- Email: [email protected]
Thermal stability and degradation mechanisms of poly(acrylic acid) and its salts: Part 1—Poly(acrylic acid)
Polymer Degradation and Stability 29 (1990) 233-246
Thermal Stability and Degradation Mechanisms of Poly(Acrylic Acid) and its Salts: Part 1 Poly(Acrylic Acid) I. C. McNeill & S. M. T. Sadeghi Polymer Research Group, Chemistry Department, University of Glasgow. Glasgow G12 8QQ, UK (Received 22 September 1989; accepted 2 October 1989) ABSTRACT The thermal degradation of poly( acrylic acid) has been studied using thermal volatilisation analysis ( T V A ) and thermogravimetry (TG). The degradation studies were carried out using two approaches. Programmed heating was carried out at lO°/min to temperatures from 170 to 500°C and the polymer was also heated isothermally at 210°C for different times, under TVA conditions. Quantitative measurements of the main product fractions have been made. The gaseous, volatile liquid and cold ring fraction ( CRF) products from the TVA degradation have been analysed by IR, M S and G ~ M S techniques. The polymer yields water and carbon dioxide as major products. The degraded polymer develops anhydride ring structures in the chain as a result of a dehydration process. At higher degradation temperatures, chain .fragments with anhydride rings and/or acid groups are formed, together with various minor volatile products. The mechanism of degradation is discussed.
INTRODUCTION The main features of the thermal degradation of poly(acrylic acid), PAA, have been known for many years, but certain aspects of the degradation products and mechanism at higher temperatures (above 300°C) still remain the subject of discussion. Otsu and Quach 1 examined the differences in the thermal degradation of head-to-head and head-to-tail PAA by using thermogravimetry. They showed that h - h PAA degrades at lower temperatures and in a different manner from h-t PAA, the degradation of which becomes 233 Polymer Degradation and Stability 0141-3910/90/$03"50~'~ 1990 Elsevier SciencePublishers Ltd, England. Printed in Great Britain
234
I.C. McNeill, S. M. T. Sadeghi
discernible at 150°C and proceeds in two steps. Eisenberg e t al., 2 who studied the dehydration kinetics and glass transition temperature of PAA have shown that anhydride formation in PAA is a first order reaction, as also is decarboxylation, the latter being much slower than the former. The thermal characterisation of PAA has been studied by Maurer et al., 3 who reported that atactic and syndiotactic PAA appear to differ significantly with regard to glass transition temperature and thermal stability. Roux e t al., 4 who studied the pyrolysis of PAA and poly(methacrylic acid), PMAA, in the range 160-240°C in a non-oxidative atmosphere, suggested that the kinetics of evolution of light products indicate a competition between dehydration and decarboxylation reactions. They believed water, carbon dioxide and monomer (PMAA only) were the products of pyrolysis of PAA and PMAA in that temperature range. No formation of carbon monoxide or acrylic acid monomer was observed. Grant and Grassie 5 showed that six-membered glutaric anhydride type rings result from intramolecular reaction of adjacent carboxyl groups in the decomposition of h-t P M A A at about 200°C. This dehydration reaction was found to be accompanied by a small amount of depolymerisation to monomer. Some investigations were extended to higher temperatures (above 250°C) in order to obtain further information on the degradation process. For example, McGaugh and Kottle 6 investigated the thermal degradation of PAA in air. They reported that in the range 25-150°C, the major reaction is the formation of anhydride groups, then at 150-275°C unsaturation begins to appear, as well as structures indicated from the IR absorption to be ester or cyclic ketone. They also noted that above 350°C, the unsaturation becomes very evident. The same workers v have also examined the thermal degradation of acrylic acid-ethylene copolymers, using TG and TVA. They reported that the degradation mechanism of the acrylic acid portions of the copolymer consisted of dehydration of the acid groups forming anhydride, followed by decarboxylation of the anhydride leading to unsaturation. Nicholson e t al. 8 - 1 ° have recently studied the thermal behaviour of partially neutralised PAA at 250°C. They showed that different chemical processes; namely, dehydration and decarboxylation, occur at this temperature, depending on the nature of the counterion. In the present work, the thermal behaviour of PAA has been investigated using TVA 11.12 and TG techniques to provide a basis for a series of studies of the degradation of PAA salts using the same experimental approach. In order to investigate the possibility of the formation of monomer and other minor products, a detailed product analysis for the less volatile liquid fraction has been carried out using the G C - M S technique.
Thermal stability and degradation of poly( acrylic acid)
235
EXPERIMENTAL
Materials Acrylic acid (Hopkin and Williams) was purified by distillation under reduced pressure (b. 60°C at 22 mm Hg) and the middle fraction was used immediately. ~-~'-Azodiisobutyronitrile, AIBN (BDH Chemicals), was recrystallised twice from AR grade methanol. A commercial poly(acrylic acid) sample (Aldrich), in powder form, with Mw 250 000, To 106°C, was used without purification, for comparison with the laboratory sample prepared as described below.
Polymerisation Polymerisation ~3 of freshly distilled acrylic acid was carried out in the presence of 0"1%w/v AIBN initiator. The product was purified by dissolution in water and isolation from the aqueous solution by freezedrying, followed by subsequent evacuation at about 40°C for 80 h. RESULTS A N D DISCUSSION
Programmed heating experiments Thermal volatilisation analysis TVA was carried out under vacuum as previously described. ~1'~2 Programmed heating was at a rate of 10°/min. The TVA curves for both polymers are shown in Fig. 1. It is clear from these curves that in each case decomposition begins at about 175°C, but appreciable rates of breakdown are observed only above 200°C. Both polymers show a three stage decomposition and from inspection of the separation of the traces, it is evident that product composition is different at each stage, the product volatility becoming greater as the degradation temperature rises. In particular, non-condensable gases are formed only above about 300°C, in which temperature region the presence of materials covering a wide range of volatilities is indicated. There is no major difference in behaviour between the commercial and laboratory samples on the basis of the TVA data. Gravimetric data were obtained under TVA conditions in experiments in which the polymer sample was placed in a thin glass, flat-bottomed sample tube, 25 mm base diameter, weighed before and after the experiment, placed inside the TVA tube and in contact with its flat base. Data for the polymers are shown in Table 1. Quantitative measurement of carbon dioxide evolution was carried out by a calibration method previously described.14
L C. McNeill, S. M. T. Sadeghi
236
PAALaloratory _
_
°
. . . . . . . . . . . . .
.
. .
. .
.
.
.
. . . . . . .
.
45
°
~
75" i00
°
196
°
PAA Aldrich
,. ,~ •
200
300
\~.;;,
400
500
lemperature,°C Fig. 1.
TVA curves for laboratory and commercial PAA samples. Heating rate 10°C/min.
Subambient Thermal Volatilisation Analysis (SA TVA) Condensable volatile degradation products from polymer samples degraded to 500°C using the TVA technique, collected in a liquid nitrogen trap in the vacuum system, were separated by the SATVA method, 15 by allowing the trap to warm up from - 1 9 6 ° to ambient temperature in a controlled manner, with continuous removal of volatile products monitored by following the pressure change using a Pirani gauge. A SATVA trace for the TABLE 1 TVA Data for Commercial and Laboratory PAA Samples, Degraded Under Programmed Heating Conditions to 500°C
Polymer
Tonset Tmax(main Wt % (°C)
PAA Aldrich PAA laboratory
reaction) (°C)
residue
190
435
12.5
190
435
9.2
Wt % CRF
Wt % volatile.fraction
Ratio of volatiles/CRF
COz
H20
Other
28"9
26.5
29.2
2.9
2"0
38.2
25-7
24.5
2.4
1.4
Thermal stability and degradation of poly( acrvlic acid)
3
237
4
a.a ¢O
-1%°C
O°C
Fig. 2. Subambient TVA for warm-up from - 196°C to room temperature of condensable volatile products of degradation to 500°C under TVA conditions of the laboratory PAA sample. See Table 2 for corresponding product assignments: peaks 1 to 3 comprise the gases and volatile liquids and peaks 4 and 5 the less volatile liquid fraction. products from the laboratory P A A sample is shown in Fig. 2. Products corresponding to each S A T V A peak were collected separately for identification.
Product analysis The degradation products were identified by infrared spectroscopy, mass spectrometry and G C - M S techniques. In this investigation, minor products from degradation of P A A to 500°C, present as a liquid fraction in the SATVA separation, were examined by G C - M S . The degradation products are listed in Table 2. Cold ring fraction (CRF) products, which collect on the cooled upper part of the TVA tube or on an inserted cold finger, were r e m o v e d using a volatile solvent. These were examined by mass spectrometry; products identified are listed in Table 3. Partial degradation under T V A conditions The polymers were partially degraded in order to permit quantitative d e t e r m i n a t i o n o f c a r b o n dioxide and water at different stages o f decomposition. P r o g r a m m e d heating at 10°/min was used and samples were heated to temperatures corresponding to the end o f stages o f reaction as shown in the T V A curves for heating to 500°C, illustrated in Fig. 1. Corresponding data are shown in Table 4. It appears likely that evolution o f carbon dioxide increases gradually due to the decarboxylation process, as the degradation temperature increases.
238
L C. McNeill, S. M. T. Sadeghi TABLE 2
Products of PAA Degradation to 500°C in the TVA System under Vacuum Using Programmed Heating Polymer
Gases and volatile liquids Noncondensable at - 196°C
Condensable at - 196°C Major
Other liquids Major
Cold ring fraction
Residue
Minor a
Minor a
PAA laboratory
CH4 CO
CO2 acetone ketene ethylene propylene butene-1 methyl vinyl ketone benzene
H20
acrylic acid toluene xylene
short chain black fragments char such as dimer, trimer, etc., including anhydride rings
PAA Aldrich
CH4 CO
CO2 acetone ketene ethylene propylene butene- 1 benzene toluene methyl vinyl ketone
H20
acrylic acid
short chain black fragments char such as dimer, trimer, etc., including anhydride rings
a These products amounted to 3% of sample weight. Thermogravimetry
The T G curves obtained under dynamic nitrogen at 10°/min heating rate using a D u Pont 951 thermobalance, shown in Fig. 3, indicate two stages o f breakdown, starting at ca. 200 and 330°C, respectively. A n initial weight loss below 100°C in the case o f the commercial sample is due to the release of water absorbed by the polymer. The first stage o f breakdown (to 330°C) accounts for about 35 and 33% weight loss, respectively, in the commercial and laboratory P A A samples; the corresponding overall weight losses (to 500°C) are 85 and 86%. The T V A and T G results are in reasonable agreement. C h a n g e s in I R s p e c t r a d u r i n g d e g r a d a t i o n o f P A A
F o r this part o f the investigation, polymer films were m a d e by dissolving
Thermal stability and degradation of poly( acrylic acid)
239
TABLE 3
Products Identified in the Cold Ring Fraction from Degradation of Laboratory PAA under TVA Conditions to Temperatures Shown
Product ~
m/e
Unsaturated and saturated anhydride dimer Unsaturated and saturated acid dimer Unsaturated and saturated acid/anhydride trimer Unsaturated acid trimer Unsaturated and saturated anhydride tetramer Unsaturated and saturated tetramer with acid/anhydride/acid sequence Unsaturated acid tetramer and pentamer Unsaturated anhydride hexamer Unsaturated hexamer with acid/anhydride/acid/ anhydride sequence Unsaturated acid hexamer and heptamer Unsaturated anhydride octamer
Temperature (~1 CRF formation 375~C
400 or 500'C
254
present present present present present
present present present present present
270, 272 288, 360 378
present present present
present present present
396 432, 504 504
present absent absent
present present present
126, 144, !98, 216 252,
128 146 200
a For corresponding chemical structures, see 'Mechanism' section of text.
