Polymer Degradation and Stability 15 (1986) 173-182
Correlation of Physico-chemical, Mechanical and Electrical Properties of Ultraviolet-degraded
Poly(Ethylene Terephthalate)
N. Ili~kovi6
Faculty of Technology, University of Banja Luka, Yugoslavia
&
M. Bravar
Faculty of Technology, University of Zagreb, Yugoslavia
(Received: 28 February, 1986)
ABSTRACT
Poly(ethylene terephthalate) (PET). f i lms have been exposed to ultraviolet radiation. Changes in physico-chemical, mechanical and eleetrical properties which occur in the films were examined. It was found that the specific viscosity, tensile strength, elongation at break, relative dielectric constant and dielectric strength decrease, while the dielectric loss factor, density, optical density of the ir spectrum band which refers to the transform of the PET molecule, as well as the quantity of carboxylie end groups, increase.
Least squares analysis and function straightening methods were used for the interpretation of the experimental data. Correlating relationships between measured parameters were established using the same methods.
Special attention was paid to the changes in the electrical properties of poly(ethylene terephthalate), because no data are available so far. Mathematical correlations with chemical and mechanical properties were established.
INTRODUCTION
The commercial applications of poly(ethylene terephthalate) (PET) have been continuously increasing. Its excellent electrical properties enable it
173 Polymer Degradation and Stability 0141-3910/86/$03.50 ~ Elsevier Applied Science Publishers Ltd, England, 1986. Printed in Great Britain
174 N. Iligkovi~, M. Bravar
to be applied widely as an insulating and dielectric material. It is, therefore, very important to know the influence of uv light on the stability of these properties, as well as on its mechanical stability and durability. These properties of the polymer depend, for example, on its chemical structure, molecular mass, conformation and crystaUinity, which change as a result of degradation during its use.
By analyses of the quality changes which occur in polymeric material exposed to artificial ageing, the course of degradation, and thus of usability, may be predicted. The purpose of this paper was to establish the common basis of the changes in different properties which occur in photo-degraded PET and to express them by mathematical functions. This could obviate the long-term experimental measurement of the various properties studied in this work. Furthermore, these mathemati- cally expressed functions could be correlated, so that experimental results could be substituted by calculated data. The present tendency is to use experiments which do not destroy the sample. The method described in this paper could be used in this way.
EXPERIMENTAL
Poly(ethylene terephthalate) film (15 pm thick) without additives was used.
The ultraviolet source was a high pressure mercury lamp (TQ 150 Z3 Original Hanau) inserted in a rotating device to ensure uniform irradiation of the films.
PET samples were irradiated for various periods up to 100 h. The time interval was chosen using the fact that the reduction of the percentage stretching could be used as a measure of the degree of polymer degradation. It was suggested 1 that an 80 % reduction of stretch could be considered to be the degradation level which represents mechanical failure of polymeric material. Thus, samples irradiated for 100 h could be considered to have failed, because their stretch was reduced by 90 relative to unirradiated film.
After some relaxation time, the following properties of the irradiated samples were measured. Chemical and physico-chemical--viscosity, quantity of carboxylic end groups, density, conformational changes. Mechanical--tensile strength, elongation at break. Electrical---dielectric loss factor, relative dielectric constant, dielectric strength.
Correlation of uv-degraded PET properties 175
Viscosity measurements (r/~p) were made in dichloracetic acid solution using an Ubbelohde viscometer. Carboxylic end groups (Kcoon) were determined by titration using a pH meter (Titriscop E-516, Methrom Herisau). The sample was dissolved in an o-cresol--chloroform mixture (7: 3) and titrated with 0.1 ~ KOH. Density (p) measurements were made by Juilfs' hanging method. 2 The liquid phase in the column was a CC14-n- heptane mixture (79:21). Conformational property changes (VoD) were observed indirectly by ir spectroscopy using a Perkin-Elmer M-377 instrument.
Fig. 1.
o-A L I so
~ - ~
I00 140
130~
120 ~"
<] 50 ~ % ~ LO 110
I00
9O I I I I I
0 20 t,O 60 80 100
Changes in the mechanical properties of PET with time of uv irradiation.
Mechanical properties (a and AL) were measured using an Instron device, Model TM-M.
Dielectric measurements (e, and tg6) were made using a Hewlett- Packard type 260A Q-meter with a Marconi Instruments TJ 155 dielectric loss test jig. Measurements were made at 500 kHz frequency. Dielectric strength measurements (Es) were made using a UPU-1 M, USSR, high voltage power supply, a Hewlett-Packard type 3435 A digital voltmeter and a Hewlett-Packard type 3411 A high voltage probe.
