State of dibutyl phthalate molecules in plasticized poly(vinyl chloride)

0932-3950/78/9601-1492507.50]0.

Polymer Science U.S.S.R. Vol. 20, pp. 1492-1498. (~) Pergamon Press Ltd. 1979. Printed in Poland.

STATE OF DIBUTYL PHTHALATE MOLECULES IN PLASTICIZED POLY(VINYL CHLORIDE)* A. ~. ~AkKLAKOV, A. A. I~¢[AKLAKOV,A. N . TEMlqIKOV a n d B. F . TEPLOV V. I. Lenin State University, Kazan V. A. Kargin Polymer Chemistry a n d Technology Research I n s t i t u t e J

(Received 22 August 1977) I t was found t h a t the rotational motion of plasticizer molecules in the polymer matrix is hindered, the degree of hindrance differing for different molecules. With " use of the data on the dibutyl phthalate (DBP) autodiffusion it" was possible to evaluate the lifetime of the PVC-plasticizer solvates. This lifetime is a function of the a m o u n t of small molecules and of temperature. The solvatation in these systems is of the dynamic type. The "free" plasticizer is absent in samples containing 30-52 wt. °/o DBP. The measurements were made b y the pulse NMR method.

SEVERAL investigations have been reported in connection with the state o f plasticizer molecules introduced into polymer [1]. However, the problem has y e t to be fully resolved. In some cases use of the N M R method allows separate evaluation of the mobility of macromolecules and molecules of a low molecular substance and it appears that the use of N M R m a y provide additional information on t h e mode of behaviour of molecules of components forming part of plasticized systems. Our aim in the present work was to shed light on the phthalate

state

of d i b u t y I

(DBP) molecules in PVC, using a pulse NMR method.

Samples were prepared b y press moulding at 180 ° from C-70 grade PVC with ~ 7~o erystallinity determined by the X-ray method, and D B P of chemically pure grade. The weight concentrations of plasticizer in the samples co1~30, 42 and 52~o. Measurement of the nuclear transverse magnetic relaxation times T~ was based either on the free nuclear induction decay, when T~<500/~sec, or on the Carr-Purcell-MayboomHfll method [2] with 250/lsec spacings between u-pulses. Spin-echo or free induction decay A (t) envelopes of the samples are complex in character, a n d were approximated over a wide range of T as the sum of two curves A (t)=A0~ exp (--t/T~)+Aobfb (t),

(1)

where T2a is the longest transverse relaxation time derived from the A (t) curve, fn(t) is a decreasing function normally differing the exponent [3]. The decay rate off,(t) was characterized as the effective transverse relaxation time T2b, in which the latter function was reduced by a factor of e. Reasons for subdividing the A (t) curve into two sections are giverl below. The n u m b e r of protons having time T2a was defined as Pa=Aoa/Aoa-~Aob. Coefficients of autodiffusion D of ~he phthalate in the samples were as previously determined in [4]. Measurements of T2,.~ were made on a laboratory pulse NMR relaxometer a t 22 Mc/s on protons in the temperature range - - 5 0 - + 1 2 0 °. * Vysokomol. soyed. A20: :No. 6, 1325-1330, 1978. 1492

State of DBP molecules in plasticized 1)VC

1493

I n the t e m p e r a t u r e i n t e r v a l e x a m i n e d the A (t) curve for p u r e D B P is described b y the first t e r m in e q u a t i o n (1), a n d for PVC, b y the second t e r m , while t h e d e p e n d e n c e for t h e plasticized systems is described b y b o t h terms. T~ v a l u e s

log T2~,b

1o

3~

&/°r

O l' ~ o o a ~

3

2

0.8o.6

l -

i

3"0

I

3.5 :FIG. 1

~

0.0

]

1

0"5

30

i

35

a,

[~,,,

z/.O

-

m~/T,°J~

1o3/r, oK

:FIG. 2

FIG. 1. Temperature dependences of T2~ (2, 3) and T2b (2", 3') for PVC-DBP systems containing 52 (2, 2') and 30 wt. % DBP (3, 3'). Curves 1 and 4 are analogous dependences of transverse relaxation times for pure DBP and PVC respectively; 2", 3"--calculations based on equation (6) for samples containing 52 (2") m~d 30 wt. °/o DBP (3"). The vertical lines indicate temperatures T*, starting at which P2~,=PT. Fro. 2. Temperature dependence of Pa/PT for plasticized PVC samples containing 52 (1), 42 (2) and 30 wt. % DBP (3). The solid curves were plotted on the basis of calculations in line with equation (4). for the pure D B P are the highest of all, those of the pure PVC are the lowest, while an i n t e r m e d i a t e position is occupied b y T2~ a n d T,.s values for the plasticized systems. An increase in the PVC c o n t e n t in the samples leads to a shorteni~g of b o t h times, T2a/T2b being n o r m a l l y ~ 10. N o n l i n e a r i t y of log T2a,b=f(1 IT) plots appears over a wide interval of T (Fig. 1). I n Fig. 2 we h a v e temperate, re dependences of Pa/PT, where PT is the n u m b e r of p r o t o n s in plasticizer molecules calculated from the composition of the systems. W e i n t r o d u c e d a characteristic t e m p e r a t u r e T*, the l a t t e r being the point after w h i c h Pa/PT=I. On lowering T ~ T * , the value of P a / P T ( I . T h e mode of r e d u c t i o n in the latter depends on ~he D B P content: a s m o o t h e r r e d u c t i o n appears for samples h a v i n g the largest (ol. Assignment of times T2a,b. The equality fla=PT in the case of T >i T * allows u s

1494

A. I. ~ A ~ O V

et al.

¢o assume that protons characterized b y time T2a pertain to plasticizer molecules, in view of which A (t) in equation (1) m a y be represented in the form of two terms, the first characterizing the behaviour of D B P protons, and the second Chat of the PVC protons, At T ~ T * the pattern is one of greater complexity, and PamPa', i.e. a fraction of the protons of plasticizer molecules are making a contribution to the shorter time T2b, and the simple assignment of times presented above is no longer valid. However, the very fact that P a ~ P T at T < : T * opens the w a y to a number of possible approaches for determination of the state of D B P molecules incorporated in PVC. Degree of hindrance of D P B molec~des incorporated in PVC. In statistical physics one normally characterizes molecular mobility b y the correlation time zc, which to a rough approximation is the time during which a molecule turns through an angle of ~ 1 rad (for rotational motion) or is displaced to an extent (distance) comparable with linear molecular dimensions (for translational motion). 'Given T 2 > 1 0 0 / l sec, which is the case for T2a for all the samples at T > T * , the relationship between T2 and rc is a fairly simple one: T2~T*, irrespective of the t y p e of motion. It follows from these considerations that values of T~a m a y be used to estimate the mobility of plasticizer molecules. Actually, the relative p l a c e m e n t of curves 1-3 in Fig. 1 at T > T * shows that the motion of D B R molecules in the polymer matrix is slower t h a n that of the pure phthalate molecules, the degree of hindrance increasing with increasing PVC content in the ,samples. It is known [5] that a liquid whose molecules have lost their individual kinetic energy (i.e. mobility) is a bound liquid. This means that the value of vc for plasticizer molecules does to some extent reflect the degree of linking of D B P molecules to the polymer matrix, and the greater the extent to which zc for D B P molecules inserted in PVC differs from re for the pure liquid, the more m a r k e d is the degree of hindrance of the plasticizer molecules in the polymeric s y s t e m . This in turn leads one to assume that the T2a/T2o ratio (here T20 is the transverse relaxation time for pure DBP) m a y be used as a quantitative index of the degree of attachment of plasticizer molecules to macromolecules [6]. Thus, a value of T2a/T2o [1] points to marked inhibition of the small molecule on incorporating the latter in polymer. Figure 1 shows t h a t the degree of hindrance of D B P molecules is higher in samples with low values of o~1; as wi rises, the mobility ,of the D B P molecules tends towards that of pure phthalate molecules. Correlation time spectrum of DBP molecules at T < T * . On examining the temperature dependences of Pa/P~' it can be said that different D B P molecules incorporated in PVC are hindered in varying degree, and form an integral correlation time spectrum. Molecules t h a t are more strongly linked to the polymer carcass will be hindered at higher T values, and vice versa. On lowering T < T * t h e mobility of D B P molecules stably linked to the PVC m a y prove to be close to that of macromolecules, and their protons will make a contribution to T~b. L e t us now substantiate the foregoing assumptions on a quantitative basis. :Let us say that there is a correlation time spectrum, and that it is fixed b y

State of D B P molecules in plasticized PVC

1495

distribution function I(~c). Now

Pa/PT= S I(vc)dvc,

(2)

o

where r~ is the maximum value of the correlation time for protons making a contribution to Tea. As a rough approximation one m a y say z~~10 -a sec [7] if Pa/PT differs only slightly from unity, though it must be that it is a little less. Let us say t h a t /(re) is the logarithmically normal distribution [7] and that

I (vc)dvc--B(zc)-* exp (--B~Z2)dZ.

(3)

Here Z = l n (re/re0), re, being the most probable correlation time, and B is a parameter characterizing the distribution width of To. Now, substituting (3) into (2) we obtain ~z~ Pa/P~,-=(n)-~' ~ exp(--X~)dX, (4) --00

where Z g : l n (z~/VCo). To obtain the temperature dependence of Pa/P~, one has to fix the shape of Zco(T). In the narrow interval of T in which Pa is affected, it was assumed t h a t ,e°~exp (E/RT). E was obtained from the log T~a=f(1/T) plot which, in the region of T where Pa changes, m a y be considered a linear dependence. Values of -Pa/PT calculated b y equation (4) are shown b y the solid curves in Fig. 2. B values corresponding to the latter curves are of the order of 0.25-0.30. Attempts to use other known I (v,) functions failed to tally satisfactorily with the experiment. Since the correlation time spectrum is sufficiently wide (low B values) it is reasonable to assert that different D B P molecules in samples do indeed differ as to the degree of hindrance of rotational motion in the polymer matrix, which means that there cannot be said to be any stoichiometric complexes in the P V C D B P system. Plasticizer molecules are randomly distributed in the PVC matrix, and are to a greater or lesser extent linked to macromolecules. Varying degrees of mobility of molecules incorporated in polymer have been substantiated b y findings of other investigators [8, 9]. The next problem to be considered is, what motions underlie and determine I(v,)? Motions determining T, at T
T~-2= T-I' ~ r + T 2-I t,

(5)

where T~r and T2~ are transverse relaxation times relating to the rotational a n d translational motions for D B P moleeule~. Let us now evaluate Tzt in the region where T>T*, in which Pa=P~' and where the coefficients of autodiffusion of D B P molecules were measured in [4]. According to relation (14), appearing in [10], we have in the case of intermittent

1496 movements

A.I.

MAx~tov

et a/.

of the molecules 27C ~'4h2N

-I

T2t - -

•5

(6)

- g2(~, co ~'/), d2~

w h e r e ? is the gyromagnetic proton ratio; h, Planck's constant; N, the number of

protons, cm3; w, the angular resonance frequency; d, the distance of the nearest approach of protons of neighbouring molecules; g~(~, co,t), a complex analytical function, values of which for different ~ values appear in Fig. 1 of [10], • t = ( ~ + 1/5)d2/D,

(7)

and ~=*/12d 2, • being the mean molecular jump length. In evaluating a it must be remembered that the value of /d for the relatively large phthalate molecules is unknown. However, analogous data are available for smaller sized molecules [11, 12]. The following are the values of */d and of the viscosity t / f o r a number of liquids at room temperature: Liquid t/, cP


acetone 0.32 1.25

toluene 0.58 1-00

benzene 0.65 0.86

cyclohexano 0.98 0.86

glycerin 1490 0-20

I t can be seen that as ,/rises, the value of */d is reduced. It is likewise reduced on lowering T [12], which is also equivalent to a rise in ,/. Accordingly, one m a y surmise that the following relationship exists between the quantities in question: as ,/rises, the value of */d decreases. Moreover; since the value of t/is higher in D B P - P V C systems than in the case of glycerin, we m a y assume that for our samples */d<~0.2. Now the value of ~ in equation (7) can be neglected, and in an approximate estimate of g2 we m a y reckon u ~ 0 . The distance in the maximal approach of protons d normally lies within limits ranging from 2/~ [12] to the diameter of the molecule [11], which for D B P is ~ 8 A. The d value selected b y us was 2.5 A, which slightly exceeds that for the double intermolecular radius of the H atom (2-34 A) in crystals [13]. Temperature dependences of T2 calculated in accordance with equation (6) are represented b y t h e fractured curves 2" and 3" in Fig. 1. On comparing the latter curves with curves 2 and 3 it can be seen that at T > T * transverse relaxation is determined b y both types of motion, i.e. the rotational and the translational. On extrapolating theoretical curves into the region of T < T * it is seen that here T u exceeds T2a b y 1-2 orders. Now, in view of equation (5) we m a y write Tea~T~r, i.e. at T < T * the observed correlation time spectrum for I(ve) characterizes hindered rotational motion of the molecules as a whole. Reorientation of the alkyl radicals of DBP, which m a y occur at these temperatures, is probably not a factor determining Tea, as it is known [9] that the rotation of alkyl radicals is practically indePendent of l he polymer content in a sample, whereas the Te~ being measured varies markedly with ~ot. Evaluation of the awrc(ae lifetime of P VC-DBP solvates. This type of evalua-

State of DBP molecules in plasticized PVC

1497

tion m a y be done using rotational [14] or translational characteristics of small molecules. However, the structural complexity of the DBP molecule (the presence of mobile alkyl radicals) rules out the use of T2r for the purpose in question, and means t h a t we must have recourse to an analysis of autodiffusion processes, As was noted on an earlier occasion [4], the measured autodiffusion coefficient of "Vs,see /

/0-9~ i 10-Io Z.5

I

3.0

m3/r, "K

FIc. 3. Temperature dependences of solvate lifetimes r8 for samples containing 30 (1), 42 (2) and 52 we. % DBP (3). the phthalate characterizes the translational motion of all the plasticizer molecules. A difference amounting to several orders (not less t h a n 2-3) exists b e t w e e n / ) values for macromolecules compared with phthalates. Thus, an experimental s t u d y specially carried out showed t h a t in a system containing 19.6 mole ~'o D B P at 140 ° D values for the phthalatc and the macromolecules are respectively 8.1 × 10 -7 and < 1 0 -9 cm2/sec. A similar difference in autodiffusion coefficients was observed by Kosfeld in a 50% solution of PS with 3 / = 4 0 0 0 in benzene: the ratio of D for benzene molecules and PS was ~-102 [15]. This is all evidence to show t h a t the translational motion of small molecules differs greatly from the motion of macromolecules. One m a y surmise t h a t the existence of the P V C - D B P solvatc m a y be limited to time rl between two regular jumps of small molecules, i.e. the solvate lifetime Ts ÷ Zs--Z~: 2D (8) To estimate the value of Zs it was assumed t h a t in the samples examined (r~)~/d~0.2, and d ~ 2 . 5 A, hence (r2/\*~0.5 A. Using the available data on D [4] and formula (8), we obtained values for Zs at T > T * which pr,)ved to be within the range 10-~1-10 -s sec (Fig. 3). The solvate lifetimes is r(duced as the temperature of the sample rises, and as the amount of DBP fl;troduced is inn
1498

A. I. I~A~'T.A~OV~ al.

I t is o n l y w i t h T ~ T ~ , i.e. in t h e glassy s ~ t e , t h a t P V C - D B P s o l v a t e s w i t h long lifetimes may be anticipated. I f t h e s y s t e m is in t h e h i g h elastic state, t h e n t h e s o l v a t e lifetime will b e s h o r t c o m p a r e d w i t h t h e t i m e scale for m a c r o m o l e c u l a r m o t i o n , as is clear f r o m a c o m p a r i s o n o f D v a l u e s for D B P a n d P V C molecules. I t is t h e r e f o r e possible t o r e g a r d t h e s o l v a t a t i o n as b e i n g a d y n a m i c s t a t e in w h i c h diffusion of D B P m o l e c u l e s in t h e P V C m a t r i x t a k e s place. T h e f a c t t h a t all t h e plasticizer molecules i n t r o d u c e d into P V C are c h a r a c t e r ized b y T2a ~nd D v a l u e s differing m a r k e d l y f r o m those for p u r e D B P [4] p o i n t s t o a n a b s e n c e o f " f r e e " plasticizer in t h e s y s t e m s u n d e r s t u d y . T h e a u t h o r s t h a n k V. D. S k i r d a n d A. G. S t e z h k o for t h e i r evah~ation o f t h e diffusion coefficient o f P V C m a c r o m o l e c u l e s , a n d Yu. V. O v c h i n n i k o v for his p a r t i c i p a t i o n in t h e discussion. Translated by R. J. A. I~-DRY REFERENCES

1. B. P. SHTARKMAN, Pla~tifikatsiya PVKh (Plasticization of PVC). p. 146, Izd. "Khimiya", 1975 2. T. FARRAR and E. BECKER, Pulse Type and Fourier NMR Spectroscopy, p. 51, Izd. "Mir", 1973 3. V. D. SKIRDA, A. I. MAKLAKOV and H. SCHNEIDER, Vysokomol. soyed. A20: No. 6, 1978 (Translated in Polymer Sei. U.S.S.R. 20: 6, 1978) 4. A. I. MAKLAKOV, A. G. STEZHKO and A. A. MAKLAKOV, Vysokomol. soyed. A19: 2611, 1977 (Translated in Polymer Sci. U.S.S.R. 19: 11, 1977) 5. A. V. DUMANSKII and R. V. VOITSEKHOVSKII, Kolloidn. zh. 1O: 103, 1948 6. L. Ya. CHENBORISOVA, V. S. IONKIN, A. I. MAKLAKOV and V. A. VOSKRESENSKII, Vysokomol. soyed. 8: 1810, 1966 (Translated in Polymer Sci. U.S.S.R. 8: 10, 1998, 1966) 7. H. A. RESING, J. Chem. Phys. 43: 669, 1965 8. T. A. ALEKSANDROVA, A. M. VASSERMAN and A. A. TAGER, Vysokomol. soyed. A19: 137, 1977 (Translated in Polymer Sei. U.S.S.R. 19: 1, 161, 1977) 9. K. E. ZYL'FUGARZADE, A. A. AKHUNOV and L. M. IMANOV, Izv. AN Azer..SSR, Seriya fiz-tekhn, i mat. nauk, No. 1, 106, 1976 10. G. J. KRt?GER, Z. Naturforsch. A24: 560, 1969 11. N. K. GAISIN, Zh. strukt, khimii 12: 324, 791,971, 1971; Izv. vuzov. Fizika, No. 3, 117, 1971 12. R. B. FIORITO and R. MEISTER, J. Chem. Phys. 56: 4605, 1972 13. A. I. KITAIGORODSKII, Molekulyarnye cristally (Molecular Crystals). p. 19, Izd. "Nauka", 1971 14. B. WILLENBERG and H. SILLESCU, Ber. Bunsenges. Phys. Chem. 77: 95, 1973 15. K. GOFFLOO and R. KOSFELD, Angew. