EuropEan CS oaTInG€¦ · european coatings JournaL 2016 BYK ADDITIVEs: Improving product characteristics and production processes. Small quantities, big properties. BYK Additives

www.european-coatings.com

EuropEan

dossierC oaTInGS

12 DispersantsCore/shell structures with encapsulated aminic pigment-affine groups

96 FluorosurFactantsCombining good eco toxicological behaviour with technical performance

aDDitivesThe best technical papers on additives for coatings published in the European Coatings Journal within the past three years.

2016

presenteD by

http://www.byk.com

e d i t o r i a l2

e u r o p e a n c o at i n g s J o u r n a L 2 0 1 6

Sour

ce: M

ashe

- Fo

tolia

.com

Tiny amount, huge effect!

Additives are indispensable ingredients of each and every formulation. They are used in tiny amounts to help modify the performance spectrum of coat-ings systems. Development work on additives never ceases and there are countless products on the market.. As a formulator, you are continually called on to be au fait with the latest advances in research and development.. Such knowledge is rarely to be found in a single package. European Coatings Jour-nal is about to change that. This thematic dossier is bursting with information on additives that we have compiled for you. In it you will find all the relevant technical papers on additives that have been published in European Coatings Journal over the last three years. Now there’s a welcome development!

Join our Group “European Coatings Industry”

Dr. sonja schulteeditor-in-chieft +49 511 [email protected]

3p o r t r a i t


BYK ADDITIVEs:Improving product characteristics and production processes. Small quantities, big properties.

BYK Additives & Instruments is one of the world’s leading suppliers in the field of additives and measuring instruments.

Additives are chemical substances which, when used in small quantities, improve product properties such as scratch resistance or surface gloss. Manufacturing processes are also optimized by the addition of additives.

The coatings, inks and plastics industries are among the main consumers of BYK additives. Yet with exploration technology oil/gas, the manufacture of care products, the production of adhe-sives and sealants, and construction chemistry, too, BYK additives improve the product characteristics and production processes.

Testing and measuring instruments from BYK can effectively evalu-ate the quality of color, gloss and appearance as well as the physi-cal properties of paint, plastic and paper products and are an impor-tant part of quality control.

As a globally operating specialty chemicals company, BYK has production sites in Wesel, Kempen, Moosburg, Schkopau and Geretsried (Germany), Deventer and Denekamp (Netherlands), Widnes (UK), Wallingford, Chester, Gonzales, Louisville (USA) and Tongling (China).

Today the company employs around 2,000 people worldwide and forms part of the ALTANA Group.

BYK-Chemie GmbH · Abelstrasse 45 · 46483 Wesel · GermanyTelephone +49 281 670-0 · Fax +49 281 [email protected] · www.byk.com

http://www.byk.com

c o n t e n t s4


EuropEan Coatings Journal 2016

6 DispErsants Designed for selective adsorption ByA.Rudolfi,M.Krohen,

F.PiestertandS.Mößmer,Byk-Chemie

12 DispErsants Maximizingdispersionstability ByV.Kilpeläinen,a.Gutierrez,S.v.d.Loon,VLCI

16 Corrosion protECtion Howdispersantsaffectcorrosionresistance ByM.Muth,A.FreytagandM.Conrad,Byk-Chemie

20 Corrosion protECtion Dispersantshelptoprotectaqueouscoatings ByPh.Favresse,F.Kleinsteinberg,

K.RolandandP.Glöckner,EvonikIndustries

24 WatErbornE Coatings Newdispersantprovidesenhancedstabilisation ByS.Onclin,H.Frommelius,

P.Gómez,E.MartinezandS.Shah,BASF28 HigH soliDs Additiveoptimisesbotcolourstabilityandvisosity ByK.Schulz,EvonikIndustries

DispErsantsMaximizingDispersion stability

12

32

32 surfaCtants New superwetting additives

solveadhesionproblems ByR.Reinartz,J.M.SnyderandY.Dai,AirProducts

38 nanotECHnology Nano-additivefortougheningwater-borneelasto-

mericroofcoatings ByD.BurgardandM.Herold,BühlerPartecGmbH

42 nanotECHnology Additionofnanoparticlescanenhanceperformance

ofwater-bornecoatings ByD.BurgardandM.Herold,BühlerPartecGmbH

46 smart Coatings Permanenthydrophobicandeasy-to-cleaneffects ByC.Dechamps,L.Bermert,H.GibbsandM.Gro-

teklaes,OMGBorcherandUniveristyofAppliedSciencesKrefeld

50 CrosslinkErs Thebestoftwoworlds ByT.Unkelhäußer,M.Mallack,H.GörlitzerandR.

Lomölder,EvonikIndustries

Corrosion protECtionNew dispersing addi-tiveshelptoprotectaqueouscoatingsagainst corrosion.

20

Sour

ce: W

oszc

zak

- Fot

olia

.com

Clau

de B

eaub

ien

- Fot

olia

.com

Sour

ce: E

voni

ksurfaCtantsNew superwetting additives solve adhesionproblemsinwaterborne coatings

5c o n t e n t s


CrosslinkErsSilane-urethanehybridcrosslink-ers create scratch-resistantclearcoats

50

HarDEnErsGettingonwithacrylics

72EmulsifiErGettingmorewithless

86

56 HarDEnEr Water-bornehardenerforzero

VOCself-levelingfloors. ByS.LiuandT.Trivedi,Incorez

60 lEvElling agEnt Hydrophilicagentenhanceslevelling

andintercoatadhesion ByM.Bessel,G.JaunkyandM.Heekeren,

Byk-Chemie

66 slip aDDitivEs Performanceofstructuallydifferent

slipagentscompared ByC.Cocucera,K.QuackenbushandO.Arseven,

DowCorningEurope

72 HarDEnErs Gettingonwithacrylics ByM.Ursai,S.A.P.I.C.I.Spa

76 CrosslinkErs Biobuildingblocks ByP.Engel,EvonikIndustries

80 EmulsifiEr Newchemistry ByC.F.Palmer,Jr.,EthoxChemicals

86 rHEology moDifiEr Newbio-basedrheologyadditive

forlow-temperatureprocess ByG.FauquetandY.Trang,Arkema

90 raDiation Curing Novelapproachestoglossreduction ByB.Tiede,Lubrizol

96 surfaCtant Fluorosurfactantflourish ByR.Friedrich,E.HönerandF.Schooren,

MerckKGaA

100 DriErs Encapsulateddriersminimise

skinningofair-dryingpaints ByJ.Horakh,H.Gibbs,F.GolandG.Kiepe,

OMGBorchers

Sour

ce: t

epto

png

- Fot

olia

.com

Sour

ce: Y

vonn

e Pr

ancl

- Fo

tolia

.com

Sour

ce: S

canr

ail -

Fot

olia

.com

D i s p e r s a n t s6


Sour

ce: N

ik M

erku

lov

- Fot

olia

.com

DesigneD for selective aDsorptionDispersants based on core/shell structures stabilize pigments in reactive systemsBy astr id rudolfi, Marcel Krohnen, Frederik piestert and stefan Mößmer

Dispersing additives with a hyperbranched core/shell struc-ture have been developed with encapsulated aminic pigment-affinitive groups. these can be used in solvent-borne and solvent-free coating systems. they provide highly effective stabilisation of various types of pigment while avoiding any undesirable interactions with reactive coating systems.

t he property profile and the associated fields of application of a coating material are essentially determined by the resin that is

used. The origins of these resins are diverse and can basically be di-vided into natural substances, modified natural substances and syn-thetics [1].In modern coating systems, synthetically produced oligomeric sub-stances used as resins play a primary role in satisfying the consider-able mechanical requirements and providing the necessary chemical and physical resistance of high-performance coating materials. Reac-tive multicomponent resins, which crosslink through chemical reac-tions, are of special importance.These include, in particular, the classic two-component systems (2-pack epoxy and 2-pack polyurethane systems), which crosslink by a polyaddition reaction at room temperature, as well as stoving systems

based on a primary resin and a crosslinking resin (usually melamine or phenolic resins), which crosslink at higher temperatures with an acidic catalyst [2].

the nature of pigment Dispersants

Polymeric dispersing additives are now widely used in order to meet the demand for high-quality visual properties such as high transpar-ency or hiding power, to ensure that colours have long-term shade stability with enhanced gloss and minimum haze, and to meet the mechanical and chemical requirements in pigmented reactive coating materials. These amphiphilic surface-active additives use either steric or electrostatic hindrance to stabilise the pigments and fillers that have been used in the coating material [3].Generally speaking, all dispersing additives are composed of low-molecular to high-molecular weight chain segments that have a sta-bilising effect, combined with at least one pigment-affinitive anchor-ing group. The interactions with the pigments enable the additive to be adsorbed onto the pigment surface by means of these anchoring groups. This is the prerequisite for effective and long-lasting stabilisa-tion of a pigment.

7D i s p e r s a n t s


Results at a glance

ű Hyperbranched core/shell structures with aminic pigment-affinitive anchoring groups were used as the basis for new dis-persing additives for solvent-borne and solvent-free reactive systems.

ű The specially designed hyperbranched core/shell structure can prevent undesirable interactions between reactive coating systems and the additive.

ű Practical tests confirmed that the use of this type of addi-tive gave effective viscosity reduction and improved stability on storage in both epoxy and 2-pack polyurethane systems.

ű Tests in a stoving system showed that the new structure performed well, whereas a linear copolymer dispersant ap-peared to interfere with the curing mechanism.

undesirable side-effects, leading to adverse effects on the property profile of the formulated systems. Table 1 provides an overview of the possible interactions between the aminic additives and various reac-tive coating systems.

three structurally Different aDDitives compareD

This application technology study compares innovative aminic dispers-ing additives based on the hyperbranched core/shell structures shown in Figure 1 with established structural types to investigate whether a high degree of pigment stabilisation can be achieved in representative reactive systems, while at the same time preventing possible interactions between the coating system and the additive.Three structure types were compared in this study. In each case, these possess both aminic pigment-affinitive groups of a similar chemical nature and basicity as well as polymeric side chains of a similar polar-ity. As shown in Figure 2, these structure types differ in terms of their basic skeleton: ą Structure type 1: linear copolymer with freely accessible aminic

anchoring groups; ą Structure type 2: branched polymer with sterically shielded aminic

anchoring groups; ą Structure type 3: core/shell polymer with sterically encapsulated

aminic core.

Figure 1: schematic representation of dispersing additives based on hyperbranched core/shell structures

Figure 2: Comparison of structural types of dispersants tested

As most pigments used in coating applications carry an acidic surface treatment, dispersing additives with aminic pigment-affinitive groups, in particular, are considered to be very effective additive structures because they can undergo stable acid/base interactions with the pigment surface.These aminic dispersing additives significantly improve pigment sta-bilisation; however, in various reactive systems they can cause some

D i s p e r s a n t s8


three Different coating chemistries stuDieD

These three differently structured dispersing additives were first used to produce solvent-borne pigment concentrates based on a universal grinding resin and different pigments (Colour Index references P.R. 122; P.R.170; P.G.7; P.B.15:4, and P.Bl.7, fine-sized). The same pigment concentrates without the dispersing additive served as a reference. After 24 hours, the viscosity of the individual pigment concentrates was measured (cone/plate geometry, cone diameter = 25 mm, angle = 1 °, temperature = 23 °C (73 °F), shear rate = 1 s-1).In a second step, these pigment concentrates were used in differ-ent solvent-borne reactive coating systems. A 2-pack epoxy sys-tem (bisphenol A/polyamide curing agent), a 2-pack polyurethane system (OH-functionalised acrylate with aliphatic-aromatic and/or pure aliphatic polyisocyanate), and an acid-catalysed stoving system (acrylate/melamine) served as let-down systems.The test systems formulated in this manner were poured onto a Poly-ester foil which was fixed at a 65 ° angle. The qualities of the pigment stabilisation in terms of optical properties were then established by measuring the gloss (at a 20 ° angle) and haze, and by visually as-

sessing the transparency, colour strength and colour shift under both transmitted and incident lighting conditions.In addition to this, measurements were made with respect to the in-fluence of the different structure types on the undesirable interac-tions (listed in Table 1) with each coating system. In the case of the 2-pack epoxy system, the pigmented base coating (epoxy component) was stored for twelve weeks at 50 °C (122 °F) and the flow time was measured at regular intervals.For the pigmented 2-pack polyurethane system, the resulting flow times were measured during the curing reaction between the acrylate and the aliphatic-aromatic and/or pure aliphatic polyisocyanate cur-ing agent. In the case of the acid-catalysed stoving systems, the influ-ence of the various structure types on the mechanical properties was determined by the König pendulum hardness test (in accordance with DIN EN ISO 1522).

two Dispersants effectively reDuce viscosity

The reduction of the millbase viscosity by the different dispersing ad-ditives gives a rough indication of their efficiency. As shown in Fig-

Figure 3: influence of the various structural types of dispersant on the viscosity of the pigment concentrates

Figure 4: influence of the different structure types on the gloss of the pigmented 2-pack polyurethane system (OH-functional-ised acrylate/pure aliphatic polyisocyanate)

Figure 5: influence of the different structure types on the transparency of the pigmented 2-pack polyurethane system (OH-functionalised acrylate/pure aliphatic polyisocyanate), here using p.G. 7

Figure 6: influence of the structure types on the storage stability of the epoxy base component

9D i s p e r s a n t s


ure 3, when compared with the reference, all three structural types significantly reduced the viscosity, regardless of which pigment was dispersed.Structure types 2 and 3 in particular showed a significant reduction in the paste viscosities, for which structure type 2 was especially good with P.R.122, P.R.170 and P.G.7, whereas structure type 3 gave the best results with P.R.170, P.B.15:4 and P.Bl.7. For all pigments tested, these two structure types gave higher efficiency than structure type 1 – quite considerably in some cases.

gloss anD transparency Differences are consiDerable

In addition to the viscosity reduction of the pigment concentrates, the three structural types were also tested in terms of their influence on the visual properties of the selected coating systems. Figure 4 shows the influence of the three structural types on the gloss of the 2-pack polyurethane system based on the pure aliphatic polyisocyanate cur-ing agent.In this case again, both the branched polymer with sterically shielded aminic anchoring groups (type 2 structure) and the additive based

on the hyperbranched core/shell structure (type 3 structure) gave the best results. For all pigments tested, these two structural types pro-duced a significant increase in gloss compared with the reference. Structural type 1 also tended to significantly improve the gloss; how-ever, this did not always occur, particularly for stabilisation of P.R.170 and P.B.15:4.This trend in results is also evident in the transparency of the pigment-ed coating systems. For example, in the 2-pack polyurethane system coloured with the pigment concentrate based on P.G. 7, structure types 2 and 3 produced an outstanding transparency in conjunction with a bluish/greenish colour shade (Figure 5), which is yet another indicator of very effective pigment stabilisation. In this case, structure type 1 with a more yellowish/greenish colour shade did not match the good results of the other two dispersing additives.The results shown for the example of the 2-pack polyurethane system were also confirmed in the tests with the 2-pack epoxy and the acid-catalysed stoving system. They can therefore be considered essentially representative for the performance specification of the three different structure types.

how structure correlates to viscosity stability

Figure 6 provides an even clearer picture of the influence of the three types of structure on the storage stability of the epoxy base component. After just a few days, the epoxy base component containing the addi-tive with structure type 1, which has freely accessible aminic anchoring groups, showed a drastically increased flow time, finally resulting in a gelling of the system.Structure type 2, which has sterically shielded aminic anchoring groups, had no medium-term influence on the storage stability of the epoxy base component. However, the viscosity did start to increase after a storage period of six weeks.Only structure type 3, which has an encapsulated core, showed no viscosity increase across the entire storage period of twelve weeks. Indeed, it provided the same storage stability as the system that did not contain any wetting and dispersing additives.In the case of the 2-pack polyurethane system, the influence of the structures on the pot-life (period of time until the initial viscosity has doubled after the two components have been mixed) the results were more variable, depending on the curing agent that was used.As shown in Figure 7, none of the additives shortened the pot-life of

Figure 7: influence of the structure types on the pot-life of the 2-pack polyurethane system. Left: OH-functionalised acrylate + aliphatic-aromatic polyisocyanate. right: OH-functionalised acrylate + pure aiphatic polyisocyanate

system reference

Component a

Component B

possible system/ additive interactions

effect on coating properties

1 epoxy poly(amido)amine

interaction with epoxy part

Major effect on storage stability

2 polyolpolyisocy-anate

catalysis of curing shorter pot-life

3 alkyd oxygen complexation of metallic dryers

Longer drying time

4.1 acrylate/ melamine

acid catalyst

partial catalyst deactivation

incomplete curing reaction

4.2 polyester/ melamine

acid catalyst

partial catalyst deactivation

incomplete curing reaction

table 1: possible interactions bet ween the coating system and the aminic dispersing additive

D i s p e r s a n t s1 0


the pure aliphatic polyisocyanate. However, compared to a system with no additives and in the case of the reactive aliphatic-aromatic curing agent, the pot-life was considerably shortened with structure type 1 and at least slightly shortened with structure type 2.Structure type 3 did not shorten the pot life. This different behaviour can be attributed to the exposed aminic functionalities in structure type 1 that catalyse the formation of polyurethane and thus shorten the pot life.

stoving system shows ‘ Differences in harDness

The differing influence of the various structure types on system prop-erties was also confirmed by the mechanical tests on the acid-cat-alysed stoving system (Figure 8). Structure types 2 and 3 showed a damping of the pendulum swing comparable to that of the reference system with no additives.Because of the significant differences in results during testing of the mechanical properties, it is assumed that structure type 1 led to in-complete curing of the system - an assumption based on the interac-tion of the acid catalyst with the freely accessible aminic functionali-ties in this molecular structure.

how the hyperbrancheD core/shell structures function

The selective interaction of the aminic pigment-affinitive core of struc-ture type 3, characterised by efficient adsorption on various pigment

surfaces, and the minimised interactions with reactive coating sys-tems are attributed to the hyperbranched core/shell structure.The prevailing mechanical shear forces in the pigment grinding pro-cess deform the flexible polymer shell of the dispersing additive, which makes the pigment-affinitive core temporarily accessible, allow-ing it to be adsorbed on the pigment surface. After the pigment has been ground, the polymer shell returns to its original state, thus re-shielding the aminic core that was temporarily accessible to the sur-rounding liquid matrix during the grinding procedure (Figure 9). This then prevents any undesirable interactions.Effective pigment stabilisation in solvent-borne and solvent-free re-active coating systems imposes great demands on the dispersing additives used. The aminic pigment-affinitive groups required for sustained adsorption on pigment surfaces can cause undesirable in-teractions with these reactive systems.Innovative dispersing additives based on hyperbranched core/shell structures can thus provide very effective stabilisation of the different pigment types without causing any undesirable interactions with the reactive coating systems. This structure shows superior performance overall to established linear copolymers with freely accessible aminic anchoring groups as well as to polymers with sterically hindered amin-ic anchoring groups.

references

[1] Brock th., groteklaes M., Mischke p., Lehrbuch der Lacktechnologie, 2nd edition, 1998, Vincentz Verlag, Hannover, p 16.

[2] as ref. 1, p 71.[3] Bielemann J., Lackadditive, 1998, Wiley-VcH Verlag gmbH, Weinheim, p

87.

Figure 8: influence of the structure types on pendulum hardness of the acid-catalysed stoving system

Figure 9: schematic representation of the working mechanism of the hyperbranched core/shell structures

BYK-2016-0010_AZ_Paint_RZ260416.indd 1

Water, wind and waves – the rough, irrepressible power of the sea is a real challenge for any surface protection. Whether it‘s a ship or drilling rig, steel or concrete, BYK Additives for Protective and Marine Coatings prove how powerful they are under extreme conditions. They simplify the application of coat-ings, enable the coating layers to be reduced, provide long-lasting corrosion protection, and reduce maintenance intervals. Positive effects to be proud of both economically and ecologically!

www.byk.com

BYK Additives Sea the protection

BYK-2016-0010_AZ_Paint_RZ260416.indd 1 26.04.16 16:11

http://www.byk.com



Sour

ce: W

oszc

zak

- Fot

olia

.com

MaxiMising dispersion stabilityHigh throughput testing optimises additive selection. By Veli Kilpeläinen, alejandro Gutierrez, sander van Loon, VLCi.and sam peel.

optimising the choice and addition rate of a pigment dis-persant can be difficult and time-consuming. a method has been developed to achieve this by using also high throughput screening. this is used to establish the Hansen solubility pa-rameters and optimum addition rates of dispersants. those with solubility parameters closest to that of the pigment can be expected to perform best.

s electing a suitable dispersant for a filler or a pigment (solid parti-cle) is not an easy task, when one has to take into account:

ą The variety and amount of dispersants available. ą The surface modification and/or the nature of the particle, both of

which can vary greatly.An effective dispersant performs the role of a protective buffer – with one portion specifically attracted to the particle (e.g. a hydrophobic tail adher-ing to the surface of the particle) and another portion specifically attracted to the solvent medium (e.g. a hydrophilic head for water compatibility).To match the dispersant with the particle in order to develop a stable dispersion in the solvent, much needs to be known about the nature of the dispersant, the particle and the solvent.

Figure 1: Over view of the Ht system

1 3D i s p e r s a n t s


Results at a glance

ű The selection of the optimum dispersant for a pigment or extender and the optimum addition rate can be a difficult and time-consuming task.

ű A composite method has been developed which provides an efficient route to achieving this; HSP, ODC and highthrough-put formulation.

ű The optimized process consists of first establishing the Han-sen Solubility Parameters (HSP) of a particle, select dispersants based on HSP and then obtain the optimum dispersant con-centration.

ű Those dispersants whose solubility parameters are closest to that of the pigment can be expected to give the best perfor-mance.

ű This procedure was applied to find the best dispersants for a hydrophobic talc in waterborne systems.

solubility paraMeters provide tHe key to analysis

The information needed can be obtained by using the Hansen Solubil-ity Parameter (HSP) of the products, with sample preparation via high throughput screening. Hansen solubility parameters (HSPs) are used to define the attractive forces Hydrogen bonding, Polar forces and Dispersion forces[1] within solvents or solvent blends, or to map the interactions of those with other materials.By the principle of like-seeks-like, solvents and solutes with similar solubil-ity parameters are more likely to form stable solutions than those with

significantly different parameters.The net difference between the parameters of two materials can be quantified using the HSP distance, which is given by the following equation, based on the three parameters mentioned above:

Dist = √((4(δD1–δD2)2)+(δP1–δP2)

2+(δH1–δH2)2) Equation 1

Thus, the smaller the HSP distance between two materials, the more likely it is that the two will be favourably associated. Ultimately it is possible to determine a solubility sphere or solubility radius: if this is determined for a resin, then a solvent (or mixture) whose parameters fall within this radius will dissolve it, while those with further distant will not.Initially this concept was applied to determine the most suitable solvent or blend of solvents to dissolve polymers; but subsequently it has been developed to allow the calculation of ‘solubility parameters’ for insoluble materials such as pigments and substrates. Applying this concept to in-teractions between dispersants and particles, once the HSPs of these match, a stable dispersion can be formed.In this work, the HSP method was verified by the Optimal Dispersant Content (ODC) method [2], a classical method to determine the right dispersant in a trial-and-error manner. These methods will be de-scribed and compared in this article to provide the selection of the best water-based dispersant to disperse the hydrophobic talc prod-uct “Finntalc M15” in a water-based coating.

HigH tHrougHput forMulation siMplifies selection

“Finntalc M15” and a broad selection of dispersants with different chem-istries were studied in relation to HSP and ODC. In order to speed up the process and make it more precise, samples were produced via the “For-max” High Throughput (HT) equipment from Chemspeed technologies.A sedimentation time method in different solvents with different HSPs was used to establish the HSPs of the talc product. Fast sedimentation means no compatibility with the solvents and slower sedimentation means better to good solvents. Good solvents are rated as 1, bad solvents as 6 and the rest are in between. Based on this rating, a dis-tinction can be made in the HSP space.The HSP can then be calculated via the dedicated “HSPiP” software (version 4.0.05, www.hansensolubility.com), by introducing this rating. All ratings are relative for each of the products.For the surfactants, the normal HSP method was applied, meaning

Figure 2: ODC analysis for the talc Figure 3: Hsp data for the talc



solvents were rated for good and bad solubility, again with rankings between 1 and 6. Samples of the products with the different solvents are made via the HT system, as it requires quite a large amount of samples for each product.

optiMising tHe dispersant concentration

An effective dispersant for waterbased applications performs the role of a protective buffer – with its hydrophobic ‘tail’ adhering to the surface of the particle and its hydrophilic ‘head’ associating favourably with the solvent medium.In addition to ionic repulsions between the particles, the dispersant-rich region that forms whenever two particles happen to come close together is subject to osmotic pressure from the surrounding solvent, which tends to force the particles apart again.As a dispersant is introduced into a pigment dispersion, there is a dra-matic reduction in the viscosity of the paste or slurry, as the particles become more mobile in the solvent. The extent of the viscosity reduc-tion is dependent on the amount of dispersant added.However, the amount must often be minimised in order to reduce costs or to prevent other harmful effects which might arise from us-ing high proportions. If the amount of dispersant used is too small, the full benefits will not be obtained. Clearly, for each system studied there is an optimum dispersant concentration (ODC).Many factors affect the ODC, including the chemistries of the dis-persant, solvent and pigment, and especially the surface area of the pigment to be tested. The ODC is the concentration of dispersant at which a plateau is reached in the pigment/solvent viscosity curve and this can be tested in order to adjust the final formulation. This experi-ment can be performed easily with HT screening on a wide range of pigments and dispersants.

principles of HigH tHrougHput screening

High Throughput (HT) is a valuable tool to automatically prepare for-mulations in parallel with high precision. In order to be flexible in various formulation preparations as in this project, an HT system was used that can add liquids, viscous and solid materials on a weight ba-

sis whilst processing, which is a unique feature.On the HT platform, different processing methods can be operated, including mixing by horizontal shaking of glass tubes in a rack and dispersing with a Cowles dissolver disc in a temperature-controlled reactor.In this HT system, the raw materials are thus brought to the glass tubes and reactors, in order to not disturb the processing, as re-quired to make proper formulations. For this project, the HT system prepared the dispersions of talc in water via the HT reactors and the HSP and ODC samples in glass tubes present in the HT stirring rack. Figure 1 provides an overview of the HT system used.

How optiMuM dispersant concentration is deterMined

In a first step, the ODC analysis on the talc paste was performed with respect to a selection of waterbased dispersants. The method for each dispersant was to prepare tubes containing 30 % of talc in dis-tilled water, which was then thoroughly mixed to a paste. The viscosity of the paste was then recorded.

Figure 4: the concept of Hsp distance including radii Figure 5: Hsp plot for talc and dispersants

Dispersant performance ODC (wt%)Viscosity reduction

(%)

“BYK 154” -- 5.0 % 26 %

“Disperbyk 190” + 7.0 % 70 %

“schwego Fluor 6536” -- 9.0 % 30 %

“efka 4580” ++ 5.0 % 85 %

“anti-terra 250” ++ 4.5 % 87 %

“BYK 348” ++ 1.5 % 83 %

“silco sperse HLD-5” + 4.5 % 75 %

“Lopon 895” - 9.9 % 42 %

table 1: summary of ODC results obtained

1 5D i s p e r s a n t s


After the viscosity had been measured, a small amount of a dispersant was weighed into the tube, which was then thoroughly mixed. Again the viscosity was measured and this process was repeated over and over until the viscosity values had plateaued, and then a little further.The viscosity of the paste with every dispersant and its concentration can be seen in Figure 2. The results of this ODC study are shown in Table 1. The profiles of the dispersants showed some interesting characteristics: the best of the dispersants displayed clear and large reductions in viscosity that remained even at low dispersant concentrations.These ‘step-changes’ in viscosity varied in their slopes and in the mass-efficiency of each dispersant, but they all showed approximately the same relative viscosity changes of around 80 % reduction relative to the origi-nal talc-water paste. The dispersants which showed this behaviour most clearly were “BYK 348” “Anti-terra 250” and “EFKA 4580”. Besides these, “Silco Sperse HLD-5” can also be considered good, but slightly less so than the others mentioned above.

sediMentation tests to Measure Hsp values

Because talc does not show solubility but dispersion characteristics in a solvent, resulting in sedimentation, the sedimentation time of talc in the different solvents with known HSPs was measured. The rating of each solvent was entered in the software, resulting in the HSP data and solubility sphere for the talc as shown in Figure 3.In order to match the HSP of the talc with a dispersant, the HSP of the dispersants was determined. The conventional HSP tests can also be extended to calculating the parameters of surfactants and disper-sants. In this case, the solubility in the different solvents is rated.Table 2 shows the experimental HSP parameters for the dispersants tested previously in the ODC studies, including the distance from the talc calculated in accordance with Equation 1. Although the HSP dis-tance between two materials is useful in itself, more understanding can be gained when the radii of the solubility spheres are also em-ployed in the analysis.Figure 4 describes how the HSP distance between the centres of two spheres is related to the radii of those two solubility spheres and the minimum distance that separates their surfaces, taking into ac-count the values obtained in Equation 1. In Table 2 these separation distances are also given (Distsep), where of course the lower the value the better.This data can be depicted nicely by plotting it on a 3D grid, along with the positions and radii of the sphere for the talc, as in Figure 5.Based on the HSPs and distance data, it can be seen that the dis-persants found to have the best performance from the ODC study are indeed inside the HSP sphere of the talc and the poor ones are

far away. In this case, “Silco Sperse HLD-5” is very far away from the HSP of talc, but is much closer to that of water than the other disper-sants. This might be one reason why this product is a good dispersant for water-based systems and therefore shows a good result from the ODC study.

Hsp and odc results coMbined

As shown, both the HSP and ODC study result in the same selection of good and bad dispersants for the talc. As a couple of dispersant HSPs are now known, next time it is a matter of measuring the HSP of a particle and plotting the dispersants in HSP space to see if there is a match or not.The best dispersants can then be selected and tested via ODC in or-der to obtain the right amount of the dispersant for the particle. HSP can reveal the best matching dispersant, so this one should have in principle the best interaction, thus the lowest amount required with the highest viscosity reduction, but the ODC method is needed to ob-tain the data.The strong point of using this HSP method in the first place is that it requires less effort than performing a full dispersant study for a particle via ODC.

an effective selection procedure in suMMary

Table 3 summarises the data for Finntalc M15, obtained from ODC test and HSP experiments, with the dispersants having the best perfor-mance shown first. The data collected for the ODC and the HSP study both show the same best-suited dispersants.In terms of the minimisation of their addition levels and the efficiency of viscosity reduction obtained from ODC, low HSP distances from Finntalc M15 are also exhibited.This leads to a useful method to select the right type and amount of dispersant to use with a solid particle: ą Use the HSP method to obtain the HSP of the particle and then

select dispersants with matching HSPs. ą Use the ODC study with these best dispersants to obtain the optimal

amount of dispersant for the particle.

references

[1] abbott s., Hansen c., Yamamoto H., Hansen solubility parameters in practice – software, eBook and datasets, www.hansen-solubility.com

[2] Bieleman J., in additives for coatings, 2000, p 65, Wiley-VcH.

Dispersant dD dp dHw radius talc distance

Distsep

“Byk 348” 18.1 11.48 11.14 15.6 3.65 -17.25

“BYK 154” 22.8 19.83 32.81 18.6 25.79 3.19

“Disperbyk 190” 12.16 17.04 12.99 14.9 13.70 -6.60

“efka 4580” 19.6 13.26 10.73 18 6.56 -6.84

“silco sperse HLD-5” 17.07 20.35 37.12 11 5.10 -3.10

“anti-terra 250” 20.36 13.4 25.86 19.35 16.83 -7.37

“Lopon 895” 23.08 13.23 35.62 17.75 25.59 4.49

table 2: Hsp data for a selection of the dispersants testedDispersant talc

distanceparticle-disper-sant separation

ODC (wt%)

Viscosity reduction (%)

“anti-terra 250” 4.25 -7.37 4.5 % 87 %

“efka 4580” 6.77 -6.84 5.0 % 85 %

“BYK 348” 3.65 -17.25 1.5 % 83 %

“silco sperse HLD-5” 29.37 -3.10 4.5 % 75 %

“Disperbyk 190” 13.77 -6.60 7.0 % 70 %

“BYK 154” 17.11 3.19 5.0 % 26 %

“Lopon 895” 15.92 4.49 9.9 % 42 %

“schwego Fluor 6536” - 9.0 % 30 %

table 3: summary of results with best dispersants shown at the top of the listing

C o r r o s i o n p r o t e C t i o n1 6


Sour

ce: A

nyka

- Fo

tolia

.com

Maintaining protectionHow dispersants affect corrosion resistance of water-borne paints. By Martin Muth, Andreas Freytag, Max Conrad, Byk-Chemie.

By utilising different test methods the impact of several dif-ferent chemical types of wetting and dispersing agents on the sensitivity of a waterborne primer towards water and mois-ture as well as anticorrosive properties has been evaluated. clear performance differences depending on the chemistry utilised have been found. However, the results obtained with the five different evaluation methods did not correlate too well. this made it evident that the test procedure has to be selected according to the final requirements that have to be improved.

i n the protective and marine coatings market, great efforts are be-ing made to reduce the amount of volatile organic compounds

(VOCs) emitted during application. Many paint customers have been switching to high-solids and solvent-free paints that combine low VOC content with the advantages of conventional paint systems.Another option to reduce VOCs is the use of waterborne paints. Be-cause of the demanding requirements and often uncontrolled applica-tion conditions, the use of waterborne systems in the protective paint market is still very limited. Indeed, in this market sector, it is estimated that globally the share of waterborne paints is around 6 %, while sol-ventborne, high solids and solvent-free systems retain 94 % [1].One key reason for this limited use is that waterborne systems restrict the temperature and humidity conditions under which the paint can be applied. Low temperature and high humidity prevent proper and sufficient curing of the coating. Additionally, the high surface tension of water can cause problems with certain substrates or impurities and lead to improper wetting, cratering and other paint defects.Next to the application challenges, the formulation of waterborne anticorrosive paints is also quite challenging. To ensure proper film formation after application, paint components must be carefully se-lected. To overcome some limitations, the use of additives is essential.

tHe iMpact of wetting and dispersing additives

Wetting and dispersing additives are an important class of additives. They provide three main features that are vital to produce coatings with good storage stability, application properties and optical appearance: ą Wetting of the pigment and filler particles; ą Dispersion of pigments and fillers; ą Prevention of the re-agglomeration of the primary particles.

These aspects are almost impossible to achieve without the aid of ad-ditives due to the high surface tension of water.Since the additive remains in the coating after drying, it may influence the overall hydrophilicity of the coating film and therefore affect cer-tain coating characteristics such as early water resistance and corro-sion protection. To study this effect on the final coating, relevant test methods must be compared.

Mode of action and general structure of dispersants

The process of the dispersion of solid particles as pigments and fillers falls into three stages: wetting of the particles, dispersion and stabili-sation. To understand the first process it is necessary to understand the circumstances under which a liquid properly wets a solid. This process is described by the Young equation:

γs – γsl = cos θ [1] γl

Where

γs = surface tension of solidγsl = interface tension solid/liquidγl = surface tension of liquidθ = contact angle liquid/solid

For complete wetting of a solid by a liquid, the contact angle becomes

1 7C o r r o s i o n p r o t e C t i o n


Results at a glance

ű Waterborne paints hold only a very small share of the mar-ket for protective and marine coatings, as they have stringent requirements for good surface preparation and drying condi-tions.

ű Wetting and dispersing agents play a critical role in the for-mulation of waterborne coatings in particular. However, they are necessarily partly hydrophilic, and this can impair the pro-tective properties of the cured film.

ű Tests are reported on the impact of several different chemi-cal types of dispersing agent on the anticorrosive properties of a styrene-acrylic based primer.

ű The best results were given by two polyacrylates and a polyurethane-based additive. However, a notable result is that there was a poor correlation between the five different tests for protective performance that were used. Thus, it is important to ensure that the tests selected are appropriate for the planned end-use.

0 °, which is the case when:γl = γs – γsl [2]

To achieve proper wetting, the surface tension of the liquid must be less than the surface tension of the solid [2]. Wetting and dispersing additives directly influence the wetting process by adsorbing on the surface of the solids and thus changing the surface tension.

How tHe additives function during dispersion

In the dispersing process the agglomerates of pigments and fillers are broken down into so-called ‘primary particles’. This process is carried

Figure 1: sample of DsC result to determine water uptake and melting peak of water

out by introducing energy to the system, for example by high-speed dispersion with an agitator disk. The amount of energy needed to separate the particles from each other and so increase the surface area that needs to be wetted by the surrounding liquid is described by the following equation:

dW = γ· dA [3]

W: introduced workγ: interfacial tensionA: interfacial surface Thus, the amount of work needed to generate a certain amount of surface area by breaking down the agglomerates into primary parti-cles is directly linked to the specific surface tension. Wetting and dis-persing additives reduce the work required by lowering the interfacial tension between liquid and solid.Stabilisation is then vital to prevent re-agglomeration of the dispersed particles. Two major stabilising mechanisms are utilised: electrostatic and steric stabilisation, which may be combined. In summary, it can be said that wetting and dispersing additives always consist of two domains, one of which interacts with the pigment surface (‘pigment affinitive group’) and one which interacts with the binder (‘binder af-finitive group’).The main currently available wetting and dispersing additives with these characteristics are based on the following chemistries: ą fatty acids ą polyurethanes ą phosphoric acid esters ą polyacrylates

Different additives based on these chemistries were included in the evaluations below.

How dispersants affect waterBorne coatings

For the dispersant to be compatible with the system and able to inter-act adequately with the liquid matrix, the binder compatible groups need to have hydrophilic properties. These properties can be ob-tained by using polar chemical domains, such as polyethers or polyes-ters, or certain hydrophilic co-monomers.Various wetting and dispersing additives use ionic salt structures to obtain sufficient solubility of the additive or its binder-affinitive do-mains in the paint matrix. Alternatively, salt structures may be in-

Water

a

d

B

Figure 2: experimental setup for water diffusion test



troduced, for example by neutralising acidic groups with bases. The chemical nature and volatility of the base also both influence the re-sistance of the final coating to corrosion stimulants and water.The effect of the additives on parameters such as early water resist-ance, water uptake and corrosion resistance must be evaluated in the search for an optimum wetting and dispersing additive.

evaluation of wetting and dispersing additives

To evaluate the influences of wetting and dispersing additives, a standard primer formulation based on a styrene-acrylic dispersion was used. Five test procedures were chosen to compare the effects of the products.To determine water uptake, the paint was applied at a specific film thickness to a polyethylene (PE) foil, dried for a fixed time interval, carefully removed from the foil and weighed. Next it was immersed in distilled water. At fixed time intervals, the paint film was removed from the water, and water droplets were removed with a cloth. The weight gain was then established and the water uptake was determined as a percentage of the original dry film weight.To check this method for reliability, some samples were evaluated by differential scanning calorimetry (DSC). After immersion in distilled water for 24 hours, a sample was cooled down to -60 °C and heated up to +80 °C. The melting peak of water and its melting enthalpy were used to calculate the amount of water inside the coating and this was compared to the values generated by the gravimetric method (see Figure 1).To measure water diffusion, a free coating film was produced by the PE foil application method, a small plastic cup was attached to its surface then filled with distilled water (see Figure 2). At certain time intervals, the IR absorption was evaluated at wavenumber 3400 cm-1 by means of an ATR (Attenuated Total Reflectance) measurement and compared to the absorption of a paint film without water contact.The salt spray test was carried out according to DIN EN ISO 9227. The test paint was applied to cold-rolled steel panels which were exposed in the salt spray chamber after drying. The panels were then checked for the amount of blistering according to DIN EN ISO 4628-2 and for delamination/rust creep according to DIN EN ISO 4628-8.Free films of the test paint were also fixed in a permeability cup. The test was carried out according to DIN EN ISO 7783. At certain time intervals, the weight lost by diffusion of water vapour through the coating was determined.

Figure 3: Water uptake after 6 hours of water immersion (gravi-metr ic and DsC test)

Figure 4: Diffusion coefficients after 24 hours of water exposure,

To determine early water resistance, the test paint was applied to cold-rolled steel panels. Four and 24 hours after application, a cot-ton wool ball drenched with distilled water was placed on the coating then covered with a small plastic beaker to prevent the water from evaporating. After 24 hours, a cross-cut test was carried out according to DIN EN ISO 2409.

forMulation, application and cHoice of additives

Several common pigments and fillers were incorporated to obtain a pigment volume concentration (PVC) of 28 %. The dry film thickness of the coating was fixed at 80 µm on cold-rolled steel panels and on PE foil to produce the free films. Drying conditions were set at seven days, at 23 °C ± 2 °C and a relative humidity of 50 % ± 5 %.As regards the choice of additives, the aim was to represent a large variety of different chemistries and variations. At least one represent-ative from the groups of phosphoric acid esters, fatty acids, acrylates, polyurethanes, and controlled-polymerisation acrylates was picked initially. The chosen additives were used at an amount of 1.5 % active substance calculated on pigments and fillers.

results in suMMary

In the early water resistance test, the best results in Table 1 are shown by two acrylate-based additives, AC-4 and AC-5, and the PU-based additive, PU-1. Their good performance can be explained by their hy-drophobic nature, although AC-5 contains a considerable additional number of hydroxyl groups. The very non-polar domains/co-mono-mers in the additive molecules make it far more difficult for water to penetrate the coating, reach the interface between coating and panel and lead to adhesion loss.In the water uptake test (see Figure 3) the parallel measurement of the water content by DSC in four samples shows that the values from both methods are the same within their variations. The tendency to absorb water can be correlated quite well with the hydrophilicity of the additives used.The reference sample without additive shows the lowest water ab-sorption, while the most polar additives AC-1, AC-3 and AC-5 (with a high content of hydroxyl groups) show the highest water uptake. The other additives range between the values of the control and AC-3, except for additive AC-4.After 24 hours of water exposure in the diffusion test, it can be seen



that AC-1, AC-2 and AC-3 show the highest diffusion and that almost all additives show a higher diffusion rate than the control, except for AC-4 and PU-1 (Figure 4). While the lower value of AC-4 can be ex-plained by its hydrophobic nature, the good performance of PU-1 is surprising because of the high number of hydroxyl groups it contains. Further investigations are needed to explain this effect, but it can be said in general that a more hydrophobic additive leads to better dif-fusion behaviour.Once again, the acrylate-based additives AC-1 and AC-2 show poor results in the permeability test (Table 1). AC-3 shows a medium level of performance, while all other additives exhibit values comparable to the control. It is hard to explain the poor result of additive AC-3. However, many factors could have increased the diffusion of water

Code neutra-lisation Water uptakeir

diffusionpermea-

bility

early water

res.

salt spray test

control

Fs-1 amine 1

ps-1 amine 2

pu-1 amine 1

ac-1 amine 3

ac-2 Base 1

ac-3 Base 2

ac-4 amine 3

ac-5 Base 2

Figure 5: over view of additives and their performance in differ-ent tests

vapour: the additive itself, poor film formation or pinholes generated by foam bubbles.Finally, in the salt spray test (see Table 1) the acrylates AC-1 and AC-2 again gave the worst results, and were removed from the test after 140 hours due to severe blistering and delamination at the scribe. The products FS-1, AC-4 and PS-1 exhibited medium performance and were removed after 384 or 500 hours. The best performance was shown by PU-1, AC-3 and AC-5, which endured the test for 700 hours.

test results sHow soMe poor correlations

Summarising all the results in one overview with a view to drawing conclusions is rather difficult, as the results of the different methods do not necessarily correlate with each other. This can be seen in Fig-ure 5.An initial observation is that only three samples show the same ten-dency in all of the tests: the control, AC-1 and AC-2. The control shows very good results in all aspects, and it could be asked why a wetting and dispersing additive should be used at all.The reason is quite obvious, as outlined in the introduction: the addi-tive is needed to ensure good storage stability, application properties and optical appearance. In fact, the control paint without any additive was barely producible due to its extreme high viscosity and bad wet-ting. It was impossible to apply it by spray, and its storage stability was very poor.This means that, although efficiency might be lost in the known pa-rameters, a wetting and dispersing additive is necessary to be able to produce a stable and usable paint. The chemistry utilised for these types of additives has to be carefully selected to obtain additives with a minimal impact on paint film properties.Additionally it can be said that the different test methods do not nec-essarily correlate with each other. This is a major finding, as it makes it very evident that, for example, a bad result in water uptake and diffusion does not necessarily indicate a bad result in the corrosion test (see additive AC-3), while, on the other hand, a good result in all other parameters can still mean a medium performance in the salt spray test (see additive AC-4). It is therefore important to carefully select the evaluation methods with regard to the subsequent use of the coating and the parameter of greatest importance.

test MetHods Must Be MatcHed to end uses

This work has shown that several testing procedures must be used to evaluate the effect of additives and to compare them with each other as there is no clear correlation, for example, between water uptake and corrosion resistance.Several structural elements that affect the overall performance of the additives could be identified. These structural parameters are range from the basic “backbone” elements via specific co-polymers to the bases used for neutralisation.Although further work on these issues is required, it can be said that by deliberate variation of structural elements, additives can be de-veloped which have only a limited impact on the critical parameters specified or which in future will even be able to improve the perfor-mance of waterborne coatings.

references

[1] Marktstudie, european protective coatings Markets, M39e-39.[2] Heilen W., additive für wässrige Lacksysteme, Vincentz, 2009.

name Chemistry

early water resistance:

cross-cut test after water exposure

permeability test (24 hours)

salt spray test (hours)

After 4 hr. drying

[Gt]

After 24 hr. drying

[Gt]r2

V [g/(m2/d)

control 1 1 0.9994 26.3 600

Fs-1 Fatty acid 5 1 0.9991 26.4 384

ps-1 phos. acid ester 2 2 -- -- 500

pu-1 polyurethane 0 0 0.9984 24.0 700

ac-1 polyacrylate 5 5 0.9949 46.2 140

ac-2 polyacrylate 5 5 0.9995 38.0 140

ac-3 polyacrylate 2 0.9979 29.1 700



table 1: summary of some results in three tests

2 0


C o r r o s i o n p r o t e C t i o n

Clau

de B

eaub

ien

- Fot

olia

.com

Tailored To endurenew dispersing additives help to protect aqueous coatings against corrosion. By philippe Favresse, Frank Kleinsteinberg, Katr in roland and patr ick Glöckner, evonik industr ies.

Graft copolymer pigment dispersants were produced by free radical and controlled radical polymerisation and their per-formance was evaluated in two pigment concentrates for waterborne coatings. Both production technologies produced similar performance, but by modifying the composition of the additive the corrosion resistance of the coatings could be en-hanced.

W etting and dispersing additives are indispensable for the pro-duction of high quality pigment concentrates with high colour

strength and adequate storage stability. However, especially with aqueous coatings, their hydrophilic structure can cause the resistance of the resulting films to deteriorate.The aim of this study was thus to develop polymeric wetting and dis-persing additives for aqueous pigment concentrates which display no significant adverse impact on the corrosion protection of aqueous coatings. The polymers were produced using various polymerisation processes and were comparatively evaluated.

WeTTinG and dispersinG addiTives enaBle a producTive dispersion process

During the dispersion process, pigment agglomerates are ground to maximise the colouring capacity of what are often expensive pig-ments and to meet optical requirements such as colour value, colour strength, chroma and transparency. Wetting and dispersing additives wet the surface of the pigment and prevent reagglomeration of the pigment particles once dispersed.In high-quality aqueous coatings, polymeric wetting and dispersing ad-ditives are used to anchor themselves to the surface of the pigment. Anchoring is accomplished by means of functional group attraction. Because of their macromolecular structure and electrostatic repul-sion, they ensure adequate stabilisation of the coating matrix [1, 2 ].Compared to low molecular weight products, coating films with poly-meric wetting and dispersing additives generally exhibit higher me-chanical and chemical resistance properties. Nevertheless, even with polymeric additives there is still an adverse effect on the water and

Figure 1: polymers with the same composition but with different constituents: (a) statistical copolymer (b) AB block copolymer

Figure 2: schematic representation of a statistical graft copolymer



Results at a glance

ű Graft copolymer pigment dispersants were produced by free radical and controlled radical polymerisation.

ű Their capabilities as wetting and dispersing additives for aqueous pigment concentrates and industrial coatings were investigated in terms of rheological, colouristic and corrosion resistance properties.

ű It was shown that both polymerisation technologies pro-duce a similar performance profile. However, the composition of the polymeric additive does affect its performance.

ű The best of these wetting and dispersing additives satis-fies the requirements for rheological and colouristic properties while providing a high degree of corrosion resistance.

corrosion resistance of aqueous coatings. This effect can be explained in part by the high polarity of the anchoring groups and is especially noticeable when part of the polymer is in the coating matrix and not fixed to the surface of the pigment.

HoW dispersanT sTrucTure can affecT performance

The efficiency of the polymeric dispersing additives is highly depend-ent on molecular weight, the choice of monomers and the arrange-ment of the monomer units within the polymer [1-3]. Dispersing ad-ditives attach themselves to the pigment surface especially well when the anchoring groups are within close proximity to one another. Fig-ure 1 shows polymers with the same composition but with different constituents: with a purely statistical arrangement (a) the monomers are distributed randomly throughout the polymer while in the block

copolymers (b) there is a distinct accumulation of monomer units in one part of the polymer. Block copolymers made from ‘anchoring’ and ‘stabilising’ monomers can therefore form highly effective wetting and dispersing additives.Graft copolymers are a special case whose anchoring groups are ei-ther arranged in the side chains of the comb-like structures or directly on the polymer backbones (Figure 2). These structures naturally lead to a particularly high number of contacts on pigment surfaces and are consequently regarded as highly effective wetting and dispersing ad-ditives. Figure 3 shows that graft copolymers can have a higher num-ber of anchoring groups and stabilising elements per unit of volume than pure AB block copolymers.

dispersanTs may Be made By several differenT rouTes

There are a variety of known processes to produce polymers with a well-defined structure. These include polyaddition as well as ionic and radical polymerisation methods [4]. With regard to the radical meth-ods, free radical polymerisation (FRP) of macromonomers to produce graft copolymers is just as well established as ‘living’ controlled radical polymerisation (CRP). This – like all living polymerisation methods – allows the number of blocks, length of the blocks, and block length distribution to be adjusted easily in the synthesis of block copolymers [5, 6].Each of the methods available has advantages and disadvantages that can be of a technological or commercial nature. For example, living controlled radical polymerisation (CRP) allows many different polymer

Figure 3: schematic representation of a pigment surface with (a) a statistical graft copolymer and (b) AB block copolymer disper-sant bonded to the surface

raw materials pr 101 pB 15:4

Deionised water 17.4 33.9

Dispersing additive 21.0 25.0

tego Foamex 810 1.0 1.0

pigment 60.0 40.0

aerosil 200 0.5 -

Biocide 0.1 0.1

total [g] 100.0 100.0

additive solids on pigment

14 % 25 %

table 1: Formulations of the pigment concentrates

Dispersant/ pigment

Ms Crp Frp1 Frp2

With pr 101

colour strength (F) 476 485 485 500

Haze 50 30 35 20

gloss 20 ° 79 81 83 83

With pB 15:4

colour strength (F) 710 740 740 745

Haze 90 85 85 85

gloss 20 ° 89 90 89 90

table 2: Colouristic properties of the 2K pU coatings based on pigment con-centrates pr 101 and pB 15:4 with different dispersing additives in full tone



architectures to be created, but processing these polymers is often time-consuming. One process that exhibits this drawback is the sepa-ration of copper-based catalysts or alkyl halides, which must be car-ried out to prevent discolouration and avoid impairing the corrosion resistance of coating films.In comparison, free radical copolymerisation (FRP) of (meth)acrylates often leads to a statistical distribution as the selectivity in attachment depends largely on the copolymerisation parameters of the respec-tive monomers towards each other [7]. The advantages of this meth-od are minimal requirements with regard to the purity of the starting materials and ease of handling on a large technical scale.

properTies of piGmenTs and addiTives sTudied

For an extensive assessment of different production processes, wet-ting and dispersing additives for aqueous coating applications were investigated. Comb-like graft copolymers produced by FRP were com-pared with those that were produced by CRP. The comparison addi-tive was a conventional copolymeric dispersing additive that can be regarded as a market standard for aqueous industrial applications.The aim of the study was to provide a wetting and dispersing additive with the best possible efficiency in terms of colorimetric properties and corrosion resistance behaviour.Four different copolymer dispersants were compared in this study: ą MS Market standard: conventional copolymeric dispersing ad-ditive produced by free radical polymerisation (statistical distribu-tion of the comonomers);

ą CRP Graft copolymer dispersing additive produced by con-trolled radical polymerisation (composition identical to FRP1 but with an alternating structure);

ą FRP1 Graft copolymer dispersing additive produced by free radi-cal polymerisation (statistical distribution of the comonomers);

ą FRP2 Graft copolymer dispersing additive produced by free radi-cal polymerisation (optimised composition).

Two pigments were studied for this evaluation: a PR 101 red iron ox-ide, and a PB 15:4 phthalocyanine blue. The formulations of the pig-ment concentrates are shown in Table 1. Pigments were milled in a Skandex shaker with 200 g glass grinding media (ø 2.5 - 2.8 mm) for a dispersing time of 1 hour (PR 101) or 2 hours (PB 15:4).

TesT coaTinGs and TesT meTHods

In this study, colorimetric and salt spray results are reported on a room temperature curing, aqueous, 2K industrial coating to which the corresponding pigment concentrate was added (PR 101: 8 %, PB 15:4: 4 %). The results of the investigation were confirmed in other formulations, such as an aqueous polyester-melamine stoving enamel system. Because the results were similar, they will not be discussed here. The test system for the investigation of water absorption was a 1K acrylate-based filler that was mixed with the PR 101 concentrate in a ratio of 7:3.The rheological properties were investigated in accordance with DIN EN ISO 3219 using a “RotoVisco1” rheometer from Thermo Electron (cone/plate type CP 35/2 ° and D = 0 - 1000 s-1). For the spectropho-tometric investigation of the colour properties, the 2K PU test coating was applied to “Leneta” cards at 150 µm wet film thickness and meas-urements were made with an X-Rite “Type SP62” spectrophotometer.The salt spray test was carried out according to DIN EN ISO 9227 (1200 h) on steel plates (2K PU test coatings). The gravimetric investi-gation of water absorption was done in relation to the starting value using a 1K filler on aluminium after the sheets had been stored in a water bath for 24 h at room temperature.

rHeoloGical properTies of THe piGmenT concenTraTes

Figure 4 shows the dynamic viscosities of the different pigment con-centrates before and after storage. The research dispersing additives (CRP, FRP1, FRP2) all produced a very good and stable viscosity profile. There is no significant difference between the measurement results on these different research products. In a trend, the viscosity-reduc-ing effect of the optimised wetting and dispersing additive (FRP2) is more pronounced.The viscosity of the blue pigment concentrate based on the market standard (MS) was slightly lower after storage. This can be explained by a wetting of the pigment subsequent to dispersion and therefore suggests slower wetting. In the case of the PR 101 concentrate with MS, the viscosity increase after storage can be regarded as undesirable.

colour sTrenGTH is enHanced

The research products tested largely produced similar colour proper-ties as demonstrated in Table 2. FRP2 tends to produce higher colour strength with better compatibility (less haze). The comparison with the market standard (MS) shows that with the organic blue pigment the research products (CRP, FRP1, FRP2) provide slightly higher colour strengths. With the inorganic red pigment, they show improved com-patibility versus the market standard and demonstrate significantly higher colour strength (by up to 5 %).It would appear that with aqueous graft copolymer dispersing additives, the distribution of the monomers throughout the polymer is less critical (comparison between CRP and FRP1). On the other hand, the composi-tion is crucial, as the comparison between FRP1 and FRP2 indicates.

Figure 4: Dynamic viscosity (D = 100 s-1 mpas) of the pigment concentrates with various dispersing additives based on pig-ments pr 101 (left) and pB 15:4 (r ight) before and after storage at 50 °C for 7 days



impacT of sTrucTure and composiTion summarised

In this study, wetting and dispersing additives for aqueous pigment concentrates produced by controlled and free radical polymerisation methods were compared. It was shown that the production method does not have much effect on the colouration or rheological proper-ties of aqueous pigment concentrates. Also, no significant differences with regard to corrosion or water resistance were found due to the varying of production methods.Therefore, whether the monomers are alternating (CRP) or statistically (FRP1) distributed throughout the graft copolymer does not appear to affect the properties evaluated. Compared to the market stand-ard, the wetting and dispersing additives evaluated have a higher per-formance spectrum. This performance spectrum could be increased even further by optimising the composition as in the test product FRP2 [9].

references

[1] evonik industries ag, the Big tego, edition 4, essen, germany, 2012.[2] Heilen et al., additives for waterborne coatings, Vincentz net work, Hano-

ver, germany, 2009.[3] Glöckner et al., the star performer, europ. coat. Jnl., no. 05, pp 26-31,

2010.[4] Odian, principles of polymerization, 3rd edition, Wiley interscience, new

York, 1991.[5] Metz N., synthese von komplexen polymerarchitekturen durch die

kontrollierte polymerisation von reaktivester-Monomeren, Dissertation, Mainz, 2009.

[6] Wittmann G., aufbau definierter polymerarchitekturen durch radikalis-che polymerisation unter atomtransfer (atrp), Dissertation, Darmstadt, 2003.

[7] Glöckner, Ritter, Macromol. rapid comm., Vol. 20, pp 602-605, 1999.[8] glöckner, Der ökologische umbau einer produktpalette, in Wässrige

Kunststoffdispersionen und ihre anwendung, 71. FH texte, aachen, pp 9-20, Jan. 2005.

[9] www.tego.de, technical data sheet tego “Dispers 757 W”.

Figure 5: Aqueous 2K pU industrial coating based on pigments pB 15:4 (top) and pr 101 (bottom) after the salt spray test (Din en iso 9227, 1200 h) on steel plates (from the left, using the different dispersing addi-tives: Ms, Crp, Frp1, Frp2)

Figure 6: Water absorption (%) of a 1K acrylate filler with the pr 101 pigment concentrate based on different dispersing addi-tives; substrate: aluminium, after being stored in water for 24 h at room temperature

resisTance To salT spray and WaTer BoTH improved

Figure 5 shows multiple aqueous 2K PU industrial coatings after the salt spray test. With both the organic phthalocyanine blue and the inorganic red iron oxide, the coatings with the standard dispersing additive (MS) exhibit the highest level of damage. Blister formation and corrosion creep can be seen, suggesting inadequate protection.The coatings containing two dispersing additives of the same compo-sition produced via controlled (CRP) and free radical polymerisation (FRP1) have the same degree of resistance. In particular, blister for-mation with these two copolymer forms is slightly better than with the market standard. The coatings with the optimised dispersing additive (FRP2) produced outstanding results. The coating remained largely unaffected throughout the entire salt spray test.The investigation of the water absorption confirms the results of the salt spray test (Figure 6). The market standard dispersing additive re-sults in the most water absorption. The coatings containing the two dispersing additives of the same composition but produced by dif-ferent methods (CRP, FRP1) provide similar results. The coating with the optimised dispersing additive, variant FRP2, has the lowest water absorption.Practically speaking, lower water absorption results in higher mechan-ical resistance during water stress as swelling, which often produces softer and more susceptible films, is minimised. Lastly, changes in appearance, such as blushing, are not observed with the optimised variant, FRP2.

2 4


Wat e r b o r n e c o at i n g sSo

urce

: Tor

sten

Loh

se -

Pixe

lio

Improved pIgment concentratesthe proper dispersion stabilisation of pigments is not only im-portant for the final properties of a coating, it also helps pig-ments to be used as efficiently as possible. peak performance on a broad range of pigments can be achieved using a dual- dispersant approach. two complementary dispersants were each designed to have an optimum performance with a se-lected group of pigments and at the same time to guarantee inter-compatibility.

g ood dispersing agents should provide excellent dispersing power and the stabilisation of different pigment classes. This results in low

mill-base viscosities and high colour strength in the final paints. When pig-ment concentrates are stored and dispensed, low viscosity and long-term stability become particularly important. To keep in line with the trend to-wards more sustainable coatings, dispersing agents must contribute to the fulfilment of the most stringent emission requirements for VOC levels, so that the paints produced can can carry eco-labels.

Figure 1: Different types of pigment stabilisation in aqueous systems

a new dispersant for waterborne coatings provides enhanced staibilisation. by steffen onclin, Harald Frommelius, Paula gómez Perea, elena Martinez and shailesh shah, basF.

2 5Wat e r b o r n e c o at i n g s


Results at a glance

ű A new dispersant, using controlled free-radical polymerisa-tion technology, has been developed for waterborne applica-tion. The new product is an electrosteric dispersant with en-hanced steric character, which provides broad compatibility and improved pigment stabilisation.

ű The new dispersing agent is particularly effective with inor-ganic pigments, including difficult to disperse pigments, such as transparent iron oxides.

ű The new product shows benchmark performance with in-organic pigments, but additionally good application results can be obtained with many organic pigments, giving it an all-round character.

ű The new dispersing agent complements the performance of a current steric stabilising CFRP dispersant which is recom-mended for many demanding organic pigments and achieves a high level of jetness for carbon blacks (5).

The correct design of dispersing agents can go a long way to achieving overall top performance in a broad range of coating systems.The classical way to stabilise inorganic pigments in aqueous systems is through electrostatic stabilisation (see Figure 1). This highly effective mechanism is mainly used in low performance applications, for exam-ple, matt interior and façade paints, because dispersing agents car-rying a high charge density can potentially have a negative influence on resistance properties. On the other hand, in the case of organic pigments, better application results are obtained when stabilisation takes place using steric hindrance and non-ionic interactions. [1] Electrosteric stabilisation, on the other hand, combines both mecha-nisms, steric and electrostatic. Recently, a new type of polymer has been developed that relies on electrosteric stabilisation with en-hanced steric character. These “enhanced electrosteric” dispersing agents are prepared by controlled, free-radical polymerisation tech-nology (CFRP) and are designed to give broader compatibility across a wide range of coating systems.

precIse desIgn

Dispersing agents need to fulfil two main requirements. Firstly, they should have a strong affinity for the pigment surface and secondly, they should provide robust stabilisation against flocculation. These demands can be fulfilled most effectively if the architecture of the polymer is controlled at the design stage.CFRP enables the precise design of polymer structures. Although non-controlled, free-radical polymerisation can lead to good dispers-ing agents, polymers with a highly defined architecture can be more efficient, resulting in improved colorant stability and broader system compatibilities [2], [3], [4]. With CFRP technology, well-defined block copolymers may be prepared that are designed optimally to fit pig-ment and resin chemistry. As illustrated in Figure 2, dispersing agents designed with CFRP consist of two defined blocks that are prepared by sequential polymerisation of monomers or monomer mixtures. Typically, a longer stabiliser block is formed initially. This needs to be compatible with the relevant paint system. The anchoring block contains functional groups which interact strongly with the pigment surface to allow for efficient and stable adsorp-tion. For demanding applications such as organic pigments, the anchoring block normally contains aminic groups, which can be further modified if required.It is preferable to use this new technology to design high-performance dispersing agents, because well-defined structures lead to anchor-ing groups with a higher efficiency. This in turn increases adhesion to the pigment surface, improves flocculation resistance and brings enhanced colour development. Additionally, it allows for precise con-trol of the polymeric backbone to achieve good stabilising properties through the optimum compatibility achieved.

enhanced electrosterIc stabIlIsatIon

A new dispersing agent was developed using CFRP technology, which shows positive results on inorganic pigments. It relies on an enhanced electrosteric stabilisation mechanism. Both the charge density of the stabiliser block and the number of steric components were studied systematically. It was found that higher levels of steric components in the stabiliser block are beneficial to the performance of the final coating.The newly developed dispersant, “Dispex Ultra PX 4575”, (referred to later as D1) complements the performance of, a current steric stabi-lising CFRP dispersant, “Dispex Ultra PX 4585” (referred to later as D2). The two dispersants were tested on a range of pigments using four commercially available products as benchmarks These are re-ferred to as: B1, B2, B3 and B4. B1, B3 and B4 and are styrene maleic anhydride-based copolymers and B2 is an acrylic block copolymer.

Figure 2: structure of dispersing agents designed with cFrP technology

Wat e r b o r n e c o at i n g s2 6


The following data illustrates test results with three different pigment types as examples: Firstly, transparent iron-oxide colorants. Figure 3 shows the mill-base viscosity of yellow transparent-iron-oxide pigment concentrates. D1 shows low viscosity at the lowest dosage level, whereas the benchmark products B2 and B4 only perform well at higher dosage levels. With benchmark product B3 and D2, it was not possible to prepare a paste that could be handled. The pigment concentrates were subsequently let-down in a 2-pack, PUR clear coat and the transparency and gloss levels were measured.Figure 4 shows lowest scattering delta E values, and therefore the best transparency, for D1. This indicates an excellent stabilisation of the finely ground pigment as well as a good compatibility with the let-down resin system.The results mentioned above are further illustrated with an applica-tion photo of a drawdown of two tinted clear coats prepared with the

Figure 3: Viscosity with transparent iron oxide yellow (30 %), dispersing agent dosage: 15 %/20 %/25 % active on pigment

Figure 4: transparency and gloss of a 2-pack PUr clear coat tinted with transparent iron oxide yellow., de* scattering: the lower the value, the higher the transparency

new dispersant and benchmark product B4 (Figure 5).Also, tests on transparent iron oxide red pigments demonstrated that D1 perfomed the best, followed by B1 and B4. The other benchmark products B2, B3 together with CFRP dispersant D2 showed some weaknesses in terms of viscosity and stability of the colorants.The second pigment type is bismuth vanadate, which is the pigment of choice to replace lead-containing inorganic pigments in high-end industrial applications. This material provides good colour saturation and durability, combined with high opacity. Optimally dispersing this pigment is important, because it makes it possible to achieve com-plete hiding at lower film thicknesses.The rheological behaviour of pigment concentrates containing 60 % bismuth vanadate was examined. Figure 6 shows that low-viscosity concentrates with good storage stability could only be achieved with the two CFRP dispersants. All the benchmark products tested were not able to produce pigment concentrates of low viscosity at the test-

Figure 5: comparison of transparencies in transparent iron-oxide yellow using dispersing agents D1 and product b4

Figure 6: Viscosity development – f resh and after storage – of colorants based on bismuth vanadate

2 7Wat e r b o r n e c o at i n g s


Figure 7: colour strength and rub-out in bismuth vanadate dispersions

Figure 8: Viscosity development, f resh and after storage of colorants based on carbon black

ed pigment concentrations. The poor rheological performance of the pigment concentrates con-taining the current industry benchmarks is also reflected in the col-our development in an industrial stoving system (Figure 7). D1 shows good colour development, while the other tested dispersants either could not be tested, because of very viscous pigment concentrates, or showed poor colour strength.Thirdly, carbon-black pigments were also tested to ascertain broad ap-plicability of the new product. Figure 8 shows the viscosities of pigment concentrates containing 35 % of carbon black. Again, the CFRP products provide low-viscosity concentrates that are minimally affected by warm storage.

broad compatIbIlIty

The compatibility and colour strength of the pigment concentrates were tested in a number of architectural and industrial systems. Figure 9 shows the results in an industrial self-crosslinking system. Already at a low dosage level, D1 provides good colour strength in combination with excellent compatibility. Only D2 and one benchmark product provide better colour development, although at higher dis-persant dosages.A range of further pigments and systems were tested to assess the application performance of D1. Such pigments include additional transparent iron oxides, a number of opaque inorganic pigments and several organic pigments. It was generally observed that the peak performance of the new product is with inorganic pigments, provid-ing low viscosities at low dispersant concentrations, combined with a broad compatibility in industrial and architectural paint systems. But also with organic pigments, D1 shows good performance, giving the product an all-round character. Best results with selected organic pig-ments, however, are obtained with D2.

references

[1] Auschra C., Pirrung F., Harbers P. E. L., FarBe unD LacK 2006, 10, 24-30.

[2] van den Haak H. J. W., J. coat. technol. 1997, 69 (873), 139-144.[3] Pirrung F. O. H., Quednau P. H., Auschra C., chimia 2002, 56 (5), 170-

176.[4] Auschra C., Eckstein E., Muhlebach ((hier fehlt das initial des Vornamens,

Fr)) et al., prog. org. coat. 2002, 45 (2-3), 83-93.[5] Onclin S., Martínez E., Huguenard S., asia pacific coatings Journal

2012, Vol. 25, no. 04, 20-22.Figure 9: colour strength and rub-out with carbon black in a self-crosslinking acrylic paint

2 8


H i g H S o l i d SSo

urce

: Mar

k H

uls

- Fot

olia

.com

HigH solids enHancementsAdditive optimises both colour stability and viscosity in 2K systems. By Kirstin Schulz, Evonik industr ies.

Figure 1: Comparison of wetting and dispersing additives: “Tego dispers 678” generates the lowest viscosities at various shear rates

a new wetting and dispersing additive achieves a stable low viscosity and consistently stable colour in two-pack high sol-ids paints. Raw material inventories can be reduced because it also gives good results in medium solids formulations.

t he use of high solids formulations is a proven way of reducing VOC emissions and meeting the European Directive 2004/42/EU.