Experimental data are presented in Figs 1 to 3.
176 N. Iligkovi6, M. Bravar
Fig. 2.
~ 8 -
£-~ 3,7" d \
3,6
d ~. 3,5
o ~> 3~ -
3,3-
o - q.sp • - Vo.o.
A- KCOOH ~
"-P ~ 0,2- • 0~6 ~.
0 • 0
"~ 0,1S
'1"
0s ~ 5~ 0,1
0~) 5
0 20 40 60 80 100 ~ [h ]
E
O~
1608
1407
1406
1405
1406
Changes in the physico-chemical properties of PET with time of uv irradiation.
Fig. 3.
% % O)
\ o ,2
300 E
>
5 ~ ~ ~ , ~ , ~ 1,1 250 "-' tn
LU
ZOO
I I I • I
0 20 40 6 0 80 100 ~[h]
Changes in the electrical properties of PET with time of uv ~irradiation.
Correlation of uv-degraded PET properties 177
RESULTS AND DISCUSSION
The main chromophore in poly(ethylene terephthalate) in the range 240-320 nm is the structure:
- - O ~ O - ~ C 6 H 4 - ~ C O - - O -
strongest absorption occurring near 31.0 nm. 3 Absorption of uv radiation causes a loss of mechanical strength, an increase in the quantity of carboxylic end groups and the formation of radicals and volatile products. Photo-oxidation causes a reduction in molecular mass, which is one of the main reasons for the deterioration in mechanical properties. Day and Wiles 1'6 -s and Wiles 4,5 have suggested the following reaction sequence, which could explain the principal volatile products, CO and CO2, as well as the carboxylic end groups:
-~PhCOOH + "CH2CH2,~
--PhCOO" + "CH2CH2--
- ~ P h " + C O 2 --F 'CH2CH2--
,__PhCOOCH2CH2,__h v a ~ ---Ph" + ' C O O C H 2 C H 2 - -
,--Ph" + CO + "OCH2CH2---
/ ---PhCO +" OCH2CH2---
These reactions also explain the molecular chain scission which is reflected in the reduction of mechanical stability (Fig. 1). Reduction in chain length is also demonstrated by viscosity changes (Fig. 2).
Rotation around the ethylene glycol residue:
~ H 2 ~ j CH2--
makes possible the existence of two rotational isomers--trans and gauche. The trans form allows more dense molecular packing, so PET can
178 N. Iligkovik, M. Bravar
exist in a partly crystalline form. Both forms of ethylene glycol residue are present in the amorphous part.9 - 13 High temperature or other forms of energy can cause a transition from the gauche to the trans form, resulting in an increase of the crystalline content. Ultraviolet radiation energy can act in this way.
Conformational changes can be observed by ir spectroscopy, because each form has its own band. According to the literature, 14 the 973 cm-1 band is a measure of the quantity of the trans form, wlajle the 632 cm- band may be used as a reference band, because it is common to both isomers. Optical density changes are shown in Fig. 2. Conformational changes also reflect on the polymer density (Fig. 2), as well as on some electrical properties of PET (Fig. 3).
PET molecules are planar and centrosymmetric in ordered regions (trans form). Increase in symmetry decreases the possibility of polarization and thus the relative dielectric constant.~ 5 Changes in the loss tangent, as well as the dielectric strength, depend upon reduction of molecular mass. A more detailed explanation of the changes in the electrical properties of PET under uv-light was given in the previous paper.16
It is important to notice that there is no linear relationship between irradiation time and the property changes which occur in poly(ethylene terephtalate) exposed to ultraviolet light. The Figures show that the longer the irradiation time, the lower the value of the specific viscosity, tensile strength, elongation at break, relative dielectric constant and dielectric strength while the loss tangent, optical density of the characteristic bands and quantity of carboxylic end groups increases. This effect is greater at the beginning of irradiation. A 'skin' effect ~ 7,18 could explain this phenomenon. As the photo-degradation of PET takes place preferentially on the polymer surface, irradiation results in the formation of a thin layer of photo-oxidised polymer, which then acts as a protecting barrier to light absorption.
