Thermal degradation of poly (methyl methacrylate) in ...

Scholars' Mine Scholars' Mine

Masters Theses Student Theses and Dissertations

1965

Thermal degradation of poly (methyl methacrylate) in solutions in Thermal degradation of poly (methyl methacrylate) in solutions in

a closed system a closed system

Vikram Pranjivandas Parikh

Follow this and additional works at: https://scholarsmine.mst.edu/masters_theses

Part of the Chemical Engineering Commons

Department: Department:

Recommended Citation Recommended Citation Parikh, Vikram Pranjivandas, "Thermal degradation of poly (methyl methacrylate) in solutions in a closed system" (1965). Masters Theses. 6764. https://scholarsmine.mst.edu/masters_theses/6764

This thesis is brought to you by Scholars' Mine, a service of the Missouri S&T Library and Learning Resources. This work is protected by U. S. Copyright Law. Unauthorized use including reproduction for redistribution requires the permission of the copyright holder. For more information, please contact [email protected].

THEJ.U;lAL DEGRADATION OF POLY (HE':eHYL 1-IETHACRYLATE)

IN SOLUr.riONS IN A CIJOSED SYSTEH

BY

VIK.Rl~H PRAHJIVALIDAS PARIKH ; ) _-/

A 115230 THESIS

submit·cecl to the faculty of the

Ui:UVERSI'.EY Ot' NISSOURI AT ROLLi\

in parJcial fulfillrt1ent of the rer.:i_uirements for the

Degree of

1'-IASTER OF SCIENCE IN CHEHICAL ENGil.'l'EERING

Rolla, .Hissouri

by

ABSTRACT

A study has been made of tllc ·chcnJal degradation of

poly (methyl meJchacryla-te) of ·t'l:lo different molecular

\1Ciglrts, 2,303,000 and 43,000, in 1,2-dichloroet-~la:ne

in a closed system. The results indica-ted that -::he poly-

mer having a molecular weight of 2,303,000 undergoes de­

gradaJcion while ·the low molecular vleigh·t polyraer 4 3, 000

undergoes reverse polymerization, and the molecular

vJeight remains cons-tant. The molecular weight of hig:1

E1olecular weight sarnples af·ter 48 hours decreased from

2,303,000 to 2,232,000 and 1,690,400 at. 50°C and a-t

98.5°C, respectively.

I.

II.

III.

TABLE OF CONTENTS

TITLE PAGE • • • • • • • • • • • • • • • • • • • 0 • • • • • • • • 0 • •

ABSTRACT

TABLE OF CONTENTS •••••o••eo•••••o••••o•og

LIST OF TABLES • • • • • • • • • • • • • • • • • • • • o e a • • o •

LIST OF FIGURES 0 0 • • • 0 • • • • • • • • • • • • • • • 0 • • • •

INTRODUCTION • • • • • • • • • • • • 0 • • • • • • • 0 • • • • • • • •

LITERATURE REVIEW e • o e • o • • o o • • o • o • • • o • • o o •

EXPERIMENTAL & • 0 • • 0 • • • 0 • 0 • • • • 0 • • • • • 0 0 • • 0 0 •

Ao

B.

c.

Purpose of Investiga·tion • • • • • • • • • • • • •

Plan of Investigation • • e • • • o • • • o • • • • o

Nethods of Procedure • o • e • • • • • • • • • • • • o

lo

2 ..

3·

4.

5·

Procedure for the Use of Cannon­Ubbelohde Dilution Viscometer ••

Calibration of viscometer 0 •••••••

Density measurements by using pycnometer e .................. o.

Viscosity, relative visco&J;;ty.;-· -in"!'~.·· trinsic viscosity and molecular \'Ieight calculations ••• o o ...... .,

Polymer samples • • • • • • 0 • • • • • • • • • • •

ii

Page

i

ii.

v

vii

1

3

21

21

21

23

23

23

23

23

28

IV.

v.

VI.

VII.

VIII.

IX.

iii

Page

6. Solvent • • • • • • e • • • • • • • • • • • • • • • • • • • 28

7. Polymer solution preparation and method of degrada-tion • • • • • .. • • • 28

DATA AND RESULTS • 0 • • • • • • • • • 0 • 0 • • • • • • • 0 • • •

A. PHMA in 1,2-Dichloroethane at 50°C· ••• 36

B. p~,iA in 1,2-Dichloroeti!ane at 98.5°C •• 37

DISCUSSION OF RESULTS 0 e e e e e e e e G e • e • • 6 e e e e

A. Thermal Degradation of P~·1A, Molecular Weight 2,303,000 ..................... 56

B. Thermal Degradation of P~:IA, Molecular Weight 43,000 ....................... 59

c. Molecular weight of Polymer After Be-gradation •••••••••••••••••••••••• 61

LIMITATIONS AND RECOMME~~ATIONS • • • • • • • • • •

CONCLUSI ObiS e • • • • • • • o • • • • • • • • • • • • • • • • • • • • • 64

BIBLIOGRAPHY e e • • e e e G e • e e e e • 0 e • e e • e • e • e e e e

APPENDICES • • • • • • • o • • • o • o • • o • • • • • • e • • • • • • • 68

Appendix I -.Procedure for the use of the cannon-Ubbelohde Dilution Viscometer •• 68

Appendix II- calibration of Viscometer .... 71

iv

Page

Appendix III - Densi·ty ~<ieasurements by Using Pycnometer ..................... o•• 73

Appendix IV - Derivation of Viscosity Equation used in this ~e,ai·s • • • • • • • • .. • 75

X. ACKNOvVLEDGID--lENTS • • • • • • • • • • • • • • • • • • • • 0 • • • • 78

XI. VI'l'A • • 0 • • • • • • • 0 • • • • • • • • • • • • • • • • • • • • • 0 • 0 0 • 79

Table

I

II

III

IV

v

VI

VII

VIII

IX

X

XI

LIST OF TABLES

Viscometer Characteristics • • • • 0 • • • • • • • • • •

Pycnometer Characteristics • e o • • • • • • o o • • • •

Intrinsic Viscosity-Molecular Weight · Relationship at 30 °C £or PI'1HA Samples

v

Page

24

25

(original) o•••························ 29

Intrinsic Viscosity of P~1A, Molecular Weight 2,303,000 in 1,2-Dichloroethane at 30°C •••••••••e••••••••••••••••••••• 30

Intrinsic Viscosity of PMMA, Molecular Weight 43,000 in 1,2-Dichloroethane at 30 °C ~2 • • • • • • 0 • • • • • • • • • • • • • • • • • • • • • • • • • • • ,

Physical Properties of 1,2-Dichloroethane at 30°C •••••••o••••••••••••••••••••••e 34

Thermal Degradation of P~~ 1 in 1,2-Di-chloroethane at 50°C •••••••••••••••••• 38

Thermal Degradation of PI:>ll•lA 2 in 1, 2-Di­chloroethane at 50°C •••••••••••••••••o 40

Thermal Degradation of PI4MA 1 in 1,2-Di­chloroethane at 98.5°C •••·••••••••••&• 42

Thermal Degradation of PHMA 2 in 1,2-Di­chloroetl1ane at 98.5°C •••••o•••••••••• 44

Intrinsic Viscosity of PI~lA 1 After De­gradation for 4 Hours at g8.5°C ••••••• 45

vi

Table Page

XII Intrinsic Viscosity of PNNA 1 After De-grada~cion for 12 Hours at 98.5°C • • • 0 • • 47

XIII Intrinsic Viscosity of PN.MA 1 After De-gradation for 24 Hours at 98.5oc •••••• 49

XIV Intrinsic Viscosity of PMMA l After De-gradation for 48 Hours at 98 .. 5oc • • • • • • 51

XV Intrinsic Vj_scosity of PNMA 1 After De-grada-tion for 48 Hours at 50°C • • • • 0 • • • 53

XVI Molecular Weight of PMMA 1 After Thermal Degrada·tion • • • 0 0 0 • • 0 • • • • • • • • • • • • • • • • • • 55

vii

LIST OF FIGURES

Figure Page

1 Theoretical reaction curve for possible poly (methyl methacrylate) depolymeri-zation mechanism o•c••·············•••o 13

2a cannon-Ubbelohde Viscometer • • • • • • • • & • • • • • 22

2b Pycnometer • • • • • • • • • • • • • • • • • • • • • • 0 • • • • • • • • 22

3 Relationship between pycnometer readings and weight of water at 30°C ••••••••••• 26

4 Intrinsic viscosi·ty of P!v'.U,·lA. 1, Ho1. Wt. 2,303~000 •e•••••e••e••••••••••••••~•• 31

5 Intrinsic viscosity of P~~~ 2, Mol. Wt. 43,000 •••••••••••••••••••••••••••••••• 33

6 Thenual degradation of P~~ 1 in 1,2-di-chloroethane at 50°C •••••••••••••••••• 39

7 Thennal degradation of PNMA 1 and PMMA 2 in 1,2-dichloroethane at 50°C ••••••••o 41

8 Thermal degradation of P~4A 1 in 1,2-di-chloroe·t.hane at 98.5 °C • • • • • • • • • • • • • • • • 4 3

9 Intrinsic viscosity of P~~ 1, after de-gradation for 4 hours at 98o5°C, Mol. Wt. 2,303,000 ••••••••o••••••••oeo 46

10 Intrinsic viscosity of Pl·1MA 1, after de-gradation for 12 hours at 98·5°C, Mol. Wt. 2,267,000 •••••••••••••••••••• 48

viii

Figure Page

11 Intrinsic viscosity of p~,m 1, after de-grada·tion for 24 hours at 98 .. 5°C, Mol. Wt. 1,939,800 •••••••••••••o••o••• 50

12 Intrinsic viscosity of P~m~ 1, after de-grada·tion for 48 hours at 98.5°C, Mol. Wt. 1,690,400 ••••o••••••o•••••••• 52

13 Intrinsic viscosity of PNliA 1, after de-gradation for 48 hours at 50°C, Malo Wt. 2,232,000 ••o•••o••••o••o••o•• 54

1

I. INTRODUCTION

Studies on the thermal behavior of polymers, partic­

ularly on their therr~al degradation, are of importance to

polymer science. Such studies help to reveal the molecular

structure, such as ti1e sequence and arrangement of re­

peating units, or monomers, and side groups in the polymer

or copolymer chain., as well as the na·ture of the cross­

links between chains. These studies also yield informa­

tion about (a) the st.rength of the various bonds holding

together the polymer molecules, (b) the kinetics of de­

polymerization, and (c) the effects of time, temperature,

pressure and o·ther variables, on the rates and products of

degradation.

Similarly, studies on the thermal degradation of

polymers are of extreme importance from a practical poin·t

of view. They not only explain the behavior of pol~ners

under conditions of high temperatures, but also help in

selecting the right kind of already existing materials for

specific uses where high temperatures are encountered.

Such.data can also suggest the design and synthesis of new

materials to meet special requirements.

2

Poly (methyl methacrylate) is of particular interest

in the study of polymer degradation because the prod.ucts

of the ther.mal degradation consist a~ost entirely of mono­

mers and also because the molecular weight of residuals

can be conveniently measured. In contrast to degradation

in the bulk phase, degradation in solution can be carried

out with various concentrations of the solute. Any visco­

sity effects which are operative in bulk phase degradation

can thus be largely eliminated. Secondary reactions are

minimized by working with very dilute solutions.

The purpose of this study was to investigate the

thermal behavior of poly (methyl methacrylate) in solution

in a closed system at elevated temperatures as a function

of time, to gather basic information about the mechanism

of break-down, and to find the molecular weight of polymer

after degradation assuming that the products of degradation

are essentially monomer units. The solvent chosen in this

work was 1,2-dichloroethane.

3

II. LITERATURE REVIEW

Studies of the thermal degradation of vinyl polymers

in solution have been reported by many investigators.

Primary studies have been made on degradation in the ab­

sence or presence of oxygen and catalysts. Alao,theories

of oxidation processes involved have been proposed by a

number of workers.

Staudinger and coworkers (1, 2) carried out a nwnber

of experiments on the thermal degrada·tion of polystyrene

in solution. In one experimen·t they found that when poly­

styrene is prepared under careful exclusion of oxygen it

is more stable thermally than when prepared in the presence

of oxygen. A sample of polystyrene, polymerized by heating

at 60°C for a period of three weeks, was pyrolyzed in nitro­

gen at 0.1 rom pressure, at 290 - 320 °C, and a.t'.va~apJleric

pressure, at 310 - 350 °C. The volatile products ~vere sepa­

rated by distillation into several fractions for identifi-

cation. The degradatj_on products \rJere found to consist,

in addition to the monomer, of a mi~~ture of the dirner and

·trimer.

4

In another experiment (3) polystyrene was dissolved

in various solven-ts and the solutions boiled in a carbon

dioxide atmosphere. The results sho~tled that the viscos-

i ·ties of the initial polymer solutions decreased wi ·th time

vlhere the extent of ·this decrease 'i.vas a function of temper­

ature and type of solvent used.

Experiments were also carried out in order to ascer­

tain whether a soJ_ution of rnonostyrene in tetralin could be

polymerized to the same average molecular weight at 200°C

as was obtained by degradation at the same temperature.

Tl1e resul·t.:s seem to indicate that an equilibrium between

polymerization and degradation does not existo

Schulz and Husemann (4) also carried out a number of

experiments on the degradation of polystyrene in solution.

Samples of a solution containing 0.5 per cent polystyrene

(molecular weight 340~000) were sealed (1) under nitrogen,

and (2) under air. These samples were heated at 132°C for

50 hours. Sample 1 showed a molecular weight of 320,000

and sample 2 showed one of 147~000. These authors assumed

that the small amount of degradation in sample 1 was due

to traces of oxygen.

5

Investigations were carried out by Jellinek and co­

workers (5) on the degradation of polystyrene in solution

in high vacuum. The object of the work was to ascertain

the role which the solvent plays in the degradation pro­

cess and whether the mechansim, as far as monomer for-mation

is concerned, is similar to that of the bulk phase. The

results indicated that monomer formation obeys a zero­

order reaction in both cases.

More detailed experiments were carried out by Jellinek

and Turner (6). These authors investigated the thermal

degradation of polystyrene in 3 per cent by weight naphtha­

lene solutions over a range of temperatures from 345 to

380°C. The percentage loss of weight as a function of

time was very similar to that found for the bulk degrada­

tion in the same temperature range. The energy of activa­

tion is similar to that of the bulk degradation. Appar­

ently the solvent has a negligible effect on the reaction.

Chen (7) studied the photodegradation of polystyrene

in benzene solutions. The solutions were exposed to ultra­

violet radiation of wavelength 2537.5 !. The intrinsic

viscosi~es decreased steadily with time of irradiation.

6

The rate and extent of degradation were found to be a

function of light intensity, as expressed in the following

equation:

where,

d{n)

dt = k :I

(n) = intrinsic viscosity

t = time in seconds

k = rate constant

I = light intensity

{1)

This relationship agrees with a random degradation process

where the mechanism does not follow a specific .law.

Cowley and Melville (8) studied the photodegradation

of poly (methyl methacrylate) exposed to ultraviolet light

of wavelength 2537 A in vacuum, and found that it degraded

rapidly to monomer at temperatures above l30°C. The re-

sults showed that once the reaction is initiated the poly-

mer chain splits off monomer units rapidly by a reverse

polymerization process until the chain is terminated by

reaction with another radical.

Mesrobian and Tobolsky (9, 10) carried out a number of

7

experiments on simultaneous polymerization and degradation

in solutions. Their experiments indicated that equilibrium

or steady state was reached between polymerization and de­

polymerization. However, later reports by JIOI!ltg-cae;y,iand

Winkler (11) and by Thompson {12) showed conclusively that

equilibrium is not involved. These workers (9, 10) used

a reversible viscometer to determine the change in visco­

sity. In using reversible viscometer the solution is

sealed off in the viscometer under air, other gases, or

vacuum: the viscometer is heated in a bath of the desired

temperature, and readings are taken periodically. In

another set of experiments, various solutions of mono­

styrene and polystyrene in toluene were placed in a flask

and refluxed at lll°C. At definite time intervals, var­

ious amounts of benzoyl peroxide were added. After same

fixed time a few milliliters of the respective solutions

were withdrawn and the relative viscosity was measured.

Apparently, the relative viscosities of all samples seemed

to converge to a steady state value. When large amounts

of benzoyl peroxide were added to the solution containing

monomer, the attainment of the steady state was retarded,

8

whereas the reverse \vas observed with polystyrene solu­

tions. The steady state viscosities obtained in the visco­

meter (reversible) were higher than those obtained under

reflux~ Similar experiments were also carried out with

methyl methacrylate. In this case, the convergence of

various samples to a steady state value was not apparent

as with polystyrene.

