7
RESEARCH ARTICLE The risk of mixing dilute hydrogen peroxide and acetone solutions The present study documents the results of a literature search and experimental work to assess the risks of mixing dilute H 2 O 2 and acetone solutions. The use of dilute H 2 O 2 to clean chemical vessels is common, but it has been shown to be potentially hazardous due to the reaction of H 2 O 2 with organic solvents to form explosive peroxides. Mixing concentrated H 2 O 2 and acetone with an acid catalyst is known to form the shock and friction sensitive explosives triacetone triperoxide (TATP) and diacetone diperoxide (DADP). A search of the chemical literature was unable to identify any directly applicable research or technical information that addressed the potential formation of explosive peroxides when mixing dilute H 2 O 2 and acetone solutions. The conclusion of these experiments is that when mixing dilute solutions, such as less than 3% H 2 O 2 and 7% acetone, the solutions are unlikely to form significant amounts of TATP or DADP. In the presence of an acid catalyst, hundreds of parts per million of organic peroxides can be formed. Although TATP is relatively insoluble in water, it is soluble at roughly the 15 ppm level and higher for acetone and H 2 O 2 solutions, thus any acetone peroxides that are formed without acid catalyst should remain soluble in the aqueous cleaning solution. By [TD$FIRSTNAME]Jimmie C.[TD$FIRSTNAME.E] [TD$SURNAME]Oxley[TD$SURNAME.E], [TD$FIRSTNAME] Joseph[TD$FIRSTNAME.E] [TD$SURNAME]Brady[TD$SURNAME.E], [TD$FIRSTNAME]Steven A.[TD$FIRSTNAME.E] [TD$SURNAME]Wilson[TD$SURNAME.E], [TD$FIRSTNAME] James L.[TD$FIRSTNAME.E] [TD$SURNAME]Smith[TD$SURNAME.E] INTRODUCTION Hydrogen peroxide mixed with organic solvents is known to form dan- gerous peroxides. Hydrogen peroxide and acetone is an especially hazardous combination that can form various explosive peroxides when mixed at high concentration while using an acid catalyst. 1,3–19 The primary explosive triacetone triperoxide (TATP) is a noteworthy reaction product, because it is insoluble in water and is highly sensitive to friction, shock and tem- perature. It is called a primary explo- sive due to its ease of detonation. Note, that other peroxide molecules can also be formed, including diacetone diper- oxide (DADP). Figure 1 provides the molecular structures and physical properties of TATP, DADP and other peroxide products that are formed according to the literature. The potential for forming explosive peroxides when mixing dilute solu- tions of H 2 O 2 and acetone has not been documented. Industrial scrub- bers are routinely used in chemical plants. Periodic cleaning to remove bio-sludge buildup inside the scrubber is required. One approach is introduc- tion of dilute aqueous hydrogen per- oxide (H 2 O 2 ). A concern is that in a scrubber used to remove acetone from process air, the combination of H2O2 and acetone could form explosive per- oxides. To evaluate this concern, these reagents were combined in acidic media and alkaline media, and in the presence of steel coupons containing welding materials. Prior to experimental work, a com- prehensive literature search was con- ducted. Wolffenstein first reported the formation of TATP in 1885 when mixing equal parts of acetone with 10% H 2 O 2 . 1 He found that TATP crystals were inso- luble in water, acids and alkalis but readily soluble in organic solvents such as acetone. Shanley and Greenspan 2 prepared a review article on the physical and chemical properties of highly con- centrated H 2 O 2 solutions in 1947. They provided a ternary diagram mapping the detonable compositions of mixtures of acetone–water–H 2 O 2 solutions. The explosive power was measured by the inspection of lead pipes containing 10 mL mixtures initiated by blasting caps. There was no identification or isolation of TATP in this study. Milas and Golubovic 3 studied the products of the reaction of H 2 O 2 and acetone with and without acid. They found that solu- tions of acetone and hydrogen peroxide in the presence of acid would predomi- nately form the trimer TATP. In the absence of acid, 2,2-dihydroperoxypro- pane was formed and slowly converted to [dioxybis(1-methylethylidene)]bis- hydroperoxide. Sauer and Edwards 4 Jimmie C. Oxley is affiliated with the Chemistry Department, University of Rhode Island, Kingston, RI 02881, USA (Tel.: 401 874 2103; fax: 401 874 2103; e-mail: [email protected]). Joseph Brady is affiliated with the Chemistry Department, University of Rhode Island, Kingston, RI 02881, USA. Steven A. Wilson is affiliated with the Eastman Chemical Company, King- sport, TN 37662, USA. James L. Smith is affiliated with the Chemistry Department, University of Rhode Island, Kingston, RI 02881, USA. 1871-5532/$36.00 ß Division of Chemical Health and Safety of the American Chemical Society 27 doi:10.1016/j.jchas.2011.07.010 Elsevier Inc. All rights reserved.