TREATMENT OF EXPERIMENTAL DATA
Least squares analysis and function straightening methods were used for the interpretation of the experimental data. Tables 1 and 2 present results obtained in this way. The polynomials in Table 1 should give a relative error of less than 10 ~o over the whole irradiation time from 0 to ! 00 h. The changes are presented in Table 2 in a more condensed form. Some
Correlation of uv-degraded PET properties 179
TABLE 1 E q u a t i o n s for C h a n g e s in the P rope r t i e s o f P E T as a F u n c t i o n o f I r r ad i a t i on T ime ,
O b t a i n e d by the Leas t Squa res Analys i s M e t h o d
t r = 1 6 0 . 3 8 - 7 . 7 3 7 3 x 10-12 + 1"4685 x 10-31. 2
AL = 115'72 - 1'481 61. + 4 '362 0 x 10- 322
r/sp = 6"5644 x 10 -1 - 3-8783 x 10-31. + 3"411 7 × 10-TZ 2 + 1"9978 x 10-7"c 3
K c o o n = 5 .5924 × 1 0 - 2 + 3 .2729 x 1 0 - 3 2 - 3 .2269 x 10-51.2+ 1.3832 x 10-7T 3
Voo = 3.3219 + 1.238 x 1 0 - 2 z - 7.11 x 10-51. 2
p = 1 4 0 4 - 1 9 5 8 + 6 . 3 3 4 9 x 10-21. - 1.7832 × 10-41. 2
e r = 1 " 2 4 0 5 - 2 " 0 5 4 x 10 31.+4"3 x 10-622
tg6 = 4"267 x 1 0 - 3 + 6 " 0 5 4 x 1 0 - 5 1 . - 4 " 0 1 9 x 10-77:2
E s = 3 1 1 " 1 4 0 - 6"058 lz +9"495 x 10-21. 2 - 5"0395 x 10-4z 3
deviations from the established equation for several functions occur at the beginning or at the end of the irradiation interval as shown in Table 2.
The dimensions of the measured properties are presented in Table 3. In this work radiation time is the common independent variable, which can connect the empirical functions. The values of the independent variable in the interval where mutually correlated functions are defined should be used. For irradiation times from 0 to 100h defining intervals [A, B] are presented in Table 3.
TABLE 2 E q u a t i o n s for C h a n g e s in the P rope r t i e s o f P E T as a F u n c t i o n o f I r r ad i a t i on Time, O b t a i n e d by the F u n c t i o n S t ra igh ten ing M e t h o d
Function Defining interval 1.[hi
tr = 158.5 × 0.995' [0, 100]
AL = 125.9 × 0-982' [0, 70]
~Lp = 0.741 3 × z - ° 1 [2, 100]
KcooH = 0"05 x z °'28 [2, 100]
Voo = 3 + 0 . 2 6 3 x z °'259 [2, 100]
p = 1 4 0 0 + 4 . 2 1 x 1.0093' [0, 70]
er = 1 + 0 . 2 5 × 0.988 ~ [0, 100]
tg6=4 x 10-3 t "°"1 [2, 100]
E s = 367.3 x 1.-o.175 [2, 100l
180 N. Ili%koviO, M. Bravar
TABLE 3 Defining Intervals of the Measured Properties
Measured Dimension Interval [A, B] Dependence o f property property
A B change on z
r/$p 1 0.457 0.'691
Mol C O O H Kc°°n kg 0.050 0.198 T
VOD OD973 3'3 3"867 I" OD632
kg 1 404.1 1 408.8 T P m-~
M N 98 159 tr m2
AL % 12 124
1 1 "081 1.239 ~,
tg6 1 4"04 x 10 -3 6'39 x 10 -3 T
kV E s 155 321.74
mm
Mutal dependences of the measured values were obtained in two ways; namely, by least squares analysis and by the elimination of the radiation time from the functions obtained by the function straightening method.
Thirty-six correlating functions are possible in each case, but correlations should be established between those parameters which can be easily measured and which cannot be changed independently of radiation time. The data from a rapid and simple measuring procedure should be used to correlate with parameters whose measurement is long term and complicated.
The measurements of dielectric strength, tensile strength, elongation at break and optical density are among the simplest and most rapid of the methods used. These simple measurements could be of use for correlation calculations, because it is known that there is a close relationship between mechanical and electrical polymer property changes 19 and because molecular motions which occur in the liquid or solid state are often
Correlation of uv-degraded PET properties 181
TABLE 4 Correlating Equations for the Parameters of Irradiated PET, Obtained by the Least
Squares Analysis
K c o o . = 0 "9554 - 2.3380,e + 1 "49 lq2~
r/$p = 1 " 1 2 6 7 - 1 ' 2766 x 1 0 - 2 o - + 6"1605 x 1 0 - 5 o .2
r/$p = 0 . 1 2 6 1 + 2 . 5 2 9 x 1 0 - 3 E s - 2 . 5 0 3 3 x 1 0 - 6 E 2
A L = - 283 + 2 . 6 E s - 4 .22 x I 0 - 3 E 2
p = 1 520.11 - 71"45VoD + 11'02Vo2D
e, = - - 4 ' 4 0 1 + 3 '388VoD -- 0 '509V2D
tg6= 10-3(2"303 + 37 '709 K c o o . - - 84"131 K2.oon)
studied by dielectric methods. 2° The measurement of the quantity of carboxylic end groups and of specific viscosity is carried out by standard methods easily performed in every laboratory, so they may also be suggested for use in correlating functions. On this basis several correlating relationships were established between certain parameters. The seven correlating functions obtained by least squares analysis are presented in Table 4, while the function straightening method was used to obtain the correlating functions shown in Table 5.
Parameter changes, as well as correlations stated in this mathematical manner, enable qualitative and quantitative mathematical analysis of the changes in poly(ethylene terephthalate) under the influence of uv-light. These established functions are suitable for the analysis and long-term estimations of the properties of the polymer under natural conditions.
TABLE 5 Correlating Equations for the P a r a m e t e r s o f Irradiated P E T , O b t a i n e d by the Function
Straightening Method
Kcoou q~
A L
P
8r
tg6
= 2 . 2 x 10 - 2 x r/~ 28
=10.401 6(2.2 - log o-) o. 1
= 2 5 " 4 x 10 3E° '57
= 125-9 x 0.982v; p = ( 3 6 7 ' 3 / E s ) 57
= 1 400 + 4"21 x 1"009 3q; q = 173 '6 (Vo0 - 3) T M
= 1 + 0.25 x 0.988 q
I~ 0.357 = 11"66 × 1 0 - 3 × " ' cooH
182 N. Ili~kovi6, M. Bravar
C O N C L U S I O N
Ultraviolet irradiation causes changes in the physico-chemical, mechani- cal and electrical properties of poly(ethylene terephthalate). The rate of change decreases as the irradiation time increases.
Empirical equat ions with a relative error of less than 10 % have been obtained which relate measured properties to irradiation time.
Changes in properties in a given sample can be successfully correlated. Mathematical correlations between specific viscosity, quant i ty of carboxylic end groups, dielectric loss factor, dielectric strength, tensile strength and elongation at break, as well as correlations between density, optical density and relative dielectric constant , have been established.
These relationships may be used for the analysis and estimation of changes in properties of polymeric material under natural conditions.
R E F E R E N C E S
1. M. Day and D. M. Wiles, J. Appl. Polym. Sci., 16, 191 (1972). 2. J. Juilfs, Forschungsberichte des Landes Nordhein-Westfalen, No. 381,
Dichte Bestimmung von Fasern, West Deutscher Verlag, K61n und Opladen (1957).
3. F.B. Marcotte, D. Campbell and J. A. Cleaveland, J. Polym. Sci., Part A-l, 5, 481 (1967).
4. D. M. Wiles, Polym. Eng. Sci., 13, 74 (1973). 5. D. M. Wiles, J. Appl. Polym. Sci., Appl. Polym. Symp., 35, 235 (1976). 6. M. Day and D. M. Wiles, Polym. Letters, 9, 665 (1971). 7. M. Day and D. M. Wiles, J. Appl. Polym. Sci., 16, 203 (1972). 8. M. Day and D. M. Wiles, J. Appl. Polym. Sci., 16, 175 (1972). 9. P. G. Schmidt and F. P. Gay, Angew. Chem., 74, 638 (1962).
10. W. W. Daniels and R. E. Kitson, J. Polym. Sci., 33, 161 (1958). 11. D. Grime and I. M. Ward, Trans. Faraday Soc., 54, 959 (1958). 12. N.A. Slovokhotova, G. K. Sadovskaya and V. A. Kargin, J. Polym. Sci., 58,
1293 (1962). 13. F. Sch6nherr, Foserforsch und Textiltechn., 21,246 (1970). 14. A. Garton, D. J. Carlsson, L. L. Holmes and D. M. Wiles, J. Appl. Polym.
Sci., 25, 1505 (1980). 15. B. S. Nady, Dielektrometrija, Energija, Moskva, 17-73 (1976). 16. N. Ili~kovi6 and M. Bravar, Poly. Deg. and Stab., 13, 139 (1985). 17. G. R. Merrill and C. W. Roberts, J. Appl. Polym. Sci., 21, 2745 (1977). 18. J. A. Dellinger and C. W. Roberts, J. Appl. Polym. Sci., 26, 321 (1981). 19. V. K. Huff and F. H. Miiller, Kolloid-Z., 153, 5 (1957). 20. B. E. Read and G. Williams, Trans. Faraday Soc., 57, 1979 (1961).
