UNCLASSIFIED

F19643

DEFENSE DOCUMENTATION CENTERFOR

SCIENTIFIC AND TECHNICAL INFORMATION

CAMERON STATION. ALEXANDRIA. VIRGINIA

Fp eproduced FromBest Available Copy

UNCLASSIFIED

NOTICE: When government or other drawings, speci-fications or other data are used for any purposeother than in connection with a definitely relatedgovernment procurement operation, the U. S.Government thereby incurs no responsibility, nor anyobligation whatsoever; and the fact that the Govern-ment may have forzialated, furnished, or in any waysupplied the said drawings, specifications, or otherdata is not to be regarded by implication or other-wise as in any manner licensing the holder or anyother person or corporation, or conveying any rightsor permicsion to manufacture, use or sell anypatented invention that may in any way be relatedthereto.

REPORT NO. RK-TR-63-20

7 30

CT') KTBTTC3S C- EPOXIDE-CARROXYLTC

I ACID REACTION

Z C/)

26 August 1963

U ARMY MISSILE COMMAND>REDSTONE ARSENAL, ALABAMA

1 C , . l?-', I J,*'j r71 P;ýjVj(l•Jz VDITIW; I'ý (111'OtrT[

DDC AVAIIABILITY NOTICE

Qualified requesters ray obtain copies of this reportfrom the Defense Documentation Center for Scientific andTechnical Information, Cameron Station, Alexandria, Virginia,22314.

DESTRUCTION NOTICE

Destroy; do not return.

26 August 1963 Report No. RK-TR-63-20

KINETICS OF EPOXIDE-CARBOXYLIC

ACID REACTION

by

Robert E. Odom

Department of Army Project No. 517-06-002

AMC Management Structur- Code No. 5010.11.58500

- Propellant Chemistry BranchPropulsion Laboratory

Directorate of Research and DevelopmentU. S. Army Missile Command

Redstone Arsenal, Alabama

ABSTRACT

The kinetics of the addition reactions of four structural-ly different epoxide compounds with glacial acetic acid in di-oxane solution have been studied at 90"C, 110 C, and 125*C.The relative order of reactivity of the epoxides is styrene ox-ide o vinyl cyclohexane monoxide 3 allyl 9, 10-epoxy stearate >1,2-epoxy-3-phenoxy propane. The reactions apparently followthe first-order-rate law with respect to epoxide and acid con-centrations. For the four epoxides whose rates were determined,loglok was found to increase linearly with 1/T according to theArrhenius equation.

ii

TABLE OF CONTENTS

Page

I-.N-TTRODUCTION . 1

A. Chemicals. . . . . . . . . . . . . . . . ... . . 2

B. Experimental . . . . . . .0. . . . . . . . . . . 2

II. RESULTS AND DTSCUSSTS1. . . . . . . . . . . . . . . 3

A. Heat end Entropy of Activation . . . . . . . . . 6

B. Base Catalyzed Epoxide-CarboxylicAcid Reactions ............... 7

III. SUMMARY .. . .. ... . . . . ... 7

LIST O ILLUSTRATIONS

Table

I Rate Constants and Arrhenius Activation EnergiesFor Epoxide-Glacial Acetic Acid Reactions

In Dioxane . . . . . . .......... 8

II Rate Constants at 90"C, •iergies, Heats, andEntropies of Activation For Epoxide-Glacial

Acetic Acid Reactions In Dioxane . . . . . 8

III Rate Constants at 1100C, Energies, Heats, andEntropies of Activation For Epoxide-Glacial

Acetic Acid Reactions In Dioxane . . . . . 9

IV Rate Constants at 125*C, Energies, Heats, andEntropies of Activation For Epoxide-Glacial

Acetic Acid Reactions In Dioxane . . . . . 9

V Carboxylic Acids - 1,2-Epoxy-3-Phenoxy PropaneReaction Rates at 125'C . . . ........ 10