The changeover from medium solids to high solids paint formulations is relatively easy for paint manufacturers and for users, as the produc-tion and application equipment are the same. In contrast, the use of water-borne or radiation curing alternatives generally involves major changes in raw materials and often also in equipment and application techniques. The resulting investment can be avoided by the use of high solids coatings. Because of its good cost/performance ratio, high solids technology has now become common.Despite the many similarities between medium solids and high sol-ids technologies, there is a marked difference between them. At ap-plication viscosity, high solids paints obviously possess a significantly higher non-volatile content. This frequently poses difficulties in ap-plication, with achieving high gloss, with drying and with the storage stability of the paints.

2 9H i g H S o l i d S


Results at a glance

ű High solids paints offer a way to meet regulatory require-ments without any need for major changes to raw materials, production or application equipment. Water-borne or radiation curing alternatives often require substantial changes in all areas.

ű However, wetting and dispersing additives used success-fully in medium solids formulations may not provide sufficient viscosity reduction, colour stability or storage stability.

ű A new additive developed specifically for two-pack high sol-ids coatings was tested in a colour-critical mixed pigment and mixed binder system and shown to provide superior viscosity reduction, colour stability in the rub-out test and storage stabil-ity. Performance in basic physical tests was slightly better than the control sample.

ű The new product’s broad compatibility allows it to be used in medium solids formulations, thus helping to limit raw mate-rial inventories.

Key RequiRements foR wetting and dispeRsing additives

All the raw materials used influence the rheology of a coating, but wet-ting and dispersing additives are of the greatest importance [1]. The correct choice of wetting and dispersing additive decisively affects the rheology of high solids paints (Figure 1). Adjusting the working viscosity of high-gloss high solids paints containing organic pigments is particu-larly difficult. The desired colour should not change during storage or

during various methods of application and drying times.Raw materials must be chosen carefully to avoid undesirable phe-nomena such as pigment sedimentation, colour shifts or increased viscosity after storage. The diverse pigment and filler surfaces must be homogeneously wetted and stabilised, particularly in the case of direct (multi-pigment) grinding, which is used when batches exceed a certain customer-specified size.The ideal wetting and dispersing additive reduces the viscosity in the paint and prevents pigment sedimentation during storage. Wetting and dispersing additives which are well established in medium sol-ids technology are still frequently used in high solids formulations. However, they no longer satisfy the increasingly stringent demands of modern high solids technology. Often, the combination of viscosity reduction with sufficient colour stability/pigment stabilisation is inad-equate in high solids paints.

peRfoRmance advantages of tHe new additive

A new wetting and dispersing additive has now been developed specifically to achieve stable, low viscosity and constant colour, par-ticularly in high-quality two-pack high solids coatings. Because it also achieves excellent results in medium solids coatings, the new addi-tive meets the coatings industry’s need for a widely-useable prod-uct that can help to limit the number of raw materials required in manufacturing.The new wetting and dispersing additive has been developed spe-cifically for the direct grinding of mixed pigment two-pack high solids coatings. The strong reduction in viscosity permits high pigment and filler loading. In applied coatings, it generates a good topcoat finish and stable colour while providing good gloss retention. Adhesion is not impaired after exposure to condensation. The outstanding stor-age stability of coatings manufactured with the new additive contrib-utes to resource efficiency.The high molecular weight polymer (50 % solids dissolved in methoxy-propyl acetate/dibasic ester) exhibits very broad compatibility both in high solids polyesters and acrylates as well as in medium solids grades of the same types of binder. The star-shaped polymer has a high density of pigment-affinic groups and stabilises inorganic pig-ments, organic pigments and carbon blacks with equal efficiency.

Figure 2: Rub-out test for colour flotation, 12 minutes after spray application, left: highly-branched polymer with non-balanced anchor groups in the molecule, centre: “Tego dispers 678”, r ight: PU chemistry, market standard Figure 3: Condensation-water test to diN EN iSo 6270-2

H i g H S o l i d S3 0


Raw material Parts by weight %

part a: Millbase

acrylate resin (70 % in butyl acetate) 36.0

“tego Dispers 678” (50 % in methoxypropyl acetate/dibasic ester)

2.0

carbon black 0.2

phthalocyanine blue 0.4

titanium dioxide 12.0

Barium sulfate 4.0

acrylate (70 % in butyl acetate) as above 13.4

Methoxypropyl acetate 4.0

Let-down

polyester (80 % in butyl acetate) 16.0

Butyl acetate 3.4

Blend of aromatic hydrocarbons 3.4

Xylene 4.3

“tego airex 990” 0.3

“tego Flow 300” 0.6

Total 100.0

PART B: Addition of hardener

HDi trimer (70 % in butyl acetate) 28.0

Voc in g/l (calculated) of useable coating 378

Property tested in two-pack PU Paint

Market standard“Tego

dispers 678”

Δe coated/rubbed surface before ageing

0.7 0.4

Δe coated/rubbed surface after ageing (two weeks, 50 °c)

1.3 0.5

Visual description of aged specimens (two weeks, 50 °c)

55 % greyish super-natant formed, no

sediment

5 % slight whitish supernatant formed, no

sediment

cross-hatch iso 2409 24 h after condensa-tion-water test

gt 1 gt 0

Degree of blistering to Din 53209

m1g1 m0g0

cross-hatch to iso 2409 on steel

gt 0 gt 0

optical appearance (levelling, gloss, body) after ageing (two weeks, 50 °c)

oK oK

Viscosity of the paint (part a) prior to ageing (mpas at shear rate 100 sec-1)

311 367

Viscosity of the paint (part a) after ageing for two weeks at 50 °c (mpas at shear rate 100sec-1)

370 384

Table 1: Test formulation (t wo-pack high solids PU) Table 2: Properties of “Tego dispers 678” in t wo-pack high solids paint

3 1H i g H S o l i d S


Rub-out test confiRms colouR stability

With mixed pigmentation, the wetting and dispersing additive has to deal with the particular challenge of reducing the millbase viscosity, wetting widely different pigment surfaces and subsequently provid-ing lasting stabilisation. Only by satisfying all three criteria can colour shifts, flooding, floating and settling be avoided.It is known that organic pigments are stabilised by aromatic anchor groups, while carbon blacks are stabilised by amine structures in the wetting and dispersing additive [2]. With mixed pigmentation, a wet-ting and dispersing additive with a customised molecular structure with differing pigment-affinic groups is required because the different surfaces must be stabilised to the same degree. To identify this struc-ture, a high solids two-pack poly-urethane test formulation as shown in Table 1 was used.The mixed pigmentation was chosen because wetting and dispers-ing additives can be easily differentiated using this sensitive tint. The wetting and dispersing additive must wet equally the various types of surfaces in this formulation and adhere to them; otherwise settling, separation, or colour differences would show up in the rub-out test. Figure 2 shows that the results of this test are clearly satisfactory.

good gloss and pHysical pRopeRties

In Part A of the formulation, a polyester was used in combination with an acrylic resin to assess the necessary broad compatibility of the wetting and dispersing additive. In two-pack topcoats for plastics substrates, it is usually necessary to improve the elastic properties by including a polyester resin. The adhesion and performance of the most promising structures were subsequently tested in a humidity chamber (Figure 3). The paints were applied to degreased “R46” cold rolled steel test panels from Q-panels. Prior to the condensation-wa-ter test to DIN EN ISO 6270-2, the painted test panels were aged for three days in an oven at 60 °C.Subsequently, they were subjected to a cross-hatch adhesion test to DIN EN ISO 2409. The edges and cross-cut surface were careful-ly masked off before the test panels were exposed in the humidity chamber. After two weeks, the coatings were examined for swelling, blushing and blistering. After 24 hours regeneration time, a further cross-hatch test was carried out.

Gloss, levelling, settling and effectiveness in traditional medium solids coatings were further criteria considered in the choice of structure for the new wetting and dispersing additive. The properties of the wetting and dispersing additive in two-pack high solids paints are shown in Table 2 and compared with those of the market standard for medium solids technology. The results of the settlement test after ageing can also be seen in Figure 4.

selecting tHe staR stRuctuRe

Technologies as PU and acrylat polymers with pigment-affine groups have been tested. To identify a successfull additive structure, a parent lacquer containing a wetting and dispersing additive was formulated. The first disqualifier was the degee of decrease in viscosity. A low vis-cosity of the parent lacquers is required in order to allow the applica-tion of the coatings via Airmix oder Airless with a high non-volatile content. Therefore, the parent lacquer was measured using cone/plate geometry. It is important that the viscosity is as low as possible and remains stable during storage.The coating has been applied before and after storage. After an evap-oration time of 12 minutes at room temperatur, a rub out was per-formed. In the following, the rubbed and non-rubbed surfaces were measured colorimetrically. An ΔE value as low as possible was used as a means to measure the stabilisation of pigments and fillers and con-stituted another exclusion criterion to find the most suitable structure of the additive. All in all, the new additive showed the best results and is distinguished by very easy and uncritical handling and storage in the paint manufac-turing process. Even at -18 °C, it is still clear and liquid.

bRoad Range of applications and benefits summaRised

The use of this new additive, “Tego Dispers 678” as a wetting and dis-persing additive for modern mixed pigment direct grinds simplifies manufacture and application. It is a general purpose product and can be used both for high and medium solid formulations. Modern high solids paints can be formulated to comply with the EU Directive. The new additive satisfies the requirements of industrial paint manufacturers because it:

ą achieves low ΔE values in situations such as the rub-out test, thus providing stable colours;

ą exhibits good results in the water condensation test, thus proving it is ideally suited for use in mechanically and chemically resistant paints;

ą lowers the viscosity of the millbase, thus permitting significantly higher non-volatile contents;

ą produces good results in terms of gloss and levelling after applica-tion, thus achieving a superior appearance;

ą possesses good storage stability, which increases process security; ą has a very low inherent viscosity (even at -18 °C) and is therefore easy to handle during production;

ą is also well suited for use in medium solids coatings and therefore reduces the number of raw materials needed in the manufacture of coatings.

RefeRences

[1] Heilen W. et al., Additives for Waterborne Coatings, Vincentz Network, Hannover,

Germany, 2009.

[2] The Big Tego, Evonik Industries AG, Essen, Germany, 2012.

Figure 4: Paints after ageing for t wo weeks at 50 °C, left: “Tego dispers 678” left: star-structure, centre: highly-branched poly-mer with non-balanced anchor groups in the molecul, r ight: PU chemistry, market standard

3 2


S u r fa c ta n t S

Sour

ce: E

voni

k

Sticking to the Subjectnew superwetting additives solve adhesion problems in waterborne coatings. By roger reinartz, Jeanine M. Sny-der and Ye Dai, air Products.

new, low-foaming siloxane superwetting surfactants have been developed to provide improved equilibrium and dynamic surface tension reduction as well as improved flow and level-ling for hard-to-wet surfaces without recoatability problems. Practical examples of their use are presented.

S urface chemistry is closely involved with wetting, flow and levelling of a coating, as well as other associated properties. When a liquid

coating is applied to a substrate, the coating should wet the substrate easily and evenly to ensure good appearance and adhesion.The coating will exhibit both an adhesive force (degree of association between the coating and the substrate) and a cohesive force within

itself. For spreading, or good substrate wetting to occur, the coating must have a stronger association with the substrate than with itself. This is shown mathematically in Figure 1. When the coating-substrate system has favourable surface chemistry, a desirable and well-func-tioning coating can be formed. If not, many defects and failure modes can occur, such as craters, crawling, picture framing, orange peel, Bé-nard Cells, cissing, pinholing, foaming, pigment flooding and floating [1-2], all of which can lead to other catastrophic deficiencies.Waterbased technology offers a means to meet various environ-mental and legislative demands, although formulation can pose many problems. Water has a high cohesive strength and high sur-face tension compared with most solvents; therefore additives are

air

Liquid coating

adhesive forcecohesive force

solid substrate

Work of adhesion Wa = γsa+ γL a-γsL

Work of cohesion Wc = 2γL a

Work of coeff icient sL/s = Wa- Wc

= γsa+ γL a-γsL - 2γL a

= γsa+ γL a-γsL

figure 1: adhesive and cohesive forces in a coatingfigure 2: Schematic structure of gemini surfactants: t wo standard surfactant molecules linked by a spacer unit

3 3S u r fa c ta n t S


Results at a glance

ű The increasing use of new resin technologies and non-fer-rous substrates to meet today’s product standards presents significant challenges for formulators. Substrate wetting, re-coatability and surface appearance are common problem areas when using waterborne coatings.

ű Many flow and levelling agents as well as wetting agents per-form well but introduce foam problems. Effective defoamers in such formulations may cause craters and can create a vicious cir-cle that is hard to break.

ű New, low-foaming optimised siloxane superwetting sur-factants have been developed to provide equilibrium and dynam-ic surface tension reduction as well as improved flow and levelling for hard-to-wet surfaces. Good results are shown in wood and plastics coatings and even a self-levelling flooring mix.

needed to reduce the surface tension of water to allow good sub-strate wetting.

gemini Structure for SurfactantS offerS advantageS

Surface active agents, or surfactants, are materials that can adsorb onto surfaces or at interfaces and lower the surface or interfacial free energies of aqueous formulated systems. Conventional surfactants have a polar or ionic, hydrophilic head group connected to a hydro-phobic, hydrocarbon tail group. Because of their amphiphilic nature, surfactants will migrate to and accumulate at interfaces and reduce surface and interfacial tensions even when used at very low concen-

trations. In contrast, a particular type of speciality surfactant struc-ture, termed Gemini, has two hydrophobic tails connected to two hydrophilic head groups on the same molecule. As shown in Figure 2, each half of the “twin” surfactant is joined together by a spacer group, forming the Gemini structure.Because of their unique molecular architectures, these Gemini sur-factants are much more surface active than their standard ‘monomer-ic’ components [3]. Gemini surfactants are often used in waterborne coating, ink and adhesive formulations for both dynamic surface ten-sion reduction and foam control.There are many different interfaces in a coating formulation where surfactants can concentrate (see Figure 3). Surfactants are still widely used to stabilise the resin emulsions used in waterborne systems and other surface active materials are used to disperse and stabilise pig-ments. However, for effective substrate wetting the surfactants must be active at the moving boundary between the liquid and substrate.

baSic ParameterS for SubStrate wetting

Many surfactants can facilitate substrate and particle wetting by re-ducing the surface tension of the formulation to a level that is equal to, or lower than, the substrate being coated. As the surface tension of water is 72 mN/m and typical coating substrates are in the region of 35-45 mN/m, surfactants must be used to decrease the surface tension of any waterbased paint to allow it to flow smoothly onto the substrate.Additionally, in order for wetting to occur, the contact angle q must be less than ninety degrees; contact angles greater than 90 ° result in beading of the coating on the substrate. Figure 4 illustrates how lower-ing the surface tension of the coating to below the surface energy of the substrate results in a final film with excellent wetting.Surface energies of some substrates may be less than 35 mN/m and present a significant problem to the applicator. When this occurs, two pathways are available. First, the substrate may be treated to raise the surface energy, making it easier for the coating to wet out the surface. When this is not a viable option, the formulator is then forced to look at a stronger set of surfactants, including superwetters, to achieve the low surface tensions needed to wet out the substrate.

chemical claSSification of SurfactantS

Surfactants can be classified in several ways. They may be non-ionic, anionic or cationic in nature and the choice of which type to use is

Adsorption at surface

CrystallisationMicelle

formation

Bilayer vesicleformation

pigment

resin

Adsorption at L/L interface Adsorption at

S/L interface

figure 3: Various surfactant interactions in a formulation

figure 4: Wetting agents facilitate substrate and particle wetting; γSa, γSL and γLa are solid-air, solid-liquid and liquid-air interfa-cial tensions

S u r fa c ta n t S3 4


often determined by the application and other components in the for-mulation. The largest and most commonly used class of surfactants, the anionics, includes chemistries such as alkyl benzene sulfonates, fatty acid soaps and dialkyl sulfosuccinates [4].While these offer excellent wetting at a relatively low cost in use, their main drawback is their tendency to foam in coatings, thus requiring the formulator to include strong defoaming agents which can lead to film defects and aesthetic issues.Non-ionic surfactants, many based on polyethoxylated materials, can be excellent wetting agents and emulsifiers [4]. This group includes the gemini surfactants mentioned earlier and their derivatives as well as siloxane-based surfactants and fluorosurfactants.The final and smallest group, the cationics, is used less often in coat-ings applications as they will react with any anionic species in the sys-tem; however, they are sometimes used as solid particle dispersants or emulsifiers [4].While all three types can offer excellent wetting, they perform dif-

ferently when subjected to dynamic conditions and exhibit different foam profiles.

imProving Performance with SuPerwetting SurfactantS

What happens when the surface tension of the substrate falls well below the typical 35 mN/m range where traditional surfactants are no longer effective? Such low-energy surfaces might include plastic, wood or even poorly prepared or oil-contaminated metal surfaces.This becomes even more challenging with lower VOC coatings, which are now formulated to well below 100 g/l and even down to 0 g/l, with less solvent to help flow, levelling, and appearance of the coatings. In such cases, the formulator may employ the use of superwetting sur-factants to reach the desired level of wetting.Superwetters are surfactants with structures that allow them to effi-ciently adsorb and pack at interfaces, resulting in low dynamic surface

figure 6: Dynamic Surface tension (DSt) comparison of aque-ous solutions containing 0.1 wt.% surfactant (measured using a Krüss “BP-II” bubble tensiometer) with new optimised siloxane based superwetting surfactants

figure 7: Equilibrium surface tensions vs dynamic surface ten-sions for select surfactants. Surface tension measured at 0.1 % water solution

no surfactant siloxane - 1 siloxane - 2

siloxane - 3 optimised siloxanesuperwetter 1

optimised siloxanesuperwetter 2

figure 8: 2K polyurethane waterborne clearcoat applied to polypropylene panels

traditional siloxane surfactant organic superwetter

figure 5: Wetting of a traditional siloxane surfactant vs. organic superwetter: 0.1 wt% aqueous solution, oily metal surface, 10 sec wetting time, 23 °c

3 5S u r fa c ta n t S


tensions, fast wetting times and low contact angles on low surface en-ergy substrates, as can be seen in Figure 5. They can be siloxane-based or organic in nature. It has been recognised that it is actually the com-pact structure of the surfactant hydrophobe that plays a major role in a molecule performing as a superwetter [5-9].

the imPortance of dynamic Surface tenSion

Many coating-related processes, such as paint manufacture (e.g., pigment dispersion), mixing and application, are high speed agitation processes, which disrupt the surfactant concentration and alignment at the surface, thus creating higher dynamic surface tension during and after agitation.Surfactants that quickly diffuse back to the interface, align and reduce surface tension offer low dynamic surface tension, i.e., they provide low surface tension during and shortly after these dynamic agitating processes. They are effective and efficient at providing good wetting and levelling, and avoiding problems such as craters, fisheyes and

other defects.While siloxane and fluoro-based surfactants can achieve very low equilibrium surface tensions, their dynamic surface tension perfor-mance is poor because they are unable to quickly migrate to the new-ly created interfaces under dynamic conditions and maintain that low surface tension state [10].Organic superwetting surfactants are able to achieve reasonably low equilibrium and extremely low dynamic surface tensions; therefore, they are often excellent alternatives to traditional siloxane and fluoro-surfactants. Also, when used in a waterborne formulation, traditional siloxane and fluoro-based surfactants will migrate to the air-coating interface, as they tend to be organophobic. This may lead to recoat-ability issues should a second layer of coating be applied.In contrast, organic superwetting surfactants maintain their low dy-namic performance even during high speed coating processes. They are also generally more compatible with most coating systems, have no or minimal foam stabilisation and present no issues regarding re-coatability. Figure 6 highlights the equilibrium and dynamic surface tension performance of some organic superwetters compared to siloxane-based and fluorosurfactants.

lateSt Siloxane technology enhanceS Performance

Many waterborne resin systems suffer from application limitations that can only be effectively eliminated by using high quality surfactants and defoamers. In particular, poor substrate wetting, edge retraction, pinholing and other defects result in high product failure rates that are costly to the manufacturer.Choosing the right surfactant that provides effective wetting and level-ling while minimising foam improves the overall performance of the coating formulation. Recently, new optimised siloxane superwetters have been developed; these products exhibit the outstanding spread-ing, flow and levelling usually associated with siloxanes as well as the low-foam, dynamic wetting benefits attributed to organic superwet-ting surfactants.Figure 6 compares the dynamic surface tension performance of the organic superwetters, a siloxane-based surfactant, a fluorosurfactant and the new optimised siloxane superwetters. Because they possess both siloxane and organic functionalities, these superwetting sur-factants provide superior flow and levelling and low-foam wetting. Ad-ditionally, they offer improved formulation compatibility and do not suffer the recoatability issues that may be seen with traditional silox-ane surfactants.When trying to coat a very hydrophobic surface such as plastic or wood, particularly when spray applying a coating, it is critical that the wetting package be able to perform while under dynamic shear and then again when the coating flows and levels across the substrate. Figure 7 illustrates how these new optimised siloxane superwetters hit the target with respect to both equilibrium and dynamic surface tension reduction.

PlaStic coatingS are a critical aPPlication area

Efficient dynamic and equilibrium surface tension reduction are key factors in affecting final wetting performance when coating substrates such as plastics with very low surface energies. As a demonstration, a two-component polyurethane coating crosslinked with isocyanate containing 0.2 wt% surfactant was prepared and spray applied to polypropylene panels and a silicone release liner at 75 μm wet film thickness.Figures 8 and 9 show that with no surfactant the system completely de-wets. Using traditional siloxane-based surfactants also results in poor wetting and defects in the film as well. By incorporating the new optimised siloxane superwetters, excellent dynamic and equilibrium

no surfactant silicone optimised siloxanesuperwetter 2

no wetting poor wetting, craters

good wetting

figure 9: 2K polyurethane waterborne clearcoat applied to silicone release liner

no surfactant organic sW1 silicone

figure 10: Polyurethane acrylic hybrid coating applied to red oak

figure 11: acrylic wood coating containing 0.2 wt% surfactant spray applied on wood

siloxane - 1



organic sW1

S u r fa c ta n t S3 6


surface tension reduction is achieved resulting in excellent wetting with no surface defects.

Performance of wood coatingS can alSo be enhanced

Wood substrates are inherently difficult to wet out due to the poros-ity of the substrate, surface contaminants such as glue or sap and irregularities in the surface roughness. As the coating flows over the surface of the wood, penetration into the substrate occurs; this can lead to surface imperfections from both inadequate wetting and foam generation.When the coating penetrates the wood and wets out the grain, air is displaced which rises to the surface and can become trapped in the dried film. Selecting a wetting agent that offers excellent flow and lev-elling as well as foam control is critical for a high quality finish. Figure 10 illustrates a waterborne wood coating based on a polyurethane-acryl-ic hybrid polymer system that has been brush-applied on red oak.When no surfactant is added, severe orange peel is seen in the coat-ing. In comparison, when incorporating Organic SW1 or a siloxane-based surfactant, the surface appearance is greatly improved. How-ever, only Organic SW1 can migrate quickly to the newly formed interfaces during application and not only lower the surface tension but also eliminate any foam generated by the wetting of the substrate. The siloxane surfactant offers excellent wetting but stabilises foam, which is still apparent in the dried coating.A self-crosslinking acrylic emulsion based wood coating was also pre-pared and spray applied onto wood at a wet film thickness of 100 µm. As shown in Figure 11, compared to the siloxane-based surfactant and Organic SW1 surfactant, the new optimised siloxane superwetters 1 and 2 show improved crack filling and foam control when coatings are spray-applied onto the wood surface.

good reSultS obtained in Self-levelling ePoxy flooring

The multifunctional benefits and broad formulation latitude of these new siloxane superwetters can be nicely demonstrated in a highly filled solvent free epoxy/cycloaliphatic amine self-levelling flooring system. The original starting formulation had poor stability and ap-pearance and suffered from pigment floating and pinholes, as can be seen in Figure 12.Various surfactants and superwetters were incorporated into the blank formulation (5 minutes at 1000 rpm) at both 0.2 % and 0.8 % dosage levels. The self-levelling flooring systems (10 x 10 cm) were ap-

plied on a plastic foil and after 15 minutes air was introduced using a notched trowel (upper half of the samples in Figure 12).As can be clearly seen, siloxane-1 has poor levelling, suffers from pig-ment flotation and has a significant amount of pinholes. Siloxane-2 increased viscosity, resulting in poor levelling performance. Of all the additives evaluated, only the new optimised siloxane superwetters gave a robust performance, independent of the dosage level and de-foamer combinations evaluated.

PromiSing additiveS for moSt difficult aPPlicationS

The application of coatings, inks and adhesives on difficult-to-wet sub-strates such as porous wood, plastics, films and oily metal presents significant challenges for coatings formulators. Selecting the sur-factant is critical to maximising wetting and minimising defects such as pinholes, edge retraction, and orange peel.New optimised siloxane superwetters have been developed to offer the formulator the ability to achieve low dynamic surface tensions with minimum foam generation while promoting excellent flow and levelling for a wide variety of coating systems.

referenceS

[1] Pierce P. E., Schoff C., coating Film Defects, Federation of series on coat-ings technology, Fsct, philadelphia, 1991.

[2] Bierwagen G. P., prog. org. coat., 1991, Vol. 19, p 59.[3] Reinartz R. et al, new gemini surfactants as paint additives, 7th nürn-

berg congress, Vol. 1, paper iii, pp 217-229.[4] Salager J.-L., surfactants: types and uses, Firp Booklet #e300-a, uni-

versidad De Los andes, 2002.[5] Hill R.M. et al, Langmuir, 1994, Vol. 10, p 1724.[6] Goddard E. D., ananthapadmanabhan K. p., chandar p., Langmuir, 1995, Vol.11, p 1415.[7] Hill R. M. et al, Langmuir, 1995, Vol. 11, p 1416.[8] Stoebe T. et al, Langmuir, 1996, Vol. 12, p 337.[9] Rosen M. J., Song L. D., Langmuir, 1996, Vol. 12, p 4945.[10] Snyder J. M., Marcella P. C., a new, environmentally friendly wetting

agent for architectural coatings, pci, april 2011.

no surfactant siloxane - 1 siloxane - 2optimised siloxane

superwetter 2

figure 12: Self-levelling solvent f ree epoxy flooring containing 0.2 % surfactant

Sour

ce: S

mile

- Fo

tolia

.com

Colourful Powder

Basis of the colour powder Gulal, which is used on holi celebrations and festivals, is usually cornmeal, corn-starch or rice meal.

Three holi faCTs

ą Traditionally, holi is a Hindu celebration of the begin-ning of spring.

ą Its is also called festival of colour or festival of love.

ą The date varies and is mostly in March or February.

Source: www.bbc.co.uk/nature/ humanplanetexplorer/events_and_festivals/holi


3 7 W o r l d o f C o l o u r

3 8


N a N o t e c h N o l o g ySo

urce

: Otm

ar S

mit

- Fot

olia

.com

A stronger shield from stormsNano-additives for toughening water-borne elastomeric roof coatings. By Detlef Burgard and Marc herold, Bühler Partec gmbh.

to fight global warming and to reduce energy costs, ir-reflec-tive roof coatings are applied in hot regions. the properties of waterborne coatings are often inadequate for this application. less than 1 % addition of a crosslinking nano-additive is shown to increase tensile strength and reduce water absorption of two waterborne elastomeric roof coatings.

d ue partly to global warming concerns, there is an increasing trend to replace black or dark grey roof coatings with white coat-

ings that have a high reflectivity and therefore contribute to cooler room temperatures. The coatings have to withstand various weather scenarios. Key issues are water resistance and a very good mechani-cal strength and high elongation at break of the dried film.When looking at the requirements that derive from these two issues, they appear contradictory at first sight. Maximum water resistance would require a non-swellable highly crosslinked polymer in the dry film. However, a high elongation at break demands a film with a low

degree of crosslinking, which in turn is susceptible to water absorption.Additionally, the requirements for this kind of roof coating are not only a high elongation at break but also high mechanical durability. This parameter is to some extent represented by the tensile strength of the film. But in many cases a high tensile strength often leads to a decreased elongation at break.As already mentioned, cool roof technology helps to reduce energy consumption. Therefore more and more IR-reflective roof coatings are applied in hot and sunny regions. In order to further reduce en-vironmental impacts, there is an increasing trend to switch from sol-vent-based coating materials to waterborne coatings.However, waterborne coatings still cannot fully replace solventborne alternatives due to their sometimes poorer performance and there is a steady need for further improvement. Using two elastomeric acrylic formulations as examples, the beneficial effect of a nano-additive on performance parameters such as tensile strength and reduced water absorption is demonstrated.

3 9N a N o t e c h N o l o g y


Results at a glance

ű In order to fight global warming and reduce energy costs, IR-reflective roof coatings are increasingly applied in hot re-gions. To further reduce environmental impacts, these coatings should be waterborne, but their properties are often inferior to solventborne types and improvement is therefore needed.

ű Using two sample elastomeric acrylic formulations, the ben-eficial effect of a nano-additive on performance parameters was demonstrated. The addition of less than 1 % additive dis-persion can reduce water absorption by more than 50 % and lower the diffusion coefficient of water to about one-third.

ű Stress-strain studies show that the tensile strength is in-creased by about 50 % while elongation at break is unchanged, leading to a coating with higher toughness and durability.

ű Non-covalent crosslinking by the nanoparticles is believed to explain these improvements.

formulAtions And test methods

To investigate the effects of the additive, two white pigmented roof coating formulations (Tables 1 and 2) were prepared based on “Primal EC 1791 E” (further named “Roof coating P”) and “Acronal NX 3250” (named “Roof coating A” respectively. Different amounts of nano-additive were added to these formulations. From these formulations samples were cast according to DIN EN ISO 527 (see Figure 1) to test the stress-strain profile.The samples were dried at room temperature for one week and test-ed on a Zwick Roell “UPM 1446” mechanical testing machine with a traction speed of 25 mm/min. The whole test procedure was set up according to ASTM D2370-98 (2010).For the determination of the water uptake, free-standing coating films were prepared by pouring samples of the formulations into PTFE cups. The samples were dried at room temperature for seven days and the coating films were peeled off and weighed. The amount of wet paint was calculated to obtain dry films with a weight of approximately 1 g. The thickness of the resulting film is in the order of 0.2 to 1 mm.The films were immersed in water and removed from the water after defined periods, dried and weighed. After the weight was determined, the films were re-immersed into the water bath. The immersion ex-periment was stopped when the water uptake of the films with and without additive reached saturation (Ms). Two series of tests were car-ried out with the blank formulation (see Table 1) and 99 % formulation with 1 % additive.The uptake of water by the polymer is assumed to proceed by a diffu-sion process that can ideally be described by Fick’s law. For the initial sorption period, a t 0,5 or √ t law can be derived [1]:

Mt = 4√ D Ms L√ π √t

(1)

In equation (1), Mt represents the amount of water absorbed at any time t, Ms the greatest amount of water in the film when the satura-tion stage is reached, L the film thickness (cm) and D is the diffusion coefficient (cm2/s). Plotting Mt/Ms versus t 0,5 or √ t provides a linear relationship that allows the diffusion coefficient D to be determined.

Ingredients Supplier kg

Water 134

“orotan 850” Dow 4

potassium tripolyphos-phate

1

“nopco nXZ” Henkel 3

“ti-pure 706” Dupont 372

“ti-pure r960” Dupont 62

“primal ec 1791 e” Dow 416

"texanol" eastman 6

aq. ammonia 10 % 2

“oxylink 3101” Bühler 0 20

table 1: elastomeric roof coating P

Ingredients Supplier kg

Water 69

1,2-propanediol 3

pigment verteiler nL BasF 11

“Byk 035” Byk 9

“Kronos 2190” Kronos 214

“DrB 8/25” imerys 138

“imercarb” imerys 9

“Microtalc it” omya 51

ammonia 10 % 4

“Byk 035” Byk 6

“nX 3250” BasF 436

Water 51

“oxylink 3101” Bühler 0-10

table 2: elastomeric roof coating a

4 0


N a N o t e c h N o l o g y

Figure 4: Water uptake plotted against film immersion time for the roof coating P with and without 1 % nano-additive (average film thickness 0.44 mm (blank) and 0.48 mm with nano-additive)

Figure 5: Water uptake plotted against film immersion time for the roof coating based on a with and without 1 % nano-additive (average film thickness 0.76 mm (blank) and 0.51 mm with nano-additive)

Figure 2: Stress-strain test results for the white roof coating P Figure 3: Stress-strain test results for the white roof coating a

FormulationDiffusion coefficient

(10-6 mm²/s)Ratio D with/ without additive

no addi-tive

With addi-tive

roof coating p (see table 1)

3.3 2.0 0.62

roof coating a (see table 2)

14.4 3.8 0.27

table 3: effect of nano-additive on water diffusion coefficient DFigure 1: typical test specimen for the stress-strain test; gr id distance is 1 cm



tensile strength And wAter resistAnce Are improved

The results of the stress-strain tests for the formulation in Table 1 are depicted in Figure 2. The test results show an improvement of the tensile strength when the additive is present in the formulation. While the tensile strength of the blank sample is 1.5 N/mm², an im-provement of > 30 % was observed, reaching saturation when 0.6 % additive was present. On the other hand the maximum elongation at break remains more or less constant at (300 ± 50) % over the com-plete sample series.The results of the stress-strain tests for the formulation based on the other emulsion can be seen in Figure 3. The test results show an im-provement of the tensile strength when the additive is present in the formulation. While the tensile strength of the blank sample is 2.4 N/mm² an improvement of > 48 % was observed, reaching saturation when approximately 0.5 % additive is present.On the other hand the maximum elongation at break again remains constant at (100 ± 20) % up to 0.8 % addition of the “Oxylink” nano-additive. Only if 1 % or more is present does the elongation at break decrease down to 80 %.The results of the water uptake of the roofing films of both formulations over time are compared in Figures 4 and 5. The results show a faster wa-ter uptake for the blank film sample when looking at the initial period of the immersion in water. Diffusion coefficients were calculated by equa-tion (1) and the results are presented in Table 3.

crosslinking explAins nAnopArticle benefits

A reason for the increase in the tensile strength of the films that incor-porate the additive can be the crosslinking of the elastomeric polymer in the formulation by the zinc oxide nanoparticles [2]. In earlier work

some indication for this crosslinking reaction with acrylic binders had also been found [3-5].No evidence has been found for a reduced elongation at break for the films with nanoparticles, which makes the additive extremely ver-satile for the crosslinking of roof coatings or elastomeric materials in general.It is assumed that the undisturbed elongation at break is due to the non-covalent nature of the crosslinking. Indications for this kind of non-covalent network bonds are also found when functionalised elas-tomers are crosslinked with zinc oxide that is bound on the surface of silica [6]. As regards the slower diffusion of water in the film and the therefore reduced water uptake in the film, it is also suggested that the increased crosslinking induced by the nanoparticles is the cause for this finding.Thus, nanoparticle-based additives such as “Oxylink” can increase the crosslinking density in 1K waterborne coating systems. As a conse-quence, various parameters including barrier properties, water resist-ance or toughness are improved. In this work, it has been shown that at the same time the flexibility of elastomeric films is maintained.

references

[1] perez c. et al, prog. in org. coat., 1999, Vol. 36, pp 102-108.[2] Zosel a., Ley g., Macromol., 1993, Vol. 26, no. 9, pp 2222–2227.[3] pilotek s. et al, the Waterborne symposium, new orleans, Feb. 18-20,

2009.[4] Burgard D., Herold M., Farbe und Lack, 2011, Vol. 117, no. 6, pp 14-18.[5] Herold M. et al, pci Magazine, august 2010, pp 24-27.[6] owczarek M., Zaborski M., KgK Kautschuk gummi Kunststoffe, 2004, Vol.

57, pp 218-223.

4 2


N a N o t e c h N o l o g y

Sour

ce: L

E BR

USQ

- Fo

tolia

.com

A little goes A long wAythe addition of nanoparticles can enhance the perfor-mance of water-borne coatings. By Detlef Burgard and Marc herold, Bühler Partec gmbh.

waterbased coatings have gained a significant market share during the last few years, while the standards in quality of these coatings have steadily improved. However, there is still a need to improve certain features to provide optimised solu-tions for different applications. one possible way to gain en-hancements is by the use of dispersed nanoparticles as perfor-mance additives.

s mall amounts of nanoparticles as additives can lead to a signifi-cant performance increase in one-component waterborne polyu-

rethane coatings, especially in terms of chemical/solvent resistance, blocking resistance drying time and hardness. The increase in perfor-mance can be achieved by a non-covalent crosslinking mechanism. There are some first indications that such nanoparticle additives can be a useful alternative to isocyanate crosslinkers in certain applica-tions, enabling increased hardness or faster curing kinetics equivalent to shorter process times. Another advantage can be seen in the fact that the tested nanoadditives did not change the 1K characteristics of the formulation. On the other hand, these nanoparticle additives can – under certain circumstances – be used together with isocyanate crosslinkers and further increase some performance parameters of the resulting coatings.As shown earlier [1-4], nanoparticle-based additives like “Oxylink” can increase the crosslinking density in 1K waterborne coating systems. As a consequence, various parameters like barrier properties, resistance against solvents and water or blocking resistance can be improved. Enhanced early properties can be achieved which may also lead to shorter process cycles in industrial applications. In these works, em-phasis focused on acrylic and styrene acrylic dispersions as resin bas-es. These nanoadditives can also be successfully used as cross linkers for PUD systems, both 1k but as well in certain 2k formulations.

MAteriAls And MetHods

A series of matt-clear coats for wood based on an aliphatic polyure-thane dispersion (Table 1) was prepared, in order to demonstrate the effect of the additive. To compare the impact of the nanoparticulate crosslinking additive on both 1K and 2K formulations coatings with and without polyisocyanate were produced. In experiments, a HDI

Ingredient Ingredient amount /g

aLX 1 aLX 2 aLX 3 aLX4

alberdingk u 9150

alberdingk Boley, germany

771.0 771.0 771.0 771.0

Hydropalat 140 cognis, germany 4.0 4.0 4.0 4.0

tego Foamex 822

evonik, germany 6.0 6.0 6.0 6.0

acemat ts 100 evonik, germany 10.0 10.0 10.0 10.0

Diethylene glycol monobutylether

40.0 40.0 40.0 40.0

Water 149.0 149.0 149.0 149.0

acrysol 2020 Dow, usa 10.0 10.0 10.0 10.0

oxylink 3102 Bühler ag, switzerland

- 10.0 - 10.0

Bayhydur 3100 Bayer Materialsci-ence ag, germany

- - 77.1 77.1

table 1: coatings and formulations



Results at a glance

ű Small amounts of nanoparticles as additives can significant-ly enhance the performance of one-component waterborne polyurethane coatings, especially in terms of chemical/solvent resistance, blocking resistance drying time and hardness.

ű First test series were done with 1K formulations ALX1 and ALX2 in order to evaluate the crosslinking effect of the nanoad-ditive, using a MEK double rub test, where is was shown that the additive nearly doubled the MEK double rub resistance from 22 double rubs (w/o additive) to 40 double rubs (w/o additive).

ű The results for the Koenig hardness measurements show that the nanoadditive has the largest impact on the 1k formula-tion. Already after one day drying, the hardness of the coating ALX 2 with 1 % additive is about 20 % higher than the hardness of the formulation ALX 1 without additive.

ű By the fast reaction of the nanoparticles, shown in Equation 1, a large portion of carboxylic groups may be blocked and are no longer accessible to undergo a reaction with isocyanates.

ű There is the potential to further improve the performance of 2K systems when the reaction kinetics of isocyanates and the crosslinking nanoadditive are carefully matched.

(“Bayhydur 3100”) was used. The dosage level for the nanoadditive was constantly at 1 %. The formulations with the polyisocyanate were prepared and applied on the glass plates and lenata foil (PVC) substrates. For detailed analy-sis of the coating performance the following tests were carried out:The degree of cross linking was determined by crock meter double rub tests with 2-butanon (MEK)-double rub resistance test (crosslink-ing) according to ASTM D-5402. The resulting parameter is the num-ber of double rubs needed to completely remove the coating from the substrate. In a parallel test the double rub resistance against 2-pro-

panol (IPA), was also evaluated. ą The formulations were applied on leneta foil (PVC) with a roll coater and approximately 60 µm wet film thickness

ą �The�samples�were�dried�at�room�temperature�(22-23� °C�/40-60�%�rel. humidity) for 16-72 hours

ą The samples were blocked for 24 hours at room temperature ą The samples were de-blocked and rated on a 0-5 scale (0=best)

For the determination of the gloss the formulations were applied on glass plates with 100 µm wet film thickness by using a doctor blade. The samples where dried overnight and the gloss was determined by measuring the reflectivity at eight different points with a gloss meter (“Elcometer�402”)�at�85°�according�to�DIN�EN�ISO�2813.�The drying time for drying level four was measured with a “B.K. Drying Recorder”. Therefore glass slides were coated with a 100 µm wet film and determined the duration until drying level d was reached (there was no visible pin trace on the coatings surface).For the determining the adhesion the formulations were also ap-plied on glass plates with 100 µm wet film thickness by using a doctor blade. The samples where dried at room temperature overnight and the film-to-substrate adhesion was determined by using the cross cut method�with�1�mm�distance�between�the�cuts�according�to�DIN�EN�ISO 2409. The adhesion was evaluated by ranking the results from 0 (best) to 5 (worst). The Ethanol resistance of the coating was tested with exposure times of�1�hour�according�to�DIN�EN�ISO�2812- 3�and�DIN�EN�ISO�4628-2.�The chemical resistance was evaluated by rating the results from 0 (best) to 5 (worse).

results

The compatibility of the additive with all formulations was excellent. The gloss of the formulations remains nearly unaltered at (40 ± 2) gloss units. The 1k formulations ALX 1 and ALX 2 have been stable at 40�°C�for�more�than�three�months�and�are�still�under�observation.�In�contrast to this, the 2K formulations ALX 3 and ALX4 have a pot life of approximately three hours.First test series were done with the 1K formulations ALX1 and ALX2 in order to evaluate the crosslinking effect of the nanoadditive, using a MEK double rub test. The additive nearly doubled the MEK double rub resistance from 22 double rubs (w/o additive) to 40 double rubs (w/o additive), as shown in Figure 1a. When applying Isopropanol (IPA) rubs

A B

Figure 1a: MeK double rub resistance with and without nanoadditive

Figure 1b: IPa double rub resistance with and without nanoadditive

A B

N a N o t e c h N o l o g y4 4


to the sample coating the effect was even more pronounced. The IPA double rub resistance, shown in Figure 1b, boosts by using the addi-tive from 19 double rubs to 143 double rubs.A reason for the increase in double rub resistance of the films that incorporate the additive can be the crosslinking of the carboxylic groups of the polymer in the formulation by the nanoparticles. In ear-lier works some indication for this cross linking reaction with acrylic binders [1-4] was also found to be the case. This result shows that these findings can be transferred to PUD systems, as long as there are accessible reactive carboxylic groups. Further evaluation of the time dependence of the cross linking effect in the ALX 1 and ALX 2 formulation by monitoring the drying times has also been carried out. The drying time is characterised by the time to reach drying grade d (“end of pin trace”). “Oxylink” causes the drying of the formula to accelerate by 15 % from 21.2 min to 17.7 min. The dry time acceleration is attributed to the early onset of the crosslinking reaction between the carboxylic groups and the nanoparticles, shown in Figure 2. The fast onset of crosslinking results in a higher film tough-ness throughout the drying process is such that a certain fastness is reached already after a shorter period of time.

The blocking of the coating was reduced by 1.5 marks from mark 4 to mark 2.5, as shown in Figure 3. Blocking of polymers and coatings is a process where parameters like the migration of substances, sur-face roughness, chemical surface structure and surface energy are involved [5-7]. Current research in this field has shown that blocking resistance of coatings can be achieved by prevention of migration of low molecular weight polymeric material between the surfaces [8]. This migration can be prevented by increased crosslinking density, for instance. In addition, anionic groups of these smaller molecules can interact with the surface of the inorganic nanoparticles of the additive and thus being immobilised or at least hindered in their migration.The radar graph in Diagram 4 provides an overview on the different performance improvements by the cross linking nanoadditive. In ad-dition to the above mentioned improvements, no negative impact of the nanoparticle additive on other coating properties like adhesion or gloss could be found in this study. Figure 4 summarises the us-age properties of the two tested coating materials and compare the overall performance of the individual formulations with and without the nanoparticle additive. A very distinct improvement of the overall performance can be achieved through the addition of nanoparticles to the respective formulation.The next step was to investigate in how far the performance of the na-noadditive crosslinker can be compared to the performance of isocy-anate crosslinkers which represent the current high quality state of the art. Results are described based on a comparison of the nanoparticu-late crosslinker with an HDI – isocyanate – that has been used in the formulations ALX3 and ALX4, shown in Table 1. In these first tests dry-ing time and hardness development was being assessed, as this is in-teresting for the application and further processing. For example, when a subsequent sanding or polishing step is required, it is important for the coating to harden early, therefore enabling the applicator to carry out further processing steps as soon as possible after the coating is ap-plied. In addition, a good final hardness improves abrasion-resistance.The results for the Koenig hardness measurements in Figure 5 show that the nanoadditive has the largest impact on the 1k formulation. Already after one day drying, the hardness of the coating ALX 2 with 1 % additive is about 20 % higher than the hardness of the formulation ALX 1 without additive. This 20 % difference in hardness is maintained over the observed time of three weeks.When looking at the 2K formulations ALX3 and ALX4, the effect of the nanoadditive on the hardness development in the presence of the HDI isocyanate is smaller but still remarkable. After one day drying,

Figure 2: Reduced drying time with nanoadditive in formulation alX2

Figure 3: Improved blocking resistance with nanoadditive cross linker in formulation alX2 0 = best; no damage at all, 5 = worst; complete failure

Figure 4: overall performance comparison for the formulations shows significant improvement by the use of the nanoadditive. : alX1; : alX2 (100: Best performing in the respective test)



the hardness of the coating ALX4 is about 12 % higher than ALX3. Whereas the hardness of ALX3 is even slightly lower than the hard-ness of the 1K formulation ALX1, the formulation which contains both nanoadditive and isocyanate (ALX4) shows about the same degree of hardness than ALX1.In both the 1K and the 2K system the presence of the nanoadditive leads to a faster hardness development, higher early hardness as well as higher end hardness. For the 1K formulation, the cross linking mechanism according to Equation 1 can be applied: Accessible carboxylic groups of the polymer are chelating to the sur-faces of the nanoparticles. This reaction is usually very fast and starts upon drying. This fast cross linking reaction is the main reason for the acceleration and increase of hardness development in coating ALX2.For the 2K formulation ALX4, there are two different cross linking mechanisms as there are two different cross linkers in the formula-tion. Here the different kinetics of the two cross linkers may play an important role. The reaction of the carboxylic groups with the nano-particles is very fast and may be faster than the undesired reaction with the isocyanate as shown in Equation 2.By the fast reaction of the nanoparticles, shown in Equation 1, a large portion of carboxylic groups may be blocked and are no longer ac-cessible to undergo a reaction with isocyanates, shown in Equation 2, so that more isocyanate can react with available OH- groups for urethane formation. In addition, less amides (and possibly CO2) as re-action product of isocyanate and carboxylate are present in the coat-ing film to influence the final film properties. There is the potential to

further improve the performance of 2K systems when the reaction kinetics of isocyanates and the crosslinking nanoadditive are carefully matched. Of course, this is still a theory and has to be further veri-fied in the future. As well, further investigations will have to be done to further evaluate the influence of the nanoparticle crosslinker on other film properties of the 2K systems, such as chemical resistance or weathering resistance.

references

[1] Burgard D., Herold M., european coatings Journal, issue 7-8/2013[2] Pilotek S., Burgard D., Herold M., Steingröver K., the Waterborne

symposium, new orleans, Feb. 18-20, 2009.[3] Burgard D., Herold M., Farbe und Lack, 117, 6 (2011) 14-18.[4] Herold M., Burgard D., Steingröver K., pilotek s., pci Magazine, august

2010, 24-27.[5] Koromminga T., van Esche G., plastics additives Handbook, Zweifel, H,

ed., Hanser, 2001.[6] Horne S. E., Suarez J. J., spe J. 1969, 25, 11, p. 34-38[6] van Esche G., Kromminga T., Schmidt A., soc. plast. eng., inc. antec,

conf proc., 1998, 2837[7] Vincent O., Osmont E., soc. plast. eng., inc. antec, conf proc., paper 131,

2002[8] Wypych G., Handbook of antiblocking, release, and slip additives, 1st

ed, chemtec

equation 1 equation 2

BA

Figure 5a: Development of Koenig hardness within the first t wo days

Figure 5b: Development of Koenig hardness within the first three weeks

A B

4 6


S m a r t c o at i n g SSo

urce

: Vic

tor

Burn

side

- F

otol

ia.c

om

Permanent hydroPhobic and easy-to-clean effects

functional surfaces with dirt repellent properties are increasingly important as it takes less effort and surfactants to clean them. existing easy-to-clean effect additives are based on fluorosilanes and polysiloxanes. their application is limited to special coating systems and their effectiveness often declines with time. in a bachelor thesis the permanent easy-to-clean effect of non-ionic monomeric and polymeric fluorinated surfactants as additives in various coating systems has been investigated. the additive developed, with additional functionalities, is universal and may be crosslinked into different coating systems. the additives were tested in a range of solventborne and waterborne coating sys-tems. the easy-to-clean effect evaluation was performed with re-sistance to abrasion stress for water- and dirt repellence and their permanence. the results show that fluorinated polymers with multiple functionalities especially generate hydrophobic surfaces with a permanent easy-to-clean effect in different coatings.

at first glance, most of the coating industry’s technical needs are well satisfied. If you take a closer look at the scientific papers, however,

you discover that functional surfaces inspired by nature, which are then used as the inspiration for future technology, are often mentioned as a theme of interest. Functional coatings ensure that the product deliv-ers a special performance [1], [2]. A study by the DFO shows that there is great interest in functional coatings in industry, especially for self-cleaning/easy-to-clean surfaces [3]. Easy-to-clean surfaces are defined as hydrophobic surfaces that could be cleaned with less effort, to which less dirt adheres and that have a contact angle with water of between 90 °and 130 °. Self-cleaning surfaces don’t need to be cleaned - water and light do the job. The dirt repellence depends on the bad wettability of the structured surface – the contact angle is higher than 130 °.

Figure 1: over view

Universal surfactants for different coatings systems. By carina Deschamps, Lada Bemert, Hugh gibbs and michael groteklaes, omg Borchers and University of applied Science Krefeld.

nature as a model

Looking at the surfaces of particular plants like the lotus, and their abil-ity for self-cleaning thanks to their microstructured superhydrophobic surface, scientist established the term Lotus-Effect in the 1970s.The self-cleaning of such surfaces is not particularly durable as the mi-

4 7S m a r t c o at i n g S


Results at a glance

ű Using fluoropolymers one can create hydrophobic, water-repellent, permanent easy-to-clean surfaces in coatings.

ű Best results are obtained using most hydrophobic modified fluoropolymers.

ű After modification, there was permanence in the coating.

ű The contact angle is high enough for an easy-to-clean effect.

ű The new coatings are easy to clean, environmentally re-sponsible and cost effective

crostructure has little resistance to mechanical stress [4-10]. The chal-lenge is to enrich the permanent cleanability of the surface. Until now there are different types of products with easy-to-clean effects. On the one hand there is the Do-It-Yourself sector, with sprays. On the other, there are industrial technical surfaces such as Polydimethyl-siloxanes and Polytetrafluoroethylene, like “Teflon®”. New binder al-ternatives are Polysilazanes for anti-graffiti applications and Sol-Gel-Coatings with fluorosilanes [4, 10, 11]. Possible applications for coatings with easy-to-clean effect include car wheel rims, furniture and floor coatings, car signs, anti-graffiti coat-ings and general industrial applications, where the cost-effectiveness could be increased with lower cleaning demand. The positive environ-mental effect is also important, as fewer cleaners and surfactants are needed. Easy-to-clean coatings can protect railways carriages, build-ings and possibly cars. Existing market products with an easy-to-clean effect are additives based on fluorosilane or polysiloxanes, non-sticking coatings like Tef-

lon in pans, or sprays with easy-to-clean effect in DIY. They have nu-merous disadvantages, such as poor adhesion, poor recoatability, and low permanence of the effect.The aim of this work is to find a universal, permanent easy-to-clean additive for different coating-systems. In addition, it addresses the question of a method for evaluating the easy-to-clean effect, as there is no industrial standard method available. An ASTM standard is un-der development. The following illustration summarises the experi-mental part in the research (Figure 1).

non-ionic addititves tested

In this work, three groups of additives were tested as to their per-formance of the easy-to-clean effect: hydrocarbon- and fluorosur-factants and low molecular weight fluoropolymers. Surfactants are surface-active agents. Owing to their hydrophilic and hydrophobic el-ements, and their amphiphilic structure, they orient themselves at the interface, for example coating/air. The surfactants used in this work are non-ionic, which means that their head-group carries no charge. Both the hydrocarbon and the fluorosurfactants have ethoxylate-groups as their hydrophilic part. Low molecular weight fluoropolymers, tested here, show high hydro-phobicity with surface activity. They allow the highest depression of the surface tension.Table 1 and 2 provide an overview over the tested additives. The fluo-ropolymers are more surface active than the others. Table 1 shows the additives tested in the coating systems.The main objective of the paper is to explain the development pro-cess. At the end, the additive with the best results with respect to competitor additives were compared. The competitor products are applicable only in certain systems, depending on the polymerisation and solvent. The overview is provided in Table 2.

how to achive Permanence

In order to achieve the permanent easy-to-clean effect, the molecule should be anchored at the surface. Modification of additives simultane-ously with different functionalities allows crosslinking of the additives into different coating systems, e.g. polyurethanes, acrylics, UV-cured

Figure 2: results of contact angle measurements with demineralised water in the 2-pack PUr-system, water-based. the additives are ordered by increasing hydrophilic properties.

S m a r t c o at i n g S4 8


systems, melamines. The functionalities, inserted into the molecular structure are vinylic- and hydroxylic, both of which are on the hydro-philic part of the additive to ensure polymerization into each coating system with the hydrophobic part towards the air. To ensure compara-bility of the effect, the concentration of the additives is kept constant.

measurement PrinciPles used

Industry standards for evaluating easy-to-clean effects from coatings are not available at present. For this reason, proprietary test methods were developed and used.Wettability is one of the important criteria for easy-to-clean effects. The value of the contact angle of the polymerised coating with water or oil provides an approximation of hydrophobicity and surface roughness. The contact angle can be measured statically and dynamically. Permanence of the effect was evaluated with the abrasion test with iso-propanol, which is often used in cleaners. If the additive is permanent the contact angle should not change after numerous cycles of abrasion with isopropanol. If the additive is washed out, the contact angle will be far lower after the abrasion. The testing procedure includes abrasion with 5 mL isopropanol on cotton with different number of cycles and a weight 300 g above the cotton pad on the coating. To evaluate perma-nency, the contact angle was measured before and after abrasion. The tests were performed on an abrasion tester. Pendulum hardness, gloss and haze and slip were also measured to see whether the surface is rougher with and without additives, or after the abrasion.To evaluate cleanability of the coating, a dirt suspension with a mix-ture of dirt from the industrial environment was created, sprayed it over the coating and washed it with water in the dishwashing ma-chine. The dirt adhesion on the coating was tested by measuring the colour values before and after the dirt treatment.At first the additives were tested in four coating systems: waterborne and solventborne two-pack polyurethane systems, conventional and waterborne UV coatings. These systems are widely used in industri-al applications and UV-cured coatings are an environmental friendly coating technology with great application especially for wood coatings. Investigations will now focus on other systems, including two pack clear coats, multi-layer coatings for automotive industry, unsaturated polyester, epoxy system and coil coatings.

how additives influence the contact angle

One of the requirements of the easy-to-clean additive is its surface activity. During polymerisation, the active substance should migrate to the surface layer. Surfactants lend themselves for such applica-tions due to their self-organising properties at the interfaces. The next question is the value of the hydrophilic-lipophilic balance of the sur-factant molecule needed for the migration tendency to the liquid-air interface. Systematic investigation was started from the well-known carbon-based fatty alcohol ethoxylates over the short-chained per-fluorinated surfactants to polymeric perfluorinated surfactants. Un-modified and modified substances were compared, whereby the modification with two functional groups should not have a large im-pact on the HLB value. Figure 2 shows the results for the contact angle with demineralised water.Since organisation of the chosen molecules was expected at the sur-face with the hydrophobic part towards the air, the contact angle was measured with demineralised water on the coating with additive. Con-tact angle provides a highly sensitive response that depends on the surface hydrophobicity. Figure 2 shows the measured contact angels in the waterborne two-pack polyurethane system; results in the other coating systems show the same tendencies. The X-axis shows the tested additives, competitor product (X) and the coating without addi-tives. The other additives are ordered from left to right by growing hy-

drophilic properties in each group. They are categorised according to three groups: fluoropolymers, fluoro- and hydrocarbon- surfactants. The blue bars are not modified (NM); the red ones are modified (M). The Y-axis shows the contact angle in degrees. The value of the er-ror bars is calculated from the repeated measurements on different places on the coating and after approximately 48 hours. Based on the results of Figure 2 and the results from the other coating systems, the following conclusions follow: ą Most additives increase the hydrophobicity of the surface. The contact angle is higher than the contact angle of the coating without additive.

ą The fluoropolymer additives are better at creating an easy-to-clean effect in coatings than the hydrocarbon and fluorosurfactants.

ą The highest contact angle generated is nearly 100° in the two-pack conventional polyurethane system.

ą In general, the modified additives show a slightly higher contact angle than the unmodified.

ą Compared to the competition, the fluoropolymers create higher contact angles.

ą Fluoropolymers need to be used in lower concentrations as they are highly effective.

ą FP1 is selected as a good active ingredient for easy-to-clean addi-tive, also due to the measurements in the other systems. It always shows good contact angle results and the modified version will be tested for coating permanency.

measuring the Permanence

Figure 3 shows the results of the abrasion stress with isopropanol in different coating systems. The X-axis shows the four coating systems after 100 cycles in the abrasion tester. The Y-axis shows the contact angle difference before and after stress with isopropanol. As the dif-ference of the contact angles increases, the permanency of the ad-ditive in the coating decreases. The unmodified fluoropolymer FP1 shows a high contact angle difference after 100 cycles. After modifica-tion, the contact angle difference of the polymer is lower than the val-ues of the unmodified polymer. The values of the competitor and the modified fluoropolymer are comparable. They are both permanent in the coating systems. The modification makes our polymer permanent in different coating systems.Looking at the sample without additive, there was a negative contact angle difference. This is the pure abrasion effect. The higher contact angle after the abrasion test results from the surface roughness. The difference, resulting from the roughness, was taken into account in

Hydrophilic properties

Hydrocarbonsurfactans c13e5 c13e7 c13e10

Fluorosurfactants Fs1 Fs2 Fs3

Fluoropolymer Fp1 Fp2 Fp3

table 1: additives in test

application area waterborne solventborne UV systems

competitor X1 X2 X3

table 2: competitor additives

4 9S m a r t c o at i n g S


the samples with additives. The measured values due to the abrasion effect were normalised according to:∆ a∙n[°] = ∆ an[°] – ∆ a0[°]with∆ a∙n – plotted contact angle difference,∆ an – measured contact angle difference from sample with additive,∆ a0 – measured angle difference from sample without additive (zero).

The contact angle difference was measured and calculated using the following formula:∆ an[°]=a’ [°]-a’’[°]whereby∆ a – contact angle difference,a’ – contact angle before abrasion,a’’ – contact angle after abrasion.

Figure 4 shows the results of the contact angle with the FP1 and the competitor in the four coating systems. The fluoropolymer is used at the same concentration as the competitor product. In this system, the modified fluoropolymer always has a higher contact angle than the competitor. However, if one uses 10 times the quantity of the compet-itor product, the contact angle is now higher. In other words, the con-tact angle depends on the quantity of the competitor additive used.

hardness and gloss are not influenced

Generally it was not possible to create oleophobic properties with our

Figure 3: resistance to abrasion stress with isopropanol in detecting the permanence of the additives.

additives or with those of the competitor.The measurement of gloss, haze, pendulum hardness and slip were not performed in evaluating the easy-to-clean effect, but were meas-ured to see whether our additives or modifications had any other ef-fects. There were no negative effects when applying our additives for pendulum hardness or gloss. The haze values were slightly better us-ing the fluoropolymer, which acts partly as a levelling agent.In further investigations the optimal concentration of the fluoropoly-mer in the coating were determined. This did not lead to the same ef-fect for increased contact angles with higher concentration as with the competitor additive. It seems that the other effect is critical; increas-ing the concentration: over a defined concentration the contact angle does not increase and is nearly constant. This is assumed to be due to the migration of the additive in lower coating layers due to the oc-cupancy of the surface with fluoropolymer molecules. In other words, one molecule of fluoropolymer takes about 3 nm2 (approximation) on the surface. Only one layer of fluoropolymer is needed to achieve the necessary effect. The concentration of active ingredient should be not higher, as since the surface area divides by 3 nm2, the residual mol-ecules will migrate into the deeper coating layers and cause no ad-ditional increasing of contact angle.

conclusion

Using fluoropolymers it was possible to create hydrophobic, water-re-pellent, permanent easy-to-clean surfaces in coatings. The best results were obtained with the most hydrophobic modified fluoropolymer. Af-ter modification, there was permanence in the coating. In all four tested coating systems, the contact angle is high enough for the easy-to-clean effect. Similar effects were achieved in all four tested waterborne and solventborne systems with one additive, but there is no universal com-petitive solution. In conclusion: fluoropolymer additives could make your coating functional – the cleaning effort decreases. It is also environmen-tally friendly as you need less effort to clean the surface. Visually, your product looks clean for longer.

references:

[1] Stenzel V., Rehfeld N., Functional coatings, 1. auflage, Vincentz net work, Hannover, (2011), 13

[2] Zhang Y. et.al., recent progress of double-structural and functional materials with special wettability, Journal of Material chemistry, 2012, 22, 799-815

[3] Roths K., Gochermann J., Forschungsagenda oberfläche, 1.auflage, DFo, neuss, (2007), 86

[4] Brock T., Millow S., schmutzabweisende oberflächen, roempp online, 3.17, georg thieme Verlag, (2009), Dokumentkennung: rD-19-06128

[5] Barthlott W. , Neinhuis C. , purity of the sacred lotus, or escape from contamination in biological surfaces, planta, 1997, 202, 1-8.

[6] Bhushan B., Jung Y. C., Koch K., self-cleaning efficiency of artificial super-hydrophobic surfaces, Langmuir, 2009, 25, 3240–3248

[7] Nosonovsky M., Bhushan B., Multiscale dissipative mechanisms and hierarchical surfaces: friction, superhydrophobicity, and biomimetics, springer, Heidelberg, 2008

[8] Nosonovsky M., Bhushan B., roughness-induced superhydrophobic-ity: a way to design non-adhesive surfaces, J. phys. condens. Matter, 2008, 20, 225009

[9] Nosonovsky M., Bhushan B., Biologically-inspired surfaces: broadening the scope of roughness, adv. Funct. Mater., 2008, 18, 843–855

[10] Barthlott W., Neinhuis C., Biologie in unserer Zeit, 05, (1998), s.314[11] Sepeur S., Laryea N., Goedicke S., Groß F., nanotechnology, Vincentz

net work, Hannover, 1.auflage, (2008), s. 40 ff[12] Mahn M., osterod F., Jung und Wirkungsvoll, Farbe und Lack 11, (2010)

Figure 4: comparison of the additive with the best contact angle in the tests before, the modified Fluoropolymer 1, with the competitor additives and with zero additive.

5 0


C r o s s l i n k e r s

Sour

ce: t

epto

png

- Fot

olia

.com

The besT of Two worldssilane-urethane hybrid crosslinkers create scratch-resist-ant clearcoats. By Tobias Unkelhäußer, Markus Hallack, Hans Görlitzer and rainer lomölder, evonik industr ies.

A range of silane-urethane hybrid crosslinkers has been de-signed to generate multifunctional, scratch resistant coatings usable on a variety of substrates. Addition of one such product to a 2K polyurethane automotive clearcoat greatly increased scratch resistance. An ambient-cure clearcoat applied over wood as well as plastic also showed very high scratch resistance.

A new and versatile product family of crosslinkers and binders for high performance coatings has been developed. The novel

approach combines the benefits of silane chemistry with the perfor-mance of polyurethanes, providing excellent scratch resistance while maintaining urethane properties in coatings. 3-isocyanatopropyltrimethoxysilane (IPMS) is the core building block of the novel technology platform and provides a broad freedom of design by creating tailor-made binders and crosslinkers. This resin concept also enables non-isocyanate (NISO) crosslinking technology that matches or exceeds polyurethane performance.Modern aliphatic 2K polyurethane clearcoats are considered as to-day’s benchmark for environmental etch, weathering and durability. They are in general well known for their excellent balance of flexibility and hardness. Furthermore, superb overall properties and good ad-hesion to various substrates are advantages which have helped the technology to set the benchmark in clearcoats for the automotive in-dustry in recent decades.Here, the automobile topcoat in particular, which is crucial to a car’s visual appearance, needs to have several properties each at the high-est level to satisfy the premium requirements of the car manufactur-ers and their customers.

CleArCoATs need To meeT very high requiremenTs

The clearcoat layer of a car (approximately 40 μm thick) is its first de-fence barrier against all kinds of mechanical, physical and chemical

level of Crosslinker M 95

0 % 10 % 30 % 50 % 100 %

Hardness (König) 1 d

171 174 164 162 143

erich-sen cupping (mm)

8.0 8.5 7.0 6.0 7.0

impact, direct (in-lb)

> 80 > 80 > 80 > 80 80

acid etch (20 % H2so4), first impact/de-stroyed

47 °c / 65 °c

48 °c / 67 °c

44 °c / 62 °c

40 °c / 61 °c

46 °c / 71 °c

MeK resistance ( double rubs)

> 150>

150> 150

> 150

> 150

Table 1: Properties of 2k PUr system modified with various amounts of Crosslinker M 95

5 1C r o s s l i n k e r s


Results at a glance

ű The “Vestanat EP-M” range is a family of silane-urethane hy-brid crosslinkers designed to generate multifunctional, scratch resistant coatings usable on a variety of substrates. Some will cure even at ambient temperature.

ű These crosslinkers and binders can be used in automo-tive OEM and refinish applications and on many different sub-strates such as wood, plastic and metals or glass etc.

ű Incorporation of these products into an existing 2K polyu-rethane clearcoat formulation had little effect on most physi-cal properties but produced a considerable increase in scratch resistance. High levels of scratch resistance were also obtained on an ambient-cure clearcoat applied over wood.

ű Resistance to suntan lotion of automotive clearcoats could also be improved.

ű Other modifications can also be incorporated, such as flu-orine, which was shown to produce a more hydrophobic and oleophobic surface, leading to easy-to-clean properties.

influences and thus has to resist many attacks – whether by UV radia-tion, abrasion caused by car wash brushes and dust, leaving micro-scratches, or aggressive chemicals such as acidic rain or suntan lotion causing swelling and discolouration (haze).The clearcoat needs to remain shiny, glossy and transparent for as long as possible – even after years it should still look like new. The quality of clearcoats has steadily improved, but nevertheless there is

still a need to significantly improve certain characteristics - particularly the scratch resistance - while also if possible adding more features such as an easy-to-clean effect [1].This trend is not just limited to automotive clearcoats. It is also true for ambient temperature curable clearcoats applied on other substrates such as wood or plastic, which are more heat sensitive.

CrosslinKing of silAne-ureThAne sysTems ouTlined

An appropriate way to gain reliable enhancements, in increasing the scratch resistances of clearcoats and to satisfy market needs and trends concerning a further integration of functions, is ensured by the use of silane-urethane hybrid crosslinkers.As shown in Figure 1, the IPMS that represents the key building block of the system can be converted with any kind of isocyanate-reactive groups (R-group) preferably with the hydroxyl groups of diols, polyols or oligomeric diols to build a urethane linked but alkoxy-silane func-tional non-isocyanate (NISO) crosslinker/binder.The choice of R-group will primarily determine the properties of the crosslinker and hence will also significantly influence the attributes of the coating. For example, the longer the backbone of the crosslinker the more it acts as a flexibiliser in the coating. In contrast, a branched and short R-group will result in higher hardness. It is also possible in principle to use diols with additional functionalities incorporated, such as fluorinated compounds, to create coatings which are even more multifunctional.But due to the fact that the urethane linkages are imparted by formu-lating with IPMS-based crosslinkers, the beneficial properties which are expected from aliphatic polyurethanes such as high chemical resistance, good adhesion and excellent mechanical properties, are retained.There are two possible reaction mechanisms by which the IPMS-based crosslinkers shown can react via their alkoxy groups in coating sys-tems. One reaction is a combination of hydrolysis and condensation to form siloxane linkages (Si-O-Si). The other crosslinking mechanism is a transesterification reaction which occurs only if a hydroxyl group, for example that of an acrylic or other polyol, is present. Each of these mechanisms can be accelerated by using appropriate catalysts which are extensively described in diverse publications [2, 3].

Figure 1: iPMs-based silane-urethane hybrid crosslinker containing urethane groups, terminated by tr i-functional alkoxy-silane structures

C r o s s l i n k e r s5 2


This paper describes a performance analysis of coatings which have been crosslinked using unique alkoxy-silane terminated crosslinkers which contain urethane structures within their backbone. The paper focuses on systems which facilitate the self-crosslinking mechanism of the alkoxy functionalities (Figure 2).The new silane-urethane hybrid crosslinkers were used both in stand-ard OEM-clearcoat 2K PUR formulations and in 1K self-crosslinking, moisture-cure systems.

TesT proCedures summArised

To analyse the impact of siloxane networks in standard 2K PUR sys-tems, clearcoats were formulated using an acrylic polyol, an HDI-trim-

er (hexamethylene diisocyanate) “Vestanat HT 2500 L” (PUR-hardener , NCO:OH ratio 1:1) and several levels of the silane-urethane hybrid crosslinker “Vestanat EP-M 95” (crosslinker M 95, low viscosity liquid, 100 % solids) where the R-group is a short chain, linear aliphatic diol. Also, to accelerate the crosslinking of the alkoxy groups a catalyst “Vestanat EP-CAT 11” (Cat 11, alkyl ammonium salt) was added [4].The clearcoat concept based on these products is shown in Figure 3. To provide a broader understanding of the influence of siloxane net-works on the resulting coating properties, the dosage of crosslinker M 95 was varied. In the figure, the 50:50 pbw blend refers to the ratio of the solids of acrylic polyol plus PU crosslinker and the silane-urethane hybrid crosslinker. In other words, the hybrid crosslinker was added at the same amount (on solids) as the resin plus PUR hardener. The

Figure 2: Hydrolysis-condensation reaction forming siloxane net works

Figure 3: example of a clearcoat formulation with 50:50 pbw (on solids) of resin/PUr-hardener and crosslinker M 95

Figure 4: residual gloss after modified scratch test of a 2k PUr system formulated with PUr-hardener and additional levels of crosslinker M 95



clearcoats were cured at 140 °C (oven temperature) for 22 minutes.

sCrATCh resisTAnCe shows signifiCAnT improvemenT

The compatibility of the hybrid crosslinker with other formulation con-stituents was excellent in general. In all cases high gloss clearcoats were obtained (82 ± 2 gloss units at 20 °)As shown in Table 1, all coatings based on the hybrid crosslinker show a well-balanced mechanical property profile which is satisfactory for many high end applications. When the concentration of hybrid crosslinker is increased, the hardness and flexibility of the clearcoat slightly decreases, while chemical resistance against sulfuric acid is maintained.It becomes apparent that the typical polyurethane characteristics such as outstanding hardness combined with flexibility (reference system = 0 % dosage hybrid crosslinker) were only slightly diluted by the addition of the siloxane network. But even in the 100 % system the mechanical performance is still at an excellent level, as confirmed by the direct impact test.The scratch resistance was tested by a modified crockmeter test (wet abrasion). The test performed consists of a bar with an abrasive textile brushing the coated surface in a detergent solution 160 times with a 2 kg weight attached to it to provide sufficient force. After this test, the gloss is measured again and the residual gloss is calculated in percent. This test is similar to scratch tests for automotive OEM re-quirements [5].The resistance against scratching increases dramatically with the ad-dition of hybrid crosslinker. Figure 4 shows that even small propor-tions of approximately 10 % show a significant effect. The residual gloss after the modified crockmeter test was significantly higher com-pared to the reference system (0 % dosage hybrid crosslinker).The scratch resistance differences of the formulations can be meas-ured but they are also easily visually detectable (Figure 5).

resisTAnCe To sunTAn loTion CAn Also be improved

The “100 %” formulation was also investigated in relation to other aspects of chemical resistance. Resistance against suntan lotion is known to be very relevant for automotive interior coatings but recent-ly this has also become relevant for automotive exterior clearcoats.The clearcoat formulation based on the 100 % dosage of hybrid

crosslinker was compared with a 2K PUR system with regard to resist-ance against an SPF 30 suntan lotion in a temperature gradient oven test for 30 minutes.While the standard 2K PUR coating shows swelling and discolouration (Figure 6) even at 37 °C, the boosted formulation with 100 % hybrid crosslinker remains stable even at higher temperatures up to 55 °C. The better resistance against suntan lotion can be explained by an increase in crosslink density via the dense siloxane network [5, 6].

silAne-ureThAne hybrids for moisTure Curing sysTems

Even in ambient curing systems, special silane-urethane hybrid prod-ucts can act as a crosslinker as well as a binder at the same time to create multifunctional scratch-resistant coatings. In the field of wood coatings in particular, the drying speed of the coating is an essential factor. For this special purpose the “Vestanat EP-MF” product family was designed based on products of the “Vestanat M” family but de-signed to be curable at room temperature (see Figure 7).“Vestanat EP-MF 201” (crosslinker MF 201) is a solvent and isocyanate free, ready-to-use version of an alkoxy-silane terminated crosslinker or binder requiring no other crosslinkers (i.e. polyisocyanates). It can be used for 1K moisture curing systems. The resulting coatings show outstanding scratch, stain and chemical resistance.2K formulations are obtained by combining the silane-terminated product with polyols. The right mixing ratio between both compo-nents must then be determined. Due to the formation of an inter-penetrated network, properties including scratch, stain and chemical resistance can be adjusted on demand. There is also no need to add polyisocyanate crosslinkers. The final coating is still a NISO system and – depending on the characteristics of the polyol – potentially low in VOC.Figure 8 shows a technology comparison between a 2K PUR (left) and a 2K formulation based on use of the silane-terminated crosslinker-binder (right). Both systems were applied on wood and were scratched using a modified crockmeter test. While many and very distinctive scratches occur on the unmodified 2K PUR, the formulation based on silane crosslinking resists the abrasive attacks keeping its shiny, initial appearance.In addition to the field of wood coatings, there is also an attractive application in terms of plastic coatings. Here the same properties will lead to scratch, stain and chemical resistant coatings. This outstand-

Figure 5: Modified crockmeter test: 2k PUr (left) vs. 2k PUr formulated with 100 % addition of hybrid crosslinker (r ight)

Figure 6: 2k PUr with additional dosages of crosslinker M 95 (left - less structured surface at 55 °C) vs. 2k PUr (r ight - highly structured surface due to less suntan resistance at 55 °C)



ing product profile is essential for sporting goods or in the automotive industry.

fluorine inCorporATion enhAnCes CleAnAbiliTy

The backbone of the novel crosslinker technology is variable, which leads automatically to a high freedom of design for creating tailor-made crosslinkers or binder with multifunctional qualities in the fu-ture.An experimental (not currently commercially available) fluorine-con-taining silane-urethane hybrid hardener was investigated. Fluorine is well known for its positive effects on surface properties, such as the improvement of water and oil repellency of coatings, also known as the easy-to-clean effect [7]. One indirect but valid method for detect-ing the hydrophobicity of surfaces is to measure the contact angle of a drop of liquid onto that surface. In this case water and squalene, a triterpene which can be obtained from shark liver oil, were used as liquids.Clearcoats were formulated both with the “Vestanat® EP-M 95” (refer-ence) and with several levels of the experimental fluorine containing silane-urethane hybrid crosslinker. The coatings were cured at 140 °C (oven temperature) for 22 minutes.The results obtained, shown in Table 2, demonstrate that even small proportions of a silane-urethane crosslinker containing fluorine can

enhance the easy-to-clean behaviour of a coating significantly. It was also noted that all clearcoats tested had in general a high abrasion resistance. The only deficit observed was a continuing decrease of hydrophobicity after several scratch test cycles.

referenCes

[1] Roths K., Gochermann J., Forschungsagenda oberfläche, 1. auflage, DFo, neuss, 2007, 86.[2] Lomölder R., Görlitzer H., Hallack M. et al., coating agent With High scratch resistance, patent application ep 2 676 982, 2012.[3] Witucki G.: J. coating. technol. 1993, Vol. 65, pp 57-60.[4] Lomölder R., Raukamp A., Nauman S. et al., adducts of isocyana toalkyl trialkoxysilanes and aliphatic, alkyl- branched Diols or polyols, patent application ep 2 641 925, us 20130244043, 2012.[5] Noh S., Lee J.W., Nam J.H. et al., prog. in org. coat. 2012, Vol. 74, pp 192-203.[6] Plueddemann E. P.: silane coupling agents, plenum press, new York, 1982, pp 141-181.[7] Gururaj N. M., Arvind R. S., Ramanand N.J. et al., prog. in org. coat. 2012, Vol. 75, pp 139-146.

Crosslinker M95 content

Fluorine crosslinker

Contact angle

Water squalene

100 % 94 33

95 % 5 % 109 72

85 % 15 % 113 72

Table 2: easy-to-clean effect of silane-urethane hybrid crosslinkers containing fluorine

Figure 7: silane-urethane-hybrid crosslinkers based on iPMs designed for creating multifunctional, scratch resistant coatings in different systems and curing conditions

Figure 8: Modified crockmeter scratch test : 2k PUr (left) vs. 2k based crosslinker MF 201 formulation (r ight)

Sour

ce: I

nPix

Kom

mun

ikat

ion

- Fot

olia

.com

Shiny SeaSon

The reflective, mirror-like properties of christmas tree ornaments are usually gai-ned by a silvering solution used as a coating on the inside. On the outside, a co-lourful lacquer over a white basecoat is applied.

Shaping Up

Traditionally, glass orna-ments were blown by hand over a bunsen burner.Nowadays, most factory-produced ornaments are shaped by compressed air blown in the mold.


5 5W o r l d o f C o l o u r

5 6


H a r d e n e r

Sour

ce: Y

ou c

an m

ore

- Fot

olia

.com

Standing on perfor-manceHigh performance waterborne hardener for zero VOC self-levelling floors. By Shuyuan Liu and Tushar Tr ivedi, Incorez Ltd.

a new water-soluble epoxy hardener for epoxy-based floor coatings has been developed by using a controlled molecu-lar architecture. the hardener is Voc-free with very low free amine monomer levels. advantages include excellent work-ability and mechanical properties, low shrinkage, long pot life yet fast cure. Highly breathable coatings permit application onto damp and green concrete.

two-pack amine-epoxy resins are widely used in applications rang-ing from coatings, adhesives and composites to civil engineering

applications such as formulations for concrete flooring, self-levelling flooring and grouts. Despite recent advances in this technology, there is a constant need for improvement in order to meet ever more strin-gent legislation and customer demands for better performance at lower cost.Self-levelling floors are an important application for epoxide coatings in that they provide relatively thin coatings of high hardness and wear resistance. These systems are applied to floor areas that demand exceptional performance and hygiene such as ‘clean rooms’, baker-ies, dairies, food processing areas, pharmaceutical production areas, motor showrooms, workshops etc. They are easy to apply by simply spreading the compound to the required thickness with a trowel.

BenefitS and limitationS of waterBorne epoxieS

Waterborne and solvent-free epoxy technology is particularly attrac-tive in floor coating applications for environmental reasons. Water-based epoxies are normally two-component systems consisting of a liquid epoxy resin and a water-based amine-functional curing agent. The properties of the amine functional curing agent are vital to coat-ing performance.Many waterborne epoxy systems are relatively low in viscosity and

thus are easy to apply and can level into a smooth film or layer after application. Water-based curing agents can eliminate or reduce vola-tile organic compounds (VOCs) in the formulations, and so reduce the odour and environmental and health risks associated with the pres-ence of such compounds.However, most amine-based hardeners have strong unpleasant amine odours and eliminating solvent alone cannot solve such a prob-lem. It is therefore desirable that the hardener itself should have a low or zero odour.Despite recent advances in WB epoxy technology, there remains a need for epoxy hardening systems and water-dispersible curing agents for epoxy resin formulations which are easy to mix, low odour, have a suitable pot life (in the range of up to 1 hour), rapid cure (even at low temperatures) and provide high hardness after curing in order

Figure 1: architectures of t wo epoxy amine adducts

5 7H a r d e n e r


Results at a glance

ű Epoxy-based technology for coating applications such as self-levelling flooring faces great challenges in combining VOC compliance, high performance, low odour and low cost.

ű A new water-soluble epoxy hardener has been developed to meet technological, economical and legislative requirements in concrete floor protection.

ű The hardener is VOC-free with very low levels of free amine monomer, allowing low hazard zero emission formulations to be produced at costs equal to or below those of existing WB epoxy hardener systems.

ű Technical advantages include very good workability, long pot life, fast cure and a very low level of shrinkage combined with very good mechanical properties. Very good breathability allows for use over damp and green concrete without blistering or delamination due to osmosis.

to provide a fast return to service.In addition, the final coatings should exhibit low shrinkage and low or no colour so that they can be used for transparent coatings without changing the base colour of the material to which they are applied. Finally, they should preferably provide good adhesion on concrete, good impact resistance, good water vapour transmission and excel-lent compressive strength.One of the biggest challenges of water-based epoxy hardeners for high build applications is shrinkage of the cured product. High shrink-age would have many adverse effects, such as gaps/cracks around the edges of the applied floor and even delamination from the surface or substrate.High tension within the network of a high shrinkage system normally leads to poor compressive strength and modulus. This also results

in low porosity and therefore poor water vapour transmission, which would lead to delamination due to the osmosis effect from damp ground.

tHe root cauSe of SHrinkage on curing

Shrinkage is caused by high tension in a cured network which has a high level of internal energy. The architectural design of the hard-ener is critical to controlling the level of shrinkage. Poor distribution of crosslinking or branching sites can lead to high localised tension in the polymer network. This is increasingly a problem in high build systems such as self-levelling formulations where the binder content is normally low.This kind of resin dilution means there is more ‘space’ between the epoxy and hardener. The evaporation of volatile components, in this case water, is relatively slow compared to thin-film coating applica-tions. Therefore, the space between the resin molecules remains little changed during cure. As a result, the size and position of the binding sites of the curing agent in relation to the size and structure of epoxy resins affects the level of tension generated from the coupling be-tween amine and epoxy groups.Further to this, the compatibility of the whole system can be another important factor in causing shrinkage. Phase separation and folding of the molecules would result in shrinkage due to progressive diffusion and migration of binder molecules during the curing process.

molecular engineering deSign of an amine/epoxy adduct

Epoxy curing agents can be made by forming amine epoxy adducts. Several factors are crucial to performance. In a very simplified scheme shown in Figure 1, the amine epoxy adduct can show distinctly dif-ferent architectures in terms of the type of building blocks and the placement of reactive amine groups. The dark green arrow blocks represent amine functional groups and the red ‘inverted’ arrow blocks epoxide groups. Yellow, blue and green bars represent building blocks between the functional groups.In Adduct A the amine functional groups are located close together around some points of the adduct backbone. When mixed in a two-pack formulation, epoxy molecules would be dispersed evenly. Only those epoxy groups close to amine functional sites can react, leaving unreachable amine and epoxy groups ‘starved’. The first impact would be slow cure as a result of starvation. The second impact would be shrinkage caused by the pulling effect when these functional groups have to move to allow coupling between amine and epoxy groups.

Property Value

amine value (mg KoH/g) 220 - 240

active hydrogen eq. weight (g/kg) 200

addition level (phr) 95 - 105

typical viscosity at 20 °c (mpa.s) 8000 – 12000

Density (g/cm3) 1.05 - 1.10

colour (gardner) <1

solids (%) 55 ± 2

Flash point (°c) >100

Table 1: Typical properties of the new epoxy hardenerFigure 2: a hardener sample (left) and the cured disc and str ip of self-levelling compound

H a r d e n e r5 8


In the case of Adduct B where the amine groups are spread evenly across the adduct backbone, the curing process and the structure of the cured network can be quite different. When dispersed evenly in a 2K self-leveller formulation, epoxy and amine groups can react easily without having to migrate, as all amine functional groups are equally accessible. This leads to relatively fast cure, a low tension network and reduced shrinkage.Clearly the architecture of Adduct B is favoured in terms of its potential for rapid cure, low shrinkage, high breathability and good mechanical properties. Meanwhile, the physical properties of the building blocks on the hardener backbone play an important role in compatibility and therefore, have a significant influence on the performance as well as shrinkage in a water-based two pack epoxy formulation.Accordingly, a new waterborne epoxy amine adduct curing agent has been developed which has been specifically engineered to have a de-sirable architecture with relatively low viscosity and low colour.

enVironmental and application adVantageS

The new hardener is a pale yellow solution (Figure 2) with a relatively

low viscosity (Table 1) and sweet, pleasant odour. The low viscosity makes general handling easier for applicators and also reduces wast-age in the drum or tank. The hardener is made without any solvent or plasticiser, so is itself zero VOC.This has been confirmed through a micro-chamber emission test on a self-levelling formulation (Table 2) at 2 mm thickness. The result (Table 3) shows minimal emission levels and is classified as zero emission un-der AgBB/DIBt (German restrictions for VOC emissions for construc-tion products).The synthesis process for the new curing agent has been optimised to consume all of the starting amine compounds. The level of free (low molecular weight) amine is not quantifiable by standard GC and HPLC analytical methods. Low hazard formulations are critical for applica-tions for schools, hospitals, food processing, homes and offices where the health and safety of people around is of high concern.In a typical self-levelling epoxy formulation (Table 2), the hardener ex-hibits excellent workability and flow characteristics as well as excellent surface appearance with matt finish and minimal pinholes. In Figure 2 (right) the cured self-levelling disc sample was made by pouring the compound directly onto a polypropylene sheet and allowing it to flow without trowelling. The smooth, even surface and thickness of the disc clearly demonstrate excellent flow behaviour.

faSt cure and low SHrinkage

As previously explained, the favourable hardener architecture and ex-cellent compatibility with epoxy facilitate fast cure on application (typi-cally at 2-3 mm thickness) even at relatively low temperatures (15 °C, 50 % RH). Under standard conditions the hardness can reach Shore D

Figure 3: debonding of impermeable coating caused by moisture in substrate (left), but no debonding for semi-permeable coating (right)

Part a Product Supplier amount

curing agent “incorez 148-700” incorez 10

Defoamer “Byk 1615” Byk 0.8

Diluent Water 4

pigment “tiona 595” cristal global 3.8

Filler “eWo” sachtleben 13

Filler Quartz powder M6 sibelco 27

Filler Quartz sand (0.1-0.3 mm)

35

thixotropic agent “Deuteron Vt 819” (3 % aq)

Deuteron 0.4

Diluent Water 6

total 100

Part B

epoxy resin “Der 358 (180)” Dow 9

total 109

Table 2: Two-pack epoxy self-levelling floor formulation

emission tests 3 days 28 days

results µg/m3

agBB requirements

mg/m3

results µg/m3

agBB requirements

mg/m3

tVoc (c6-c16) 188 0 ≤ 10 7.2 0 ≤ 1.0

Σ SVOC (C16-C22) 0 none 0 0 ≤ 0.1

r dimensionless 0.439 none 0.003 0 ≤ 1

total Voc 11.2 none 0 0 ≤ 0.1

total carcinogens 0 0.0 ≤ 0.01 0 0.000 ≤ 0.001

Table 3: emission evaluation according to agBB/dIBt (note the different units in test and requirement columns)

5 9H a r d e n e r


60 in 24 hours and 70 within 48 hours.This is particularly beneficial for applicators, as the floor can receive heavy loads soon after applying, minimising any delays during con-struction or refurbishment work, thus increasing productivity for the end user. In addition to this, the smart hardener architecture means minimal shrinkage can be achieved, typically below 1 %.It can be seen in Figure 2 that after full cure the self-leveller sample made on a polypropylene sheet shows minimal curling, indicating a low tension network. This was also demonstrated using a cured self-leveller strip cast from a 435 mm long silicon mould – no bending or curling, again illustrating very little shrinkage after full cure. This will assist long term adhesion and prevent delamination, so ensuring the long service life of the flooring system.

pHySical and cHemical propertieS SummariSed

Water vapour diffusion rate through a sample of cured

product (using the formulation in Table 2) has been test-ed according to EN ISO 7783-1/-2. At a film thickness of 2 mm the water vapour diffusion rate is measured at 2.85 g/(m2.d) corresponding to an equivalent air layer thickness of 8.2 m, which indicates good breathability of the coating.It is well known [1-8] that moisture in concrete can cause blistering and delamination of coatings that are water vapour impermeable (Fig-ure 2) by osmotic pressure build-up from water beneath the coating trying to evaporate. Good water vapour permeability can therefore prevent debonding of coatings from damp concrete substrates (Figure 3), prolonging service life.Table 4 shows some physical properties of the formulation in Table 1 after full cure (23 °C and 50 % relative humidity for 28 days). These properties are typical for this type of system and meet the require-ments for wide range of applications for concrete protection.As industrial coatings can be subjected to a variety of conditions and chemicals during their service life, a sample of the cured material also underwent a 24 hour spot test with a range of chemicals and stains. Figure 4 shows a comparison between the new hardener and a com-mercial equivalent. The resistance was rated according to the level of residual marks from heavily damaged, marked as 0, to no effect at all as 5. It can be seen that the new hardener system offers the better performance.

all-round performance BenefitS

This summary shows that the architectural design of a new amine epoxy hardener by way of molecular engineering is able to meet in-creasingly stringent market demands on legislation compliance, opera-tor and environmental impact, and high productivity whilst showing an excellent cost-performance balance in self levelling epoxy systems.

referenceS

[1] Guenter M., Hilsdorf H. K., stresses due to physical and chemical actions in polymer coatings on a concrete substrate, proc. adhesion bet ween polymers and concrete, chapman and Hall, London, 1986, pp 8-21.

[2] Raupach M., Wolff L., Durability of adhesion of epoxy coatings on con-crete, causes of delamination and blistering, concrete repair, rehabilita-tion and retrofitting ii, 2009, pp 921-928.

[3] Warlow W. J., Pye P. W., osmosis as a cause of blistering of in-situ resin floorings on wet concrete, Mag. concrete research, 1978, Vol. 30, no. 104, pp 152-156.

[4] Wisser S., Behaviour of solvent-free epoxy resin systems on water absorbent concrete surfaces, Kunstharz nachrichten (Hoechst).

[5] Defects in epoxy resin flooring, Bre technical File no. 14, July 1986, pp 32-34.

[6] Pye P. W., Blistering of in situ thermosetting resin floorings by osmosis, international colloquium, ‘industrial Floors’, esslingen, 1995.

[7] Stenner R., Magner J., influence of moisture from substrate to blistering, international conference on polymers in construction, 1995.

[8] Pfaff F. A., Gelfant F. S., osmotic blistering of epoxy coatings on concrete, JpcL, Dec. 1997, pp 52-64.

Figure 4: Chemical and stain resistance test, in graphical form and visually, where Sample a represents the new curing agent

Test results

Bond strength by pull-off-test (en 1542) 220 - 240

Bond strength by pull-off-test (en 1542 200

Bond strength by pull-off-test (en 1542) 95 - 105

typical viscosity at 20 °c (mpa.s) 8000 – 12000

Density (g/cm3) 1.05 - 1.10

colour (gardner) <1

solids (%) 55 ± 2

Flash point (°c) >100

Table 4: Physical properties of fully cured self-levelling epoxy system

6 0


L e v e L L i n g A g e n t

Primed for imProvementHydrophilic agent enhances levelling and intercoat adhesion. By Michael Bessel, guillaume Jaunky and Mark Heekeren, Byk-Chemie.

Acrylate levelling agents have little effect on the surface ener-gy of coatings, while silicone or fluoro-based additives reduce it. However, a high surface energy is desirable in primers to increase intercoat adhesion. A new branched polyacrylate lev-elling additive designed to increase surface energy has shown positive results in waterborne automotive primer surfacers.

m any consumer goods used in our daily life are coated to achieve a variety of features and beneficial effects. For instance, pack-

aging coatings protect foodstuffs from exposure to oxygen and the metal of a can from aggressive acids in beverages or tinned fruit. Pa-per coatings on the pages of a magazine help to protect the print-ed art and allow the pages to be easily separated. Wooden floors are coated to withstand humidity or to achieve scratch and stain resistance.In the automotive sector, excellent coating performance in terms of durability, scratch resistance, levelling and layer-to-layer adhesion is sought after. An OEM coating consists of a cathodic electro-deposi-tion (CED) coating on a metal substrate for protection against corro-sion, then a primer surfacer layer followed by a pigmented basecoat and clear coat. The preparation of each of the coating layers requires the use of a variety of additives for good wetting of the substrate, sta-bilisation of pigments or surface features such as carwash resistance.

tHe imPortAnce of surfAce energy in coAtings

Levelling additives are an inherent component of many coating for-mulations, helping to minimise differences in surface tension during the curing process and provide a smooth surface appearance.A standard polyacrylate levelling additive has a polymer backbone with a linear molecular structure (see Figure 1). It has a relatively low Tg and is slightly incompatible with the coating matrix. The polyacrylate concen-trates on the surface of a wet film and keeps the surface tension of the coating constant during the curing process.In order to obtain low surface energy of the coating or to coat contam-inated substrates that exhibit areas of low surface energy, additives based on modified polydimethylsiloxanes or fluoro-based chemistry are used (see Figure 1). Low surface energy is often desirable in coat-ings when water repellency, water resistance, a low coefficient of fric-tion and easy-to-clean properties are required.Due to regulatory restrains and environmental considerations, a trend towards developing more waterborne coatings has been observable over the last few decades in the automotive industry. In general, an automotive coating system consists of three to four layers, as mentioned above.Within such a multilayer coating system, low surface energy can have an adverse impact on the recoatability of a coating in relation to the other layers. Therefore, the first coating layer on the substrate must have the highest possible surface energy and, at the same time, a surface ten-

Sour

ce: C

ytec

6 1L e v e L L i n g A g e n t


Results at a glance

ű Levelling additives are widely used in coatings to improve flow during cure and create a smoother surface. Standard poly-acrylates have little effect on surface energy, while siloxane or fluoro-based products reduce it.

ű While low surface energy is desirable in finishing coats to improve water repellency and reduce surface friction, it can in-terfere with recoatability.

ű A branched polyacrylate additive has now been designed to increase the surface energy of cured primer coats and improve application of subsequent coats.

ű While primarily designed for use in waterborne systems, it can also be effective in any other coating with which it is slightly incompatible. Optimum dosage is affected by a number of vari-ables including curing temperature.

ű It was confirmed that in a waterborne primer surfacer formulation the new additive provided better levelling than a standard polyacrylate while also improving wetting by the subsequent basecoat.

sion that is low enough to allow good wetting of the substrate. It is for this reason that a polyacrylate levelling agent that significantly increases the surface energy of cured coatings and paints has been developed.

structure of tHe HydroPHilic surfAce Additive

A polyacrylate levelling additive modified with certain polyether grafts has been developed to increase the surface energy of cured coatings. A special chain transfer polymerisation technology was applied to cre-ate a branched polyacrylate backbone with very hydrophilic polyether graft structures, as shown in Figure 1.The polyacrylate backbone is slightly incompatible with common coat-ing matrices and this, with its branched structure, drives the polymer to migrate to the surface. The polyether structures are responsible for the increase in surface energy of the cured coating. The hydrophilic polyether side chains can exhibit a linear or hyperbranched structure to reduce the tendency of the additive to crystallisation.The polyether side chains are incorporated into the polymer back-bone in the form of polyether macromonomers (or macromers for short). These are specific polymeric structures that contain one reac-tive group at one chain end. This reactive group can be an acrylate, methacrylate or vinyl group, for instance, and can be subjected to radical polymerisation reactions with other conventional monomers.This branched polyether-modified polyacrylate levelling additive (also referred to below as a hydrophilic additive) is an excellent levelling agent which provides its distinctive properties in a broad range of coatings. In order to achieve the hydrophilic surface effect, the additive has to be slightly incompatible with the liquid coating matrix in which it is used (see Figure 2). Therefore, the hydrophilic additive is most effec-tive in waterborne paints and coatings and in coatings with a very non-polar formulation.

Figure 1: Structures and features of standard levelling additives compared with the new product

testing tHe effects of Additives on surfAce energy

A waterborne primer surfacer 1K OEM stoving system based on “Seta-qua B E 270” (a saturated polyester resin) and “Bayhytherm 3246 XP” (an aliphatic self-crosslinking urethane resin) was chosen as a refer-ence formulation and test results are shown in Table 1. The version of the additive tested is the one sold as “Byk-3560”. The surface energy of the cured coating was determined from the contact angles with five standard liquids: water, glycerine, ethylene glycol, n-octanol and n-dodecane. The polar and disperse part of the coating’s surface energy were also calculated according to the Owens-Wendt-Rabel-Kaelble method.As shown in Table 1, standard polyacrylate additives do not influence the surface energy of the cured coating (see entries 2 and 3). However,

L e v e L L i n g A g e n t6 2


Figure 2: images of water drops on different cured films modified with an hydrophilic additive: the hydrophilic effect is obser ved when the additive is slightly incompatible with the coating

Figure 3: Optimum dosage of the hydrophilic additive in a primer surfacer test formulation is reached at 0.3 wt % on total

Figure 4: Relationship bet ween the contact angle of water and curing temperatures of the test primer surfacer formulation

6 3L e v e L L i n g A g e n t


the total surface energy of the film was increased to 58.5 mN/m by ap-plying 0.3 wt % of the hydrophilic additive (see entry 5). The disperse part of the coating’s surface energy was reduced by around 50 % to 10.9 mN/m compared to the control without additive. In contrast, the polar part of the surface energy was greatly increased to 47.6 mN/m.The hydrophilic effect provided by the new hydrophilic additive is always accompanied by an increase in the polar part of the surface energy and a decrease in the disperse part, both to varying extents. It should be noted that the absolute values of surface energy obtained vary with the type of coating, the polarity of its formulation, its curing procedure and the amount of additive used.It is important to add that the hydrophilic additive does not influence the surface tension of the liquid film or the wetting performance of the liquid coating on the substrate.

severAl fActors Affect oPtimum Additive concentrAtion

As mentioned above, the concentration of the hydrophilic additive in-fluences the absolute values of the cured coating’s surface energy. Figure 3 shows a series of increasing dosages of the additive in a wa-terborne primer surfacer coating. The optimum effective concentra-tion was obtained using 0.3 wt % hydrophilic additive on total formula-tion, which thus corresponds to saturation.In many solvent-based paints, a higher dosage of around 1.5 wt % of the hydrophilic additive is needed to provide the best results with respect to the increase in surface energy.The activity of the hydrophilic additive has been found to be influ-enced by the curing temperature, as in the case of the reference primer surfacer formulation. The curing procedure was designed as follows: after application of the film on CED pre-treated panels a flash-

off time of 10 minutes at room temperature was applied. Afterwards, the films were cured for 10 minutes at 80 °C and then for 20 minutes at a temperature between 120 °C and 180 °C.As shown in Figure 4, the hydrophilic effect obtained with the additive reaches its maximum at 0.3 % dosage at 160 °C, which is also the optimum recommended curing temperature for the primer surfacer formulation. This implies that a lower curing temperature requires a higher amount of the hydrophilic additive. Conversely, it is assumed that both the coating and the additive suffer from degradation at 180 °C and long exposure times of 20 minutes.The hydrophilic effect was found to be virtually independent of dry film thickness. The contact angle with water changed only from 32° to 35° as film thickness increased from 10 µm to 45 µm. With no ad-ditive, the contact angle was also essentially constant at 87-88°. The optimum curing conditions for the waterborne formulation were cho-sen based on the results shown in Figure 4: 10 min. flash-off time at room temperature after application with a wire bar, 10 min. at 80 °C and 20 min. at 160 °C.The hydrophilic effect has also been observed to be dependent on the substrate. As shown in Table 2, the contact angle of water on the cured coating indicates a stronger hydrophilic effect on the CED panel compared to a glass panel. It is assumed that the hydrophilic addi-tive interacts with the glass surface rather than floating to the air/film interface.

good wetting confirmed in A wAterborne coAting system

The performance of the hydrophilic additive was then investigated in a waterborne polyester-melamine primer surfacer formulation. Op-timised levelling, as demonstrated by a significant reduction of Wa

no Waterborne formulation based on “Setaqua B e 270/

Bayhytherm 3246 XP”

Calculated surface energy of cured coating in mn/m

substrate Dosage of hydrophilic

additive

contact angle of

water in °

surface energy

Dis-perse

part

polar part

1 glass panel

no additive 87 27.0 22.0 5.0

2 glass panel

0.2 wt % on total

78 27.0 18.9 8.1

7 ceD panel

no additive 88 26.8 21.2 5.6

8 ceD panel

0.2 wt % on total

36 46.6 11.1 35.5

table 2: How the substrate modifies the performance of the hydrophilic additive

no eStA application, addi-tive concentrations ap-plied as recommended

Contact angle in °

Calculated surface energy of cured coating in mn/m

Waterborne coating based on “setaqua B

270” and “Bayhytherm 3246 Xp”

Water surface energy

Disperse part

polar part

1 control without additive 89 27.4 21.8 5.6

2 0.6 % polyether-modi-fied polyacrylate

89 25.0 21.4 3.6

3 0.6 % polyacrylate with polar modifications

88 26.5 22.5 4.0

4 1.2 % silicone-free surfactant

84 26.9 21.5 5.4

5 0.3 % hydrophilic additive

16 58.5 10.9 47.6

table 1: Surface energy of a 1K OeM primer-surfacer calculated according to the Owens-Wendt-Rabel-Kaelble method from con-tact angles of five standard liquids determined with a Krüss “g2” instrument

L e v e L L i n g A g e n t6 4


and Wb surface smoothness values in comparison to standard poly-acrylates, was achieved in horizontal and vertical electrostatic spray applications (ESTA) with the use of 0.2 wt % of the hydrophilic additive (see Figure 5). In addition to the very good levelling performance, the hydrophilic ad-ditive increased the surface energy of the primer surfacer layer. In comparison to a standard polyacrylate for waterborne formulations, the use of 0.2 wt % of the additive increased surface energy of the cured coating by 18 mN/m (see Figure 6).It was then confirmed that high surface energy makes for better wet-ting of the primer surfacer layer by a waterborne basecoat. A (red) pigmented basecoat was applied to a panel coated with the additive-modified primer surfacer (Figure 7). The thickness of the basecoat was increased from 0 µm at the top end to 25 µm at the bottom end of the panel. The panel coated with the additive-modified primer surfacer exhibits much better wetting and spreading with the red waterborne OEM basecoat at low film thicknesses than the control panel without additives.