Thompson (12), and Montgomery and Winkler (11), found

that the nature of the sol vent had a marJ~ed effect on the

efficiency of degradation. Benzene and carbon tetrachloride

were more effective than toluene. However, carbon tetra­

chloride seemed to be much less effective in Thompson's

experiments than in those reported by Montgomery and

Winkler. Thus, Thompson's results for solutions containing

0.5 gram of polystyrene and 0.1 gram of benzoyl peroxide

in 11 ml of solvent were as follows: initially {n) = 0.97;

after 48 hours at 100°C, the intrinsic viscosities were

0.77, Oo61, and 0.61 in toluene, benzene, and carbon tetra­

chloride, re-.pectively. Hydroquinone was found to retard

degradation. From these results they concluded that the

role of the solvent may be that of a transfer agent.

9

Radicals derived from benzoyl peroxide collide with solvent

molecules thus producing solvent radicals, and these in

·turn might react \o'li th polymer molecules_, causing chain scis­

sion. The efficiency of this process is expected to be a

function of the type of solvent used.

Taylor and caverhill ( 13) studied ·the thermal degrada­

tion of polypropylene promoted by organic halogen compounas.

~1e results indicated that a high degree of chlorination

was necessary for proQotion of degradation. Additional

work by these 1:..rorkers ( 13) suggested ·tha·t bromine compounds

were more effective than chlorine compounds; the results

obtained with dichlorodifluoromethane suggest in turn_, that

chlorine compounds are more effective than fluorine corn­

pounds.· By heating at lower temperature (259°C) it was

found that carbon te-trachloride was more effective than

chloroform. They also found that a variety of halogenated

hydrocarbons -v.rere active in reducing ·the molecular \veight

of polypropylene heated in the absence of air in the temper­

ature range 240 to 280°C. In the chlorinated methanes_, the

activities rank as follows:

.. CC14 > CHCls > CH2Cl2 > CF2Cl2

10

Bamford ( 14) and co\vorkers found t!1e sarne order when ·these

compounds were used as chain transfer agents in vinyl poly­

rnerization.reactions.

Harrison (15) investigated the effect of air and iron

salts on solutions of polyvinyl acetate, polystyrene, and

pol~ethyl methacrylat~. The viscosity of these solutions

decreased in the presence of air a·t a ·temperature of 60 °C.

'1"'11e viscosities of solutions of polyvinyl acetate in bis

(2-chloroethyl) ether, kept at 60°C in the presence of

0.003 per cent anhydrous ferric chloride and 1 nun Hg of

air pressure, increased "'.viJch time, and gelling ·took place.,

1~1en the same experiments were carried out in the presence

of larger amounts of air, the viscosity first decreased

before a gel was finally obtained. The general effect was

·that oxygen shows a degrading ac-tion, whereas iron salts

lead to cross-linking.

Mechanisms of degradation of poly (methyl me·thacryla·:.:e)

have been proposed by many investigators. Kuhn's (16)

interpretation was based on the random breaking theory that

linear polymers degrade entirely by random scissions of c-c

bonds in chains.

11

Simha (17) suggested that the mechanism of for.mation

of monomers is due to stepwise breaking away of monomers

at chain ends. Later Blatz and Tobolsky (18) considered

that depolymerization takes place by a stepwise breaking

off of monomers, and is a reverse process of addition

polymerization. In fact, this is exactly the same

mechanism as suggested by S~a.

An extensive investigation of the ther.mal degradation

of poly (methyl methacrylate) was studied by Grassie and

his associates (19, 20). Since the volatile products con­

sist of the monomer which has a high vapor pressure at

roam temperature, the pressure method was used in measuring

the rates. According to Grassie, the theory of random

scissions of chain bonds proposed by Kuhn (16) to explain

the hydrolytic degradation of high polymers is insufficient

to explain the thermal degradation of vinyl polymers, par­

ticularly in cases where the pyrolyzate contains a large

amount of monomer. They considered the various possible

models of molecular weight change due to depolymerization

or polymerization. It was postulated that appreciable

amounts of monomer appear in a random scission process in

12

large molecules \<vhere the average molecular weight of re-

sidual is reduced to a small fraction of its initial value.

Thj_s behavior is represented by line AD in Figure 1.

Another sugges-tion was that ·tile bonds joining ·the end

units to the res-t o£ the chain micrht. be particularl v vulncr-__, ~

able and thus e:,clusively become broken.. Such a process_,

known as stepvlise depolymerization, v:ould result in a oe-

crease in molecula.r \'leight directly proportional ·to the

amount of monomer produced. This behavior is represen:ted

by line AC in Figure 1.

A third possible depolymerization mechanism is the

exact reverse o£ polj(meriza-tion. Initiation would consist

of chain scission resulting in the production of radicals.

This would then rapidly lose monomer uni·ts until they had

completely disintegra-'ced. If these degrading radicals had

life ·times of the order of those of the growing radicals in

polymerizing systems (approximately 10-3 - 10 sees) virtu-

ally all the monomer units would exis-'c a.t any instant as

free monomer or unchanged polymer. No large change in the

molecular weight of the residue \vould therefore be expected

13

A B 100

n m ~ -~ ~ -~ ~ 0 ~ 0

• ~ ~

• n 0 ~

~ ~ m u ~ m ~

0'~------------------------------------------------------~ c 0 100 Percent Degradation to Monomer

Figure 1. Theoretical reaction curve for possible poly (methyl methacrylate) depolymeriza­tion mechanism~ {20)

14

throughout the '¥Thole course of reac·i:.ion. This mechanisra

is represented by line AB in Figure 1.

Grassie and Nel ville ( 20) s-tudied Jche effect of pyro­

lysis on the molecular weight of bulk poly (methyl

methacrylate) san1ples. The results ob·i:.ained in these e.:;:­

periments showed that for the pol:;{:raer of molecular v-1eig~1t

44,300, the molecular weight of the residue remains con­

stant through 65 per cent degrada.·tion. The mechanism of

this reaction con1c1 conceivably be u. :.:cverse pol~;merization.

This series of e:::periments, however, gives no indication

how the chain breaks initially; v1~1e·t:i.~er, for example, eacJ:1

bond in the chain has the same possibility of initial rup­

ture or whether the ends of the chains are the v-.reak points.

As the molecular v-1eight of the ini·tial polymer is raised,

the mechanism ceases to be stric·tly reverse polymerization.

In the later s·tages of the reaction the molecular weig"i1t

falls. The degradat:ion of polymer of molecular Y.leigh·t

94,000 was a reverse polymerization up to about 30 per cen·t

degradation. The molecular weight of t'l1e 179,000 pol~nner

falls after about 10 ·to 15 per cent degradation while ·the

molecular weight of the 725,000 pol~~er appears to suffer

~n immediate fall.

15

Har-t (21) also studied the chan9e of molecular ~Teigll·l:

of the residue "Vli th respect to the extent of degradation,

using two grades of poly (methyl methacrylate). Polymer A

\:las prepared in ·tl1e presence of 0. 6 per cent. benzoyl per­

o:-cide and had an average molecular \'leigh·t of 150,000.

Polymer B was prepared by placing a pure-grade monomer

wi-thout promoters in an evacuated tube and keeping the tube

at -25 ·to -35 °C. r-c had an average molecular weight of

5,100, 000 as de-i:.erm.ined by ·the light-sca-ttering method.

The results showed that the higher the molecular ·t;~eigh-'c -'che

easier the degradation.

The results o£ Grassie (20) and Hart (21), led to a

general agree.uen-'c -'cha-'c the higher -the molecular weigi:rt of

·the original pol:;lmer, the more dras-tic is tne drop in ·the

molecular weight o£ t.he residue. This fact was explained

by Madorsky (22) as sho~m in the follo"VTing mechanism.. De­

gradation takes place by random scissions of chains. Be­

cause of steric hindrance of the CH3 and COOCH3 groups on

alternate carbon atoms, tl1ese scissions cannot be accom­

panied by a hydrogen transfer at the site of scission ..

Therefore, a scission results in the £ormation of t~1o free

radicals:

CH7. H CH3 H CH'2 H I v I 4 I I v I

--- c- c-c ------ c- c- c ~ I I I I I 1 C=O H C=O H C=O H I I : I OCH3 OCH3 .~ QCH3

• ~ YH3 ~ + c-c -c-----1 I I H C=O H

I OCH3

16

~i~hc :Eree radicals further dissocia-'ce to form monomers and

O@~ller free radicals:

H CH3 I H CH3 I I 1. I I ....-.-c-c --+-c-c· I I I I I H C=0 1 H C=O

1 I I OCH3

1 OCH3

H CH3 I I • ----l......... ---c-c I I H c=o

I OCH

3

~ YH3 + C=C

I I H C=O

I OCH3

, e·tc.