VI Effect of Added Cetyl Dimethyl Benzyl AmmoniumChloride On The Reaction Half-Life of

l,2-Epoxy-3-Phenoxy Propane and GlacialAcetic Acid At 90"C ......... 10

iii

LIST OF ILLISTRATIONS (Concluded)

Figure Page

1 Comparison of Epoxide-Carboxylic Acid Reactionsat go"C .. .. .. .. .. ... . ... .i

2 Comparison of Epoxide-Carboxylic Acid Reactionsat 1IO*C . . . . . . . . . . . .*. . . . . . . . 12

3 First Order Plot of 1,2-Epoxy-3-Phenoxy Propane-Glacial Acetic Acid in Dioxane . . . . . . . . . 13

4 Arrhenius Plot of Epoxide-Glacial Acetic AcidReactions In Dioxane .1............ 14

5 l,2-Epoxy-3-Phcnoxy Propane-Carboxylic AcidReactions at 125*C . . ............ 15

6 Reaction of Equimolar Quantities of 1,2-Epoxy-3-Phenoxy Propane and Glacial Acetic Acid

With Added Cetyl Dimethyl BenzylAmmonium Chloride at 90C. . . . . . . 16

iv

KINETICS OF EPOXTDE-CARBOXYTIC

ACID REACTION

I. INTRODUCTION

The rapid growth of epoxide compounds into areas otherthan the protective coating field has directed considerableattention to a field of chemistry in which there has been areal lack of published data. This is particularly true inthe field of solid propellant binder systems where a widevariety of epoxide compounds are available for use as curingagents. This interest in epoxide compounds in the area ofpropellant binder systems has emphasized a fundamental need foradditional information on the reactivity of epoxides as well asa simplified method for obtaining kinetic data on reactions in-volving these materials. This report is concerned with a studyof the fundamental nature of these reactions.

Relatively little work in this field was reported priorto 1950; and in most cases, little or no speculation was madeconcerning the kinetics of these reactions. The most signifi-cant work prior to 1950 was reported by Br8nsted, Kilpatrick,and Kilpatrick (Reference 1). Br~nsted, et al. investigatedthe hydrolysis of ethylene oxidi in dilute aqueous solution ofacids. On the basis of their studies, the authors concludedthat the reactions lead to the 1,2-glycols and follow the first-order-rate law, in which the specific rate constant, kl, isdirectly proportional to the hydrogen ion concentration.

The most recent work in this field has been reported byShechter, Wynstra, and Kurkjy (References 2,3, and 4) in whichthe addition reactions of glycidyl ethers with acids, phenols,and alcohols were studied. The authors reported that undercondition of base catalyzed reactions the order of reactivityis alcohoi > phenol > acid.

The object of the present work was to study quantitativelythe addition reactions of four structurally different epoxides--vinyl cychohexane monoxide (I), styrene oxide (1I), 1,2-epoxy,3-phenoxy prorane (I11), and Allyl 9, l0-epo:%y stearate (IV)-

1I

with glacial acetic acid to obtain a quantitative comparison ofthe effects of structure and temperature on the reactivities ofthe epoxide compounds.

Ila 0 0H-CH- Q.C -R..CI2 CH2 -CH-CH2

II III

IV

A. Chemicals

Vinyl cyclohexane.monoxide and styrene oxide (98 and990) purity by epoxide analysis, supplied by Union CarbideChemical Company, were used as received. Allyl 9,10-epoxystearate was a carefully refractionated Union Carbide commercialproduct.-Glacial acetic acid (99.59 purity) and 1,2-epoxy-3-phenoxy propane (99ý), supplied by Fisher Scientific Company,were used as received. The dioxane, used as the solvent in thereaction, was purified by the conve"tional method of refluxingover sodium followed by distillation.