5 mg of PAA in A R methanol, casting a thin film on a Teflon sheet and drying in a vacuum oven at 40°C for 24 h. The polymer was partially degraded under vacuum at 10°/min heating rate to 170, 200, 210, 230, 260, 290, 350, 380 and 440°C, respectively. A 5 mg film was heated progressively to each of the above temperatures and the IR spectrum was recorded at each stage. The main bands of interest are those arising at about 1805, 1760 and 1040cm- 1 due to conversion of some of the carboxyl groups to anhydride rings, 4- 6 as shown in Fig. 4. There is no significant indication of anhydride formation at 170°C, but this structure increases gradually from 200 to 290°C. Anhydropoly(acrylic TABLE 4
Quantitative Data for Major Volatile Degradation Products at Different Extents of Degradation in the TVA System under Programmed Heating (10°/rain)
Stage (~]" degradation
PAA (Aldrich) T(°C)
1st 2nd 3rd Total Wt %
275 320 470
PAA (Laboratory)
Wt % C O 2 Wt ~YoH 2 0 5.5 8.5 12-5
! 2.8 3"8 12.6
26.5
29.2
T( C) 285 470 --
Wt % COz Wt % H20 12.9 !4.7 --
14.0 16-6
27.6
30.6
L C. McNeill, S. M. T. Sadeghi
240 1.0
-~\X\PAA
Laboratory
0.8!
4J tU L~-
0.6
4J
= 0,4
"o
0
Fig. 3.
I
i
100
200
i
1
300
400
500
Temperature, °C T G and D T G curves for laboratory ( - - - ) and commercial ( - - ) PAA samples. Dynamic nitrogen atmosphere, heating rate 10°C/rain.
,, ~ ,°
'~° o
undegrade~ 170°, 200 °C 210 °C 230 °C 260 °C 290 °C
~
."~.o ',," SY
....
° ° i "°~
° 0 O0
~f~'/O
for',#~,f//".'--
?,g
c\ / ",,.,"
380 °C 440 °C
f
1900
q
1800
i
I
1700
). < ,,.., I
1600
1200
I
I
I
1100
1000
900
Wavenumber, cm -I
Fig. 4.
Infrared spectra of undegraded commercial PAA and films degraded to various temperatures in the TVA system under programmed heating.
Thermal stability and degradation of poly( acrylic acid)
undegraded
r ~Q'e .~\.~ ~
2 ~i~
- -
10 rain
....
40
miP
©
o\--°:',J"l° ,. o\'i ~ o ,,_.,o
.....
=
2f/o°
. 2 o ~ . ~ . ' n'~ • . ~0 <>~ o \,~ .'1o
20 min 30 rain
241
=
°o o° 0
70 rain
90,
110 rain
' ' ' ' ~'~
, 1900
T 1800
i
i
1700
1600
Wavenumber,
Fig. 5.
56
I
I
1100
1000
900
cm - I
Infrared spectra of undegraded commercial PAA and films degraded for various periods at 210°C in the TVA system.
acid) is the main product obtained from heating at 250°C. Above this temperature, the increase in carbon dioxide evolution already noted means that the anhydride structures formed are beginning to decompose. At temperatures above 290°C, the IR absorptions due to anhydride decrease. In the same temperature region, ketone and hydrocarbon minor products can be detected and the film, which is still white at 290°C, becomes yellow at 350°C and brown at 440°C.
Isothermal experiments A 5mg film of the polymer was degraded isothermally at 210°C under vacuum in the TVA system for different periods. The IR spectrum was recorded in each case: the results are shown in Fig. 5. The most notable feature is again the appearance of strong anhydride absorptions. It is evident that the anhydride concentration reaches a maximum after 10min heating at this temperature. The rapid rise and subsequent slow decline in anhydride concentration as a function of time, at 210°C, as shown by the intensities of absorption at 1805 and 1040cm-1, respectively, are illustrated in Fig. 6. M E C H A N I S M OF D E G R A D A T I O N Three main processes may be discerned in the decomposition of poly(acrylic acid); namely, dehydration, decarboxylation and chain scission.
242
L C. McNeill, S. M. T. Sadeghi
100
g 4~ ~4 (:3 Ul
~"
~ o
cm-1 50
0
40
80
120
Time, min
Fig.6. Change in anhydride absorption intensities at 1805 and 1040cm-1 in commercial PAA film with time of heating at 210°C.
Dehydration This could occur by intra- or intermolecular reaction of carboxyl groups. The nature of the reaction has been discussed. 2A'5'1° It appears that the reaction temperature is of major importance, temperatures above 200°C being required for intermolecular reaction to occur, probably between C O O H groups left isolated after r a n d o m intramolecular dehydration involving adjacent m o n o m e r units has occurred at lower temperatures. The intramolecular reaction leads to the formation of six-membered glutaric anhydride type rings:
""CH2 # H 2 CH I COOH
\CH~ I COOH
""CH2 /CH2\ ,
CH-\ CH I i o//C~ojC%o
In this study, the TVA data indicate that water evolution commences at about 170°C and reaches a m a x i m u m rate a r o u n d 250°C, under programmed heating conditions.
Decarboxylation Decarboxylation is evident at temperatures above 200°C and has been found to become increasingly important above 250°C. Carbon dioxide
Thermal stability and degradation of poly( acrylic acid)
243
evolution due to anhydride decomposition can be envisaged as occurring as follows: / C \H 2
""CH2 CH2
J
CH I
CH-- --~--CH2--C.H I
\O /
%
/ CH \ z
CH-- ~ > --CH2--CH CH-I \C /
.cII O
,,
O
(1)
At high temperatures, the postulated intermediate species (I) can be regarded as the source of minor products of degradation amounting to about 3% of the sample weight, such as carbon monoxide, ketene, ketones and unsaturated compounds, in reactions in which it undergoes fragmentation. Chain scission
The formation of significant amounts of cold ring fraction, consisting of dimer, trimer, etc., at high temperatures is possibly due to release of fragments with short sequences of acrylic acid units, previously isolated between anhydride rings. Chain scission may occur at points adjacent to the anhydride rings, followed by further reaction of the macroradicals so formed: [CH2
"'q~H 2 CH 2
CH2 CH 2 CH 2 \ / \ / \
CH E
CH I
/
CH 2 \
CH I
CH-i
homolysis
S-o ,o J-o % n>l [CH2
..--,.CH 2 C H 2
CH 2
\CH C
og
C
/ .F\
\o" ~ o [ d
OH
•CH2 CH 2 CH2 CH2 \ / \ Y \ / \ CH CH CH C H - + I I I I
Ss.
n--I
(m)
(II) ~-CH 2 CH 2 \ /\.
CH CH I i
dC,o~C,,,o Ov)
or
J J,,o
~CH 2 CH 2 CH 2 \V\V\. CH CH CH
o ~ C~o ~C~o COOH (v)
+ oligomer or monomer
244
L C. McNeill, S. M. T. Sadeghi
The other fragments of the CRF shown in Table 3 may form in reactions in which macroradicals (III), (IV) and (V) undergo homolysis and hydrogen transfer. In the case of radicals (III) and (IV), the CRF products are purely anhydrides: v'-'-CH2 C H 2
"CH 2 C H 2 C H 2 CH 2 \ VNY \/ \
\CFI \CH I
CH CH
L
I
J rOv)%
I
j,o
I (III)
~
homolysis
~iCH 2 CH 2 \6 /
CH
I
C
I
C o ~ \ o " %o ,
C
CH
CH \. 2
CH
I
CH
I
I
C C C o ~ \ o " %o o ~ \ o " %o
\
CH2 CH2
at a
•C \H 2/ C H ~. 2/ C H 2\ /
\CH 2
I
I
,o o~c ",o~c ~o
[ intramoWcular ~ transfer
b
a
CH C H I ~
__ \io,
in,....,--r'
~
CH 2 CH2/CH 2
%C / \ C H 2 I I S\o/C%o
saturated .... annyorlae 'dimer'
.C.H2
~.
%C / \ C H \ -CH - / ' - ~CH - 2 saturated "~ I I J I o//C\ o/C%o o / / C \ o C % o 'tetramer' anhydride
saturated anhydride Mimer'
unsaturated anhydride 'tetramer'
Similar reactions of radical (V), however, lead to the CRF products containing both anhydride rings and carboxylic acid structures: ~ i C H\ 2 / C \H 2/ \C H 2
•C H H 2/ C ~H. 2 \ 2/ C \
CH CH (~H I
I
I
o//C\ o/C%o COOH (v)
scission )
CH
CH
CH
P
I
I
o~C\o/C%o COOH intramolecular/ transfer1
CH2
CH 2 CH 2
unsaturated acid/anhydride 'trimer'
ntermolecular bstraction
saturated anhydride/acid 'trimer'
Thermal stability and degradation of poly(acrylic acid)
245
CONCLUSIONS The thermal degradation of poly(acrylic acid) commences with a dehydration reaction occurring by intramolecular cyclisation of adjacent m o n o m e r units to give six-membered anhydride ring structures. Decarboxylation to give carbon dioxide as a product becomes important under programmed heating conditions at about 250°C and both water and carbon dioxide continue to be evolved on heating up to 500°C. A number of other volatile products, including monomer, are formed, but only in trace amounts. Under isothermal heating at 210°C, it is found that the anhydride concentration builds up rapidly during the first 10min of heating, but reaches a maximum after which there is a slow decline due to decarboxylation. Above 300°C, the elimination of water and the decarboxylation process seem consistent with the occurrence of both intra- and intermolecular reactions. The latter processes may involve acrylic acid units left in the chains between anhydride ring structures. Above 350°C, the polymer residue decomposes mainly to cold ring fraction products, volatile at degradation temperature under vacuum but not at room temperature. These consist of short chain fragments derived from two or more original repeat units; these may contain only anhydride rings or both rings and carboxylic acid structures. Small amounts of other volatile products, including carbon monoxide, are also formed at these higher temperatures.
ACKNOWLEDGEMENT We would like to thank the Ministry of Culture and Higher Education of the Islamic Republic of Iran and also the University of Tehran, College of Environmental Health, for the provision of a scholarship.
REFERENCES 1. Otsu, T. & Quach, L., J. Polym. Sci., Polym. Chem. Ed., 19 (1981) 2377. 2. Eisenberg, A., Yokoyama, T. & Sambalido, E., J. Polym. Sci., Part A1, 7 (1969) 1717. 3. Maurer, J. J., Eustace, D. J. & Ratcliffe, C. T., Macromolecules, 20 (1987) 196. 4. Roux, F. X., Audebert, R. & Qurvoron, C., Europ. Polym. J., 9 (1973) 815. 5. Grant, D. H. & Grassie, N., Polymer, 1 (1960) 125. 6. McGaugh, M. C. & Kottle, S., Polymer Letters, 5 (1967) 817.
246 7. 8. 9. 10. 11. 12. 13. 14. 15.
L C. McNeill, S. M. T. Sadeghi McGaugh, M. C. & Kottle, S., J. Polym. Sci., Part A1, 6 (1968) 1243. Nicholson, J. W. & Wilson, A. D., Br. Polym. J., 19 (1987) 67. Nicholson, J. W. & Wilson, A. D., Br. Polym. J., 19 (1987) 449. Nicholson, J. W., Wassan, E. A. & Wilson, A. D., Br. Polym. J., 20 (1988) 97. McNeill, I. C., J. Polym. Sci., Part A1, 4 (1966) 2479. McNeill~ I. C., Europ. Polym. J., 6 (1970) 373. Carraher, C. E., Jr & Piersma, J. D., J. Appl. Polym. Sci., 16 (1972) 1851. McNeill, I. C. & Rincon, A., Poly. Deg. and Stab., 24 (1989) 59. McNeill, I. C., Ackerman, L., Gupta, S. N., Zulfiqar, M. & Zulfiqar, S., J. Polym. Sci., Polym. Chem. Ed., 15 (1977) 2381.
Thermal Stability and Degradation Mechanisms of Poly(Acrylic Acid) and its Salts: Part 1 Poly(Acrylic Acid) I. C. McNeill & S. M. T. Sadeghi Polymer Research Group, Chemistry Department, University of Glasgow. Glasgow G12 8QQ, UK (Received 22 September 1989; accepted 2 October 1989) ABSTRACT The thermal degradation of poly( acrylic acid) has been studied using thermal volatilisation analysis ( T V A ) and thermogravimetry (TG). The degradation studies were carried out using two approaches. Programmed heating was carried out at lO°/min to temperatures from 170 to 500°C and the polymer was also heated isothermally at 210°C for different times, under TVA conditions. Quantitative measurements of the main product fractions have been made. The gaseous, volatile liquid and cold ring fraction ( CRF) products from the TVA degradation have been analysed by IR, M S and G ~ M S techniques. The polymer yields water and carbon dioxide as major products. The degraded polymer develops anhydride ring structures in the chain as a result of a dehydration process. At higher degradation temperatures, chain .fragments with anhydride rings and/or acid groups are formed, together with various minor volatile products. The mechanism of degradation is discussed.
INTRODUCTION The main features of the thermal degradation of poly(acrylic acid), PAA, have been known for many years, but certain aspects of the degradation products and mechanism at higher temperatures (above 300°C) still remain the subject of discussion. Otsu and Quach 1 examined the differences in the thermal degradation of head-to-head and head-to-tail PAA by using thermogravimetry. They showed that h - h PAA degrades at lower temperatures and in a different manner from h-t PAA, the degradation of which becomes 233 Polymer Degradation and Stability 0141-3910/90/$03"50~'~ 1990 Elsevier SciencePublishers Ltd, England. Printed in Great Britain
234
I.C. McNeill, S. M. T. Sadeghi
discernible at 150°C and proceeds in two steps. Eisenberg e t al., 2 who studied the dehydration kinetics and glass transition temperature of PAA have shown that anhydride formation in PAA is a first order reaction, as also is decarboxylation, the latter being much slower than the former. The thermal characterisation of PAA has been studied by Maurer et al., 3 who reported that atactic and syndiotactic PAA appear to differ significantly with regard to glass transition temperature and thermal stability. Roux e t al., 4 who studied the pyrolysis of PAA and poly(methacrylic acid), PMAA, in the range 160-240°C in a non-oxidative atmosphere, suggested that the kinetics of evolution of light products indicate a competition between dehydration and decarboxylation reactions. They believed water, carbon dioxide and monomer (PMAA only) were the products of pyrolysis of PAA and PMAA in that temperature range. No formation of carbon monoxide or acrylic acid monomer was observed. Grant and Grassie 5 showed that six-membered glutaric anhydride type rings result from intramolecular reaction of adjacent carboxyl groups in the decomposition of h-t P M A A at about 200°C. This dehydration reaction was found to be accompanied by a small amount of depolymerisation to monomer. Some investigations were extended to higher temperatures (above 250°C) in order to obtain further information on the degradation process. For example, McGaugh and Kottle 6 investigated the thermal degradation of PAA in air. They reported that in the range 25-150°C, the major reaction is the formation of anhydride groups, then at 150-275°C unsaturation begins to appear, as well as structures indicated from the IR absorption to be ester or cyclic ketone. They also noted that above 350°C, the unsaturation becomes very evident. The same workers v have also examined the thermal degradation of acrylic acid-ethylene copolymers, using TG and TVA. They reported that the degradation mechanism of the acrylic acid portions of the copolymer consisted of dehydration of the acid groups forming anhydride, followed by decarboxylation of the anhydride leading to unsaturation. Nicholson e t al. 8 - 1 ° have recently studied the thermal behaviour of partially neutralised PAA at 250°C. They showed that different chemical processes; namely, dehydration and decarboxylation, occur at this temperature, depending on the nature of the counterion. In the present work, the thermal behaviour of PAA has been investigated using TVA 11.12 and TG techniques to provide a basis for a series of studies of the degradation of PAA salts using the same experimental approach. In order to investigate the possibility of the formation of monomer and other minor products, a detailed product analysis for the less volatile liquid fraction has been carried out using the G C - M S technique.
Thermal stability and degradation of poly( acrylic acid)
235
EXPERIMENTAL
Materials Acrylic acid (Hopkin and Williams) was purified by distillation under reduced pressure (b. 60°C at 22 mm Hg) and the middle fraction was used immediately. ~-~'-Azodiisobutyronitrile, AIBN (BDH Chemicals), was recrystallised twice from AR grade methanol. A commercial poly(acrylic acid) sample (Aldrich), in powder form, with Mw 250 000, To 106°C, was used without purification, for comparison with the laboratory sample prepared as described below.
Polymerisation Polymerisation ~3 of freshly distilled acrylic acid was carried out in the presence of 0"1%w/v AIBN initiator. The product was purified by dissolution in water and isolation from the aqueous solution by freezedrying, followed by subsequent evacuation at about 40°C for 80 h. RESULTS A N D DISCUSSION
Programmed heating experiments Thermal volatilisation analysis TVA was carried out under vacuum as previously described. ~1'~2 Programmed heating was at a rate of 10°/min. The TVA curves for both polymers are shown in Fig. 1. It is clear from these curves that in each case decomposition begins at about 175°C, but appreciable rates of breakdown are observed only above 200°C. Both polymers show a three stage decomposition and from inspection of the separation of the traces, it is evident that product composition is different at each stage, the product volatility becoming greater as the degradation temperature rises. In particular, non-condensable gases are formed only above about 300°C, in which temperature region the presence of materials covering a wide range of volatilities is indicated. There is no major difference in behaviour between the commercial and laboratory samples on the basis of the TVA data. Gravimetric data were obtained under TVA conditions in experiments in which the polymer sample was placed in a thin glass, flat-bottomed sample tube, 25 mm base diameter, weighed before and after the experiment, placed inside the TVA tube and in contact with its flat base. Data for the polymers are shown in Table 1. Quantitative measurement of carbon dioxide evolution was carried out by a calibration method previously described.14
L C. McNeill, S. M. T. Sadeghi
236
PAALaloratory _
_
°
. . . . . . . . . . . . .
.
. .
. .
.
.
.
. . . . . . .
.
45
°
~
75" i00
°
196
°
PAA Aldrich
,. ,~ •
200
300
\~.;;,
400
500
lemperature,°C Fig. 1.
TVA curves for laboratory and commercial PAA samples. Heating rate 10°C/min.
Subambient Thermal Volatilisation Analysis (SA TVA) Condensable volatile degradation products from polymer samples degraded to 500°C using the TVA technique, collected in a liquid nitrogen trap in the vacuum system, were separated by the SATVA method, 15 by allowing the trap to warm up from - 1 9 6 ° to ambient temperature in a controlled manner, with continuous removal of volatile products monitored by following the pressure change using a Pirani gauge. A SATVA trace for the TABLE 1 TVA Data for Commercial and Laboratory PAA Samples, Degraded Under Programmed Heating Conditions to 500°C
Polymer
Tonset Tmax(main Wt % (°C)
PAA Aldrich PAA laboratory
reaction) (°C)
residue
190
435
12.5
190
435
9.2
Wt % CRF
Wt % volatile.fraction
Ratio of volatiles/CRF
COz
H20
Other
28"9
26.5
29.2
2.9
2"0
38.2
25-7
24.5
2.4
1.4
Thermal stability and degradation of poly( acrvlic acid)
3
237
4
a.a ¢O
-1%°C
O°C
Fig. 2. Subambient TVA for warm-up from - 196°C to room temperature of condensable volatile products of degradation to 500°C under TVA conditions of the laboratory PAA sample. See Table 2 for corresponding product assignments: peaks 1 to 3 comprise the gases and volatile liquids and peaks 4 and 5 the less volatile liquid fraction. products from the laboratory P A A sample is shown in Fig. 2. Products corresponding to each S A T V A peak were collected separately for identification.
Product analysis The degradation products were identified by infrared spectroscopy, mass spectrometry and G C - M S techniques. In this investigation, minor products from degradation of P A A to 500°C, present as a liquid fraction in the SATVA separation, were examined by G C - M S . The degradation products are listed in Table 2. Cold ring fraction (CRF) products, which collect on the cooled upper part of the TVA tube or on an inserted cold finger, were r e m o v e d using a volatile solvent. These were examined by mass spectrometry; products identified are listed in Table 3. Partial degradation under T V A conditions The polymers were partially degraded in order to permit quantitative d e t e r m i n a t i o n o f c a r b o n dioxide and water at different stages o f decomposition. P r o g r a m m e d heating at 10°/min was used and samples were heated to temperatures corresponding to the end o f stages o f reaction as shown in the T V A curves for heating to 500°C, illustrated in Fig. 1. Corresponding data are shown in Table 4. It appears likely that evolution o f carbon dioxide increases gradually due to the decarboxylation process, as the degradation temperature increases.
238
L C. McNeill, S. M. T. Sadeghi TABLE 2
Products of PAA Degradation to 500°C in the TVA System under Vacuum Using Programmed Heating Polymer
Gases and volatile liquids Noncondensable at - 196°C
Condensable at - 196°C Major
Other liquids Major
Cold ring fraction
Residue
Minor a
Minor a
PAA laboratory
CH4 CO
CO2 acetone ketene ethylene propylene butene-1 methyl vinyl ketone benzene
H20
acrylic acid toluene xylene
short chain black fragments char such as dimer, trimer, etc., including anhydride rings
PAA Aldrich
CH4 CO
CO2 acetone ketene ethylene propylene butene- 1 benzene toluene methyl vinyl ketone
H20
acrylic acid
short chain black fragments char such as dimer, trimer, etc., including anhydride rings
a These products amounted to 3% of sample weight. Thermogravimetry
The T G curves obtained under dynamic nitrogen at 10°/min heating rate using a D u Pont 951 thermobalance, shown in Fig. 3, indicate two stages o f breakdown, starting at ca. 200 and 330°C, respectively. A n initial weight loss below 100°C in the case o f the commercial sample is due to the release of water absorbed by the polymer. The first stage o f breakdown (to 330°C) accounts for about 35 and 33% weight loss, respectively, in the commercial and laboratory P A A samples; the corresponding overall weight losses (to 500°C) are 85 and 86%. The T V A and T G results are in reasonable agreement. C h a n g e s in I R s p e c t r a d u r i n g d e g r a d a t i o n o f P A A
F o r this part o f the investigation, polymer films were m a d e by dissolving
Thermal stability and degradation of poly( acrylic acid)
239
TABLE 3
Products Identified in the Cold Ring Fraction from Degradation of Laboratory PAA under TVA Conditions to Temperatures Shown
Product ~
m/e
Unsaturated and saturated anhydride dimer Unsaturated and saturated acid dimer Unsaturated and saturated acid/anhydride trimer Unsaturated acid trimer Unsaturated and saturated anhydride tetramer Unsaturated and saturated tetramer with acid/anhydride/acid sequence Unsaturated acid tetramer and pentamer Unsaturated anhydride hexamer Unsaturated hexamer with acid/anhydride/acid/ anhydride sequence Unsaturated acid hexamer and heptamer Unsaturated anhydride octamer
Temperature (~1 CRF formation 375~C
400 or 500'C
254
present present present present present
present present present present present
270, 272 288, 360 378
present present present
present present present
396 432, 504 504
present absent absent
present present present
126, 144, !98, 216 252,
128 146 200
a For corresponding chemical structures, see 'Mechanism' section of text.
5 mg of PAA in A R methanol, casting a thin film on a Teflon sheet and drying in a vacuum oven at 40°C for 24 h. The polymer was partially degraded under vacuum at 10°/min heating rate to 170, 200, 210, 230, 260, 290, 350, 380 and 440°C, respectively. A 5 mg film was heated progressively to each of the above temperatures and the IR spectrum was recorded at each stage. The main bands of interest are those arising at about 1805, 1760 and 1040cm- 1 due to conversion of some of the carboxyl groups to anhydride rings, 4- 6 as shown in Fig. 4. There is no significant indication of anhydride formation at 170°C, but this structure increases gradually from 200 to 290°C. Anhydropoly(acrylic TABLE 4
Quantitative Data for Major Volatile Degradation Products at Different Extents of Degradation in the TVA System under Programmed Heating (10°/rain)
Stage (~]" degradation
PAA (Aldrich) T(°C)
1st 2nd 3rd Total Wt %
275 320 470
PAA (Laboratory)
Wt % C O 2 Wt ~YoH 2 0 5.5 8.5 12-5
! 2.8 3"8 12.6
26.5
29.2
T( C) 285 470 --
Wt % COz Wt % H20 12.9 !4.7 --
14.0 16-6
27.6
30.6
L C. McNeill, S. M. T. Sadeghi
240 1.0
-~\X\PAA
Laboratory
0.8!
4J tU L~-
0.6
4J
= 0,4
"o
0
Fig. 3.
I
i
100
200
i
1
300
400
500
Temperature, °C T G and D T G curves for laboratory ( - - - ) and commercial ( - - ) PAA samples. Dynamic nitrogen atmosphere, heating rate 10°C/rain.
,, ~ ,°
'~° o
undegrade~ 170°, 200 °C 210 °C 230 °C 260 °C 290 °C
~
."~.o ',," SY
....
° ° i "°~
° 0 O0
~f~'/O
for',#~,f//".'--
?,g
c\ / ",,.,"
380 °C 440 °C
f
1900
q
1800
i
I
1700
). < ,,.., I
1600
1200
I
I
I
1100
1000
900
Wavenumber, cm -I
Fig. 4.
Infrared spectra of undegraded commercial PAA and films degraded to various temperatures in the TVA system under programmed heating.
Thermal stability and degradation of poly( acrylic acid)
undegraded
r ~Q'e .~\.~ ~
2 ~i~
- -
10 rain
....
40
miP
©
o\--°:',J"l° ,. o\'i ~ o ,,_.,o
.....
=
2f/o°
. 2 o ~ . ~ . ' n'~ • . ~0 <>~ o \,~ .'1o
20 min 30 rain
241
=
°o o° 0
70 rain
90,
110 rain
' ' ' ' ~'~
, 1900
T 1800
i
i
1700
1600
Wavenumber,
Fig. 5.
56
I
I
1100
1000
900
cm - I
Infrared spectra of undegraded commercial PAA and films degraded for various periods at 210°C in the TVA system.
acid) is the main product obtained from heating at 250°C. Above this temperature, the increase in carbon dioxide evolution already noted means that the anhydride structures formed are beginning to decompose. At temperatures above 290°C, the IR absorptions due to anhydride decrease. In the same temperature region, ketone and hydrocarbon minor products can be detected and the film, which is still white at 290°C, becomes yellow at 350°C and brown at 440°C.
Isothermal experiments A 5mg film of the polymer was degraded isothermally at 210°C under vacuum in the TVA system for different periods. The IR spectrum was recorded in each case: the results are shown in Fig. 5. The most notable feature is again the appearance of strong anhydride absorptions. It is evident that the anhydride concentration reaches a maximum after 10min heating at this temperature. The rapid rise and subsequent slow decline in anhydride concentration as a function of time, at 210°C, as shown by the intensities of absorption at 1805 and 1040cm-1, respectively, are illustrated in Fig. 6. M E C H A N I S M OF D E G R A D A T I O N Three main processes may be discerned in the decomposition of poly(acrylic acid); namely, dehydration, decarboxylation and chain scission.
242
L C. McNeill, S. M. T. Sadeghi
100
g 4~ ~4 (:3 Ul
~"
~ o
cm-1 50
0
40
80
120
Time, min
Fig.6. Change in anhydride absorption intensities at 1805 and 1040cm-1 in commercial PAA film with time of heating at 210°C.
Dehydration This could occur by intra- or intermolecular reaction of carboxyl groups. The nature of the reaction has been discussed. 2A'5'1° It appears that the reaction temperature is of major importance, temperatures above 200°C being required for intermolecular reaction to occur, probably between C O O H groups left isolated after r a n d o m intramolecular dehydration involving adjacent m o n o m e r units has occurred at lower temperatures. The intramolecular reaction leads to the formation of six-membered glutaric anhydride type rings:
""CH2 # H 2 CH I COOH
\CH~ I COOH
""CH2 /CH2\ ,
CH-\ CH I i o//C~ojC%o
In this study, the TVA data indicate that water evolution commences at about 170°C and reaches a m a x i m u m rate a r o u n d 250°C, under programmed heating conditions.
Decarboxylation Decarboxylation is evident at temperatures above 200°C and has been found to become increasingly important above 250°C. Carbon dioxide
Thermal stability and degradation of poly( acrylic acid)
243
evolution due to anhydride decomposition can be envisaged as occurring as follows: / C \H 2
""CH2 CH2
J
CH I
CH-- --~--CH2--C.H I
\O /
%
/ CH \ z
CH-- ~ > --CH2--CH CH-I \C /
.cII O
,,
O
(1)
At high temperatures, the postulated intermediate species (I) can be regarded as the source of minor products of degradation amounting to about 3% of the sample weight, such as carbon monoxide, ketene, ketones and unsaturated compounds, in reactions in which it undergoes fragmentation. Chain scission
The formation of significant amounts of cold ring fraction, consisting of dimer, trimer, etc., at high temperatures is possibly due to release of fragments with short sequences of acrylic acid units, previously isolated between anhydride rings. Chain scission may occur at points adjacent to the anhydride rings, followed by further reaction of the macroradicals so formed: [CH2
"'q~H 2 CH 2
CH2 CH 2 CH 2 \ / \ / \
CH E
CH I
/
CH 2 \
CH I
CH-i
homolysis
S-o ,o J-o % n>l [CH2
..--,.CH 2 C H 2
CH 2
\CH C
og
C
/ .F\
\o" ~ o [ d
OH
•CH2 CH 2 CH2 CH2 \ / \ Y \ / \ CH CH CH C H - + I I I I
Ss.
n--I
(m)
(II) ~-CH 2 CH 2 \ /\.
CH CH I i
dC,o~C,,,o Ov)
or
J J,,o
~CH 2 CH 2 CH 2 \V\V\. CH CH CH
o ~ C~o ~C~o COOH (v)
+ oligomer or monomer
244
L C. McNeill, S. M. T. Sadeghi
The other fragments of the CRF shown in Table 3 may form in reactions in which macroradicals (III), (IV) and (V) undergo homolysis and hydrogen transfer. In the case of radicals (III) and (IV), the CRF products are purely anhydrides: v'-'-CH2 C H 2
"CH 2 C H 2 C H 2 CH 2 \ VNY \/ \
\CFI \CH I
CH CH
L
I
J rOv)%
I
j,o
I (III)
~
homolysis
~iCH 2 CH 2 \6 /
CH
I
C
I
C o ~ \ o " %o ,
C
CH
CH \. 2
CH
I
CH
I
I
C C C o ~ \ o " %o o ~ \ o " %o
\
CH2 CH2
at a
•C \H 2/ C H ~. 2/ C H 2\ /
\CH 2
I
I
,o o~c ",o~c ~o
[ intramoWcular ~ transfer
b
a
CH C H I ~
__ \io,
in,....,--r'
~
CH 2 CH2/CH 2
%C / \ C H 2 I I S\o/C%o
saturated .... annyorlae 'dimer'
.C.H2
~.
%C / \ C H \ -CH - / ' - ~CH - 2 saturated "~ I I J I o//C\ o/C%o o / / C \ o C % o 'tetramer' anhydride
saturated anhydride Mimer'
unsaturated anhydride 'tetramer'
Similar reactions of radical (V), however, lead to the CRF products containing both anhydride rings and carboxylic acid structures: ~ i C H\ 2 / C \H 2/ \C H 2
•C H H 2/ C ~H. 2 \ 2/ C \
CH CH (~H I
I
I
o//C\ o/C%o COOH (v)
scission )
CH
CH
CH
P
I
I
o~C\o/C%o COOH intramolecular/ transfer1
CH2
CH 2 CH 2
unsaturated acid/anhydride 'trimer'
ntermolecular bstraction
saturated anhydride/acid 'trimer'
Thermal stability and degradation of poly(acrylic acid)
245
CONCLUSIONS The thermal degradation of poly(acrylic acid) commences with a dehydration reaction occurring by intramolecular cyclisation of adjacent m o n o m e r units to give six-membered anhydride ring structures. Decarboxylation to give carbon dioxide as a product becomes important under programmed heating conditions at about 250°C and both water and carbon dioxide continue to be evolved on heating up to 500°C. A number of other volatile products, including monomer, are formed, but only in trace amounts. Under isothermal heating at 210°C, it is found that the anhydride concentration builds up rapidly during the first 10min of heating, but reaches a maximum after which there is a slow decline due to decarboxylation. Above 300°C, the elimination of water and the decarboxylation process seem consistent with the occurrence of both intra- and intermolecular reactions. The latter processes may involve acrylic acid units left in the chains between anhydride ring structures. Above 350°C, the polymer residue decomposes mainly to cold ring fraction products, volatile at degradation temperature under vacuum but not at room temperature. These consist of short chain fragments derived from two or more original repeat units; these may contain only anhydride rings or both rings and carboxylic acid structures. Small amounts of other volatile products, including carbon monoxide, are also formed at these higher temperatures.
ACKNOWLEDGEMENT We would like to thank the Ministry of Culture and Higher Education of the Islamic Republic of Iran and also the University of Tehran, College of Environmental Health, for the provision of a scholarship.
REFERENCES 1. Otsu, T. & Quach, L., J. Polym. Sci., Polym. Chem. Ed., 19 (1981) 2377. 2. Eisenberg, A., Yokoyama, T. & Sambalido, E., J. Polym. Sci., Part A1, 7 (1969) 1717. 3. Maurer, J. J., Eustace, D. J. & Ratcliffe, C. T., Macromolecules, 20 (1987) 196. 4. Roux, F. X., Audebert, R. & Qurvoron, C., Europ. Polym. J., 9 (1973) 815. 5. Grant, D. H. & Grassie, N., Polymer, 1 (1960) 125. 6. McGaugh, M. C. & Kottle, S., Polymer Letters, 5 (1967) 817.
246 7. 8. 9. 10. 11. 12. 13. 14. 15.
L C. McNeill, S. M. T. Sadeghi McGaugh, M. C. & Kottle, S., J. Polym. Sci., Part A1, 6 (1968) 1243. Nicholson, J. W. & Wilson, A. D., Br. Polym. J., 19 (1987) 67. Nicholson, J. W. & Wilson, A. D., Br. Polym. J., 19 (1987) 449. Nicholson, J. W., Wassan, E. A. & Wilson, A. D., Br. Polym. J., 20 (1988) 97. McNeill, I. C., J. Polym. Sci., Part A1, 4 (1966) 2479. McNeill~ I. C., Europ. Polym. J., 6 (1970) 373. Carraher, C. E., Jr & Piersma, J. D., J. Appl. Polym. Sci., 16 (1972) 1851. McNeill, I. C. & Rincon, A., Poly. Deg. and Stab., 24 (1989) 59. McNeill, I. C., Ackerman, L., Gupta, S. N., Zulfiqar, M. & Zulfiqar, S., J. Polym. Sci., Polym. Chem. Ed., 15 (1977) 2381.