The risk of mixing dilute hydrogen peroxide and acetone solutions

Embed Size (px)

Citation preview

Page 1: The risk of mixing dilute hydrogen peroxide and acetone solutions

RESEARCH ARTICLE

The risk of mixing dilutehydrogen peroxide andacetone solutions

Jimmie CChemistRhode IUSA (Tefax: 401e-mail: j

Joseph BChemistRhode IUSA.

Steven AEastmansport, TN

James LChemistRhode IUSA.

1871-5532

doi:10.101

The present study documents the results of a literature search and experimental work to assess the risks ofmixing dilute H2O2 and acetone solutions. The use of dilute H2O2 to clean chemical vessels is common, but ithas been shown to be potentially hazardous due to the reaction of H2O2 with organic solvents to formexplosive peroxides. Mixing concentrated H2O2 and acetone with an acid catalyst is known to form theshock and friction sensitive explosives triacetone triperoxide (TATP) and diacetone diperoxide (DADP). Asearch of the chemical literature was unable to identify any directly applicable research or technicalinformation that addressed the potential formation of explosive peroxides when mixing dilute H2O2 andacetone solutions. The conclusion of these experiments is that when mixing dilute solutions, such as lessthan 3% H2O2 and 7% acetone, the solutions are unlikely to form significant amounts of TATP or DADP. Inthe presence of an acid catalyst, hundreds of parts per million of organic peroxides can be formed. AlthoughTATP is relatively insoluble in water, it is soluble at roughly the 15 ppm level and higher for acetone andH2O2 solutions, thus any acetone peroxides that are formed without acid catalyst should remain soluble inthe aqueous cleaning solution.

By [TD$FIRSTNAME]Jimmie C. [TD$FIRSTNAME.E] [TD$SURNAME]Oxley [TD$SURNAME.E], [TD$FIRSTNAME]Joseph [TD$FIRSTNAME.E] [TD$SURNAME]Brady [TD$SURNAME.E],[TD$FIRSTNAME]Steven A. [TD$FIRSTNAME.E] [TD$SURNAME]Wilson [TD$SURNAME.E], [TD$FIRSTNAME]James L. [TD$FIRSTNAME.E] [TD$SURNAME]Smith [TD$SURNAME.E]

INTRODUCTION

Hydrogen peroxide mixed withorganic solvents is known to form dan-gerous peroxides. Hydrogen peroxide

. Oxley is affiliated with thery Department, University ofsland, Kingston, RI 02881,l.: 401 874 2103;874 2103;[email protected]).

rady is affiliated with thery Department, University ofsland, Kingston, RI 02881,

. Wilson is affiliated with theChemical Company, King-37662, USA.

. Smith is affiliated with thery Department, University ofsland, Kingston, RI 02881,

/$36.00

6/j.jchas.2011.07.010

and acetone is an especially hazardouscombination that can form variousexplosive peroxides when mixed athigh concentration while using an acidcatalyst.1,3–19 The primary explosivetriacetone triperoxide (TATP) is anoteworthy reaction product, becauseit is insoluble in water and is highlysensitive to friction, shock and tem-perature. It is called a primary explo-sive due to its ease of detonation. Note,that other peroxide molecules can alsobe formed, including diacetone diper-oxide (DADP). Figure 1 provides themolecular structures and physicalproperties of TATP, DADP and otherperoxide products that are formedaccording to the literature.

The potential for forming explosiveperoxides when mixing dilute solu-tions of H2O2 and acetone has notbeen documented. Industrial scrub-bers are routinely used in chemicalplants. Periodic cleaning to removebio-sludge buildup inside the scrubberis required. One approach is introduc-tion of dilute aqueous hydrogen per-oxide (H2O2). A concern is that in ascrubber used to remove acetone fromprocess air, the combination of H2O2and acetone could form explosive per-oxides. To evaluate this concern, thesereagents were combined in acidic

� Division of Chemical Health

media and alkaline media, and in thepresence of steel coupons containingwelding materials.

Prior to experimental work, a com-prehensive literature search was con-ducted. Wolffenstein first reported theformationofTATPin1885whenmixingequal parts of acetone with 10% H2O2.

1

He found that TATP crystals were inso-luble in water, acids and alkalis butreadily soluble in organic solvents suchas acetone. Shanley and Greenspan2

preparedareviewarticleonthephysicaland chemical properties of highly con-centrated H2O2 solutions in 1947. Theyprovideda ternarydiagrammappingthedetonable compositions of mixtures ofacetone–water–H2O2 solutions. Theexplosive power was measured by theinspection of lead pipes containing10 mL mixtures initiated by blastingcaps. There was no identification orisolation of TATP in this study. Milasand Golubovic3 studied the products ofthe reaction of H2O2 and acetone withand without acid. They found that solu-tions of acetone and hydrogen peroxidein the presence of acid would predomi-nately form the trimer TATP. In theabsence of acid, 2,2-dihydroperoxypro-pane was formed and slowly convertedto [dioxybis(1-methylethylidene)]bis-hydroperoxide. Sauer and Edwards4

and Safety of the American Chemical Society 27Elsevier Inc. All rights reserved.

Page 2: The risk of mixing dilute hydrogen peroxide and acetone solutions

[(Figure_1)TD$FIG]

HO OH +

O

OOHHOO O O OOHHOO

OO

O O

OO

O

OO

O

1,1'-(1-methylethylidene)bis-hydroperoxide(2,2-Dihydroperoxypropane)

CAS# 2614-76-8[dioxybis(1-methylethylidene)]bis-hydroperoxide

M.P. = 34°C (3)CAS# 1471-09-6

3,3,6,6-tetramethyl-1,2,4,5-tetroxane(Diacetone Diperoxide, DADP)

CAS# 1073-91-2M.P. = 133 - 135 °C (17)

Vapor Pressure (298 K) = 17.7 Pa (17)3,3,6,6,9,9-hexamethyl-1,2,4,5,7,8-hexoxonane

(Triacetone Triperoxide, TATP)CAS# 17088-37-8M.P. = 94 °C (17)

Vapor Pressure (298K) = 7.8 Pa (17)

Figure 1. Reaction Products of Hydrogen Peroxide and Acetone.

studied the formation of 2,2-dihydro-peroxypropane at various pH values.The work provided reaction kineticsparameters, but there was no mentionof TATP or DADP. They later studiedthe formation of higher adducts, butonly proposed that TATP is formed bya reaction between acetone and [dioxy-bis(1-methylethylidene)]bis-hydroper-oxide.5 Evans et al.6 described, in aforensic science case study, how easilyTATP is formed by mixing acetone, 6%H2O2 and concentrated sulfuric acid. Areview by Mackenzie7 addressed thehandling and use of H2O2, and stressedsafety precautions necessary at variousstrengths. There was no mention ofTATP, but he did warn against thebuildup of gases, including oxygen, asH2O2 degraded. In addition, there werewarnings that after the reaction oneshouldbe cautious during process steps,such as distillation, that would concen-trateanyunstableperoxidecompounds.A 1991 review by Mackenzie8 provided

28

safety considerations for processdesigns that use H2O2 and organic che-micals. It highlighted various risks,including the dangers from decomposi-tion of H2O2, as well as a variety ofproduction hazards. U.S. patent5,003,1099 describes a process ofdestroying acetone peroxides formedduring phenol synthesis by heating,decreasing pH, and adding a coppercompound. The work implies that thechemical destruction of acetone perox-ides would require conditions difficultto achieve during industrial processing.In1999,Bellamy10 foundthatonecouldchemicallydestroyTATPby refluxing inethanol containing tin(II) chloride.They also report that TATP is almostas powerful as trinitrotoluene (TNT),but is extremely sensitive to frictionand impact sensitivity tests. TATPis quite soluble in chloroform(42.7 wt%), toluene (28.6 wt%), acet-one (17.2 wt%), and hexane(14.4 wt%). Jiang et al.11 described the

Journal of Chem

preparation of tetrameric acetone per-oxide (CAS #74515-93-8). They claimthat SnCl4�5H2O or SnCl2�2H2O couldact asacatalyst to form a tetramer.Theirresults indicate that no reaction takesplace when treating acetone with 30%H2O2 and no catalyst, but when0.5 mmol of catalyst was added, a>19% yield of tetrameric acetone per-oxidewasformed. In2002,Oxleyetal.12

studied the decomposition of TATP andconcluded that the predominantdecomposition product in the gas phaseismethylacetate. Inacondensed-phase,or within proton-donating solvents,TATP breaks down predominately intoacetone and water. Widmer et al.13

developedanLC/MSmethodtoreplaceGC/MS methods, stating TATPdegrades GC columns. This paper alsoobserved that TATP has two structuralisomers, both stable at room tempera-ture. Their method was sensitive to10 ng per 100 mL (or 0.1 ppm).Schulte-Ladbeck et al.14 developed a

ical Health & Safety, March/April 2012

Page 3: The risk of mixing dilute hydrogen peroxide and acetone solutions

[(Figure_2)TD$FIG]