Figure 6: Compared to a standard polyacrylate for waterborne systems, the hydrophilic additive increases the surface energy of the cured coating (poly ester-melamine primer-surfacer)

Figure 7: the red OeM basecoat shows much better wetting on the primer surfacer modified with the hydrophilic additive than on the control panel

new Additive HAs A broAd rAnge of APPlicAtions

The increasing usage of waterborne formulations requires new solu-tions for levelling and surface additives, especially in multilayer coat-ing applications. The newly developed hydrophilic additive exhibits outstanding levelling performance and very good compatibility.In addition, this additive increases the surface energy of a cured coat-ing and therefore improves the wettability and recoatability of the film. Its use is not limited to waterborne formulations; it has also been successfully applied in a number of solvent-based paints, UV-curable systems and powder coatings. However, a slight incompatibility of the additive in the paint is required to observe the hydrophilic effect.The uses of the hydrophilic additive are likely to be expanded into much wider fields of application than those currently investigated.

Figure 5: Levelling performance of the hydrophilic additive used in a waterborne poly-ester-melamine primer surface on horizontally and vertically coated panels (determined with a Byk “Wave-scan dual”)

Sour

ce: p

lato

ngko

h55

- Fot

olia

.com

Coloured Trees

50,000 litres of protecti-ve coatings were used on Singapore‘s Gardens by the Bay. This attraction consists of the world‘s largest clima-te-controlled glass houses as well as 18 of the pictured supertrees.

MIMICkIng naTure

Not only the consevatories of Gardens by the Bay but also the trees are home to many tropical plants. They provide air intake and mimic the ecological functions of real trees.

6 5W o r l d o f C o l o u r


6 6


S l i p a d d i t i v e S

Sour

ce: M

ario

Bea

ureg

ard

- Fot

olia

.com

Smoothing the Surfaceperformance of structurally different slip agents compared. By Juan Carlos Corcuera, Kim Quackenbush and Ogul arseven, dow Corning europe Sa

additives that improve surface slip properties have many ben-efits in different applications. tests are reported on various silicone and wax slip additives. a high molecular weight PDmS material, supplied as an emulsion in special surfactants, is shown to be very effective at low dosages. in some cases, com-bining it with waxes may be beneficial.

Silicone polyether technology has been available to the formula-tor to enhance wetting, levelling and slip performance for some

years. The typically low molecular weight and high hydrophilicity - due to polyoxyalkylene side chains - give excellent compatibility in coat-ings, paints or inks, but performance can vary significantly from one formulation to another.These materials lack the very high degree of slip performance that higher molecular weight silicones can bring, and the resulting hand-feel is not luxurious or valued by the end consumer. This paper will show how moving to higher molecular weight (MWt) sili-cone technology will allow the formulator to improve slip, hand-feel and abrasion resistance to a degree that the end customer will appreciate.

Structure of Different tyPeS of Silicone aDDitiveS

In this study a variety of silicone based additives were tested in paints, inks and coatings formulations, evaluating the impact of the molecular

weight and chemistry of the silicone on the final performance prop-erties of the cured films. The degree to which the molecular weight affects the slip (in terms of the Coefficient of Friction, CoF), hand-feel and scratch resistance of the cured film was evaluated, together with any potential adverse impacts the different chemistries may have on surface levelling, wetting and recoatability. The active silicone families within this study differ in their chemistry. Type 1 materials are polydimethyl siloxanes (PDMS) that have high to very high molecular weight. These are delivered as low viscosity emulsions, meaning they are easy to incorporate despite the relatively high viscosity of the active component. Type 2 are based on silicone polyethers (SPE) and are classed as having low to medium molecular weight. Typical commercial products on the market today are classed as having low molecular weight. Typical structures of these materials are illustrated in Figure 1.The SPE materials differ from the PDMS based materials in two ways: ą The degree of hydrophobicity. Ethylene and/or propylene oxide

Figure 1: typical molecular structures of polydimethylsiloxane (pdMS) (top) and silicone polyether co-polymer (Spe) (bottom)

Figure 2: Friction reduction in acrylic based flexographic ink with different MWt silicone additives

6 7S l i p a d d i t i v e S


Results at a glance

ű Additives are widely used in the coatings and inks industries to improve surface slip properties. The reduced coefficient of friction (CoF) allows for faster processing and protects coating surfaces from mechanical damage during processing, transport and use. Tactile properties may also be improved.

ű Tests are reported on a number of silicone and wax ad-ditives designed to improve slip properties. High molecular weight polydimethylsiloxane (PDMS) materials are not normally used, as they are hard to incorporate and may cause surface defects. However, a material of this type has been developed using special surfactants to avoid compatibility problems.

ű In a range of tests, the high MWt PDMS additive was shown to provide better slip (lower friction) properties at lower addi-tion rates than the more widely used silicone polyethers and waxes.

ű In some cases, further benefits could be obtained by com-bining waxes with this additive.

groups on the silicone backbone impart some hydrophilic charac-ter that typically improves their compatibility.

ą The degree of polymerisation is typically far lower in SPE chemistry than in PDMS based chemistry and as a result the viscosity of the active silicone material is orders of magnitude lower than that of the PDMS based materials.

The use of pure PDMS-based materials is practically zero in paints, coatings and inks due to their incompatibility with organic resins and their tendency to cause surface craters. For the PDMS materials eval-uated in this study, the delivery system is also critical to their perfor-mance. The compatibility of these high MWt PDMS-based emulsions is dependent on the use of a non-traditional surfactant technology that

acts as a compatibiliser in the coating or ink.A number of wax additives, used to obtain similar property improve-ments to silicones, were also evaluated. The full list of materials tested can be found in Table 1.

the critical imPortance of SliP ProPertieS

Numerous coatings applications require a reduction in the coefficient of friction (CoF). In the furniture coatings market, for example, an im-provement in slip properties can improve resistance to damage during international shipment, construction and in use. In the printing indus-try, slip allows for faster production speeds, insurance against paper jams and of course, sensory benefits for the end user. The molecular weight of the silicone additive can vastly impact its slip performance, and this is seen consistently across multiple coating and ink formulations. Figure 2 shows the static and dynamic CoF of a waterborne acrylic based flexographic ink with some of the different silicone additives listed in Table 1. This shows a direct relationship between increased silicone MWt and higher slip performance.The two CoF values for the control without any additives are 0.865 and 0.803. While these are re-duced by all the silicone additives, the highest reduction by far is seen with the silicone emulsion additive, with values of 0.167 and 0.117 respectively with 0.5 % dosage.Figure 3 shows the static and dynamic CoF of a waterborne acrylic based wood coating with two different dosages of silicone polyethers and silicone emulsion, 0.2 % and 0.4 %. Again the biggest reduction in CoF is with the silicone emulsion additive, so a lower dosage could be used to achieve the same desired slip.Figure 4 shows the impact on the dynamic CoF of an overprint varnish formulation and the effects of dose rate. This is quite significant up to 0.5 %, then increasing to 1 % the reduction is not as significant as before. Experience suggests that 1 % should be the highest dosage if the additive is well dispersed in the formulation, having on average great benefits be-tween 0.1 % and 0.5 % in this type of formulation, inks and wood coatings.

SiliconeS can be more coSt-effective than waxeS

In many instances waxes are used to lower the coefficient of friction. Whilst these can appear efficient on a price per kg basis, this compari-

Figure 3: acrylic based wood coating with different MWt silicone additives

S l i p a d d i t i v e S6 8


hand-feel’, having the control as reference with a value of zero. The 2 % HDPE wax emulsion sample was also rated as zero. It can be seen in Figure 6 that all samples with 0.1 % of the silicone emulsion perform significantly better for a silky hand-feel than the samples with 2 % wax emulsion alone.The impacts on gloss of a combination of additives on this high gloss clear coat, measuring the gloss at an angle of 20°, are shown in Figure 7. The silicone emulsion at the low recommended use level of 0.1 % has little impact on gloss. The dots in blue indicate the samples con-taining the silicone emulsion. In some cases, the silicone emulsion + wax combinations (PE wax emulsion (1), paraffin wax emulsion, and HDPE wax emulsion (1)) have higher gloss values than the sample containing silicone emulsion

Figure 4: the dynamic coefficient of f r iction shows the impact of silicone emulsion on slip in a waterborne acrylic based overprint varnish

son often looks quite different when these materials are compared to silicone-based materials on a cost-in-use basis. In Figure 5 the CoF of a waterborne acrylic wood coating is evaluated when using synthetic waxes versus competitor’s grades of silicone polyethers and the high MWt silicone emulsion additive in waterborne acrylic based wood coating formulation.Using the axis on the left hand side of the graph, the green line shows the dynamic coefficient of friction. Abrasion resistance was measured by change in gloss. Using the axis on the right hand side of the graph, the blue and red bars on the graph show the initial gloss and the gloss after 1400 cycles in a reciprocating abraser using felt pads with a weight of 15 N. The calculated percentage reduction in gloss is re-corded at the top of the two bars.In this case the emulsion has by far the biggest impact on CoF reduc-tion compared to silicone polyethers and waxes, at vastly reduced ad-dition levels by weight of the total coatings formulation: 3.5 % against 0.1 % if slip values are compared. So the highest MWt silicone emul-sion has biggest impact in CoF reduction with a superior slip com-pared to waxes, and to achieve this, the addition level required for the silicone emulsion is very low.

Synergy benefitS of combining Silicone with waxeS

The effects of combining the silicone emulsion with different types of wax emulsions will next be examined. The impact on surface modification, hand-feel (something desirable in many applications), gloss, slip and abra-sion in a waterborne high gloss acrylic wood clear coat are examined.The silicone emulsion was combined with the wax emulsions used in the work described above, where silicone emulsion is tested alone, but with a lower wax dosage of 2 %, instead of 3.5 %. A second poly-ethylene wax of interest for wood coatings was also evaluated.The testing was carried out using a high gloss waterborne acrylic wood clear coat. The panels prepared with waxes, silicone emulsion and the combinations were all similar in appearance and were rated by touching the surface and assessing what may be called the ‘silky

Figure 5: impact of silicone additives and waxes on slip and abrasion resistance in waterbased acrylic wood coating

6 9S l i p a d d i t i v e S


alone. Conversely, the PE wax emulsion (2) had a significant impact on gloss, reducing it from the 109.5 of the control sample to 82.In all cases, silicone emulsion plus wax showed a significant improvement in slip over samples with waxes alone (see Figure 8). Likewise, in all cases, the addition of 0.1 % of the silicone emulsion + wax showed a significant improvement in abrasion over the wax-only sam-ples. The data was obtained with a reciprocating abraser using felt pads with a weight of 15 N.

product Chemistry Relative silicone active MWt

silicone polyether 1 silicone polyether low



competitor additive 1 silicone polyether low

competitor additive 2 silicone polyether low

silicone emulsion pDMs Very high

pe wax 1 polyethylene wax emulsion

26 % solids polyethyl-ene wax emulsion

HDpe wax High density polyethy-lene wax emulsion

35 % non-ionic emul-sion based on an

oxidised HD polyethyl-ene wax

paraffin wax paraffin wax emulsion 35 % non-ionic emul-sion based on a modi-

fied paraffin wax

pe wax 2 polyethylene wax emulsion

42 % non-ionic polyethylene emulsion

table 1: Nature of paint additives tested

Using the axis on the left hand side of Figure 8, the green line shows the coefficient of friction. Abrasion resistance was measured by change in gloss. Using the axis on the right hand side of the graph, the blue and red bars on the graph show the initial gloss and the gloss after 1400 cycles. The calculated percent reduction in gloss is recorded above the two bars.

why the Silicone emulSion giveS gooD Performance

The presented additive is an ultra-high molecular weight silicone dis-persion having an average of 64 % solids and hydroxyl (silanol) func-tionality (see Table 2 for general properties). The typical viscosity is 4000 centipoise. This additive can be added directly, typically in post addition, but preferably pre-diluted with water to help dispersion of the actives to achieve the highest benefits with lowest dosage, de-pending on the level of surface modification required.So why does the higher MWt silicone emulsion give superior slip, hand-feel and abrasion resistance performance compared to the oth-er materials? The actives in the silicone emulsion of this study have a low surface tension and therefore migrate to the air/liquid interface. The siloxane backbone then orients to the air/polymer interface, thus

property "dow Corning 52 additive"

Description ultra-high molecular weight silicone dispersion in water

reactive group oH-functional, silanol

appearance smooth, milky white liquid

non-volatile content 62-67 %

Viscosity at 25 °c (77 °F) 3,000-5,000 cp

table 2: General properties of the high MWt pdMS additive

Figure 6: Rating of the impact of wax and silicone emulsion combinations on hand-feel property

S l i p a d d i t i v e S7 0


Figure 7: impact of wax and silicone emulsion combinations on gloss

Figure 8: impact of wax and silicone emulsion combinations on slip and abrasion

forming a very thin lubricating layer on the coating surface to reduce the coefficient of friction.This high molecular weight silicone emulsion brings the paint formu-lator the chance to differentiate their coatings, paints and inks in the market. It provides excellent slip properties and helps differentiate products by creating a sensation that the end user will see, feel, ap-preciate and value.Inks, industrial wood coatings, trim and joinery, together with the interior and exterior paint markets are already benefiting from this high MWt emulsion technology. With inks, increased slip means that

printed pieces do not stick together or scratch. Printers can then im-prove productivity.For trim and joinery, the main goal is durability and a better abrasion re-sistance is always welcome. In wood coatings, slip and hand-feel makes the initial difference for the end user, with better durability thanks to higher abrasion resistance. Finally, a different surface appearance and feeling is of some interest for both external and internal paints.No matter which benefits are considered most significant, the silicone emulsion is highly effective when used alone, but further benefits may be achieved by combining it with waxes.

Sour

ce: V

esna

Cvo

rovi

c - F

otol

ia.c

om

Fresh From the printers

Around 11.3 million litres of printing ink are used for the world‘s inkjet printers over the course of a year.

olympic proportions

This amount of ink could fill 4,5 olympic swimming pools or 15 million wine bottles. It also equals half an hour of rainfall over the UK.

Source: www.neowin.net

7 1w o r l d o f c o l o u r


7 2


H a r d e n e r sSo

urce

: Yvo

nne

Pran

cl -

Foto

lia.c

om

GettinG on with acrylicsa new hardener compatible with acrylic resins. By Mauro Usai, s.a.P.I.C.I spa.

Polyurethane resins are a useful tool in the coatings industry. they can be combined with polyols. however, the hardeners needed for curing often show incompatibility with acrylic poly-ols. a new hardener has been developed to overcome this.

the market sector for aromatic-aliphatic hardeners is an established one, but a weakness is the compatibility. It is not possible to combine all acrylics with this class of hardener. The goal was to produce a fully compatible, aromatic-aliphatic hardener for acrylic system. This work is a comparative study of a new aromatic-aliphatic hardener with other products on the market of the same family from various companies as a reference. The first step, a look at the reactivity, helps to position the product. A second step, the application in a standard cycle in comparison to an aliphatic hardener, shows a new way to think acrylic polyurethane coating. The aim was not to compete with aliphatic on topcoat, but rather to demonstrate that formulators can choose something different from aliphatic hardener to combine with acrylic resins.

reactivity

The starting point was the evaluation of the reactivity. This was carried out by monitoring the evolution of viscosity during the standard pot life. In order to avoid any external influence, each hardener was mixed with the same oxydrilated resin to reach the same degree of crosslink-ing and then diluted to the same solid content. The resulting curves of the different mixes depicted in Figure 1 show that the initial behav-

P olyurethane is one of the most versatile technologies as regards application in the coatings sector. This does not just make a prod-

uct for general use, but rather gives it the possibility of covering the performance requirements of sectors as varied as wood, metal, plastic, leather, etc. At the same time, it meets the specific need for chemical-physical resistance. The possibility of combining polyurethane with other technologies makes it one of the materials mostly used in sol-vent-based systems. Polyurethane has better performance and can be combined with a polyol and a hardener for curing. Polyols consist of acrylic, alkyd and polyester resins and curing takes place with an aromatic, aliphatic or aromatic-aliphatic hardener. Obviously, an aro-matic-aliphatic hardener possesses intermediate properties between the first two. Its main characteristics are faster curing than aliphatic materials and less yellowing than aromatic. Therefore, it can cover all those applications in which colour drift is not so important such as general undercoats (primer, sealer etc.) or a substrate such as wood, the properties of which change more than the coating itself. In general,

7 3H a r d e n e r s


Results at a glance

ű Polyurethane technology for 2K applications is often com-bined with polyols. The hardeners are often incompatible with acrylic resins.

ű The new product has better compatibility and full solubility in aromatic solvent.

ű It shows no haze when mixed with acrylic resin and it shows low yellowing.

ű The material has similar properties to current, commercial-ized products with the advantage of compatibility with acrylic resins.

ű The new hardener can replace other products in several ap-plications with the advantage of faster curing.

ű The material reduces the coating raw material price with di-rect impact on the final cycle.

one was the new aliphatic-aromatic product. Once this first parameter was established, the evaluation of the final formulation was carried out in order to underline the influence of reactivity in an actual ap-plication situation. But prior to this, it was important to verify compat-ibility when standard products were used in the formulation. The work was focused on standard filmogen and aromatic solvents, because these are the products that can have a more important affect on the final transparency. It was not possible to cover the complete range of additives on the market, but all the formulations developed using the new aliphatic-aromatic hardener contained the most commonly used wetting, defoaming, levelling and dispersing agents from a range of producers. The evaluation was made on the hardener itself in combi-nation with increasing quantities of filmogen solution. The level of ad-ditions reached a percentages beyond those encountered in real-life formulations in order to investigate the limit of the compatibility of the different aromatic-aliphatic hardeners. The solutions were CAB (CAB 381/0.5 20 % - 40 % MEK - 40 % MPA), Nitro (NC E14 20 % - Xy 5% - BA 75 %), Vinylite (30 % VROH- 80 % BA). As can be seen from the Figures 2 and 3, the new aliphatic-aromatic hardener is the material with the best compatibility and is fully soluble in aromatic solvent.Once the above behaviour had been established, the materials were evaluated in combination with acrylic resin. Clearly, it was impossible to make a comparison with all the products on the market, but the major materials used in the wood and metal markets were compared to all the hardeners and a larger test was carried out on the new al-iphatic-aromatic hardener. In the case of the new aliphatic-aromatic hardener, both products with different properties (OH number, vis-cosity) and those with same properties, but from different producers were tested. None of the mixtures showed any haze. To do this, the correct quantity of hardener was added to each acrylic resin to reach 100 % of crosslinking. Figure 4 shows one of the tests of comparison: only two samples were transparent, the one with aliphatic and the one with the new aliphatic-aromatic hardener. It was impossible to measure the degree of haze, because the value for those with haze was outside a measurable range. As regards the combination of the new aliphatic-aromatic hardener with the different acrylic resins, the values were between 1.1 and 1.4.

yellowinG resistance

The other key factor taken into consideration was the yellowing be-tween the different aromatic-aliphatic isocyanurates. It was not pos-

Figure 1: Comparison of the evolution of viscosity of a variety of hardeners

Figure 2: Comparison of the compatibility of various hardeners

iour of all the hardeners was similar due to solvent evaporation. After this, the curves developed different slopes. The steeper the slope, the higher the reactivity. This confirmed the expected performance of Hardener 1, a fast curing product. Hardener 2 and a competitive product had the same, slower reactivity and at the end the slowest

H a r d e n e r s7 4


sible to compare them with the full aliphatic as is discussed later, but the materials had to guarantee good light resistance for a number of applications. The evaluation of yellowing was made on a metal plate coated with white, non-yellowing sealer and subsequently coated with pigmented and transparent high-gloss topcoat 170 g/m2. The topcoat was formulated as set out in Table 1 and mixed with different quantities of hardener in order to achieve the same degree of crosslinking. The formulations were not based on acrylic resin, because of the problem of compatibility of traditional products with respect to the new aliphatic-aromatic hardener. This material needed to be combined with an acrylic resin and a special, absolutely non-yellowing alkyd resin, in order to reach compatibility. The different samples were exposed to the QUV ac-celerated weathering test for 100 hours and subsequently the yellowing was evaluated using a colorimeter. Table 2 shows the value for the two different topcoats. There is a big difference between them. The reason

is firstly that pigment hides the change in colour and secondly the trans-parent system absorbs the direct light and the reflection from the white sealer. The difference between each set of values, and that of the final aspect, are quite similar (Table 2).

lowest viscosity

Drying and resistance tests were also performed on these formula-tions. The drying test is important, because in the final high-gloss for-mulation it is not possible to detect differences in the dust-free or in the touch free formulation. The evidence of a longer curing time is more visible when a mat topcoat is formulated, and it is reflected more by the higher gloss of the new aliphatic-aromatic hardener than from the longer time needed to detect it discussed earlier. The new aliphatic-aromatic hardener is the lowest in viscosity, so at the end

Figure 3: The solubility of various hardeners in xylene Figure 4: degree of haze in mixtures with acrylic and aliphatic resins

Pigmented topcoat

Component a

“re xin 2268” 22.5 %

“Domacryl 546” 30.0 %

t io2 32.0 %

But yl acet ate 9.0 %

Mpa 4.0 %

MiBK 1.5 %

“Byk 141” 0.5 %

“Byk 306” 0.5 %

100

Component B

Mixed isocyanurate* 50.0 %

ethyl acet ate 25.0 %

But yl acet ate 25.0 %

100

*competitor/oKD/oKDs/oK-Hp

Table 1: Formulations of topcoats with different quantities of hardener

Gloss clear topcoat

Component a

“rexin 2268” 62.0 %

Butyl acetate 12.0 %

Xylene 9.0 %

ethyl acetate 7.0 %

Mpa 3.0 %

MeK 6.0 %

“Byk 141” 0.5 %

”Byk 306” 0.5 %

100

Component B

Mixed isocyanurate* 58.0 %

ethyl acetate 21.0 %

Butyl acetate 21.0 %

100

*competitor/oKD/oKDs/oK-Hp

7 5H a r d e n e r s


it is possible to increase the matting agent in the final formulation without an increase in final viscosity, a change that can affect the ap-pearance. With regard to the chemical physical resistance, all the tests performed and the differences observed are within the margin of er-ror for each evaluation. The only test that differs and gives the new aliphatic-aromatic hardener higher hardness is the Sclerometer test.The new aliphatic-aromatic hardener performs in a similar way to products well known on the market, with the additional advantage of the possibility to combine it with acrylic resin. This property makes it worth comparing the new material with aliphatic hardeners, to date the only hardener that are fully compatible with acrylic resins. This does not necessarily mean that the aliphatic-aromatic technology should be recommended for the application in pigmented non-yel-lowing topcoats. Figure 5 shows the results of QUV tests made on a yellowing substrate such as wood and in a sealer for pigmented systems. The behaviour on light and dark wood shows that in both cases what yellows most is the substrate. Thus, there is no advantage, apart from the compatibility question, to use aliphatic. Talking about the pigmented system, the evaluation of colour change starts from a screening of the possible application on different substrates, i.e. the testing of the adhesion and flexibility on a general pigmented for-mula. The conclusion is that the new aliphatic-aromatic hardener can be used in a similar way to the aliphatic material on different flexible substrates, from plastic to metal, each time modifying the formulation of component A (oxydrilic component) and using known methods of achieving good adhesion. Generally speaking, with respect to aromat-ic, the idea is to have a rigid and brittle system. The new aliphatic-aro-matic hardener is designed in order to reach the maximum flexibility: 7.5 at the conical mandrel, the full scale.

∆e [100 h]

PTCH * CTCH **

competitor 2.2 11.5

“polurene oK.D” 2.4 14

“polurene oK.D.s.” 2.3 11

“polurene oK-Hp” 2.1 14

* pigmented top coat high gloss, ** clear top coat high gloss

Table 2: degree of yellowing of t wo different topcoatsFigure 5: results of QUV tests made on aliphatic and aliphatic-aromatic coating cycle for dark and clear wood

same toPcoat

In conclusion, it seems that the evaluation of yellowing, a feature considered only at the end of the work, can modify the idea and the planning of the pigmented cycle made with 2K acrylic and isocyanate systems. The proposal is to use the new aliphatic-aromatic hardener in the sealer rather than the topcoat, as it is known that the yellowing of the sealer can have an effect on the final finish. Thus it is essential to evaluate this effect by applying in parallel two cycles with two coats of sealer, both with the same component A, but with a different hard-ener, i.e. the full aliphatic and the aromatic-aliphatic. To do this both samples were coated with the same aliphatic topcoat and then ex-posed at the QUV. As can be seen from Table 3, in one cycle there was a yellowing of the basecoat, but this had no effect on the final finish. The values were calculated taking the full aliphatic as a reference to ensure that there would be a greater difference between the differing coats of sealer. However, as can be seen from the value ΔE=0.3, there was no influence on the final appearance.Compared to previous formulations, the new product has a further advantage. Products using the new aliphatic-aromatic hardener are touch-free at room temperature after three hours. With aliphatic, the time needed is six hours. Both have a pot life in excess of eight hours. It is clear that this reduces the curing time and/or temperature, with a direct impact on environment and energy saving.Finally, the new aliphatic-aromatic hardener can help also to reduce the impact of the coating raw material on the final price – a subject that, in this time of crisis is a sensible one for everyone.

dL da db ∆e

1st coat sealer -0.58 0.72 4.25 4.3

2nd coat sealer -0.48 -0.6 3.62 3.7

ptcH 0.07 0.05 0.25 0.3

Table 3: degree of yellowing after QUV of t wo different cycles

7 6


C r o s s l i n k e r s

Sour

ce: E

voni

k In

dust

ries

Bio Building BlocksA method for the economic biological production of new industr ial amines as crosslinkers. By Philip engel, evonik industr ies.

Figure 1: reaction cascade to produce amines from alcohols

industrial amines have a broad range of applications in the chemical industry, such as polyamides, surfactants and crosslinkers. For crosslinkers in coating applications, amines in epoxy resins play a major role. To overcome the limitations of petro-based crosslinkers, a new technology platform that allows the synthesis of amine building blocks based on renew-able resources such as glucose and isosorbide has been devel-oped.

T he global market of biobased chemicals reached $9.1 billion in 2011 [1]. Demand is continuously growing with above average

growth rates. More and more biochemicals have been launched into the market, that are not only biobased but also produced by a bio-process (biocalaysis or fermentation). Well-known examples for large scale products include 1,3-propan-ediol, that was launched in 2006 by DuPont [2]. Furthermore, other products are just now reaching commercial scale, e.g. succinic acid (Succinity, BioAmber, Myriant, Reverdia) [3], 1,4-butanediol (Genom-atica) [4], and isobutanol (Gevo).



Results at a glance

ű To overcome the limitations of petro-based crosslinkers, a new technology platform that allows the synthesis of amine building blocks based on renewable resources has been devel-oped

ű Amines are one of the central building blocks in the chemi-cal industry, used in the production of polymers, agrochemi-cals, emulsifiers and in epoxy resins for crosslinker applications

ű For the biological production of amines from alcohols in one step, different types of enzymes have to be coupled to a reaction cascade

ű The reaction cascade was successfully applied for a range of alcohol educts and showed excellent selectivity and conver-sions for a number of the tested alcohol educts

ű After the redox self-sufficient reaction system was estab-lished for the in-vitro reaction, the enzyme cascade was trans-ferred into a living whole cell system, (an E.coli host), a step required for the economic production of amines as chemical building blocks.

ű The formation of the respective a,w-diamins and amino-alcohols was analysed and it was shown the developed enzyme cascade could be successfully transferred from an in-vitro sys-tem to a whole cell system

The most common resource for biotechnological fermentation pro-cesses are carbohydrates. Today the majority of industrially used car-bohydrates originate from sugar cane and sugar beet. In particular with respect to recent biofuel initiatives, a lot of research efforts focus on sugar production from non-food lignocellulosic biomass such as wood, straw of other biological residues. However, in the near term it cannot be expected that sugars from non-food biomass will be avail-able in large volumes [5]. For the production of biobased chemicals employing a bioprocess, carbohydrates can be directly used as fermentation raw material, e.g. for 1,4-butanediol or succinic acid. Furthermore, biochemicals can be produced from other sugar derived intermediates, e.g. isosorbide is a bycyclic furofuran compound produced by catalytic hydrogenation of glucose to sorbitol and subsequent dehydratisation to isosorbide [6, 7].Biological production of industrial aminesAmines are one of the central building blocks in the chemical indus-try. They are used, for example, in the production of polymers, agro-chemicals, emulsifiers and in epoxy resins for crosslinker applications.

The vast majority of the amines are today produced from petrochemi-cal resources by chemical conversion. For example, carbonyl com-pounds, aldehydes or ketones, can be effectively reductively aminated [8]. However, the chemical conversion of alcohols to amines is not as established. The conversion of alcohols to amines was described by a “borrowing-hydrogen” methology, eliminating hydrogen as cosub-strate and using ammonia as nitrogen source [8]. Despite promising first results, these approaches are not economically viable yet and further development concerning the catalyst synthesis, recycling, and immobilisation is needed.Enzymes are another established technology to produce amines from carbonyl groups. Transaminases (TA) are a group of enzymes that can produce enantiomerically pure amines with very high selectivities. Due to the high stereoselectivity of transaminases they already play an important role in the synthesis of pharmaceutical intermediates, where enantiopure products are needed [10]. As an amine donor for the transaminase reaction amino acids are most common. The platform developed here provides a new approach to biological production of different industrially relevant amines directly from alco-hols. A major aim is to show the perspective for a cost reduction, since amine building blocks for the chemical industry are substantially lower in prices compared to pharmaceutical amines.

developing enzyme cascades For conversion oF alcohols To amines

For the biological production of amines from alcohols in one step, different types of enzymes have to be coupled to a reaction cascade. First, the alcohol group has to be converted to an aldehyde or ke-tone by an alcohol dehydrogenase (ADH). This enzyme requires the cofactor NAD (Nicotine adenine dinucleotide) or NADP (Nicotinamide adenine dinucleotide phosphate) for electron transfer. This reaction is coupled with a second enzyme, a transaminase that converts the carbonyl group to the amine by transferring nitrogen from an amino acid, e.g. alanine, as amine donor. The amino acid amine donor is con-verted to a keto acid, e.g. pyruvate.Overall the combination of alcohol dehydrogenase and transaminase requires two cosubstrates, NAD and amino acid. For the biological production of large volume chemicals it is by far too expensive to provide these co-factors stoichometrically to the reaction, particularly NAD is very expensive. Therefore, the reaction was extended with a third enzyme, an alanine dehydrogenase, to create a redox self-suffi-cient reaction system that regenerates the needed NAD cofactor. The

Figure 2. reaction f rom isosorbide to the aminoalcohol [7]

Figure 3: successful conversion of different a,w-diols to di-amines in a whole cell system. 10 mM is equal to 100% concversion. For reaction details, please refer to [13]



alanine dehydrogenase regenerates the keto acid using ammonia and at the same time converts NADH, formed by the alcohol dehydroge-nase, to NAD. Thereby, both cofactors can be effectively regenerated and only cheap ammonia as nitrogen source has to be added. The reaction cascade is illustrated in figure 1.The reaction cascade was successfully applied for a range of alcohol educts and showed excellent selectivity and conversions for a number of the tested alcohol educts, shown in table 1. The reaction was also successfully performed for diols as educts and it was found that for bi-functional molecules with two alcohol groups the reaction occurs se-quentially, i.e. first one alcohol group is converted to an amine and the subsequently the second alcohol group is converted to an amine [11].As a sugar derived and particularly bulky molecule, the amination of isosorbide to the aminoalcohol, as shown in Figure 2, and further, to the respective diamine was investigated in depth [12]. After extensive enzyme screening and engineering it, suitable enzyme system com-prised of an alcohol dehydrogenase, a transaminase and an alanine dehydrogenase was identified to aminate isosorbide. For isosorbide as educt the bifunctional amination can lead to three different ste-reoisomers, diaminoisoidide, diaminoisosorbide and, diaminoiso-mannide. The respective amionoalcohol was the main product of the reaction and reaching concentrations of 1-2 g/L. Additionally, the re-spective diamine was also identified in this study [12].

producTion oF amines using enzymes in living whole cell sysTems

After the redox self-sufficient reaction system was established suc-cessfully for the in-vitro reaction, the next step was the transfer of the enzyme cascade in a living whole cell system. This step is required for the economic production of amines as chemical building blocks, as the need for expensive co-factor addition hampers the economics of the process even though the developed cascade is self-sufficient. By transferring the reaction in a microbial system no addition of external co-factor is needed, but supplied from the natural cell metabolism, using glucose as cheap and abundant co-substrate for energy supply.The enzymes successfully applied in the in-vitro system were trans-ferred in an E. coli host. After growing the cells, they were transferred in a aqueous buffer system and contacted with different a,w-diols. The formation of the respective a,w-diamins and amino-alcohols was analysed (see figure 3) [13].

For different primary alcohols excellent conversions to the diamine of 80-100 %, with nearly 100 % selectivity for the diamine, were achieved within only a few hours reaction time. Therefore, it was clearly shown that the developed enzyme cascade could be successfully transferred from an in-vitro system to a whole cell system.

conclusions and ouTlook

The new technology platform developed provides a novel way for the production of industrial amines, e.g. for coating applications. It was shown that the individual reactions can be performed enzymatically and that the combination of different enzymes to a redox self-suffi-cient cascade is possible. Furthermore, the reaction was successfully transferred in a whole cell system, which is an essential milestone to-wards the economic biological production of new amines.

acknowledgmenTs

this work was funded by the BMBF project “Bioxamine” (grant no. 0316044B). We would like to thank all project partners, particularly the groups of prof. W. Kroutil, uni graz; prof. a. skerra, tuM and prof. V. Wendisch, uni Bielefeld for providing the data shown.

reFerences

[1] L uc int e l home page . w w w.luc int e l .com /r e por t s /c he mic al _ compo s it e s /

global _ polyme r_ indu s t r y _ 2012 _ 2017_ t r e nd s _ f or eac a s t_ june _ 2012.

a s px (acce s s e d F e b 4, 2014)

[2] nak amur a c . and W hit e d g. , Me t abol ic e ngine e r ing f or t he mic r obial

pr oduc t ion of 1, 3 - pr opane diol , cur r. opin. Biot e c hnol . , 2003, Vol . 14, pp.

454−459.

[3] s c holt e n e . , re n z t. , t homa s J. , cont inuou s c ult i v at ion appr oac h f or

f e r me nt at i v e s ucc inic ac id pr oduc t ion f r om c r ude glyce r ol by Ba s f ia s uc-

c inic ipr oduce n s DD1, Biot e c hnol . L e t t. , 2009, Vol . 31, pp. 1947−1951.

[4] Y im H. , Ha s e lbe c k r . , niu W. , e t al . , Me t abol ic e ngine e r ing of e s c he r ic hia col i

f or dir e c t pr oduc t ion of 1,4- but ane diol , nat. c he m. Biol . , 2011, Vol . 7, pp.

445−452.

[5] Bior e f ine r ie s r oadmap, f e de r al go v e r nme nt of ge r many , 2012.

[6] c l ime nt M . , cor ma a . , ibor r a s. , con v e r t ing c ar bohydr at e s t o bulk c he mic al s

and f ine c he mic al s o v e r he t e r oge ne ou s c at aly s t s, gr e e n c he m ., 2011, Vol .