B. Experimental

The general experimental method consisted of holding agiven set of reactants at a predetermined temperature, samplingfrequently, and determining the amount of epoxide remaining atregular time intervals. More specifically, the analysis wascarried out as follows: Equimolar quantities of glacial aceticacid and epoxide were weighed separately and diluted to volumewith dioxane. The two solutions were then combined and mixedthoroughly. Carefully measured 10-milliliter aliquots of thereaction-Tifitlre were placed in 20 x 150 millimeter test tubeswhich had previously been drawn to a narrow neck for conveniencein sealing. After addition of the mixture, the tubes were evac-uated, cooled in liquid nitrogen, and sealed with a flame. Thesealed samples were allowed to warm to room temperature and then

2

placed in constant temperature baths, carefully regulated at 90 +0.10, 110:0.1i, and 125± O.l 0 C. The reaction was followed bywithdrawing samples at regular time intervals, depending on thereaction rate, and titrating the epoxides remaining in the seal-ed tube with atanlard hydrogen bromide solution. Crystal violetwas used as an indicator.

II. RESULTS AND DISCUSSION

Theoretically, the following epoxide-carboxylic acid reac-tions are possible.

A. Reaction of the carboxyl group with the epoxide to forma monoester and a hydroxyl group:

0 0% if I IRI C H + --C- -- R, C-O-C-C-Of W i

I I I I

B. Reaction of the carboxyl group with the newly formedhydroxyl group to form a diester with the elimination of water:

0 0 0

RCOH*HO-C-C -- C-R,-p"- RC-O-C-C-O-C-R 1 +Rao. (2)

C. Reaction of the epoxide with the hydroxyl group with-out the elimination of water:

0, O , , 0-c - C-+HO-C--O-8-R, -0l HO-C-C-O-C-C-O-C-R . (3)

This sequence of reactic-' was investigated by using vinylcyclohexane monoxide as a model epoxide. Equimolar quantitiesof the epoxide and glacial acetic acid (1 molar in dioxane)were thoroughly mixed and placed in a constant temperature bath,regulated to 1100l lC for 7 days. Aliquots of the reactionmixture were withdrawn at regular time intervals and the absorp-tion spectra recorded.

The fcllowing changes in the absorption were noted: Asignificant increase in absorption at 3450 cm"1 , which can beassigned to the hydroxyl group. Accompanying this increase wasa significant decrease in absorption at 1235 cm"1 , which can beassigned to the epoxide group. A simultaneous decrease in theabsorption at 1724 cm-1 and an increase in the absorption at1198 cm"1 which caa be associated with the disappearance of thecarboxylic acid groups and the formation of the monoester groups,respectively.

3

To support the above observations, quantitative data on thereaction products of vinyl cyclohexane monoxide and acetic acidwere obtained. The reaction mixture was carefully vacuum-dis-tilled and the following fractions were obtained:

(1) b.p. 25-30o/17 mm(2) b.p. 31-35o/17 mm(3) b.p. 72-75o/17 mm

Fraction (1) consisted essentially of acetic acid anddioxane, while (2) was identified by epoxide analysis to belargely unreacted epoxide. The greater part of fraction (3)boiled at 740/17 mm and gave a negative oxerane test. Fraction(3) was further concentrated at 73-74*/17 mm and identified as2-acetoxy-2-cyclohexyl ethane-l-ol by alkaline hydrolysis saponi-fication equivalent (186), hydroxyl group determination by acet-ylation in pyridene followed by titrating the excess anhydrideafter hydrolysis (9.15*), and molecular weight determination bythe freezing point depression of cyclohexanol (184).

The results of holding equimolar amounts of 1,2-epoxy-3-phenoxy propane, allyl 9,10-epoxy stearate, vinyl cyclohexanemonoxide, styrene oxide, and glacial acetic acid at 900 and 1100C are shown in Figures 1 and 2, respectively. A comparison ofthe slopes of these rate curves indicate that the order of reac-tivity is styrene oxide > vinyl cyclohexane monoxide > allyl-9,10 epoxy otearate > l,2-epoxy-3-phenoxy propane. This decreasingorder of reactivity shows that the rates of epoxide-carboylicacid reactions are influenced significantly by the nature of theepoxide substituents.

The order of reactivity may be explained by assuming thatthe reaction proceeds through a carl"onium ion intermediate com-plex. Addition of a proton to the o'irane oxygen to form anintermediate-complex results subsequently in the rupture of thecarbon-oxygen bond with the formation of a carbonium ion:

-CH - CHa+H -0- -CH C-12 -0- -CHOH-CHa . Slow (4)

Attack of the carbonium ion by an acid:

-CHOH-CHg+HAc -0- -CHOH-CHaAk+H(R. Fast (5)

If the ring opening reaction (4) is the rate determiningstep, the relative order of reactivity of the epoxide compoundsshown in Figures 1 and 2 is dependent upon the relative ease of

'4

formation of the carbonium ion. In the case of styrene oxide,the addition of a proton to form the intermediate complex isfavored by the close proximity of the aromatic and oxiraneoxygen rings, resulting in resonance stabilization of thecarbonium ion intermediate. While in the case of 1,2-epoxy-3-phenoxy propane, where no appreciable reaction occured at 900C, apparently the ether linkage to which the phenyl group isattached acts as an electron withdrawing force resulting indecreased reactivity toward the proton.

Kinetic treatment of the data given in Figures 1 and 2indicates that the reactions follow the first-order-rate lawwith respect to the epoxide and acid to approximately one-halflives. A representative Flot is shown in Figure 3. If therate determining step is that shown in Equation 4, then thekinetics would follow the expression:

H ]+1-c- ](6)"="

An examination of the results shown in Table I, indicatesthat the kinetics of the reaction apparently follow this ex-press ion.

For the four epoxides whose rates were determined, log10kwas found to increase linearly with lI/T according to the Ar-rhenius equation:

k = PZ 1,"Ea/RT (7)

where:

k = specific rate constant in sec-PZ = the probability collision factorEa = the activation energy

R = the molar gas constantT = temperature in degrees absolute

The excellent straight line agreement of the experimentalpoint given in Figure 3 appears to conclusively establish thefirst-order kinetics.

To test the mechanism postulated by Equations 4 and 5,the effect of carboxylic acids of higher strength (Ki) on therate of reaction was investigated at 1250 C. Equimolar quan-tities of l,2-epoxy-3-phenoxy propane and benzoic acid in onecase and monochloroacetic acid in another were investigated

5

I

in dioxane solution. An examination of the results summarizedin Table V and Figure 5 indicate that as the available protonsare increased, a corresponding increase in the rate of reactionis observed. These data lend substantial support to the car-bonium ion intermediate complex concept.

The activation energies (Ea) given in Table I were obtain-ed by plotting the logarithms of the specific rate constantsdetermined at the experimental temperatures against the re-ciprocal of the absolute temperature according to Equation 7.The slope of the line (Figure 4) is equal to -Ea/R and theintercept is the logarithm of PZ, the probability collisionfactor.

The activation energies, (Ea) shown in Table [, vary in themanner predicted by the carbonium ion intermediate complex. Theactivation energies vary in the manner determined by the relativeease of formation of the carbonium ion intermediate. In the caseof styrene oxide where resonance stabilization of the carboniumion is possible, the rate of reaction is approximately 10-foldfaster than 1,2-epoxy-3-phenoxy propane where resonance stabili-zation of the intermediate complex is unlikely.

A. Heat and Entropy of Activation

The Arrhenius parameters, heat and entropy of activa-tion, are related to the specific rate constant by the expression:

k = K'T t= .L (8)

where K' is equal to k/h--the Boltzmr-', and Planck constantsrespectively, T, the temperature in degrees absolute; AS*, theentropy of activation, and AV, the enthalpy of activation. Theenthalpy or heat of activation is a quantity closely related tothe experimental energy of activation by the expression:

... AHV = AEa-nRT (9)

The small difference in tne entropy and heat of activationfor each compound as shown in Tables II, III,and IV suggests thatthe reaction product diabribution will not vary significantlywith temperature. The large negative entropy values are indic-tive of the large decrease of entropy accompanying the formationof the activated complex.

6

B. Base Catalyzed Epoxide-Carboxylic Acid Reactions

The data presented in Table I indicate that the epoxide-carboxylic acid reactions are rather sluggish and an appreciableinterval of time is required to reach the reaction half-life.Tertiary amines and amine salts represent an important class ofcatalyst for the epoxide-carboxylic acid reactions.

To investigate the effect of a quantary ammonitu salt onthe reaction rate, cetyl dimethyl b-nzyl ammonium chloride wasadded to the reaction mixtures of equimolar amounts of 1,2-epoxy-3-phenoxy propane and glacial acetic acid in dioxane solution.Kinetic--studies were made at several different concentrations ofcetyl dimethyl benzyl ammonium chloride. The results of thisstudy are summarized in Table VI and Figure 6. These data indicatethat the reaction proceeds more rapidly when the quaternary ammoni-um salt is added to the reaction mixture. A measure of the cat-alyzing effect of the quaternary ammonium chloride on epoxide-carboxylic acid reactions is given by Table VT. Non-catalyzed,the reaction half-life is approximately 53 days, while in thepresence of 0.10 mole/liter of cetyl dimethyl benzyl ammoniumchloride the reaction half-life is reduced to less than 3 hours.These data indicate that the base catalyzed reaction is apparentlyfirst order with respect to the epoxide concentration.

III. SUMMARY

The kinetics of the addition reaction of four structurallydift~efft-epoxides compounds with carboxylic acids in dioxanesolution have been investigated at 900, 1100, and 125'C. Therelative order of reactivity of the epoxides with glacial aceticacid is styrene oxide > vinyl cyclohexane monoxide > allyl 9.10-epoxy stearate > l,2-epoxy-3-:ienoxy propane. A reaction mechanismfor these reactions was postulated. The experimental data in-dicated that the rate of reaction is determined to a large extentby relative ease of formation of the carbonium ion intermediatecomplex.

The Arrhenius parameters: energy, heat, and entropy ofactivation were determined and correlated with the structure ofthe epoxid. compounds.

Kinetic treatment of the experimental data indicated theepoxide-carboxylic acid reaction apparently follow the first-order-rate law.

The second phase of this investigation will make use ofthese data in the study of the cure reactions of the PBAA-ERLbinder system.

7

Table I

RATE CONSTANTS AND ARRHENIUS ACTIVATIONS ENERGIESFOR EPOXDES - ACETIC ACID REACTIONS IN DIOXANE

IO7 k sece"Cccup d 9c c 1 25 C Ea kcal

i,2-epoxy-3-pbenoxy propane 1.50 2.73 5.49 11.04

Allyl 9,10-epoxy stearate 4.18 7.17 15.80 10.99

Vinyl cyclobexane monoxide 6.87 14.95 23.20 10.05

Styrene oxide 13.64 27.50 48.20 10.29

Table TI

RATE CONSTANTS AT 90eC, ENERGIES, V12ATS AND ENTROPIES OF ACTIVATIONSFOR EPOXIDE-ACETIC ACID ktACTIONS IN DIOXANE

Compound 90PC 107k sec-4 AH* kcal AS* E. U. Ea kcal

1,2-epoxy-3-pheno;.y propane 1.50 10.32 -61.74 11.04

Allyl 9,10-epoxy stearate 4.18 10.26 -59.86 10.99

Vinyl cyolobexane monoxide 6.87 9.33 -61.44 10.05

Styrene oxide 13.64 9.57 -59.42 10.29

8

Table III

RATE CONSTANTS AT 110 OC, ENERGIES, HEATS AND ENTROPIES OF ACTIVATIONSFOR EPOXIDE - ACETIC ACID REACTIONS IN DIOXANE

Compoumd 11CPC 107k see-& AH$ kcal AS* E.u. Ea kcal

1, 2-epoxy-3-pbenoxy propene 2.73 10.28 -62.25 11.04

AlIyl 9,10-epo stearate 7.17 10.22 -60.47 10.99

Vinyl cyclobexane monoxide 14.95 9.29 -61.44 10.05

Styrene oxide 27.50 9.53 -59.61 10.29L

Table IV

RATE CONSTANTS AT 125 C, ENERGIES, .HEATS AND ENTROPIES OF ACTIVATIONSrFOREPOXIDE - ACETIC ACID REACTIONS IN DIOXANE

Compound 1250 C 10'k sec-4 AH* keal AS*E.U. Ea kcal

1, 2-epoxy-3-phenoxy propane 5.49 9.64 -61.97 11.04

AIlyl 9,10-epoxy stearate 15.80 10.20 .60.01 10.99

Vinyl eyelohexane monoxide 23.20 9.26 -61.59 10.05

Styrene oxide 48.20 9.50 -59.53 10.29

9

Table V

CARBOXYLIC ACIDS - 1,2-EPOxY-3-PHENCKY PROPANE

REACTION RATES AT 1250 C

Acid 103 k Hrs-I T. J Hr.

Acetic (Glacial) 1.98 350.6

Benzoic 5.10 135.9

Monochloroacetic 34.10 20.32

Table VI

EFFECT OF ADDED CETYL DIMETHYL BENZYL AMMONIUM CHLORIDE

ON THE REACTION HALF-LI.E OF 1,2-EPOXY-3-PHENOXYPROPANE AND GLACIAL ACETIC ACID AT 90*C

Catalyst 104k Hre-" T. J, Hr.

Mole Liter

0.0 5.4o 1283.3

O.3139 829.1 d.36

0.0507 1512.0 4.583

0.1012 2379.6 2.9123

10

-- 2

C

C

0

-

H

0- -- --- eq

-4UH

C

U H

:5 �rz�

_ -a0

H

2C.,

- CWIa

- - �����0 'I 2.1

rz�

- - - - - -_ �

I �% �. L�

- � In

- - --- D��OI-� F � p---* V� N -: C

zs11�/Ss1oM '�niwj sPixodi

12.

0

w LA 0

A Z 0

o u 0

> V)

'puno0 opxd

-12'_ __

0 0

-P) ol9 -

13 .

--Up

oAS

to~

144

0

-Go

UU

-44

1 ha

0

- i U - --1 5

fTTE-4

* U 6E-4

~ ~. p.

- 000-

U 6 00-0-

OOP U ' oU ý 'AC

op,

&Uj/*o'uo U Uod

16-'U

REFERENCES

1. J. Br3nsted, M. Kilpatrickand M. Kilpatrick, Journal ofAmerican Chemistry Society 51, 4L42, (1929).

2. L. Shechter and J. "iynstra, Industrial Engineering Chemistry48, 86, (1956).

3. L. Shechter, J. Wynstra, and R. Kurkjy, Industrial Engineer-ing Chemistry 48, 94 (1956).

14. L. Shechter, J. Wynstra, and R. Kurkjy, Industrial Engineer-ing Chemistry L9, 1107, (1957).

17

26 August 1963 Report No. RK-TR-63-20

APPROVED:

p w . JAMES

' irector, Propulsion laboratory

DISTRIBUTION

Copy

Solid Propellant Mlailing List,dated 1 June 1962 1-136

AMSMI-R, Mr. McDaniel 137-RBL 138-142-RFS 143-RR 144-RK, Mr. James 145-RK, Dr. Odom 146-150-RE 151-RG 152-RL 153-RS 154-RT 155-RAP 156-RK= 157-160"-RFE 161,162

41l 0

X u U" 0=1 ~Z4 -u A I41 .r z a

U~r o E tt 01 - a

u . 0 01. 0

.2 2 .2 - - -

o u AJu t.a> ~~ ~ ~ ~ s

01 u 0, '. 2

U~' 0 : W

IV 0. -. a

'04. I7u :2u -. _ 0 A -. 00 0 , t A1 m 1

r- -5~ =0 U 4

0 41-0 0 0

r 01 '.41 0.

Eu u

S. :3U 2

~~IVm C6

U"00 ~ aS4~ 00

OnI