Calibration Curve for TATP in Dichloromethane

R2 = 0.9987

0

175000

350000

50403020100

Concentration TATP (mg/L)

Peak

Are

a

Figure 2. Calibration Curve for TATP by GC/mECD.

trace analysis method for TATP. Themethod uses liquid chromatography toseparate the analytes followed by post-column UV irradiation and derivativi-zation-based fluorescence detection.Denekamp et al.15 further describedthe two structural isomers of TATPand explained that a high transitionenergy step was the cause of the twoconformations being stable at roomtemperature. In 2004, Oxley et al.16

developed procedures for training dogsto detect TATP. The work highlightsthat loss of TATP from surfaces is dueto its high rate of sublimation. In 2009,Oxley et al.17 studied the vapor densityof TATP. This work determines vaporpressures of both DADP and TATP(17.7 and 7.8 Pa at 25 8C). Dubnickovaet al.18 studied the decomposition ofTATP and determined that it is not anenthalpy-based explosion, but rather,an entropic burst in which four gasmolecules are created from one TATPmolecule. Eyler19 explored the solventeffects on the thermal decomposition ofTATP (135–1708 C) and found that theratewaspositively influencedbysolventpolarity.

EXPERIMENTAL

General Experimental Set-Up

A ‘‘mix’’ was defined as a combinationof hydrogen peroxide (initially50 wt%) and acetone (initially 100%)added to water to give hydrogen per-oxide concentrations that varied from30 wt% to 12 wt% to 3 wt% and theacetone concentrations of 50 wt% or7 wt%. Catalysts included various con-centrations of sulfuric acid, aqueoussodium hydroxide (50 wt%) or metalcoupons. The ability of the ‘‘mix’’ toform TATP was judged by gas chroma-tography (GC) analysis equipped witheither a mass selective detector (MSD)or an electron capture detector (ECD).If, after 48 hours, white solid precipi-tate was observed, the solids were col-lected, dried and analyzed by GC/MSto confirm their identity. If, after48 hours, no solid had formed, thesolutions were left for an additional 3to 6 days and extracted with chloro-form or dichloromethane, and theorganic extract was analyzed by gaschromatography.

Journal of Chemical Health & Safety, March

An Agilent 6890 gas chromatographcoupled with an Agilent 5973i Massselective detector (GC/MS), and anAgilent 6890 gas chromatographcoupled with a micro-electron capturedetector (GC/mECD), were employedto separate, identify and quantify TATP.The GC/MS was fitted with a 6-meterAgilent DB-5MS capillary column witha nominal diameter of 250 mm and filmthickness of 0.25 mm. The carrier gaswas UHP helium at an initial flow rateof 2.5 mL/min and an average linearvelocity 129 cm/sec. The inlet was runwith a 5:1 split at 170 8C at a pressure of1.09 psi. The oven was held at 50 8C for2 minutes, ramped to 190 8C at 20 8Cper minute and the transfer line to theMSD was held at 275 8C. To ensure thatall materials had passed through thecolumn, a 1 minute post run at 280 8Cwith a flow rate of 8 mL/minute wasused.

The GC/mECD was fitted with a 6-meter Restek RTX-200 capillary col-umn with a nominal diameter of530 mm and a film thickness of1.50 mm. The carrier gas was UHPhelium at a rate of 15.8 mL/minutewith an average linear velocity of117 cm/sec. The inlet was held at170 8C in the splitless mode for0.50 minutes at 2.37 psi before theinlet purged at 40.0 mL/minute. Themakeup gas to the mECD was UHPnitrogen run in the constant flow modeat a rate of 30 mL/minute and thedetector was held at 280 8C. The ovenwas initially held at 40 8C for 2 min-utes, ramped at 20 8C/minute to150 8C and held for 1 minute, followedby a 0.50 minutes post run at 250 8C.

The presence and quantity of TATPwas determined by matching the reten-

/April 2012

tion times and mass spectra of samplesto authentic standard solutions. A cali-bration curve for TATP was developedby dissolving a known amount ofauthentic TATP in dichloromethaneand preparing concentrations thatbracketed the amounts that wereexpected during an analytical run. Atypical set of concentrations for ananalytical run on the GC/MS was100, 200, 300, 400 and 500 mg/L ofTATP. A typical set of concentrationsfor an analytical run on the GC/mECDwas 0.5, 1, 2.5, 5, 10, 25, and 50 mg/L.A linear dependence was determinedby measuring the peak height or inte-grating the area under the character-istic peak for TATP and plotting thearea or height versus concentration(Figure 2).