13, pp. 520–540.

[7] s c haf f e r s . , Haa s t. , Bioc at alyt ic and f e r me nt at i v e pr oduc t ion of

a,w - bif unc t ional polyme r pr e c ur s or s, or g. p r ogr e s s r e s . , 2014,

Vol . 18, pp. 752-766.

[8] p agnou x - o z he r e lye v a a . , p anne t ie r n . , Mbaye M . , gai l lar d s . , re naud r . ,

K nölk e r ’s i r on comple x : an e f f ic ie nt in s i t u ge ne r at e d c at aly s t f or r e duc t i v e

aminat ion of alk yl alde hyde s and amine s, a nge w. c he m., 2012, Vol . 124,

pp. 5060–5064.

[9] imm s . , Bähn s . , Z hang M . , e t al . , impr o v e d r ut he nium - c at aly ze d aminat ion

of alcohol s w it h ammonia: s ynt he s i s of diamine s and amino e s t e r s, a nge w.

c he m., 2011, Vol . 123, pp. 7741–7745.

[10] Höhne M . , Bor n s c he ue r u. , Bioc at alyt ic r out e s t o opt ic al ly ac t i v e amine s,

c he mcat c he m, 2009, Vol . 1, pp. 42- 51.

[11] s at t le r H. , F uc h s M . , taube r K . , e t al . , re dox s e l f - s uf f ic ie nt bioc at aly s t ne t -

w or k f or t he aminat ion of pr imar y alcohol s, a nge w. c he m ., 2012, Vol . 124,

pp. 9290 –9293.

[12] L e r c hne r a . , a c hat z s. , r au s c h c . , Haa s t. , s k e r r a a . , couple d e n z ymat ic

alcohol - t o - amine con v e r s ion of i s o s or bide u s ing e ngine e r e d t r an s ami -

na s e s and de hydr oge na s e s, c he mcat c he m, 2013, Vol . 5, 3374-3383.

[13] K lat t e s . , We ndi s c h V. , r ole of L- alanine f or r e dox s e l f - s uf f ic ie nt aminat ion

of alcohol s, Bioor g. Me d. c he m., 2014, Vol . 22, pp. 5578–5585.

no. substrate Conversion % Aldehyde % Amine %

1 1-hexanol >99 <1 >99

2 1-octanol 50 <1 50

3 1-octanol 57 <1 57

4 1-decanol 2 <1 2

5 1-decanol 25 <1 25

6 1-dodecanol 10 <1 10

7 benzyl alcohol >99 13 87

8 cinnamyl alcohol >99 30 70

9 3-phenyl-1-propanol >99 <1 >99

Table 1: Amination of different primary alcohols (for details please refer to [11])

Sour

ce: S

hane

Phot

o - F

otol

ia.c

om

Ancient PAints

Popular pigments used for cave paintings were iron oxi-de, maganese oxide and char-coal, mixed with water, spit or fat. Researchers assume that other raw materials were ore, feldspar, blood, limestone, re-sin, milk and herbal juices.

From Brushes to sPrAy PAinting

The paint was applied with branches, stamps or

fingers. In addition to this, paints were also spray

painted using pipes made from bone.



8 0


E m u l s i f i E rSo

urce

: jpr

amir

ez -

Foto

lia.c

om

New chemistry for reactive emulsifiersBenefits found in production and performance of aqueous alkyds. By Charles f. Palmer, Jr., Ethox Chemicals.

the conversion of alkyds to waterborne form has a number of problems. very high viscosities may occur during the invert emulsion process, while the surfactants used may impair the properties of the dried film. a new emulsifier that crosslinks during drying can improve final properties, while allowing emulsification to be completed before the solvent is stripped from the original alkyd

w hile they are no longer the most widely used binders in coat-ings, alkyd resins are still of major importance since they re-

main the most commonly used resin or binder system in oil-based and solvent-based coatings.Alkyd coatings are relatively inexpensive and perform well, often with fewer film defects than other coatings. They are used in many indus-trial and architectural applications. The hydrophobic nature of the alkyd polymer makes them good choices when water repellency is important.However, formulating alkyds into low VOC (volatile organic compound) waterborne coatings results in a number of problems, which can be minimised by the use of a new reactive emulsifier as described below.

the geNeral Nature of alkyds

Alkyd resins are polyesters generally prepared from a polyol, phthalic anhydride and unsaturated vegetable fatty acids such as linseed, soy,

material % by weight Amount used

“Beckosol 6422-K3-75” (25 % methyl n-propyl ketone solvent) short oil alkyd (reichhold)

49.4 % 741 g

reactive a 4.0 % 60 g

ammonium hydroxide 0.74 % 11.1 g

Deionised water 45.86 % 687.9 g

solids content 41 %

Table 1: Composition of emulsified alkyd

8 1E m u l s i f i E r


Results at a glance

ű Alkyd resins continue to be used in large volumes in sol-ventborne coatings. However, the conversion of alkyds to water-borne form has a number of problems.

ű Waterborne alkyds may be prepared either by grafting onto an emulsion polymer or directly by the invert emulsion process. The latter is more flexible, but the surfactants used lead to very high viscosities during production and may impair the physical properties of the dried film.

ű A new type of emulsifier based on the styrenated phenol hydrophobe and containing allyl groups crosslinks during dry-ing, thus improving final properties. It also allows emulsification to take place before the solvent is stripped from the original alkyd, reducing viscosity problems.

ű The reactive emulsifier produced a more hydrophobic film which resisted water penetration better than a standard waterborne alkyd. Gloss and adhesion to metal also showed some improvement

or tung oil (see Figure 1). The inclusion of the fatty acid confers a ten-dency to form a flexible coating. Alkyds are often categorised as long, medium or short oil, based on the amount of vegetable oil in the alkyd; thus, long oil alkyds have more fatty acid content than short oils.Fatty acids known as drying oils have multiple double bonds which will air cure to give a hard coating. This curing reaction crosslinks the oligo-meric alkyd chains to increase molecular weight and improve durability and other properties.Alkyds are sometimes modified with other radical reactive monomers or polymers for a number of reasons. These are included (for example)

to speed curing, improve water compatibility or solubility or reduce viscosity.Solvents are employed in traditional alkyd manufacture to reduce the viscosity of the oligomeric polyester and often to help remove by-product water formed in the synthesis. These solvents include xylene or ketones.

role of alkyds iN the shift to waterborNe coatiNgs

The legislation in many countries to reduce VOC content in coatings is limiting the usage of alkyds in many applications, since most traditional alkyds have a significant level of solvent. The solvent is necessary to reduce the viscosity of the alkyd resin and to allow easy application.A number of new technologies have been developed recently to ren-der solventborne alkyd coatings more environmentally acceptable by replacing some or all of the solvent with water. Alkyd resins are con-verted into useable waterborne products by one of two methods.One method is to graft an alkyd resin onto an emulsion polymer. This gives the coating some properties of both polymer types. The emul-sion polymer and the surfactant additives that are used to manufac-ture the emulsion render the emulsion/alkyd polymer dispersible in water, while the grafted alkyd imparts alkyd-type properties such as toughness and resistance against various chemicals to the coating.Additional processing steps are needed to make these hybrid products. The alkyd polymer needs to be of a particular type and structure in order to make a viable alkyd/emulsion coating. In some cases, the durability of these products is superior to those of emulsion polymeric coatings. However, the alkyd resin chemistry must be altered to maximise the benefits from the grafting of alkyd resin onto the emulsion backbone. The additional processing steps increase costs. This process also em-ploys additives such as coalescing solvents to improve properties such as the gloss and flexibility of the coating.Emulsification of the alkyd resin into water is the other method to re-move some or all of the VOCs in the emulsified product. A key advan-tage of the emulsification process is that the alkyd resin used does not necessarily need to be altered to prepare the emulsion, so long as the proper surfactant and emulsification process is used to make the product.The proper surfactant is one with the proper molecular weight, struc-ture, and HLB (hydrophilic/lipophilic balance). Nonionic surfactants

figure 2: Generic chemical structures of styrenated phenol-based surfactants (‘reactive A’)figure 1: Generic scheme of alkyd synthesis

E m u l s i f i E r8 2


generally outperform anionic surfactants in making stabilising emul-sions - as demonstrated by improved water sensitivity, better colloidal stability and lower foam profile when compared to an emulsion made with an anionic surfactant. However, in some cases, a small amount of anionic surfactant is used in conjunction with the nonionic surfactant to provide some charge stabilisation.

limitatioNs of the emulsificatioN process

However, there are a number of technical difficulties and limitations with the emulsification process as described in the patent literature. Usually what is termed an invert emulsion process is preferred since it does not require high shear mixing and produces less foam.In this process, the undiluted alkyd with little or no solvent is heated

to a high enough temperature to reduce its viscosity to a manageable level. The solvent used to make the alkyd is usually removed prior to emulsion preparation and the alkyd is kept at elevated temperature to keep it from solidifying.The surfactant package of choice is then added to the molten alkyd, followed by gradual addition of hot water. As the water is added, the mixture forms a water-in-oil emulsion However, as the water content increases and the emulsion nears the inversion point (‘flipping’ from a water-in-oil emulsion to an oil-in-water emulsion) the viscosity often becomes unmanageably high.The temperature is maintained as high as possible to reduce the vis-cosity, but one problem is that surfactants, especially nonionic sur-factants, have lower water solubility at high temperatures. Once the inversion point is crossed, forming the final oil-in-water emulsion, the viscosity decreases.These alkyd emulsions contain a few percent of surfactants that re-main in the formulation and thus are present in the coating after appli-cation. There they can cause a number of problems. When surfactant-coated micelle spheres stack on a surface during the drying process and start to coalesce, the concentration of surfactant molecules at the intersection of micelles can become relatively high.If the surfactant molecules remain unbound and free to migrate, when the dried coating is later exposed to water some of the surfactant mol-ecules can dissolve. This reduces the surface tension of the water, im-proves its wetting of the coating surface and promotes its penetration through the coating.Damage to the substrate to be protected becomes much more likely. On metal substrates, this can lead to corrosion and loss of adhesion. In addition, extraction of the surfactant can lead to pitting and degrade the integrity of the coating. Free surfactants can also plasticise the al-kyd and prevent it from reaching maximum hardness upon curing.

solviNg the problems created by surfactaNts

Many of the problems that are caused by extractable surfactants can be reduced if the surfactant can co-cure into the alkyd coating. This locks it into place so that it cannot be extracted by water. A number of patents claim reactive surfactants that can do this. They are usually based on fatty acid alkoxylates in which the fatty acid has a number of double bonds similar to the drying oil fatty acids.However, most of the known ones have some significant drawbacks. They are nonionic surfactants and so have lower solubility in hot water, which limits their ability to make inverse emulsions with manageable viscosity. Nonionics also cannot take advantage of charge stabilisation mechanisms to reduce particle size and improve emulsion stability.Most nonionic surfactants used for alkyd emulsions have long ethoxy-late chains to give them the high HLB value which is required to make a stable emulsion. These nonionics have a strong tendency to retard drying, probably due to complexation of metal ion drying agents. Long chain nonionics also have a tendency to plasticise alkyd coatings, re-ducing their hardness.A new class of reactive surfactants has now been developed, based on the styrenated phenol hydrophobe and containing allyl groups, as shown in Figure 2. Instead of curing through fatty acid double bonds, these cure through the pendant allyl groups. Styrenated phenol is a polyaromatic hydrophobe, so these surfactants have good affinity for the aromatic groups in the alkyd. They are also inherently lower foam-ing than most linear surfactants.The alkoxylate chain is relatively short so that it has less affinity for chelating metal ions. Curing times have not been affected with the dri-er packages tested. The anionic character gives it charge-stabilisation ability. This surfactant chemistry is also available as a nonionic and as a phosphate ester.

reactive surfactant formula

Non-reactive surfactant formula

“Beckosol 10-539” long oil alkyd

38.2 38.2

reactive a 1.8 --

poe 20 Dsp -- 1.8

ammonium hydroxide

0.6 0.6

Water 58.8 58.8

catalyst blend 0.6 0.6

Drawdown properties after 2 weeks

cross-cut adhesion 0B (100 % fail) 0B (100 % fail)

pencil hardness F 2B

60 ° gloss 117 113

Water contact angle 84.5 ° 79.0 °

appearance of water drop on surface of dried film

Table 2: Comparisons bet ween reactive and non-reactive surfactant based alkyd coatings

8 3E m u l s i f i E r


emulsioN preparatioN procedure outliNed

The materials used in alkyd emulsion preparation and their addition levels are shown in Table 1. The procedure used is as follows. Charge the solventborne short oil alkyk, (‘Reactive A’), and ammonium hydrox-ide to a reactor flask equipped with nitrogen sparge, thermocouple, overhead mixing and condenser. Once the mixture is homogeneous, begin adding water at ambient temperature to the mixture in a slow stream so that the water is incorporated evenly into the mixture and does not pool on the surface.As the water is added, the viscosity will rise until the inversion point is met, then the viscosity will begin to drop. When all the water is charged, allow the emulsion to mix for several minutes and then check the parti-cle size. This emulsion was prepared at 41 % solids.After ensuring that the particle size is satisfactory, raise the tem-perature of the emulsion by heating to 70 °C with a nitrogen sparge. Gradually increase the temperature until vapours are detected in the condenser and allow the emulsion to strip until all of the solvent is removed.After cooling, the particle size, percentage solids and viscosity were checked. This emulsion was 47.9 % solids and contained 0.4 % residual solvent. The viscosity was not measured but was quite low.

As Figure 3 shows, the particle size of the alkyd emulsion remained very low, actually decreasing from the solvent-containing emulsion stage. The solids content was significantly higher. These results were not opti-mised; the solids concentration could probably have been significantly increased further.This result shows that the use of these anionic surfactants can over-come the viscosity problem inherent in the invert emulsion proce-dure by leaving the solvent in the mixture and then removing it after emulsion formation. The anionic sulfate has sufficient water solubility to remain an effective emulsifier even at the elevated temperatures needed for practical solvent stripping. This procedure will allow coat-ing manufacturers to convert existing low-cost solventborne alkyds to waterborne products containing no VOC.

how differeNt surfactaNts affect coatiNg properties

Two more waterborne emulsions were prepared using the proce-dure as given above, but with a long oil solventborne alkyd. Identical amounts of reactants were used. The only difference was the emulsi-fier. Both emulsifiers (“RS-1618” referred to as ‘Reactive A’ and POE 20 DSP (poly(oxyethylene) with 20 POE units)) are distyrenated phenol based, have nearly the same ethoxylate chain length, and are both sul-

figure 3: Alkyd emulsion particle size with solvent (left) and after solvent str ipping (r ight)

E m u l s i f i E r8 4


fates. The only significant difference between them is that Reactive A is reactive, while POE 20 DSP contains no reactive allyl groups.After the emulsions were prepared, a cobalt-based drying catalyst package was added to the emulsion and drawdowns were made on steel “Q-panels”. After two weeks, the adhesion, pencil hardness, gloss, and water contact angle were checked. Results are shown in Table 2.Review of the data in Table 2 shows that the alkyd coating with the reactive surfactant has a higher water contact angle. This indicates that the coating is more water repellent, evidence that the reactive surfactant is cured into the alkyd and not available to dissolve into the water and reduce its surface tension.The coating with the reactive surfactant is also significantly harder than the other. Presumably, this is due to the plasticisation of the coating by the unbound surfactant. The gloss of the coating is also higher with the reactive surfactant. This suggests that the surface has fewer defects. This might be due to a more homogeneous particle size distribution.

adhesioN to metal aNd rust protectioN are improved

Water drops were placed on each of these drawdown panels and allowed to stand covered so that they would not evaporate. The water drop on the alkyd coating with the non-reactive surfactant had significantly wet the coating and had spread out to cover a much larger area than the drop on the reactive surfactant coating. After two days, rust on the panel was clearly visible on the panel with non-reactive surfactant (see Figure 4). The coating with the reactive surfactant resisted the ingress of water and pro-

tected the steel much better than did the non-reactive one. In another test of adhesion, drawdowns were made of an aqueous short oil alkyd emulsion prepared with the difunctional Reactive A surfactant accord-ing to the first example above and a commercially available waterborne short oil alkyd (assumed not to contain a reactive surfactant).Both were catalysed with the same drier package. After 24 hours, a crosshatch pattern was scribed onto the alkyd surfaces. The attempt to scribe the crosshatch pattern on the commercial alkyd failed, while it was successful on the alkyd with reactive surfactant (see Figure 5). This failed the tape pull, however.The reactive surfactant emulsion thus had significantly better adhe-sion to the steel panel. It is not known what the mechanism for the improved adhesion is.

beNefits of reactive surfactaNts summarised

The reactive surfactants described here allow the viscosity problems found in the invert emulsion process to be reduced. They give a more hydrophobic coating (as measured by water contact angle) than struc-turally similar non-reactive surfactants. Similarly, they produce a more water-resistant coating (as measured by rust resistance) than structur-ally similar non-reactive surfactants.Both gloss and adhesion to metal can be improved by the use of reac-tive surfactants. Further work to develop and understand this technol-ogy is therefore under way.

figure 5: Crosshatch adhesion test on steel panels coated with aqueous alkyds

figure 4: Outlines of water drop on alkyd coating drawdown on steel panels. left: Coating with reactive surfactant. right: Coating with non-reactive surfactant

Sour

ce: p

hanu

wat

nand

ee -

Foto

lia.c

om

Like CLoCkwork

Big factories can coat up to 100,000 pencils a day. The pencils are pushed in a pipe-like case with paint. Surplus coating is then wiped off when they are extruded from the container. The process is repeated up to seven times.

Less is More

Only small amounts of coa-ting are needed to coat a pencil – less than one gram is used apiece. The pencil has a long history – its first precusors were produced in 1565 in England.

Source: Deutsches Lackinstitut



8 6


R h e o l o g y M o d i f i e R

Source: Scanrail - Fotolia.com

GettinG more with lessNew bio-based rheological additives for low-temperature process. By germain fauquet and yohann Trang, Arkema.

recently, a new generation of thixotropic amides (slt) has come onto to the sealant market. it reduces cost and cycle times and also improves reliability and rheological perfor-mance. more than 90 % of its ingredients are bio-sourced and bio-renewable. it is also 100 % solid and therefore VoC free. Furthermore, this development will enable formulators to produce sealants at lower temperatures (cold process), thus reducing energy consumption and cycle time and leading to lower manufacturing costs.

r heology modifiers are normally used in adhesives and sealants to improve their properties during the three different stages of

their life, i.e. during storage, in the application phase and finally during the use of the cured sealant. During storage, a good rheological additive will provide anti-settling properties, i.e. the fillers and pigments will remain homogeneously dis-tributed in the liquid phase without any syneresis or phase separation. It will also enhance long-term stability by preventing advanced curing and viscosity variations after several weeks or even months of storage.The ideal rheological additive shows excellent shear-thinning behav-iour. This will enable to play a major role by providing a good extrusion rate and workability during the application phase. This is important,

figure 1: Amide fibres 3d net workfigure 2: Processing time comparison bet ween the products SlX and SlT

8 7R h e o l o g y M o d i f i e R


Results at a glance

ű The new rheological additive presented in this article (SLT) is a bio-based material which allows adhesive and sealant formulators to use a cold manufacturing process.

ű This fact enables them to reduce their energy cost and in the same time to shorten their production time.

ű SLT also exhibits higher rheological performance (higher viscosity, good shear thinning behaviour, good pseudoplastic properties) and it is by 10 to 25 % more efficient than older ge-nerations.

ű This new development therefore extends the existing range of additives and offers various solutions to the formula-tor depending on processing conditions – from cold to hot pro-cess – and systems, including hybrids, epoxies, 2K-PU, silicones, acrylates, rubbers, polychloroprenes etc.

ű These products embody the principle of achieving more by using less while contributing positively to the protection of the environment.

because the relatively fast recovery of the viscosity will lead to non-slumping properties and will prevent the sealant from flowing once applied. The additive should not modify the other properties of the sealant such as curing time, adhesion, tack-free time, green adhesion.Finally the properties of the sealant after curing should not be com-promised. The rheological additive should no longer act at this stage of the life of the sealant by affecting the mechanical properties, the weather-ability, etc.

existinG teChnoloGies to aChieVe Good meChaniCal properties

Over the years, several technologies have been developed to achieve these different properties. These are used nowadays by sealants and adhesives manufacturers. For example, on one hand inorganic tech-nologies such as fumed silicas, precipitated calcium carbonates and organoclays, and on the other, organic materials such as hydrogen-ated castor oil and polyamides.Inorganic rheology modifiers, for example, fumed silicas, are widely used today, because they are activated at low temperature, have a short processing time and give high viscosity build. However, they cre-ate dust, are difficult to handle (low density) and have a major impact on mechanical properties.Precipitated calcium carbonates are easy to process at low tempera-tures, easy to handle, but act as fillers. Much higher concentrations are needed leading to a significant impact on mechanical properties.Organoclays are easy to handle and to process, but their efficiency is system dependent. They do not behave so well in solvent-free sys-tems and the polarity of the final system is a key factor. Furthermore, they can have an impact on the final colour of the sealant and may require pre-gel preparation.Looking at organic rheology modifiers, castor-oil derivatives were historically the first generation of this family of rheological additives. They were developed in the early 60’s and are still used, because of their low cost and ease of activation at low temperatures (40 °C to 70 °C). They are simple to handle, give a high viscosity build and show a good extrusion rate. One of the main limits of this technology is that the temperature must be well controlled and the stability of the viscosity upon storage can be an issue. This is due to their sensitivity to processing conditions, high temperatures and temperature varia-tions.The amide technology is clearly the one that has grown significantly during the last 10 years. The move from amide-modified castor oil to pure amides has enabled adhesive and sealant manufacturers to reach a level of performance which was not possible with the previous technologies.By forming a dense 3D-network of fibres (Figure 1), amide derivatives provide a high body, excellent shear-thinning behaviour and very good pseudoplastic properties. Amides are easy to handle and offer very good storage stability. They are also more temperature robust and have no effect on mechanical properties and can be used in both low modulus and high modulus systems.On the other hand, the first generations of amides need a high ac-tivation temperature (between 70 °C and 110 °C depending on the grades and systems where they are used) and a relatively long pro-cessing time to develop a proper fiber network. The first generations of amides have significant advantages compared to inorganic technologies and even to castor-oil-based products. They are less dependent on solvent polarity and more resistant to severe condi-formulation

“Ms polymer s203 H“ 14.9

“Ms polymer s303 H“ 10

plasticiser 17.4

rheological additive 3.7

caco3 fillers 49.51

tio2 pigments 2.5

uV stabilizer 0.25

Light stabilizer 0.25

Dehydration agent 0.74

adhesion promoter 0.5

Hardening catalyst 0.25

100

Table 1: MS Sealant recipe

Product Type incorporation conditions

Fumed silicas inorganic 3.7 %, 2500 rpM, 30 min, no external heating

precipitated calcium carbonates (pcc)

inorganic 3.7 %, 2500 rpM, 30 min, no external heating

"crayvallac sL" amide (1st genera-tion)

3.7 %, 2500 rpM, 30 min, 100°c

"crayvallac sLX" amide (2nd genera-tion)

3.7 %, 2500 rpM, 30 min, 80°c

"crayvallac sLt" amide (3rd genera-tion)

3.7 %, 2500 rpM, 30 min, no external heating

Table 2: evaluated rheology modifiers

R h e o l o g y M o d i f i e R8 8


tions. Furthermore, the first amide generations are also more efficient and allow formulators to improve the viscosity stability of their products.A major disadvantage is the need to activate the additive at a high temperature (between 90 °C and 110 °C) under high shear during the dispersion of pigments and fillers.Most developments during the last fifteen years were aimed at ena-bling sealant producers to work at lower temperatures (70 °C to 90 °C for an earlier product and less than 60 °C for the new material).

workinG at low temperature

The idea behind the development of this new bio-based rheology modifier was to develop an amide which allows the adhesive and seal-ant manufacturers to work at low temperatures and achieve a shorter processing time thereby reducing energy consumption and therefore being more environmentally friendly. At the same time, all the benefits of the amide technology needed to be maintained. That is to say, ex-cellent shear-thinning behaviour providing ease of extrusion, ease of application and excellent workability, together with the feature that the pseudoplastic properties prevent slumping and flow once applied.

This new generation of amide has been developed to allow the activa-tion step to take place between 45 °C and 60 °C depending on the formula. It has a direct impact on the processing time of the sealant.In Figure 2, the orange curve represents a regular processing time for a sealant using an amide which requires heat activation at 80 °C. The full process is relatively long and takes about two hours. The main time-consuming steps are the activation and the subsequent cooling.Using the new generation of amide (Figure 2, blue curve) allows the activation process to be carried out at a temperature of 50 °C. This means that external heating is no longer necessary, since this tem-perature is easily and quickly achieved by shear. The full activation also becomes easier and faster. Only 20 to 40 minutes are necessary to fully develop the 3D network of fibres. The result is that the new amide generation enables the sealant manufacturing time to be al-most halved.

aChieVinG more by usinG less

This is not the only advantage of this new product. The market trend is to develop materials that can do more with less and to design mol-

figure 3: Molteni lab mixer

figure 4: Brookfield viscosities

figure 5: Storage stability

figure 6: optimum activation conditions for SlT product

8 9R h e o l o g y M o d i f i e R


ecules which are more effective at the same content level and thereby save energy, raw material costs and also waste. The following case study shows that, in that respect, this new amide fulfils all necessary requirementsThis case study involved a comparison of various additives in a silane terminated polymer. Several rheology modifiers were evaluated and compared in a low-modulus, hybrid-sealant formulation (Table 1) at the same addition level, using a Molteni lab mixer (Figure 3). Incorporation conditions were adjusted for each additive in order to achieve the opti-mum performance of each technology.The different rheological additives (Table 2) were added at 3.7 % in weight on the total formulation and were compared in terms of pro-cessing time, ease of use, gunnability, viscosity, storage stability and me-chanical properties.Viscosities at low shear were measured three days after production using a Brookfield viscometer equipped with a Helipath system (usu-ally used for very thick materials). Results are shown on Figure 4.All the additives tested exhibited a shear-thinning behaviour. PCC did not perform so well and showed poor viscosities at this level. Actually, PCC require a higher addition level to achieve a significant rheologi-

figure 7: Comparison of Brookfield viscosities of different technologies

cal effect. Usually, the recommended level of PCC is around 20 %. All amides and fumed silica provide a good viscosity and a high body.Nevertheless, the level of fumed silicas would need to be increased in order to reach similar properties to the amide additives.

improVed perFormanCe, VisCosity and struCture

This latest amide development shows improved performance, viscos-ity and structure. Performance is significantly improved. As this addi-tive is more efficient, the level at which it is used can even be reduced compared to the older generation of amides.Viscosity stability is also an important parameter for sealant manufac-turers. Especially for amide additives, this data also makes it possible to verify that the activation of the rheological additive has been fully achieved during the sealant manufacturing process. Figure 5 shows the viscosity stability after 15 days at 50 °C compared to the viscosity measured three days after production.The earlier amide (shown as SL) exhibits a significant increase of vis-cosity. This means that the additive was not fully activated during the manufacturing process. A higher activation temperature, or a longer dispersion time, is needed to achieve this. This is the main disadvan-tage of the first generation of amides due to the fact that the neces-sary activation temperature is difficult to control at such a high level.The viscosity of fumed silicas also increases on storage. This is normal with this technology.On the other hand, the latest generations of amides, SLX and SLT, show very good viscosity stability. Full and correct activation has taken place. These two products are also more robust against variations of the incorporation conditions.Another test has been performed in order to define the optimum ac-tivation conditions of the new additive, SLT. Therefore, it was activated at three different temperatures (30 °C, 40 °C and 50 °C) during 15 minutes. Then, the rheological performances were measured (Figure 6).It is clear that SLT is properly activated as from 40 °C. Performance changes little between 50 °C and 40 °C. At 30 °C, activation is incom-plete. The additive would require a much longer dispersion time, around 45 to 60 minutes.

ComparinG additiVes in epoxy adhesiVes

A second case study was carried out to compare various additives in epoxy adhesives. The results were as follows. In the past, amide-based rheological additives were often only con-sidered suitable for use in silane based sealants, but in fact they are compatible with many different adhesives and sealants. The following results were obtained from tests in an epoxy system. In this case the different additives were compared in the epoxy part of a 2K epoxy ad-hesive. SLX was measured against a previous generation of material and with fumed silicas. The different additives were added at a level of 1.5 %. The activation step took place under the same conditions for all samples, i.e. at a temperature of 50 °C at 2.000 RPM for one hour. The results obtained are set out in Figure 7.Both amide derivatives gave good results. Viscosity at low shear was excellent and the shear thinning behaviour was good, thus guarantee-ing excellent storage stability and application properties. Fumed silicas possess lower viscosity at low shear than amides. Thus more fumed silicas are needed to reach the same level of performance. The best result is obtained with the product containing the SLX.As shown in Figure 8, this family of amide powders is more than 90 % bio-sourced. More specifically, the latest generation of amide, SLT, is bio-sourced and bio-renewable at 98 %.

figure 8: The bio- renewable content of the family of additives discussed

9 0


R a d i at i o n c u R i n gSo

urce

: dem

arco

- Fo

tolia

.com

The hearT of The maTTernovel approaches to gloss reduction in uV coatings. By Björn tiede, Lubrizol.

matting of 100 % solids UV curable coatings can be difficult and film properties may also be impaired. Tests are reported on two new products which reduce gloss in different ways from established additives. Compared with waxes and silicas, per-formance is less dependent on film thickness but more de-pendent on oligomer reactivity.

D emands on coating materials in terms of cost efficiency, opti-mised cycle times and emission of volatile organic compounds (VOC)

have changed significantly during the last ten years. Industry has found

several approaches to meet today’s demands on new coating materials.Radiation curable coatings show great potential, especially in terms of short cycle times, reduced VOC emissions and outstanding film proper-ties. This technology offers coating manufacturers the option to formu-late highly versatile coatings for almost every field of application.Compared to conventional coatings (i.e. waterborne

Figure 1: Representation of different matting mechanisms: top, particle based; middle, internal reflection; bottom, excimer pattern

Figure 2: gloss development at different addition rates of three different additives

9 1R a d i at i o n c u R i n g


Results at a glance

ű Matting of UV curable coatings can be difficult, especially in the case of 100 % solids systems. Film properties may also be impaired if monomer levels have to be increased when the matting agent increases viscosity.

ű A new type of matting additive produces controlled inter-nal film crystallisation during cure, so that the light scattering required for gloss reduction occurs inside the film instead of on the surface.

ű Another new additive also creates a wrinkled surface by modifying the surface flow of the coating during cure. This is more efficient than the internal crystallisation method alone.

ű With both new additives performance is almost inde-pendent of film thickness, whereas efficiency of waxes or silicas can decrease at high film thicknesses. However, the new ad-ditives are more affected by differences in oligomer reactivity.

‘concentrate’ the matting agent during physical drying. During evaporation of volatile compounds, the applied film starts to shrink. This shrinkage can vary from 30 % up to 60 % of the wet film volume depending on volume solids. Compared to this, 100 % solids UV coatings show film shrinkage of between 3 % and 8 % (depending on the resins and monomers used).In the case of 100 % solids UV coatings the absence of volatile solvents leads to various problems for the formulator. To lower the viscosity of the resin, especially with higher amounts of matting agents, the use of low viscosity monomers is unavoidable.As these monomers will crosslink with the resin they will also influence the final film properties. The lower the gloss level of the final coating, the more monomer is needed to obtain an acceptable viscosity. This can lead to impaired film properties, flow and levelling issues, reduced reactivity and other problems. Solventbased systems can overcome this problem due to the thinning ability of the volatile solvents.Another important factor is curing speed. Convention-al coatings usually show open times between five and 30 minutes, whereas the curing process of UV coatings takes place in seconds. This short time frame for curing in UV coatings also prevents adequate orientation / floating of the matting agents and lowers their ef-ficiency. The result is a much lower efficiency of the matting agents, which needs to be compensated by higher addition rates.Below, the use of classic matting agents (silica and wax based) is com-pared with new approaches to matting UV curable coatings. The main focus of this article is to compare different matting technologies (com-mercial and experimental ones) and show up the individual strengths and weaknesses of the different technologies.

ConVenTional maTTing oUTlineD

The traditional way of reducing the gloss of a coating is the use of a solid matting agent. In general these matting agents are silica or wax powders. (In many applications silica is preferred due to its higher efficiency.) Using a classic silica or wax as matting agent means using a particle to add light scattering to the final film.In this approach the matting agent particles lead to a structured coating surface that no longer reflects light directionally. The final structure can vary between fine grain structure and a texture-like effect and is influ-enced by particle size, porosity and chemical nature of the matting agent.Moving away from the classic particle-based matting approach means looking for alternative mechanisms to scatter light in a coating film. Two novel ways to achieve light scattering will be discussed and are illustrat-ed in Figure 1.

inTernal sCaTTering proViDes maTTing in Clear CoaTs

To obtain low gloss values it is necessary to scatter the light reflected by the film. Most matting approaches focus on scattering the light on

Figure 3: appearance of films when using different matting additives at different addition rates: left, Matting agent a; middle, Matting agent B; r ight, Matting agent c

1 pack or 2 pack systems) the field of radiation curable coatings bears different challenges on the formulation and curing side, many caused by the curing mechanism.The formulation of matted clear coats in particular can become an ex-tremely tough challenge to the formulator. In the case of 100 % solids coatings it can be complicated to find the balance between: ą Acceptable viscosity for the targeted application method; ą Targeted gloss level; ą Desired film properties (e.g. scratch resistance, hardness, chemical resistance, etc.)

The naTUre of The maTTing meChanism

To understand the difficulty of matting 100 % UV coatings, the classic mat-ting process must be understood in order to clarify the differences from conventional coatings containing volatile compounds.Due to solvent evaporation, conventional coatings start to orientate and

R a d i at i o n c u R i n g9 2


the coating surface. In the case of clear coats, significant amounts of visible light will pass through the coating film and will be absorbed and reflected by the substrate.By creating light scattering structures such as defined crystal struc-tures inside the coating film, the light that passes into the film will be scattered inside and the film appears matt. In the case of this new matting additive, the matting performance is achieved via controlled crystallisation during cure. This new additive technology is based on the synergistic effect between an acrylic-functional carrier and a spe-cial polymeric solid phase.The internal light scattering behaviour can be seen by looking inside the coating with a laser microscope. Inside the coating a refractive pattern similar to those that are sometimes seen in a swimming pool becomes visible. Each of these structures corresponds to a crystalline border in the film. Due to the interaction of the crystals with light the film becomes matt.

alTernaTiVe ways To aChieVe sUrfaCe sTrUCTUring

By influencing the surface curing of a radiation curable coating, the final film can be forced to form a heavily structured layer. This highly wrinkled surface scatters the incoming light, which can lead to very low gloss values. In general this effect can be obtained by utilising spe-cial curing equipment such as excimer devices in combination with specially formulated coatings.Here this processing equipment will not be examined, but a different way to achieve comparable structuring patterns will be considered. Basically, the structuring effect can be caused by influencing the sur-face flow of a coating during the curing process.In case of the novel chemistry, individual particles cause the coating to flow away from them. This flow process is related to polarity changes of the coating during the curing process and generates micro-scale waves during curing.

Raw material addition %

oligomer 65.0

“irgacure 500” 5.0

uV matting agent 5.0

DpgDa 25.0

total 100.0

table 3: Basic formulation scheme for oligomer compatibility tests

Figure 4: gloss development at different film thicknesses

Matting agent a Matting agent B Matting agent c

general nature proprietary blend of

polymers

proprietary blend of polymers

untreated silica (the classic mat-ting approach)

Main matting mechanism

internal light scattering

internal light scattering plus

surface structuring (without particles)

surface structuring (particles)

Mean particle size

6 µm 9 µm 9 µm

table 1: over view of matting agents used

Raw material addition level %

description

“Laromer Lr 9004 (BasF)” 40.0 polyester acrylic

“Laromer Lr 8986 (BasF)” 31.0 epoxy acrylic

“Laromer DpgDa (BasF)” 24.5 thinning monomer

“Byk a-530 (Byk chemie)”

0.5 air release agent

“irgacure 186 ” (BasF) 4.0 photoinitiator

total 100.0

table 2: Formulation used for comparison of matting agent performance

9 3R a d i at i o n c u R i n g


oVerView of The TesT proCeDUres

The matting processes described above are examined to show spe-cific advantages and disadvantages of each process. The emphasis lies on: ą General efficiency at different use levels ą Stability of matting effect at various film thicknesses ą Compatibility with various oligomers.

The materials used in this comparison are briefly described in Table 1.Due to its high reproducibility and the ease of use, all application tests were carried out by using a drawdown application on contrast charts. The film thickness is indicated in the different segments individually.For all microscopic pictures, a Keyence “VK-X210” laser scanning micro-scope was used. Confocal laser scanning microscopy is used to analyse surface structure effects at a very high resolution. In addition, the tech-

nology is able to effectively look into a coating film to determine the internal light scattering behaviour of a coating.As coating viscosity and formulation costs are important drivers for choosing the right matting agent and setting up the final formulation, the matting agents should show maximum efficiency to keep addition rates at a minimum.To check the efficiency development of the different matting approach-es, the matting agents were incorporated at varying addition rates into a common UV curable coating system based on epoxy- and polyester acrylates. For this test the addition rates varied between 1 % and 15 % matting agent (supply form based on total formulation weight).

resUlTs wiTh The Two new aDDiTiVes CompareD

Matting agents were compared in the formulation shown in Table 2, and results are shown in Figure 2. With Matting Agent A, even at low addition rates the internal light scattering shows a significantly lower gloss compared to a particle-based approach. Interestingly, the matting efficiency shows a maximum at approximately 5 % addition level. After reaching this maximum the matting performance becomes worse when matting agent content is further increased.This phenomenon is related to the matting approach itself. The in-ternal light scattering is caused by controlled crystal formation in the coating film. Increasing the addition level of matting additive increases the number of crystals and the matting performance thus gets better.At a certain point a maximum density of crystalline borders in the film is reached. In this specific system this point is reached at 5 % matting agent content. Adding more matting agent to the system forces the crystalline borders to overlap and finally to disturb each other which results in less efficient light scattering – the gloss goes up (see Figure 3).Matting Agent B utilises a multipurpose approach to reduce the gloss of the final film. The gloss development is comparable to Matting Agent A but the overall efficiency is higher. This similarity in matting behaviour at low addition rates between Matting Agents A and B is related to internal light scattering.Due to the multi-functional approach of using more than one matting mechanism the gloss increase at higher addition rates is significantly lower compared to Matting Agent A. (The loss of performance caused by overlying crystals can be compensated by surface structuring.)On the microscopic pictures in Figure 3 it can be seen that at 1 % ad-dition level the formation of microscopic ‘craters’ occurs around each particle. This texture effect is related to flow effects that occur during the curing process. The coating around each particle moves away and leaves a small dent with the particle in the centre.The size of these dents is approximately 20-30 µm in diameter and 3-4 µm deep. By increasing the amount of matting agent the texture comes closer till it starts to interconnect and form a uniform texture comparable to an excimer pattern.The classic particle-based approach leads to an almost linear matting curve. The more silica is added to the coating the rougher the surface becomes. This can also be seen by looking at the microscopic pic-tures taken from the coating surface, again in Figure 3. In addition the surface structures obtained with silica tend to be more ‘rough’ with less uniformity and a more irregular height profile compared to the texture obtained by Matting Agent B.

how film ThiCkness moDifies maTTing effeCTs

In the field of UV curable coatings, film thickness is another important topic for discussion. Usually the matting efficiency depends strongly on the particle size of the matting agents. In general this means that increas-ing the particle size of silica makes matting efficiency better. This also means that changing the applied film thickness could have a significant influence on the final matting performance.

advantages Limitations

Matting agent a High matting power at low addition levels.no influence on coating viscosity.superior surface feel.Mechanically stable matting effect (as the matting appears inside the film it is not polish-able).uniform matting.

no linear correlation between addition level and gloss values.ideal addition range varies depending on the formulation details.efficiency depends on the formulation details (reactivity, curing speed, raw materials).Limited usability in pigmented systems.

Matting agent B High matting power at low addition levels.Low influence on coat-ing viscosity (compared to standard silica).superior surface feel.Broad compatibility in different oligomers.Very uniform matting with better aesthetics.

no linear correlation between addition level and resulting gloss values.efficiency depends on the formulation details (reactivity, curing speed, raw materials).

Matting agent c efficiency develops almost linearly – easy and reliable gloss control.standard matting method with formulators having long experi-ence.

High viscosity build upindex of refraction prevents any light scattering inside the coating film.results may be dependent on film thickness.

table 4: Property comparison of matting agents tested

R a d i at i o n c u R i n g9 4


In this comparison the influence of film thickness on the efficiency of the different matting mechanisms was tested. The same coating was therefore applied at different film thicknesses.As expected, the pure particle approach shows an almost linear cor-relation between film thickness and gloss. Once the mean particle size of the chosen silica (9 µm) is exceeded, the gloss goes up significantly (see Figure 4).On the contrary, the particle approach with Matting Agents A and B shows an almost stable performance over all film thicknesses tested. Once a sufficient film thickness is reached for internal light scattering to occur, it does not matter if the film thickness changes. This leads to a very consistent performance over a wide film thickness range.

performanCe TesTeD in VarioUs oligomers

As well as efficiency, broad compatibility is a key criterion for a matting additive. To compare the efficiency of the new matting approaches a broad test series was set up including several acrylic oligomers. The main idea behind these trials was to formulate a coating that was as simple as possible (see Table 3) and check different additives in it.The results presented graphically in Figure 5 show that the perfor-mance of Matting Agent A correlates with the reactivity of the oligom-ers. The internal light scattering is based on precipitation processes that occur during the curing process.Thus, the whole matting mechanism is influenced by curing speed and most importantly by the gel time. It is clearly observable that highly reac-tive systems with short gel times prevent the coordinated formation of the matting crystals. The combination of different matting mechanisms

Figure 5: comparison of matting performance in different oligomers

with Matting Agent B results in two major advantages: ą broad compatibility with lower influence of oligomer reactivity; ą higher efficiency than pure internal and pure surface light scattering.

As a significant portion of the matting performance is still linked to struc-tural and crystalline formations, the oligomer reactivity does influence the final matting performance.The silica shows a consistent performance with most of the oligom-ers tested. Reactivity slightly influences the final matting performance, but not as significantly as with Matting Agents A and B. The matting efficiency of the silica particles is influenced more by wetting and flow properties of the oligomers. The final surface roughness that gives the matting efficiency is in most cases almost identical, and so is the gloss.

BenefiTs anD limiTaTions sUmmariseD

Utilising mechanisms other than the traditional ones often means changing the way a formulation is set up to get optimum results. The matting agents discussed above are a first step into developing new ways of controlling the light scattering properties of radiation curable coatings.These additives are the first and second generation of this approach and show that the basic principles already work well but still need further refinement.In general it can be said that a combination of different matting mech-anisms shows superior performance, whereas it appears that surface scattering still gives more consistent and stable matting results. The benefits and limitations of each approach to matting are summarised in Table 4.

Sour

ce: S

tefa

n Sc

hurr

- Fo

tolia

.com

LightWeight

Only 30 grams of paint is used on bikes for the Tour de France. This correlates to merely one sixth of the weight of a coating of a normal bike which weighs around 180 g.

A Long Journey

The Tour de France has been held every July since 1903, with only a couple of inter-missions during World War I and II. The tour, that is among the world‘s most prominent sports events, is regarded to be the hardest bycicle race.



9 6


S u r fa c ta n tSo

urce

: Cm

on -

Foto

lia.c

om

FluorosurFactants Flourishcombining good eco toxicological behaviour with technical performance. By reiner friedrich, Elisabeth Höner and fanny Schooren, Merck KGaa.

the control of the surface energy is a major task in high perform-ing coating industries. During the coating process the properties like wetting and leveling are governed by the interfacial forces within the substrate, the coating and the air. Fluorosurfactants provide low surface tension. a new class of short chain fluori-nated surfactants show superior technical performance in com-bination with a better environmentally friendly profile.

i n high-performance coating technologies control of surface energy is key. Many of the important parameters of a coating like wetting,

leveling, (anti)foaming, adhesion, recoatability, stability of dispersions and so on are governed by the interfacial forces within the substrate/coating/air system. Coating additives are often specially designed to retard at the boundary of these interfaces to influence surface energy in the desired way. To measure the effectiveness of such surface ac-tive substance (= surfactant), a variety of different analytical instru-ments are used. In general they are based on the concept of surface tension or contact angle measurement [1]. For the very obvious effect of wetting, Figure 1 depicts three common situations of how a substrate surface interacts with a coating system: a) perfectly wettable, b) wettable and c) unwettable.

figure 1: three stages of wetting – a) perfectly wettable, b) wet-table and c) unwettable. the contact angle Θ is measured at the meniscus of the liquid/solid interface. the spreading behaviour is governed by the interfacial forces of the surface tension of the liquid gLa , the surface energy of the substrate g Sa and the interfacial tension gSL bet ween solid and liquid. Spreading coef-ficient S is the sum of the interfacial vectors gLa, gSa and gSL. Spreading occurs only if S > 0.

9 7S u r fa c ta n t


Results at a glance

ű In high-performance coating technologies where control of surface energy is key fluorosurfactants provide low surface tension at smallest concentration – an additional feature of ear-lier versions are its extreme stability at high temperatures and under harsh chemical conditions.

ű However, these fluorosurfactants with chain length >C6 were found to be bio accumulative and toxic.

ű Shorter chain C2 and C3 branched fluorosurfactants have been developed exhibiting remarkable surface tension reduc-tion.

ű Characteristics include super spreading and good dewet-ting resistance, which are beneficial for many coating applica-tions

ű Importantly, the new NIO surfactants are beneficial for cus-tomers and the environment as they perform like the older C8 ma-terial in terms effectiveness and efficacy (surface tension reduction and working concentration) and are less harmful than existing C6 materials.

Good wetting is only achieved when the spreading coefficient S be-comes positive (S>0) when gSA>(gSL+gLA). In other words the surface energy of the substrate has to exceed the combined surface ener-gies of the Liquid/Air and Liquid/Solid interface. In practice this means either to pretreat the substrate surface (rise the surface energy e.g. though plasma treatment) or to lower the surface tension of the coat-ing by adding suitable surfactants.

Applying waterborne coating systems on surfaces like PE or PP is a challenging task, because of the huge discrepancy in surface energy of both systems. Only powerful additives like fluorosurfactants are able to bridge this gap sufficiently and provide reliable and cost efficient solutions. Fluorosurfactants provide low surface tension (< 20 mN/m) at smallest concentration (0.1-0.01 w%). This characteristic is due to the fact that the perfluorinated alkyl chains exhibit almost no interaction with other components. An additional feature of this perfluorinated molecule part is its extreme stability at high temperatures and under harsh chemical conditions [2]. The C-F bonds in fluorosurfactants are not susceptible to biodegradation and therefore regarded to be persistent in the en-vironment. Additionally, fluorosurfactants with chain length >C6 were found to be bio accumulative and toxic. This is why this group of sur-factants is in the focus of regulatory affairs and NGOs [3]. In many areas C6 Fluorosurfactants are now used as C8 replacement. Environmental and toxicological profiles depend on the length of the perfluorinated alkyl chain; the smaller the better. A decline in chain length normally leads to a lower technical performance and higher us-age concentration. A new class of short chain fluorinated surfactants which exhibit superior technical performance in combination with a better environmentally friendly profile than existing C6 materials have been developed.The results were published in two recent articles about the develop-ment of branched short chain fluorosurfactants [4,5]. This article aims at expanding on application results.

c2 anD c3 brancheD FluorosurFactants – a synergistic couple

For anionic surfactants the best solution arose from combining environ-mental friendliness and high technical performance in a branched struc-ture containing three fluorinated alkyl chains C2 or C3. Both versions are now commercially available and are offered as 30% solutions, based on C2-chains or either on C3-chains. Despite their extremely short fluorinated chains the surface tension reduc-tion is quite remarkable. Figure 2 shows the dynamic surface tension profile of a 0.1w% solution of the C2 and C3 surfactants and mixtures thereof.

figure 2: Dynamic surface tension measurement of 0.1 w% aqueous solutions of pure c2 and c3 surfactants and mixtures thereof measured with a bubble pressure tensiometer. the mixtures are indicated as weight % of c3 and c2, e.g. 0:100 in-dicate 0 % c3 and 100 % c2; 20:80 indicate a mixture of 20 % c2 and 80 % c3, etc. Blow up: Minimum surface tension at long bubble life times (~1 minute) of selected mixtures.

a) Water

b) C6 Fluorosurfactants c) C3 Fluorosurfactants

figure 3: Wetting behavior on PtfE 60 sec after deposition of a droplet. a) Pure water does not wet PtfE; b) 0.1 % aqueous solution of a common c6 fluorosurfactants shows improved wet-ting. this effect is not time dependent (the contact angle does not become smaller over time); c) 0.1 % aqueous solution of tivida fL 2500 shows super spreading behavior on PtfE. this effect is time dependent. the solution spreads further and until the amount of evaporated water becomes significant (not shown here).

S u r fa c ta n t9 8


The pure C2 material (100:0) shows a very steep decline in surface ten-sion over the first 100 ms, which indicates that this fluorosurfactant has a high mobility. As bubble life time becomes longer the curve flattens out and approaches a constant level. The surface tension at bubble life times of 1 minute can be regarded as “quasi” static and present therefore the lowest achievable value. For the C2 material this level is about 21 mN/m. The pure C3 material (0:100) shows a different behavior. Up to 700 ms bubble life time the surface tension reduc-tion lag behind the C2 values, indicating less mobility for the C3 mol-ecule. On the other hand a final surface tension reduction to a value of 16.6 mN/m is an outstanding performance which is hardly reached by most C6-fluorosurfactants.If both materials are combined synergies in speed and surface ten-sion reduction can be seen. If only 20 % of C2 is replaced by C3 (see Figure 2, line 80:20) the final surface tension is as low as 18.8 mN/m without sacrificing the mobility of the molecules. This allows tuning of speed and surface tension in a broad range and reach optimum solu-tions for many applications.

super spreaDing on ptFe

As explained in Figure 1 the surface energy of the substrate gSA is of great importance to the spreading of a coating. For “critical” surfac-es (surfaces with low surface energy) only powerful surfactants like fluorosurfactants can provide satisfactory solution. This becomes even more relevant when the coating-system is waterborne. Table 1 lists the surface energy of some common polymers. The substrates which are most critical for coating applications exhibit a surface energy of 30 mN/m and below (PP, PE, FEP and PTFE). PTFE, for example, is used as non-stick surface finish as Teflon, because the low surface energy will prevent water or even grease to adhering to it.The C3 based product exhibits a very unusual behavior on these ex-tremely low energy surfaces. Figure 3a shows a water droplet on a PTFE surface. There is a high contact angle between liquid and surface due to the negative spreading coefficient. If a 0.1 w% standard C6 fluoro-surfactant is applied, the contact angle is much lower (~ 45°) indicating a good wetting of the surface (Figure 3b). 0.1 w% of the C3 surfactants shows a different picture. Here the droplet spreads further apart lead-ing to a complete spreading of the water film on the PTFE surface (Fig-ure 3c). This phenomenon is referred to as “super spreading”. Super spreading is only common in trisiloxane surfactants, which are used in coatings, cosmetics and agrochemical [6]. The mechanism

behind the rapid spreading is still not fully understood, but it seems to be correlated with a branched structure [7,8] the ability to form bilayer aggregates [9] and the Marangoni effect [10-12]. In the case of low surface energy substrates (<30 mN/m) the spreading rate of trisiloxane solutions is substantially reduced [13]. There is no example of super spreading on low energy surfaces like Teflon reported in the literature [14, 15].

reDuce Dewetting DeFects

Another feature which is important for general coating application is to prevent dewetting effects of surfaces. Especially on rough or un-even surfaces the lacquer has to level irregularities leading to an inho-mogeneous thickness of the coating. This remains a major challenge in industrial wood coating industry. Surfaces here do not only differ among the types of wood used, but also vary within one wood species due to natural fluctuations. To demonstrate the efficacy of a coating to withstand dewetting at various thicknesses the following test was applied. A white aqueous low VOC alkyd enamel dispersion was formulated with 0.1 % with the C2-base product and 0.1 % C-6 fluorosurfactants respectively. A panel was dipped into the dispersion and slowly pulled out. Immediately a gradient in film thickness over the substrate oc-curs, as shown in Figure 4. If the panel is kept in the upright position the dispersion continues to drip from the panel leaving behind a thin layer coating. Good dewetting resistance is assured if, over time, no “black spots” occur on the panels. The film thickness of the coating increases from 0 to 100 µm (top to bottom). The images in Figure 4 show that the C2-based product is far superior to common C6 fluorosurfactants in resisting the tendency of the coating to dewet, particularly at low film thickness.

new class oF non ionic surFactants

The concept of short chain branched surfactants can also be applied to non ionic fluorosurfactants. The design of these surfactants con-sists of two fluorinated alkyl chains -C2, C3 or C5 resp. - and a polygly-col chain. Through the length of the polyglycol chain the hydrophilic-lipophilic balance of the molecule can be tuned effectively. For the investigation ethoxylation degrees of 8, 12 and 17 repeating units yielding nine different surfactants (see Figure 5) were used. The CMC

figure 4: Images of black panels coated with a white aqueous low VOc alkyd enamel dispersion via dip coating. Immediately after pulling out the panel a photo was taken (time = 0 min) and the panel was kept in an upright position. after 5 minutes the second picture was taken. Surfactant concentration 0.1 w% (if applicable).

9 9S u r fa c ta n t


(critical micelle concentration) of these compounds was estimated to compare them with existing C6 material and the “old” C8 level. Figure 5 shows that C2-surfactants have the lowest impact on surface tension reduction and also show a high CMC (for C2-EO-12 and C2-EO-17 no precise CMC could be determined, but seems to be > 1 g/l). It seems that the existence of only two C2 groups per molecule is not sufficient enough to design high performing surfactants. C3 groups seem to be more suitable, showing a CMC in the range of existing C6 materials and a surface tension between 22 and 24 mN/m. Best performance show branched NIO surfactants with C5 chains having a very good surface tension reduction in combination with very low CMC values (C5-EO-8; -12; -17). Since CMC reflects the ef-ficiency of a surfactant, the branched C5 design outperforms exist-ing C6 materials because the working concentration for applications will be much lower. Additionally these materials show a much lower tendency to foam than existing C6 material. Therefore they are much more suitable e.g. for high thoughput printing application were foam is a very critical parameter. Last but not least the new NIO surfactants are beneficial for custom-ers and the environment; they perform like the “old” C8 material in terms effectiveness and efficacy (surface tension reduction and work-ing concentration) and are less harmful than existing C6 materials.

conclusions

Branched short chain fluorosurfactants are a new class of environ-mentally friendly, high performing surfactants, which show a variety of benefits in multiple technical applications. The main characteris-tics of this surfactants group are: fast acting, high performing and low foaming. Additional to the technical benefits this material class can be considered as the most environmentaly friendly fluorosurfactants on the market due to its outstanding toxicity and non hazard profile.

figure 5: comparison of the surface tension of branched short chain fluorosurfactants with standard c6 and “old” c8 mate-r ial. the surfactants are grouped according to their fluorinated chains. the crosses above the bars indicate the cMc of the par-ticular compound in g/l. nomenclature: E.g. c2-EO-8: Surfactant containing t wo c2-groups and a polyglycol chain with 8 Ethoxy-repeating units. c6-1 to 3 present three different, commercially available c6 nIO surfactants.

Polymer abbreviation Surface energy at 20 °c [mn/m]

polyethylene tereph-thalate

pet 42

polymethylmeth-acrylate

pMMa 41

polyvinyl alcohol pVa 37

polystyrene ps 34

polyethylene pe 30

polypropylene pp 30

Fluorinated pe/pp Fep 20

polytetrafluoroethylene ptFe 19

reFerences

[1] couper c., in “physical Methods of chemistry, 2nd ed., rossiter B.W. and Baetzold r.c., eds., Vol. iX John Wiley sons , new York (1993)

[2] Kissa e., Fluorinated surfactants and repellents, surfactants science series 2001, 97; 80ff

[3] environmental protection agency (epa), 2010/2015 pFoa stewardship program

[4] Jonschker g. et al, europ. coat. Jnl, 2010, no. 7-8, pp 24- 27[5] schellenberger s. et al, europ.coat. Jnl, 2012, no. 11, pp 32-36[6] ivanova n., starov V., Johnson D., Hilal n., rubio r., Langmuir., 2009, 25, 3564-3570[7] ananthpadnamabham K. , goddar e. D., chandar p., colloids surf.,1990,

44, 281–297.[8] shen Y., couzis a., Koplik J., Maldarelli ch., tomassone s., Langmuir.,2005,

21, 12160–12170.[9] Venzmer J., Wilkowski s. p., trisiloxane surfactants – mechanisms of

spreading and wetting. pesticide Formulation and application systems: 18th Volume,astMstp1347; nalewaja J. D., goss g. r., tann r. s., eds.; american society for testing and Materials: West conshohocken, pa, 1998; pp 140-151

[10] chengara a., nikolov a., Wasan D., colloids surf., a 2002, 206, 31.[11] nikolov a. D., Wasan D. t., chengara a., Koszo K., policello g. a.,

Kolossvary i., adv. colloid interface sci. 2002, 90, 325.[12] chengara a., nikolov a. D., Wasan D. t., ind. eng. chem. res. 2008, 47,

3639–3644.[13] Kumar n., couzis a., Maldarelli c. J., colloid interface sci. 2003, 267, 272.[14] Zhu s., Miller W. g., scriven L. e., Davis H. t., colloids surf., a 1994,90,

63–78.[15] svitova t., Hill r. M., smirnova Yu., stuermer a., Yakubov g., Langmuir.,

1998, 14, 5023

table 1: Surface f ree energy of some commercial plastics [source: WWW]

1 0 0


D r i e r s

Sour

ce: i

mag

o pu

bblic

ità -

Foto

lia.c

om

Cutting ketoximesencapsulated driers minimise skinning of air-drying paints. By Jörg Horakh, Hugh Gibbs, Franjo Gol and Gaby Kiepe, OMG Borchers.

Air-drying alkyds require anti-skinning agents to prevent premature curing in the can. there is regulatory pressure in europe to reduce exposure to these agents to levels which are difficult to meet. An iron drier encapsulated so that it is protected from oxygen during storage but released during ap-plication produces minimal skinning. good to know that they don’t reduce the coatings performance.

A ir-drying alkyd resins play a major role in DIY and professional paints as well as in some industrial applications. This is mainly due to their

outstanding performance in terms of application properties including gloss, levelling, pigment wetting and drying under adverse conditions.Additionally, autoxidatively curing alkyd resins allow one to formulate

single-component paint systems which crosslink at ambient tempera-tures. The introduction of high solids binders as well as water-based alkyds has been a major contribution to reduced VOC emissions of paints, providing eco-friendly alternatives for paint manufacturers.The high content of naturally based raw materials in alkyd resins pro-vides a promising alternative to fossil sources. Therefore, the increas-ing demand to use renewable raw materials will be a main driving force towards future developments of alkyd based paints.

Anti-skinning Additives fACe regulAtory ChAllenges

Paint manufacturers and their suppliers are currently facing chal-lenges caused by the pending threat of hazard labelling of two key

Figure 1: seM picture of capsules and particle size distr ibution of a pilot batch of the encapsulated siccative

1 0 1D r i e r s


Results at a glance

ű Skin formation during manufacture and storage of air-drying alkyd-based paints has historically been resolved mainly by the use of MEKO and cyclohexanone oxime, which exhibit very similar toxic effects.

ű There is pressure in Europe to reduce exposure to these compounds to levels which are difficult to meet. Fu-ture regulations may also require replacement of cobalt dri-ers.

ű A solution is to deactivate driers in the paint can but make them active on application. An iron drier was encapsu-lated so that it is protected from oxygen during storage, but when the paint is applied the capsules break and release the drier.

ű This system was shown to reduce the required lev-els of anti-skinning agents to very low values, with no reduc-tion in drying performance or physical properties compared to the non-encapsulated form of the same drier.

ingredients of most alkyd paints: methyl ethyl ketoxime (MEKO) and cobalt carboxylates.Historically, the issue of skin formation during manufacture and stor-age of air-drying alkyd-based paints has been addressed mainly by the use of MEKO and cyclohexanone oxime, which exhibit very similar toxic effects. There is pressure in Europe to reduce exposure to levels which are difficult to meet. Current concerns are [1]: ą Actual working and workplace exposure levels, especially for pro-fessional uses and the possible need for further risk management measures.

ą If MEKO becomes classified as carcinogen 1B, (H350 labelling) it will affect sales to the general public and the substance would fulfil the criteria for identification as SVHC (Substance of Very High Concern)

according to Article 57(a).Additionally, anti-skinning agents can slow drying after application, by reducing the activity of the metal driers and if their own evaporation is too slow. In order to overcome this skinning issue, one solution is to find ways of deactivating driers in the paint can but making them active on application.

regulAtions mAy require replACement of CobAlt driers

Cobalt carboxylates are the catalysts (siccatives or driers) most fre-quently used for the oxidative curing of alkyd resins. But their use is under pressure due to their probable classification as carcinogens.Consequently, there is great interest in possible alternatives for Co with no technical drawbacks. Only Mn- and Fe-based driers have been proved in commercial paints to be viable alternatives. In particular the iron-based catalyst containing the polydentate ligand FeLT [2] in the “Borchi Oxy-Coat” product range shows no drawbacks with respect to paint drying activity at extremely low dosage levels. Additionally, the notorious yellowing of alkyds is minimised with FeLT-based driers [3] and they work exceptionally well in cold and humid weather.

drier enCApsulAtion CAn resolve skinning problems

In order to overcome the skinning issue, it was necessary to find ways of deactivating driers in the paint can but making them active on appli-cation. The solution is to encapsulate the iron drier so it is protected from oxygen during storage.When the paint is applied, the capsules break and release the drier so it can react with oxygen and function as a drier without loss of the drying performance in oxidatively curable alkyd-based formulations.A microencapsulated catalyst solution is prepared by an interfacial polyaddition process and suitable microcapsule wall materials are selected from polyureas, polyurethanes and combinations of these materials.Microencapsulation by interfacial reaction is carried out by initially preparing an emulsion in which the disperse phase is an aqueous solution of the siccative and the continuous phase is an aliphatic sol-vent. The polymer is formed at the phase interface of the dispersed spherical droplets and forms the shell of the microcapsule.The result of the microencapsulation process is a dispersion of micro-capsules in an organic solvent in which each microcapsule comprises

Figure 2: Drying times of a high solid clearcoat comparing cobalt driers and FeLT with and without encapsulation

Figure 3: skin formation of a high solids clearcoat with 0.2 % anti-skin agent during four weeks at ambient conditions

D r i e r s1 0 2


an aqueous solution of the metal siccative as the core within a polymer wall [4]. Figure 1 shows a typical particle size distribution of a pilot batch.

AppliCAtion tests summArised

The microcapsule dispersion (containing 1 % of the complex, which corresponds to 0.09 % of iron metal) was added to oxidatively curable paints at concentration levels of 0.25 to 0.5 wt% on formulation. The concentrations of the unencapsulated FeLT were always at the same level for comparability reasons.Additionally the product was compared with standard cobalt driers. Calcium and zirconium were used as secondary drier metals. Wet paint films were applied at 100 µm wet layer thickness in order to evaluate drying times and hardness development. Pendulum hardness was de-termined according to König and drying times using BK drying record-ers. Skin formation was evaluated visually (using grades from 0: no skin to 6: complete skin) after four weeks storage at 40 °C and at ambient conditions.

drying speed And skin formAtion tests

Drying times of the encapsulated FeLT were slightly higher compared to cobalt and to the FeLT drier without encapsulation, as shown in Figure 2.Drying properties were investigated in alkyds with variations in oil lengths as well as variations of the constituent fatty acids composi-tion of the alkyds without revealing any significant drawbacks in drying times for the encapsulated drier.Skin formation of paint films was evaluated visually. Figure 3 shows skin formation of a high solids clearcoat with 0.2 % anti-skin agent dur-ing four weeks at ambient conditions. Encapsulated FeLT showed no skinning, whereas samples containing the cobalt and unencapsulated FeLT- displayed medium to strong skin formation.The use of small amounts of anti-skinning agents is recommended, in order to prevent small amounts of skin formation which could even be observed for alkyd resins without any siccative added.

enCApsulAtion does not AffeCt hArdness or yellowing

Comparison of pendulum hardness developments of standard driers

and the encapsulated FeLT-drier showed no influence of the encapsu-lation on the hardness development.Neither oil length nor variations of the fatty acid composition of the alkyd resin revealed any significant influence of the encapsulation on the hardness development of the applied films, indicating that the catalyst in the film is fully available for the curing process. Figure 4 shows the results for hardness development of linseed oil-based al-kyd paint films.Dark-yellowing of white pigmented paint films (Figure 5) was evaluated by colour measurements using an “X-rite 8200” spectrophotometer. Encapsulated FeLT has a significantly lower influence on long term yel-lowing of the system compared to other standard driers such as co-balt or manganese, which are noted for their strong tendency to pro-mote discolouration [5]. The influence of the encapsulated drier on yellowing of alkyds is comparable to the non-encapsulated product.

meChAniCAl And thermAl stAbility Are both good

Thermal stability of the encapsulated drier dispersion was tested by storing samples for three weeks at -18 °C, 8 °C, 23 °C and 40 °C. Eval-uations of drying and skin-forming properties in paint formulations after storage showed no significant deviations. In the case of paints containing encapsulated FeLT the encapsulated drier showed no ad-verse influence on the paint stability (3 weeks at 40 °C and at ambient conditions).Dispersions of the capsules as well as various paints containing the dispersion were dispersed for up to one hour at high shear rates im-parting high shear stress on the capsules but resulting in no change of the drying or skin-forming properties.

polAr solvents CAn however disrupt enCApsulAtion

As a straightforward preliminary test of the stability of the encapsu-lated FeLT, the dispersion was mixed with various solvents in the ra-tio 10:1. Addition of polar, protic solvents yielded brownish sample colours as visible in Figure 6. Table 1 summarises colour changes and the viscosity increases. The observed increases in viscosity depend strongly on the solvents and are apparently caused by swelling of the microcapsules.Addition of polar solvents with OH- and NH- functionalities results in FeLT and encapsulated FeLT having equal drying behaviour and equal

Figure 4: Comparison of pendulum hardness development dur-ing five weeks after application of linseed oil-based alkyd paint films containing Co-, FeLT- and encapsulated FeLT-drier

Figure 5: Yellowing of white pigmented paint films of a long-oil alkyd with Co-, Mn- and Fe-driers; measured after six weeks storage of the applied films

1 0 3D r i e r s


skin formation properties.This is clearly because migration of small polar molecules with OH- or NH- functionalities results in destruction of the capsule walls and re-lease of the drier. Formulations containing these substances will there-fore not benefit from the capsules’ advantages in reducing skin forma-tion.

benefits And limitAtions summArisedThe encapsulated FeLT catalyst offers a sustainable alternative to the use of cobalt and MEKO in alkyd-based paints: ą Drying properties of encapsulated FeLT are similar to FeLT, cobalt and manganese driers;

ą The tendency to skinning is strongly reduced even without anti-skinning agents;

ą MEKO is not necessary as an anti-skinning agent; low amounts of anti-skin are sufficient, having a low impact on drying time;

ą Encapsulation has no influence on hardness development; ą FeLT from capsules is, however, readily available for extraction in low molecular solvents with OH- and NH-functionalities, yielding drying times and skinning properties comparable to FeLT without encapsulation;

ą No CMR-classification; ą Significantly less discoloration compared to cobalt or manganese driers

referenCes

[1] Results of the evaluation of 2-Butanone oxime (MEKO) published October 2014

on the ECHA website: substance evaluation conclusion document, as required by

REACH Article 48 for Butanone oxime (EC No 202-496-6; CAS No 96-29-7).

[2] Hage, R., Wesenhagen P. V., Liquid Hardening, Patent WO/2008/003652.

[3] de Boer J. W. et al, The quest for cobalt-free alkyd paint driers, Eur. Jnl. Inorg.

Chem., 2013, pp 3581-3591.

[4] Gibbs H. W. et al, Encapsulated Catalysts, Pat. Applic. WO/2015/011430.

[5] Soucek M. D., Katthab T., Wu J., Review of autoxidation and driers, Prog. in Org.

Coat., 2012, Vol. 73, pp 435-454.

Figure 6: samples with 10 % of various solvents on the encapsulated FeLT-dispersion showing discoloration

solvent (10 %) Viscosity increase & colour change

Xylene no

Mpa no

Water yes

ethanol yes

Buac no

White spirit no

propylene glycol yes

DMea yes

DpM no

MeK yes

Table 1: effect of adding 10 % solvent to encapsulated FeLT-dispersion: significant viscosity increase and colour change when polar molecules with OH- and NH-functionalities are used

Documents

EuropEan CS oaTInG€¦ · european coatings JournaL 2016 BYK ADDITIVEs: Improving product characteristics and production processes. Small quantities, big properties. BYK Additives