17

Grassie and Melville (20) also found that the activa­

tion energy increases with extent of degradation of poly­

(methyl methacrylate). They suggested that this might be

due to some process of strengthening of the bonds which

results in a greater activation energy. However, the in­

crease in acti va·tion energy may be caused by the fact ·ti.1a:::

the weak links are eliminated in tJ.1e initial stage of de­

gradationo Thus, the remaining bonds in the pol~ner are

the regular c-c bonds and are stronger than the wealc links 0

These mechanisms are not adequate to explain that the pro­

ducts of degrada-tion generally con·tain monomers.

Straus and nadorsky (23), and Lehmann, Brauer, et al,

(24), analyzed the volatile products from the pyrolysis of

poly (methyl methacrylate) and sho\Ied that on degradation,

in the temperature range of about 150 to 500°C, poly -

(methyl methacrylate) yields almos·t 100 per cent monoraer.

ii.·t a higher temperature ( 525 °C), ·tile products, as analyze()

by gas chromatography, \-lere shown to be 96.2 per cent of

monomer, 3.6 per cent of gaseous products, and 0.2 per cent

of a residue.

Bywater and his associates (25, 26) investigated the

thermal degradation of poly (methyl me-thacrylate) in

18

diphenylether, o<-methylnaphthalene, and 1,2,4-tichloro­

benzene solutions. The kinetics of depolymerization of

)01y (methyl methacrylate) was studied in the three sol­

vents to assess the effect of solvent on the reaction.

The results show that the mechanism is the same in all sol­

venas. The rate constant for chain initiation changes

little with solvent. The observed variations in rate are

primarily dependent on the magnitude of the chain transfer

constant which increases with the solvent series shown as

follows:

trichlorobenzene < diphenylether < o<-methylnaphthalene.

Grassie (27) was able to explain their-data asing the theory

that d.epolymerization at low temperature is initiated at

unsaturated end-groups produced with termination by dis­

proportionation of macro-radicals.

Chen (28} studied the ther.mal degradation of poly

(methyl methacrylate) of different molecular weights in

toluene, ethyl acetate, and chloroform. Four poly (methyl

methacrylate) samples having molecular weights of

2,210,000, 7,3,000, '20,000, and 4,,000 respectively

19

were studied, each at two concentrations, 0.5 ; w/v and

1.0 ~ w/v. Using a constant temperature bath for heating,

he refluxed the polymer solutions for 10, 20, 30, and 45

hours at their normal boiling points, lll°C for toluene,

78°C for ethyl acetate and 64°C for chlorofor-m. The re­

fluxing condenser was open to the atmosphere.

Molecular weight changes of the refluxed samples were

followed by viscosity measurements. ~s results showed

that the polymer having a molecular weight of 2,210,000

undergoes degradation in all solvents, with the loss of

solution viscosity increasing with increase in reflux

time. The most degradation occured in the chloroform and

toluene solutions, and the least in ethyl acetate. Thus,

degradation appeared to be more depend.ent on the nature of

the solvent than on the temperature.

For molecular weights of 7)3,000, 320,000, and 43,000

his results indicated that the viscosities of polymer

solutions increased with increase in reflux time up to

45 hours. He explained this last result as a case of re­

verse polymerization. However, Chen's results showing an

increase in viscosity for low molecular weight samples

also could be explained by a loss of solvent during re­

fluxing in his open system. This possibility could be

eliminated by using a closed system.

20

21

III. EXPERIMENTAL

A. Purpose of Investigation

The purpose of this study was to investigate the

thermal behavior of poly (methyl methacrylate) in solution

in a closed system at elevated temperatures as a function

of t~e, to gather basic information about the mechanism

of break down, and to find the molecular weight of polymer

after degradation assuming that the products of degrada­

tion are essentially monomer units. The solvent chosen in

this work was 1,2-dichloroethane. Hereafter, poly (methyl

methacrylate) will be referred to as PMMA.

B. Plan of Investigation

When degraded ther.mally, the viscosities of polymer

solutions decrease with time where the extent of this de­

crease is a function of temperature and type of solvent

used. The change in viscosities and molecular weight of

polymer as a function of reaction time were determined in

this investigation.

A cannon-Ubbelohde Dilution Viscometer, and a pycno­

meter, as shown in Figure 2, were used for measuring the

rrf

(a)

• A B

c D

E

F

G

H

1

8 8

6 6

4 4

z l.

0 0

(b)

Figure 2 • (a) Cannon- Ubbelohde Viac:ometer

(b) Pycnometer

• See Appendix I

22

II

23

viscosities and densities of the solvent and the polymer

solutions at a constant temperature of 30°C. A glycerol

bath was set up for refluxing the polymer solution in

sealed Pyrex test tubes at elevated temperatures.

c. Methods of Procedure

1. Procedure for the use of the cannon-Ubbelohde

Dilution Viscometer is described in Appendix I.

2. Calibration of viscometer is described in

Appendix II and shown in Table I.

3. Density measurements by using pycnometer is

described in Appendix III and shown in Table II.

4. Viscosity, relative viscosity, intrinsic visco-

sity and molecular weight calculations.

Viscosities of solvent and solutions were obtained

as described in Appendix I. Relative viscosity is the

ratio of viscosity of solution and solvent:

= nsolution I nsolvent

Specific viscosity is defined as:

- - (nsolution - nsglventl nsolvent

(I·)

(J)

TABLE I

Viscometer CharacJceris·tics

Viscometer Viscome·ter Constan·ts A.Y.- B*

cannon-Ubbelohdc 0.00804 2 .. 30981

* A and B are defined by the equation as wri t·ten up in Appendix II.

24

25

TABLE II

Pycnometer Characteris·i:ics

Reading* Weight of Water Volume of 'tva ter

(gm) (ml)

5-2801 5-3029

5-3167

5-3194 5 .. 3424

13 .. 30 5-3624

* Readings re:Eer to calibration marks at 30°C for water., Density of wa·i:er at 30°C = 0.9957 gm/ml.

~ 01\

J.f Q) .fJ AS ~

'1-1 0

.fJ ~ Ol -rl Q)

~

5.40r------.------r------.------.-----~------~------

5-36

5-32

5.28

5.24~----~------~----~------L-----~------~----_J 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0

Pycnometer Reading

Figure 3. Relationship between pycnometer readings and weight of water at 30°C.

1\)

0\

then calculate Jche reduced viscosity from the following

equation:

= (4)

The intrinsic viscosity, (n), is independent of concen-

tration by virtue o£ extrapolaJcion_, c = 0, but is a func-

tion of solvent usede Intrinsic viscosity, (n), can be

obtained from "'che plot of nsp /c versus c, and extra-

polating to c = zero; ·the intercep-t is equal to (n).

Also, the correlation bet\veen in·trinsic viscosity a.nd

molecular weighJc of linear polymers is expressed in the

equation: (89,30)

(n) (5)

where K and a are constants deterrnined from a double

logari-thmic plot of intrinsic viscosi·ty and molecular

weight.. Both K and a are functions of the solven-t as

vvell as of the polymer type. K and a used in this

paper are shown in Table IIIe

The viscosity equation used in this paper is:

B t

(I)

28

Derivation of equation (6) is described in the appendix

section, Appendix IV.

5. Polymer samples - Two different samples of PMMA

having molecular weight of 2,050,000 and 43,000 were used

throughout this work. All samples were supplied by

Research Lab., Rohm & Haas co., in the for.m of powders.

The molecular weights were checked from dilute solution

viscosity data using ethylene dichloride and following

formula:

(n} =

The results are shown in Table III, and Figures 4 and 5,

pages 29, 31, and 33, respectively.

6. Solvent - Chemically pure reagent, ethylene di­

chloride was used for preparation of polymer sample solu­

tions. The physical properties of the solvent are given

in Table VI.

7. Polymer solution preparation and method of de-

gradation - Ten grams of polymer was dissolved in 1,000 cc

of solvent to make a 1.0 per cent w/v polymer solution.

Portions of this solution, approx~ately 30 cc each, were

29

TABLE III

Intrinsic Viscosi·::y-Ivlolecular Weight: Relationship

at 30 °C for PM.lflA Sarnples (Original)

Polyrner

PI1r-'1A 1

PHNA 2

Molecular Weight (Rohm & naas co.)

2_,050_,000

43_,000

(n)*

3.84

0.196

* calculated from equation:

(n)

4.20

(n) a = K H- in 1,2-dichloroethane where,

(n} -K -a -

X<1 -

intrinsic viscosity 5o3 x lo-5 0.77 raoJ_ecular weight.

** As reported in this. paper.

.J:-iol e cul ar Weight**

2,303,000

43,000

30

TABLE IV

Intrinsic Viscosi-ty of PNMA, Molecular Weight 2,303,000

Concentration

c

gm/100 m1

0.3484

0.3283

Oo2819

0.2292

0.1945

0.1929

0.1584

in 1,2-Dichloroethane at 30°C

Relative Viscosity

nR

3.1161

2-9539

2.6183

2.2345

2 .. 0274

2 .. 0075

1.8124

Specj_fic Viscosity

2.1161

1.9539

1.6183

1.2345

1.0274

1.0075

0.8124

Reduced Viscosi·ty

nso /c ""' 100 ml/gm

6o0738

5-9516

5-7413

5.3854

5.2823

5.2222

5.1287

Intrinsic viscosity • 4.20 100 ml/gm from Figure 4.

a ~ ...... E! 0 0 ...... ~

0

' g. s::

. 6.2

5.4

s.o

4.6

4.2

Intrinsic viscosity intercept) • 4.20

(from l.OO ml./gm

31

3.8--------~------~------~------~ 0 0.1 0.2

Concentration, c, gm/1.00 ml

Figure 4. Intrinsic viscosity of PMMA 1, Mol. Wt. = 2,303,000.

0.4

32

TABLE V

In-trinsic Viscosi i.:y of PWlA, Holecular \'leight 43,000

Concen tra·ti on

c

in 1.,2-Dich1oroethane at 30°C

Relative Viscosity

s::.:)ecific Viscosity

Reduced Vis cosi ·ty

nsp /c

gm/100 m1 100 rnl/ gm

0.8181 1.1624 0.1624 0 .. 1958

0.6533 1.1248 0.1248 0.1911

0.6221 1.1205 0 .. 1205 0 .. 1937

0.5047 1.0988 0.0988 0.1958

0.4978 1 .. 1017 0.1017 0 .. 2043

0.4124 1.0771 0 .. 0771 0.1869

Xntrinaic viscosity = 0.195 100 ~/gm from Figure 5.

0.30--------------~------~---------------

~ 0.20 ~ <o· . 00 0 -0 r-i a

0 0 r-i

.. ()

0.10 r . ......... Intrinsic viscosity (from intercept)

~ = 0.195 100 ml/gm s::

OoOO~------~------------~~------------~ 0 0.2 0.4 0.6 0.8 1.0

Concentration, c, gm/100 ml

Figure 5. Intrinsic viscosity of PMMA 2, Mol. Wt. = 43,000.

\)J \)J

TABLE VI

Physical Propert.ics of l_, 2-Dichloroethane at 30 °C

Boiling Point

Density

Viscosity

1 .. 2404 gm/ml

0. 7160 cen·ti­poise

34

transferred to Pyrex teat tubes which were then sealed.

The tubes were placed in a constant temperature glycerol

bath kept at 50°C. After 4 1 12, 24, and 48 hours inter­

vals samples were withdrawn from the bath, cooled to room

temperature, and then opened.

The density and the viscosity were then measured.

Viscosity and density measurements were made in a constant

temperature bath held at 30f0.2°C. The relative viscosity,

the intrinsic viscosity 1 and the molecular weight of each

sample were then calculated as described before.

Procedure NUmber 7 was repeated at a constant bath

temperature of g8.5°C.

IV. DATA AND RESULTS

Results are reported for the thermal degradation of

PMMA in 1,2-dichloroethane. All data were obtained as

9utlined in the experimental section, pages 21 to 35.

The results which are presented in both tabular and

graphical form, show the effects of initial molecular

weight, reaction time and temperature upon the thermal de­

gradation of PMMA. The data and results are given sepa­

rately for each temperature.

A. PMMA in 1,2-Dichloroethane at 50°C

The results for PMMA having initial molecular weight

of 2,303,000 are given in Tables VII, XV, and ~' pages

38, 53, and 55, respectively. Figures 6 and 13, pages 39

and 54, respectively, are.graphical representation of

these results. The results for PMMA having an initial

molecular weight of 43,000 are presented in Table VIII,

page 40, and in Figure 7, page 41.

:57

B. PHHA in 1, 2-Dic~rloroethane a·l: 98.5 °C

The resul·l:s for Pf"lMA having ini·tial molecular '\veiqht

of: 2,303,000 are given in Table IX and Figure 8, pages. 42

and 43, respectively; and, in Tables XI ·to XIV and

Figures 9 to 12, pages 45 to 52, respectively; and, in

Table XVI 1 page 55. The resul·ts for PI:~1A having an initial

molecular weight of 43,000 are presented in Table X,

page 44.

Sample :CJo.

1 1a 2 2a -;;:, -""

3a 4 4a 5 5a

'i'ABLE VI :t:

~i.1c:cual Dcsrada·tion o:C P!-'J.-:J\ l in

l, 2-D.icllloroe·thanc ut. 50 °C

(1 grrun Pol~ner in 100 m1 Solvent)

Re&CiiUon Ef£1ux Densj_·ty Time Time a·c 30°C

t ~ hrs sec nn/-rll ';;;J L l( --

0 840.0 l. 24J_ 0 840.3 1.241 4 829.0 1.240 4 833-7 1.240

12 816.0 1.240 12 816.0 1.240 24 807-3 1.240 24 807.8 1.240 48 777.6 1.240 48 776.1 1.240

38

Viscosit.y a·c 30 oc

·.n

centipoise

8.375 8.378 8.262 8.307 8.135 8.135 8.047 8.047 7-750 7-734

:59

J I

'

• u

0

I 0 It'\

.q -

co ~

..,. "' • .: "' ~ • 0

-~

k 0 roof

-6 ...t ry Q

l "' -~

roof

.: •

...t

~

..... .. I

1 -

..,. ! •

Oil ..

..., roof

0

I .: 0

• ...t

I ....

• •

+'

-\0

1

• .....

tJ k

...t i'

.....

" '0

r-f

"' <J

-• e • ~

4p •

\0

~

I I

I •

........ 0

k 0

0 0

0 0

::s •

• •

• •

~

0\

GO

t-

\0

In

...t

88lOdl~U80

"=>. <>' ..

~· u

40

TABLE VIII

Therr;lal Degradation o£ Pl·:ll·ll-1. 2 in

l, 2-Dichloroethane a-t 50 °C

( 1 gra1n Polyrner in 100 ral Sol vent)

Sample No.,. Reaction Efflux Density Viscosi-cy '.Ein1e 'l'in1e at 30°C at 7..0o,...

./ "' t ~ n

11rs sec gm/ml centipoise

1 0 89.5 1.240 0 .. 860 2 4 89.5 1.240 0.860 2a 4 89.5 1.240 0.860 3 12 89.0 1.240 0.855 3a 12 89.5 1.238 0.859 4 24 89.5 1.236 Oe858 4a 24 89.5 1.240 0.860

5 48 89.5 1.240 0 .. 860 5a 48 89.5 1.240 0.860

• • -g

10.0------~-----.-----,------.------.-----.----~

PMMA 1

~ 6.0 ..., I: • u ~

CJ 0

~ - Duplicate aaaple

~ 4.0 ..., ., I:

2.0

PMMA 2

8 16 24 '2 40 48

'l'ime, bra

Pigure 7. Ther.mal degradation of PMMA 1 and PMMA 2 in 1,2-dichloroethane at 50°C •

. . ····-···------- ·-··-··-·-·. ·····-·.- ··--·-···-·· ----------------

... ....

42

TABLE IX

Thermal Degradation of P11L,1A l in

1,2-Dichloroethane ai: 98.5°C

( l gram of PolyYaer in 100 ml Solven-t)

Sample No. Re&a.td.-on Efflux Density Viscosity Time Time aJc 30oC at 30°C

t ~ n

hrs sec ga/rnl centipoise

1 0 840.0 1.241 8.375 la 0 840 .. 3 1.241 8.378 2 4 818.3 1.240 8.153 2a 4 818.5 1.240 8.159 3a 12 783.6 1.240 7.808 4 24 720.0 1.238 7.164 4a 24 725 .. 0 1.240 7.225 5 48 565-3 1 2~9 5.626 -· ./

5a 48 566 .. 2 1.240 5 .. 635

Q) Ctl

.,-! 0 ~

.,.J .jJ d Q.) u

0\

u 0

~ .jJ RS

d

9-0r-----~-----T------.-----~-----T------~-----

fLo~ 6. - Duplicate sample

~

~ 7.0

6.0

0

s.o ~---:----:-:--~----L--..J__----'--_j 0 8 32 40 48 16 24

Time, hrs

Figure 8. Thermal degradation of PMMA 1 in 1,2-dichloroethane at 98.5°C. ~ \)J

44

TABLE

Therrnal Degradation of PH.l'iA 2 in

1,2-Dichloroetilane at 98.5°C

( 1 gram Polymer in 100 rnl Sol vent)

sample No., aeaeuon Efflux Density Viscosi·ty Time Time at 30°C a~c 30oC

t ~ n

l1rs sec gm/ml centipoise

1 0 89.5 1 .. 240 0.860 2 4 89.5 1 .. 238 0 .. 859 2a 4 89.5 1.240 0.860 3 12 90.1 1.240 0.866 3a 12 89 .. 5 1.240 0.860 4 24 89 .. 5 1.240 0 .. 860 4a 24 89 .. 5 1 .. 240 0.860 5 48 89.5 1.240 0.860 5a 48 89.5 1.240 0.860

45

TABLE XI

Intrinsic Viscosi·ty of P.t11-1A l After Degradation

Concentration Relative S:;_Jeci:Eic Reduced Viscosity Visoosj_ t.y Viscosi·ty

c nR nco nsp /c o_,_

gm/100 m1 100 m1/gm

0.3543 3 .. 1267 2.1267 6 .. 0024

0.2222 2.1680 1.1680 5-2567

0.1797 1.9183 0.9183 5.1114

0.1518 1 .. 7585 0.7585 4.9968

0.1247 1 .. 6080 0.6080 4.8759

Intrinsic viscosity = 4.2 100m1/gm from Figure 9.

7-0

6.0

0

Xntrinsic viscosity (from intercept)= 4.2 lOOml/gm

0.1 0.2 0.3

Concentration~ c~ gm/100 ml

0.4

Figure 9. Intrinsic viscosity of PMMA 1~ after degradation for 4 hours at 98.5°C, Mol. Wt. 2,303,000.

46

47

TABLE XII

Intrinsic Viscosity of Pl~IA l After Degradation

ConcenJcra·tion

c

grn/100 m1

0.3624

0.3351

0.2428

0.2240

0.1975

0 .. 1766

0 .. 1618

0.1295

for 12 Hours at 98.5•c

Re1a·tive Viscosi·ty

3.2163

3.0077

2.3299

2.2061

2.0313

1.9127

1.8208

1.6507

S?ecific Viscosi-ty

2.2163

2.0077

1 .. 3299

1.2061

1.0313

0 .. 9127

0 .. 8208

0.6507

Reduced ViscosiJcy

n 8 p /c

100 m1/gm

6 .. 1148

5 .. 9913

5 .. 4776

5 .. 3833

5 .. 2219

5.1690

5 .. 0734

5.0244

Intrinsic viscosity • 4.15 100 ml/gm from Figure 10.

48

7.0

6 - Dup1icate samp1e

6.0

~ ...... s 0

5.0 0 ...... ~

0

' 0. tQ

s:: Intrinsic viscosity (:from 4.0 intercept) - 4.15 l.OO ml./gm

0 0.1 0.2 0.4

Concentration, c, gm/100 m1

Figure 10. Intrinsic viscosity of PMMA 1, after degradation for 12 hours at 98.5°C, Mo1. Wt. = 2,267,000.

49

TABLE XIII

Intrinsic Viscosity of PMI1A l Af·ter Degradation

for 24 Hours at 98.5°C

Concentration Relative Specific Reduced Viscosity Viscosi-ty Viscosity

c nR nsp nsp /c

gm/100 ml 100 ml./gn

0.4447 3-7287 2.7287 6.1364 0.4073 3-3558 2-3558 5-7844 0.3215 2 .. 7037 1 .. 7037 5-2988 0.2848 2.5100 1.5100 5-3023 0.2695 2.3881 1.3881 5-1508 0.2214 2.1076 1.1076 5-0032 0.2018 1.9517 0.9517 4.7153 0.1839 1.8902 0.8902 4.0409 0.1410 1.6621 0.6621 4 .. 6960

Intrinsic viscosity • 3.68 100 ml./gm from Figure 11.

7-0r-----.------.------~----~------

6.- Duplicate sample

6.0

~ r-f a 0 ~ s.o

"' 0

' ~ s:: 4.0

L

Intrinsic viscosity (from intercept) = 3.68 100 ml/gm

o.s 3.0 ::--~-=---.1.---..1.--__j_--_j

0 0.2 0.4 0.3 0.1

concentration, c, gm/100 ml

Figure 11. Intrinsic viscosity of PMMA 1, after degradation for 24 hours at 98.5°C, Mol. wt. = 1,939,800. \Jl

0

51

TABLE XIV

Intrinsic Viscosity of Plv:II'-iA 1 After Degradation

for 48 Hours at 98.5°C

Concen·tration Relative Specific Reduced Viscosity Viscosity Viscosi-'cy

c nR nsp n 8 p/c

gm/100 m1 100 ml/gm

0.4560 3.2431 2.2431 4.9196 0.4453 3-1744 2 .. 1744 4.8830 0.3210 2.4326 1.4326 4.4629 0.3170 2.4037 1.4037 4.4279 0 .. 2541 2.0590 1.0590 4.1679 0.2483 2.0384 ]_.0384 4.1829 0.2108 1.8529 0.8529 4.0470 0.2020 1.8166 0.8166 4.0433 0.1552 1. 6036 0.6036 3.8897 0.1486 1.5711 0.5711 3.8442

J:ntrinsic viscosity • 3.31 100 m1/gm from Figure 12.

6.0 -------r------,r-----r------r---.--1

~ r-1 s.o a 0 0 r-1

.. ~

0. 4.0 ~Uj

6 - Duplicate sample

Intrinsic viscosity (from intercept) a 3-31 100 ml/gm

3.0~----~----~------~----~----~--~ 0 0.1 0.2 0.3 0.4 o.s

concentration, c, gm/100 ml

Figure 12. Intrinsic viscosity of PMMA 1 after degradation for 48 hours at 98.5°C, Mol. wt. = 1,690,400.

\J1 1\)

53

TABLE A.'V

Intrinsic Viscosi·ty of PMiwiA l Af·ter Degradation

ConcenJcration

c

gm/100 m1

0 .. 4053

0 .. 3138

0.2697

0 .. 2218

0.1990

0.1672

0 .. 1645

0 .. 1330

£or 48 Hours at 50°C

Relative Viscosity

3.4626

2.7832

2.4831

2.1806

2.0073

1.8222

1.8193

1.6436

Specific Viscosity

2.4626

1.7832

1.4831

1 .. 1806

1 .. 0073

0.8222

0.8193

0.6436

Reduced Viscosi·ty

n 5 P /c

100 m1/gm

6.0758

5.6828

5-4992

5 .. 3237

5.0629

4.916l

4.9814

4.8375

Intrinsic viscosity • 4.10 100 m1/gm from Figure 13.

7 oO ~--r-----,-----,-----r----

6.- Duplicate sample

6.0

~ r-1 a 0 0 s.o r-1

~

u

"' g. Intrinsic viscosity (from intercept) = 4.10 100 m1/gm s:: 4.0

3.0 0 0.1 0.2 0.3 0.4 0.5

Concentration, c, gm/100 ml

Figure 13. Intrinsic viscosity of PMMA 1 after degradation for 48 hours at 50°C, Mol. Wt. = 2,232,000.

\Jl ~

Temperature

oc

50.0

98o5

98-5

98o5

98-5

55

TABLE A.'VI

Nolecular Weight of PlvlHA 1 After

Thermal Degrada~ion

Heating Time

hrs

48

4

12

24

48

Molecular Weight

2,232,000

2,303,000

2,267,000

1_,939,800

1,690_,400

56

V. DISCUSSION OF RESULTS

The results of this investigation are discussed in

the same order as the data presented in the preceeding

section.

A. Thermal Degradation of PMMA, Molecular Weight 2,303,000

The effects of degradation of high molecular weight

PMMA on the viscosity of the polymer solutions in 1,2-di­

chloroethane are shown in Figures 6 and 8, pages 39 and 43,

respectively. The viscosity of polymer solutions decreased

with increasing reaction time. When the reaction time was

kept constant, the viscosity of polymer solution decreased

with increasing temperature of reaction. Generally it can

be stated that the higher the molecular weight of a polymer

the higher the viscosity and the longer the chain length.

It has been postulated that the double bonds at the chain

ends, as a result of chain termination, are weak links (26,

27, '~). A possible mechanism is shown on page 57· It

has been generally accepted that the C=C bonds are more

subject to thermal and chemical attack than normal c-c

bonds. under certain conditions, where solvent radicals

may be present at elevated temperatures, they are probably

57

the only bonds which are attacked. In this case the exte~~

of clegracla tion o£ P.L•lMA with increasing reaction· tL.-~e c'~n be

explained by the following mechanism:

(a) The degradation at a given temperature stops at

some chain length, corresponding to a stage of degradation

where all links have been attacked. The viscosities of

polymer solutions will then approach a constant value. A

possible mechanism is:

H CH3 H CH3 H ·cHs

-c-c ---c-6 c=~ I I ........ I ~ I

H C = 0 tH\ C = 0 ?' C = 0 0CH3 "'"~,, OCHs / / OCHs

Long chain, weak links

r--/ H CH Ff yHs

---~-£ s-0-C==~=O 6cH3 OCHs

+ H CH., I l ._.

c=c I I

H c=O OCH~

y.i:l3

Y\. CH2 = C I C=O 0CH3

CHs CHs I

Monomers H CHs Ff -6-6--c=

I I H c=O

6 I C=O + -m CH2=y

C=O OCHs OCHs

I .

OCHs

Short1 stable chain ... \ .... . ..

58

The mechanism given on page 57, vnLile generally accepted,

is slightly misleading. A closer look at this mechanism

indicates that it is not the C=C bond tl~at is attacked

but rather the c-c bond alpha to the double bond which is

the weak link. Actually, C=C bonds are not \'leaker than

c-c bonds, they are more polarizable and therefore more

reactive. A more detailed version of the above mechanism

is shown belo\•1:

---/ ~

\'T-t, CH3 ,; -, titl:s._ ).. • -------- y ..._'f. - ... -·

H y=O H =

yH3 r y = 0

OCHs OCH3

! ------'-"-' c -==

!r

CH3 I c c===-o I OCHs

+ H CHs 9 ==-¢ H y=O

0CH3

(b) The degradation will proceed until almost all

polymer is converted to monomer. The viscosity of polymer

solution will decrease to the viscosity of methyl

59

methacrylate monomer.

If the above mechanisms are valid, the question arises

as to why the initial molecular weight of the polymer should

nave any influence on the degradation& ~~is study does not

give sufficient information to answer this question. How­

ever, the low molecular weight PMMA does not behave in the

same mannero At this point it can be only postulated that

the high molecular weight PMMA degrades until a statistical

balance is obtained between the degraded polymer chain and

the newly generated monomer units, at which time it appears

reasonable to a.ssurne that a reverse polymerization mechanism

prevails. Such mechanism is discussed in the following

section.

B. Thermal Degra~ation of PMMA, t-iolecular Weight 43,000

It was found in the experiments with PMMA of molecular

weight of 43,000 that the viscosities of these solutions

remain fairly constant with increasing reacUon:: :~,..:;apto

48 hours, and temperatures upto 98.5°C in 1,2 dichloro­

ethaneG This result can be explained as follows. The

chain-transfer mechanism is of the usual hydrogen-transfer

type (27). In the absence of transfer, and assuming

60

te~ination by disproportionation~ each molecule (radical)

must~ therefore~ be te~inated at one end by a catalyst

fragment.

~=free radical of catalyst

Initiation: CH3 H H CH3 I I I I •

R• + c - c ,... R-C -c I I I I

C=O H H C=O I J 0CH3 OCH3

Propagation: H CHs I I

I-I CH3 H CH3 H CHs I l.e R-C-C c= c __.R

1 h. ---f- c1 - c1

• y-T I I I I

H C=O I

OCHs

I I H C=O

I OCHs

H y=O H y=O OCH3 0CH3

'T\

Te~ination by disproportionation:

H yHs CHs H

+ .I I c-r· ,.., ,_,. ,.,_ ;ty ....... -.-:1-. I

rH' ~=0 / C=O H -<:.._ / I

/ OCH3 ......_ CHs ----- -! H CHs CHs ¥ I I + I

H-C <f .........

---- ---c=~ I

=0 y=O H

6cHs OCHs

61

In the above mechanism a free radical abstracts a

hydrogen atom from a carbon atom in the backbone of an­

other chain which can become a polymer molecule with a

saturated end group as seen in the disproportionation re­

action. For low molecular weight polymers, the polymer

chain is initiated at the end by either becoming activated

or by breaking off of a monomer unit. These monomers can

then add to a pol~uer radical or an activated monomer in

rapid succession, as in the propagation reaction during

polymerization. As a result the degradation process will

be a reverse polymerization for a low molecular weight&

This means the molecular weight of residue remains constant

upto some extent. It would be expected that if the reaO*ion

time and temperature were extended sufficiently that a de­

crease in viscosity would occur and that the mechanism of

decomposition would be the same as that for the high

molecular weight polymer.

c. Molecular Weight of Polymer After Degradation

The molecular weight of the high molecular weight

samples after 48 hours decreased from 2,303,000 to

2,232,000 and 1,690,400 at 50°C and 98.5°C, respectivelyo

62

There has not been any change observed in the visco­

sity of the low molecular weight solution for different

temperatures, and reaction times, which indicates that the

molecular weight of the low molecular weight polymer re­

mains constant in the experimental range studied in this

thesis.

VI. LIMITATIONS AND RECOMMENDATIONS

1. The ther.mal degradation of poly (methyl

methacrylate) in solution in a closed system was studied

at 50°C and at 98.5°C. At a higher temperature, viz. 150°C,

the sealed samples always exploded. This may be due to the

high vapor pressure of 1,2-dichloroethane at that tempera­

ture developed inside the sealed test tube. This may be

avoided by sealing the tube under vacuum. Thus, the pres­

sure developed may not shatter the sealed test tubes.

Thermal degradation could then be studied at higher temper­

atures.

2. It is suggested that the degraded polymer be frac­

tionated and then analyzed for the products of degradation.

It would then be possible to obtain the rate data on the

different degradation processes.

3. The Cannon-Ubbelohde viscometer used for measuring

the viscosities was open to the atmosphere during testing.

There may be some error involved in the circulations due to

small amounts of solvent evaporation. Such an error can

possibly be eliminated by making the viscosity measurements

in a reversible viscometer.

64

VII. CONCLUSIONS

As a result of this study the following conclusions

were made:

1. The rate of thermal degradation of PMMA in solu­

tion is a function of polymer molecular weights, tempera­

ture, and reaction time •

. 2. PMMA having an average molecular weight of

2,303,000 degrades when its solutions in 1,2-dichloro­

ethane are heated for same time at 50 or 98.5°C.

3. The role of the solvent on these degradation

processes may be that of a transfer agent.

1.

2.

3-

4.

5-

6.

7-

8.

g.

10.

11.

12.

65

VIII. BIBLIOGRAPHY

Staudinger, B., Brunner, M., Frey, K., Garbash, P., Singer, R., and Wherli, s., Ber., 62B, 241 (1929)., Ann., 468, 1 (1929): cf. Chern. Abstr. ~~ 2949 (1929).

Staudinger, H., and Steinhol~r, A., Ann., 5!I, 35 (1935): cf. Chem. Abstr. ~' 4336 (1935}.

Staudinger, H., Frey, K., Garbash, P., and Wherli, s., Ber., 62B, 2912 (1930); cf. Cham. Abstr. 24, 1564 (1930).

Schulz, G. v., and Busemann, E. z., Physik. Chem. 40, 524 (1948); (Reference No. 31, p. 181). --

Jellinek, H. B. G., and Spenaer, L. B., J. Polymer Sci. §., 573 ( 1952).

Jellinek, H. H. G., and b~:ner;,K. J., J. Polymer Sci. ll, 353 (1953).

Chen, s. w., J. Phys. & Colloid Chern • .5,2., 486 (1949).

cow1ey,~~P. R., and Melville, H. w., Proc. Royal Soc. (London} A210, 461 (1952).

Mesrobian, R. B., and Tobo1sky, A. v., J. Polymer Sci. a, 463 (1947).

Mesrobian, R. B., Metz, D., and Tobolsky, A. v., J. Am. Chem. Soc. ~ 785 (1945).

Montgomery, D. s., and Winkler, c. A., can. J. Research sg8, 407, 416, 429 (1950).

Thompson, J. o., J. Phys. & Colloid Chem. ~ 338 (1950).

caverhill, A. R., and Taylor, G. w., Polymer (London) ~ 19} (1965).

66

14. Bamford, c. H., Barb, w. G., Jenkins, A. D., and Onyon, P. F., .,Kinetics of Vinyl Polymerization by Radical Mechanisms", p. 239, Butterworths: London, (1958), (Reference No. 13, p. 195).

15. Morrison, J. A., Holmes, J. M., and Mcintosh, R., can. J. Research B24, 179 (1946): (Reference No. 31, p. 201).

16. Kuhn, w., Ber., ~~ 1503 (1930): (Refereace No. 22, p. 176).

17. Simha, R., J. Appl. Phys. ~, 569 (1941).

18. Blatz, P. J., and Tobolsky, A. v., J. Phys. Chem. ~ 77 (1945).

19.

20.

21.

22.

23.

24.

26.

Grassie, N., and Melville, H. w., Faraday Soc. Disc., g_, 377 ( 1947).

Grassie, N., and Melville, H. w., Proc. Royal Soc., Al99, 1 ( 1949).