It should be noted that TATP chro-matograms exhibit two peaks, one foreach conformational isomer. Underthe GC conditions used in Figure 3,the isomeric peaks eluted �0.9 min-utes apart. When each peak was exam-ined by GC/MS, the mass spectra wereidentical, confirming that the twopeaks are isomers as reported by Dene-kamp et al.15 (Figure 3).

Solubility of TATP in Acetone andAqueous Acetone Solutions

The solubilities of TATP and DADP inpure acetone were measured as151 mg/mL (0.191 mg/mg) and43 mg/mL (0.0548 mg/mg) respec-tively. TATP was found to have a solu-bility of 200 mg/mL (0.298 mg/mg) inchloroform. These values were deter-mined by adding solvent drop wise tothe solid peroxide while stirring in around-bottom flask. Qualitative solu-bility tests in aqueous acetone and

29

Page 4: The risk of mixing dilute hydrogen peroxide and acetone solutions

[(Figure_3)TD$FIG]

Figure 3. Chromatograms of Authentic TATP Standard Solutions. From Top toBottom: 50, 25, 10, 5, 2.5, 1 mg/L TATP Standard Solutions.

aqueous hydrogen peroxide are sum-marized in Table 1. Solubility wasdetermined by visual inspection.

Mixing under Acidic Conditions

Solutions for the acidic media condi-tions were prepared as follows; thewater necessary for dilution wasplaced in an Erlenmeyer flask, andthe flask was immersed in an ice bath.Hydrogen peroxide (0.10 mol) andthen acetone (0.10 mol) were addedto the chilled water in proportionssuch that the final concentration ofthe mixture mimicked the combina-tion of 30%, 12%, or 3% (w/w) hydro-gen peroxide with 50% or 7% (w/w)

Table 1. Solubility of TATP in Various Solu

TATP (mg) H2O Amount (kg) Aceton

24.7 1.800025.6 1.600024.5 1.0000 1025.3 0.9999 1051 2.0001 1025.4 1.0001 1524.2 1.0000 1525.2 0.2504 7924.8 1.0002 2050.2 0.9999 1375.8 1.9999 1025.6 1.0316 9725 0.473725.2 0.400024.8 0.5000

30

acetone. Each of the six possible com-binations was prepared in triplicateand either 0, 1 or 5 drops of concen-trated sulfuric acid was added to eachof the samples. The average mass ofone drop of concentrated sulfuric acidwas found to be 50 mg.

The flasks were sealed with Parafilmand left at room temperature for 2 to 7days. If a precipitate formed after 2days, it was removed by filtration,washed with water, dissolved in acet-onitrile, and analyzed by GC/MS. If noprecipitate was observed after 7 days, a10 mL aliquot of the aqueous solutionwas extracted twice with 10 mL ofdichloromethane, the organic extracts

tions.

e (ppm) H2O2 (g) H2O2 (wt%) Dis

000 72 h0 72 h0 72 h9 72 h0 72 h0 96 h5 96 h9 96 h

0.585 0.12% 96 h24 12% 24 h30 12% 96 h

Journal of Chem

were combined, rinsed twice with10 mL of deionized water, dried overmagnesium sulfate and that solutionwas analyzed by GC/MS (Tables 2and 3).

Mixing with Welded Steel Coupons

Three types of steel coupons were eval-uated to determine the catalytic effectsassociated with formation of TATP.The three types were a coupon withouta weld, a coupon with a ‘‘dirty’’ weld,and a coupon with a ‘‘clean’’ weld. Asingle coupon of each type was placedin the six acetone/hydrogen peroxidemixtures. The reaction vessels weresealed and left at room temperaturesfor approximately 100 hours. An ali-quot was removed and analyzed byGC/MS as described above. Only the50%/30% and 50%/12% mixtures ofhydrogen peroxide/acetone indicatedthe presence of TATP by GC/MS.Table 4 shows the configurations thatwere examined and the concentrationof TATP observed in these solutions.

Alkaline Conditions

The most concentrated mixture inves-tigated in the presence of acid was re-investigated in the presence of base(50% acetone/30% hydrogen perox-ide). In a 125 mL Erlenmeyer flask,water, hydrogen peroxide and acetonewere combined so that acetone was50 wt% and hydrogen peroxide was30 wt%. Drops of sodium hydroxidesolution, 0, 1, 2, or 10, were added.

solved Not dissolved TATP (ppm)

100 hours 13.7100 hours 16100 hours 24.5100 hours 25.3

ours 25.5ours 25.4ours 24.2ours 101ours 24.8ours 50.2ours 37.9ours 24.8ours 52.8ours 63ours 49.6

ical Health & Safety, March/April 2012

Page 5: The risk of mixing dilute hydrogen peroxide and acetone solutions

Table 2. Mixes Prepared to Test TATP Formation Under Acidic Conditions.

Acetone (%) H2O2 (%) H2SO4 (d.)

TATP ppt in 48 hours 50 30 550 12 550 30 150 12 1

TATP in solution after 7 days 50 3 550 3 150 30 050 12 07 30 57 12 57 3 57 30 17 12 17 3 1

No TATP after 7 days 50 3 07 30 07 12 07 3 0

Each solution was prepared in dupli-cate. The flasks were sealed and leftundisturbed at room temperature for 8days. The mixtures were twiceextracted as described above and theextracts were analyzed for TATP byGC/mECD. The concentration of theNaOH solution was determined byrecording the mass necessary to fill a2 mL volumetric flask and repeatedthree times. The density was found tobe 1.522 g/mL. By comparing thisvalue to literature values, the solutionwas found to be 48.58% NaOH bymass. The mass of a drop of the NaOHwas determined to be 38 mg by record-

Table 3. TATP Found in Solutions Without

Acetone Peroxide Acid(drop

7% 30% 57% 30% 17% 12% 5

50% 3% 550% 12% 07% 12% 17% 3% 57% 3% 1

50% 3% 150% 3% 07% 30% 07% 12% 07% 3% 0

Journal of Chemical Health & Safety, March

ing the mass of 10 individual drops andtaking the average. After 8 days, GC–mECD analysis of the organic extractsof the alkaline solutions showed nodetectable TATP had been generated,while two control solutions (acetone,hydrogen peroxide, no NaOH) werefound to contain TATP in excess of50 mg/L.

Examination of Dilute Acids in Acetone/Peroxide Mixture

To determine if more dilute solutionsof acetone in hydrogen peroxide couldform TATP, solutions of acetone (7%and 0.04%) were mixed with hydrogen

Precipitate after 7 Days.

s)Mass %Acetone

Mass %Peroxide

6.2% 3.6%6.2% 3.6%5.2% 3.1%4.7% 2.7%

14.6% 8.5%5.2% 3.1%3.0% 1.7%3.0% 1.7%4.6% 2.7%4.7% 2.7%6.1% 3.6%5.2% 3.0%3.0% 1.7%

/April 2012

peroxide solutions (12% and 3%). The7% acetone solutions were combinedwith the hydrogen peroxide solutionssuch that the mixture contained0.1 mol of each reactant, theoreticallyyielding 0.033 mol TATP. The 0.04%acetone solutions were combined withhydrogen peroxide such that each mix-ture contained 0.01 mol of each reac-tant, theoretically yielding 0.0033 molTATP. After acetone and hydrogenperoxide were combined, one drop ofsulfuric acid (18 M, 13 M, 5 M, or0.1 M) was added, and the mixtureswere sealed and left at room tempera-ture for �75 hours. The mixtures fromthe 12% and 7% acetone wereextracted with 20 mL dichloro-methane twice. The organic extractswere combined and rinsed with10 mL of water to remove any residualacid or acetone and dried over magne-sium sulfate. The mixtures from the0.04% acetone solutions had a totalvolume of approximately 1,560 mLand were extracted twice with 50 mLdichloromethane and rinsed once with20 mL water and dried over magne-sium sulfate (Table 5).

RESULTS AND DISCUSSION

Solubility of TATP in Water

TATP is practically insoluble in water,but is soluble in acetone. Solubilityvalues of dilute aqueous solutions ofacetone have not been published. Inthis study, 25 mg of TATP was added tovarious aqueous solutions of acetone

Mass %Reactants

Mass TATP insolution (mg)

9.8% 10,0829.8% 6,2248.3% 5,9687.4% 5,914

23.1% 5,3228.3% 4,3624.7% 4,1364.7% 3,7207.3% 3,3437.4% Not found9.8% Not found8.3% Not found4.7% Not found

31

Page 6: The risk of mixing dilute hydrogen peroxide and acetone solutions

Table 5. TATP Found in Solutions Prepared with Dilute Acid.

InitialAcetone

InitialH2O2

AcidConc.

Mass TATP(mg)

pH Mass %Acetone

Mass %H2O2

Mass %Reactants

7% 12% 18 M 4,993 2.41 5.27% 3.06% 8.33%7% 12% 18 M 2669 2.64 5.21% 3.10% 8.32%7% 12% 13 M 2,803 2.65 5.22% 3.07% 8.29%7% 12% 5 M 1,632 2.86 5.21% 3.09% 8.30%7% 3% 18 M 279 2.86 2.96% 1.74% 4.70%7% 3% 18 M 170 2.63 2.97% 1.73% 4.70%7% 3% 13 M 154 2.76 3.13% 1.74% 4.87%7% 3% 5 M 103 2.98 2.96% 1.73% 4.68%7% 3% 0.1 M 0 3.11 2.99% 1.73% 4.72%0.04% 3% 18 M 0 – 0.04% 0.02% 0.06%0.04% 3% 13 M 0 – 0.04% 0.02% 0.06%0.04% 3% 5 M 0 – 0.04% 0.02% 0.06%0.04% 3% 0.1 M 0 – 0.04% 0.02% 0.06%7% 12% – – 3.84 – – –– 50% – – 1.24 – – –– 12% – – 3.44 – – –

Table 4. TATP Found in Solutions with Steel Coupons.

Acetone,Initial conc.

HP, Initialconc.

Weldtype

Mass %Acetone

Mass %Peroxide

Mass %Reactants

Mass TATP(mg)

50% 30% C 25.3% 14.9% 40.2% 22,63650% 30% D 25.2% 15.0% 40.2% 26,53650% 30% N 25.3% 14.8% 40.1% 23,77050% 12% C 14.6% 8.5% 23.1% 4,94750% 12% D 14.6% 8.6% 23.1% 4,15350% 12% N 14.4% 8.4% 22.8% 4,48150% 3% C 4.7% 2.7% 7.4% 050% 3% D 4.6% 2.7% 7.3% 050% 3% N 4.7% 2.7% 7.4% 07% 30% C 6.1% 3.6% 9.8% 07% 30% D 6.2% 3.6% 9.8% 07% 30% N 6.2% 3.6% 9.8% 07% 12% C 5.2% 3.0% 8.3% 07% 12% D 5.2% 3.1% 8.3% 07% 12% N 5.2% 3.1% 8.3% 07% 3% C 3.0% 1.7% 4.7% 07% 3% D 3.0% 1.7% 4.7% 07% 3% N 3.0% 1.7% 4.7% 0

Weld type: C = cleaned weld, D = dirty weld, N = no weld.

and visually inspected to determine ifthe solid dissolved. Table 1 is a compi-lation of the results. The TATP solubi-lity in neat water was determined to bebelow �15 mg/L while solubility in a0.1% acetone solution exceeded25 mg/L for TATP. TATP solubilityin the two H2O2 solutions screenedwas found to be greater than 50 mg/L. The slight solubility of TATP inwater is slightly increased by the inclu-sion of acetone and H2O2.

32

Influence of Concentration and AcidCatalyst on TATP Formation

This experiment was conducted todetermine if TATP would be formedat concentrations that might beencountered in an industrial process.Table 2 lists the results of this experi-ment. Note that the concentrations ofH2O2 were initially 30%, 12%, and 3%,while the concentrations of acetonewere initially 50% and 7%. The higherconcentrations were included to pro-

Journal of Chem

vide levels known to make TATP withthe purpose of verifying the experi-mental methods of observing or mea-suring TATP. Solid TATP was seen inthe highest concentrations with 50%acetone and 30% or 12% H2O2, withthe sulfuric acid catalyst, and smallamounts of precipitate were observedin the 50% acetone, 30% H2O2 mix-ture. After 7 days, the remaining solu-tions were extracted and analyzed byGC/mECD. TATP was detected in all

ical Health & Safety, March/April 2012

Page 7: The risk of mixing dilute hydrogen peroxide and acetone solutions

solutions in which sulfuric acid cata-lyst had been added. Note that for thesolutions with no acid, TATP was seenonly in the 50% acetone and 30%H2O2 solution; thus, the presence ofan acid catalyst appears to be a keyingredient for making TATP in dilutesolutions.

Influence of 304L Stainless Steel andWelds on TATP Formation

To evaluate if stainless steel surfaces orstainless steel welds might catalyze theformation of TATP, coupons wereobtained to serve as examples of the304L stainless steel metal commonlyused to build an industrial scrubber.Coupons were also obtained whichrepresented clean and dirty welds on304L stainless steel to evaluate theinfluence of weld seams. The solutionsfrom the above experiments, with noacid catalyst present (Table 2), were re-evaluated using the metal coupons aspotential catalyst materials. The resultsare shown in Table 4. TATP wasdetected only in the 50% acetoneand 30% H2O2 solution at nearly iden-tical concentrations to those withoutthe coupons. This experiment suggeststhat 304L stainless steel does not cat-alytically influence the formation ofTATP.

Influence of Acidity on TATP Formation

The formation of TATP is an acid cat-alyzed reaction, but the amount of acidrequired to generate the peroxide hasnot been published. Solution acidity isan important consideration for estab-lishing a safe boundary for the use ofH2O2 in the presence of acetone. H2O2

solutions are often sold and storedunder acidic conditions for increasedstability. Table 5 contains the results ofan evaluation of the influence of acid-ity. No TATP was detected at any acidlevel for the 0.04% acetone solutions.There was also no TATP detected forthe 7% acetone and 3% H2O2 with0.1 M acid solution added. Theseresults highlight the need for sufficientacid concentration in forming TATP.

