a certain amount of swelling of polymers, so the plastic materialsused with a d-limonene system must be chosen with caution. Viton is the best seal to use injoints, and nylon-braided PVC seems to be the most acceptable material for hoses. As withany solvent, gloves should be worn when working directly in the solution. The best mate-rial for these is nitrile latex.

CONCLUSION

d-Limonene is an extremely effective and relatively safe cleaner and solvent for use inmany industries. It can be used in a wide variety of applications and in most cases will per-form better and longer than the classic solvents. Although it is not perfect, it is a goodoption to be considered when choosing a cleaning system or looking for an effective sol-vent replacement.


REFERENCES

1. Dykema, K.J., and G.R. Larsen. 1993. Shifting the environmental paradigm at Martin MariettaAstronautics, Pollut. Prev. Rev., Spring, 205.

2. Brown, L.M., J. Springer, and M. Bower. 1992. Chemical substitution for 1,1-trichloroethane andmethanol in an industrial cleaning operation, J. Haz. Mater., 29:179–188.

3. Wisconsin Department of Natural Resources, Case Study: GE Medical Systems; Replacing 1,1,1-Trichloroethane with Citrus-Based Solvents, PUBL-SW-168 92, Hazardous Waste MinimizationProgram (SW/3), Madison, WI.

4. Wisconsin Department of Natural Resources, Case Study: Northern Precision Casting; Replacing1,1,1-Trichloroethane (TCA) with Citrus-Based Solvents, PUBL-SW-161 92, Hazardous WasteMinimization Program (SW/3), Madison, WI.

5. Fennell, M.B., and Roberts, J.M., Naval Aviation Depot: hazardous minimization—saving time,money , and the environment, in Proceedings of the Aerospace Symposium, Lake Buena Vista, FL,1993, 39–46.

6. Terpene Cleaning of Surface Mount Assemblies, Aqueous and Semi-Aqueous Alternatives forCFC-113 and Methyl Chloroform Cleaning of Printed Circuit Board Assemblies, EPA/400/1–91/016, June 1991, 51–60.

7. Terpene Cleaning of Printed Circuit Board Assemblies, Aqueous and Semi-Aqueous Alternativesfor CFC-113 and Methyl Chloroform Cleaning of Printed Circuit Board Assemblies, EPA/400/1–91/016, June 1991, 61–62.

8. Kohler, K., and A. Sasson, Case studies: multi-industry success stories to reduce TCA use in Ohio,Pollut. Prev. Rev., Autumn, 407–409, 1993.

9. Currie, W.T., Vice President, Facilities and Environmental Affairs, Bureau of Engraving, Inc.,Industrial Division, 500 South Fourth Street, Minneapolis, MN, 55415, MnTAP, 1993 Governor’sAwards for Excellence in Pollution Prevention.


Documents

1655_01.pdf