Hart, v. E., J. Research Nat '1 Bur. Standards, .5.Q., 67 {1956).

lladorsky, s. L., .. Thermal Degradation of Organic Poly­mer", p. 179, John Wiley & Sons, Inc., New York (1964).

Straus, s., and Madorsky, s. L., J. Research Nat'l Bur. standards 66 A-B, 401 (1962).

Lehmann, F. A., and Brauer, G. M., Anal. Chem. 22, 673 (1961).

Grant, D. H., and Bywater, s., Trans. Faraday Soc. ~ 2105 (1963).

Bywater, s., and Black, P. E., Polymer Reprints, 2 372 (1964).

27.

28.

29.

JO.

31.

Grassie, N., "Chemistry of High Polymer Degradation Processes", p. }0, Interscience Pub., Inc., New York ( 1956).

Chen, Y. c., "Thermal Degradation of Poly (methyl methacrylate) in Solution in various Solvents", M. s. Thesis, University of Missouri at Rolla, (1965).

67

Billmeyer, F. w., "Textbook of Polymer Chemistry", p. 131, Interscience Pub., Inc., New York (1957}.

Tompa, H., "Polymer Solutions", p.273, Academic Press Inc., Butterworths Pub., London (1956).

Je.llinek, H. H. G., "Degradation of Vinyl Polymers", p. 29, Academic Press Inc., New York (1955).

68

IX. APPENDICES

Appendix I

Procedure for the Use of the cannon-Ubbelohde Dilu­

tion Viscometer.

1. Clean the viscometer using suitable solvent

(such as acetone) and dry by passing clean, dry filtered

air through the instrument to remove the final traces of

solvento Periodically, traces of organic deposits should

be removed with a chromic acid-sulfuric acid cleaning

solutiono

2o If there is a possibility of lint, dust, or other

solid material in ~che liquid sample, filter the sample

through a sintered glass filter.

3. Place the sample in a constant temperature bath

at 30°C for 15 minutes. Charge a measured amount of sample

(about 10.0 ml) directly from the pipette through tube G

(Figure 2, page 22) in to the lower reservoir of the visco-

meter.

4. Place the viscometer into the holder and insert

it into the constant temperature water bath at 30°C• Verti­

~ally align the viscometer in the bath.

69

5. Allow approximately 5 minutes for the sample to

reach the bath temperature.

6.. Place a finger over tube B and apply suction to

tube A until the liquid reaches the cen·ter of bulb c.. Re­

move finger from tube B, and inunediately place it over tube

A until the sample drops from the lower end of the capillary

into bulb I. Then remove finger and measure the efflux

time.

7. To measure the efflux time, allow the liquid

sample to flow freely down past the e·tched mark D, measur­

ing the time for the meniscus to pass from mark D to marJc

F, to the nearest 0.1 second (use stop watch}.

8. Without recharging the viscometer, make check

determination by repeating steps 6 and 7 until the efflux

time are nearly the same (approx. 0.1 second difference).

9. To calculate intrinsic viscosity of solution,

charge measured amount of solution (about 5 to 15 gram)

directly from the weighing buret through tube G into the

lower reservoir of ·the viscometer. Dilute samples by add­

ing measured amount of solvent from the weighing buret.

Mix the original sample and the solvent by applying slight

pressure to tube B several times, and shaking the I.:.::.. s-::

70

viscometer.

10. Additional dilution may be made, if necessary,

by repeating steps 5 to 8.

71

Appendix II

Calibration of Viscometer.

For gravity type viscometers, the viscosity equation

is usually written as follows:

where,

_lL t

'V' = kinematic viscosity in stokes or centipoise

n - viscosity of solution in poises or centipoise

~ - density of solution in gm/ml

A - viscometer constant

t = efflux time in seconds,

(6)

B/t is called the l<.inetic energy correction. If the co-

efficient of tl1e kinetic energy correction is constant, it

may be said that B is constant.

The viscometer constant(the same at all temperatures)

A can be determined by using two or more standard oils

whose kinematic viscosities and densities at elevated

temperatures are known. The viscometer constant B can be

~imilarly determined. Pure water or pure solvent whose

viscosity and density are known can be used. Follow the r

72

procedure described in Appendix I observing the efflux

time, t. Substitute t into tl1e viscosity equation, equa­

tion (·)~ to calculate the viscometer constants.

73

Appendix III

Density Measurements by Using Pycnometer.

Density of the solvent and the polymer solutions were

measured using a pycnometer as follows:

1. Clean the pycnometer using acetone, then dry by

passing dry ~iltered air or by using vacuum pumpe

2. Weigh the pycnometer to the nearest 0.0001 gram.

3. Charge the pycnometer with water (above 5 ml) by

applying pressure (using a rubber bulb). The water to be

used herein, is placed in a constant temperature bath at

30°C for 15 minutes.

4. Place the pycnometer into constant temperature

bath maintained at 30°C.

5. Allow approximately 5 minutes for the contents of

the pycnometer to come to bath tempe~ature.

6. Record the reading marks from the pycnometer.

7o Determine the total weight of water and pycno­

meter to obtain the weight of water in the pycnometer.

8. charge the pycnometer with different volumes of

water (approxo 5 ml), repeat steps 3 to 7•

74

9. Plot the relationship between the reading marks

of pycnometer and weight of watero

lOo Charge the cleaned pycnometer with the samples

to be measured, and repeat steps 3 to 7.

llo The density of the sample under consideration,

at 30°C, is obtained from the weight of the sample (as

determined from the above-mentioned steps) divided by

the weight of wa-ter a-t the satae rcaaing.

75

Appendix IV

Derivation of Viscosity Equation used in this ~·'•·

The viscosity equation used in ·this paper '\vas derived

as follows:

If an energy and material balance is made about a

capillary viscome·ter, equation ( .. ""'~) is obtained:

(.8:')

where,

x1 - x2= vertical dis·tance bc·c\'lcen two menisci in viscometer

F = friction in capillary f

Fe= fricJcion due to strcar.1 contraction

F -e- fric-tion due to stream expansion.

F .c can be calcula·te<J by Poiseuilles law: I

where,

F f

L = capillary length

8LV_A ~ 9 1'( r4 ·t

u - average velocity of solu·tion in capillary

).{= absolute viscosity of solution

( ~,-.)

76

g - gravi ta·tional constant

~ - densi·ty o£ solution

D = capillary diameter

v - efflux volt-une

t = efflux time

and, r - capillary radius ..

Since both l<..,c and Fe have been correlated as a func>cion o£

kinetic energy, they can be added such that:

(m u2 I g) (18)

where m is the kinetic energy correction coefficient.

Equations ( ,.) and (10) can now be subs·tituted into equa-

tion ( 81) yielding equa·tion (11):

xl - =\{2 - 8LVA + m u2

.T( S g r4 t g (llia)

(nr~ tj 2

8LVA + m Xl - X2 =

n. ~ g r4 t g (11.)

Solve equation (11) for ..-L( /3 , kinematic viscosity, and

get equation (ll)s

.Ttg r4 t (xl - ~) 8 LV

m V (12) 8L.rct

77

Equa·tion (11-) is viscosity equation and is usually written

as follows:

n = At B

t (6)

X. ACKNOWLEDGEHENTS

The· author wishes to express his sincere appreciation

and gratitude to Dr. tvouter Bosch, Dean of Graduate School,

and Professor of Chemistry, for his guidance and advisory

during this research \vork.

Sincere thanks are extended to Drs. K. G. Nayhan,

J. L. Zakin, ana S. B. Hanna for thei~ ~1elpful adv:Lse ana

,.,,.lc':'c~~·;- · rl··10. .,,~-;ng J_-ll;S research WOr}C. t:)J,... :j .J - \--·- '--· .L ._. u \...L.- .1- '- .J..

Special thanks go to the Rohm and Haas Company,

Philadelphia, Pa., \'lho furnished the poly (methyl

methacrylate) fractions. used in this research work.

79

XI. VITA

The author, son of Pranjivandas and Sarojben Parikh,

was born a·t A11Iaedabad, India, on October 27, 1942.. He

a·t·tended Th.e Tutorial High School, A~u·:~edabad, IncJia, anCi

<;Jraduated in 1958. rn 1958, he joined Gujarat Univerisi~cy

and obtained his Bachelor's degree in Hathematics and

Physics in 1962.

He carne to ·t:1is country in Sep·tember 1962 and v1as

enrolled at the University of N.issouri at Rolla. He

received the degree of Bachelor of Science in Chemical

Engineering in June 1964.

Also, in January 1964, he was dually enrolled as a

graduate studen~c in the Department of Chemical Engineer­

ing.