Journal of Chemical Health & Safety, March

Alkaline Conditions

The literature concerning the forma-tion of TATP focuses on the use of acidcatalysts and there is no mention onthe effects of alkalinity. To evaluatethese conditions, test solutions wereprepared with three different concen-trations of NaOH. No TATP wasdetected in these experiments, indicat-ing that the formation of TATP is notbase catalyzed and, furthermore, thatthe presence of trace amounts ofNaOH inhibits the formation of TATP.

CONCLUSIONS

Oxidation of acetone with hydrogenperoxide produces TATP. While thisreaction is relatively gentle, the pro-duct, TATP, is highly sensitive to shockand friction that results in explosivedecomposition. Situations whereTATP accumulates are highly hazar-dous. This study was initiated to assessthe possibility that TATP accumulationcould result from cleaning processesfor industrial scrubbers which usedilute hydrogen peroxide and acetone.Hydrogen peroxide solutions aremildly acidic and even solutions con-taining only a few percent of acetoneand hydrogen peroxide can formTATP. Without additional catalyticactivity, the formation of small quan-tities of TATP in dilute H2O2 and acet-one in aqueous solutions are likely toremain soluble in the reaction media.The presence of 304L stainless steeldid not catalyze the formation ofTATP. The presence of trace amountsof sodium hydroxide in solutions thatnormally form TATP inhibits the for-mation of TATP in all the configura-tions screened here. The cyclic dimer,DADP, was not found in any of theconfigurations, leading one to con-clude that the formation of DADPnecessitates a relatively high concen-tration of acid catalyst. Acid catalyzedmixtures of acetone and hydrogen per-oxide are capable of producing hazar-dous amounts organic peroxides (i.e.,

/April 2012

hundreds of ppm) which could accu-mulate. This study shows that the sim-ple mixing of dilute solutions ofhydrogen peroxide and acetone doesnot appear to accumulate TATP inhazardous amounts. The solubility ofTATP in pure water is about 15 ppmand somewhat higher when acetoneand peroxide are present. The smallamount of TATP produced withoutacid catalysis is likely to remain solublein the aqueous cleaning solution.

REFERENCES1. Wolffenstein, R. Ber. Dtsch. Chem.

Ges. 1885, 28(1459), 2265.2. Shanley, E. S.; Greenspan, R. P. Ind.

Eng. Chem. 1947, 39, 1536.3. Milas, N. A.; Golubovic, A. J. Am.

Chem. Soc. 1959, 81, 6461.4. Sauer, M. C. V.; Edwards, J. S. J. Phys.

Chem. 1971, 75(19), 3004.5. Sauer, M. C. V.; Edwards, J. S. J. Phys.

Chem. 1972, 76(9), 1283.6. Evans, H. K. ; Tulleners, F. A. J.;

Sanchez, B. L.; Rasmussen, C. A. J.Forensic Sci. 1981, 31(3), 1119.

7. Mackenzie, J. Chem. Eng. 1990, 97(6),84.

8. Mackenzie, J. Plant Oper. Prog. 1991,10(3), 164.

9. Constantini M. Destruction of AcetonePeroxides. US Patent 5,003,109(Rhone-Poulenc Chimie), 1989.

10. Bellamy, A. J. J. Forensic Sci. 1999,44(3), 603.

11. Jiang, H.; Chu, G.; Gong, H.; Qiao, Q.J. Chem. Res. 1999, 28, 288.

12. Oxley, J. C.; Smith, J. L.; Chen, H. Prop.Exp. Pyro. 2002, 27(4), 209.

13. Widmer, L.; Watson, S.; Schlatter, K.;Crowson, A. Analyst, 2002, 127, 1627.

14. Schulte-Ladbeck, R.; Kolla, P.; Karst, U.Trace analysis of peroxide-based explo-sives. Anal. Chem. 2003, 75, 731–735.

15. Denekamp, C.; et al. Org. Lett. 2005,7(12), 2461.

16. Oxley, J. C.; Smith, J. L.; Moran, J.;Nelson, K.; Utley, W. E. Proc. SPIE,vol. 5403. 2004, .

17. Oxley, J. C.; Smith, J. L.; Luo, W.; Brady,J. Prop. Exp. Pyro. 2009, 34(6), 539.

18. Dubnickova, F.; et al. J. Am. Chem.Soc. 2005, 127, 1146.

19. Eyler, G. N. J. Phys. Org. Chem. 2006,19(11), 776.

33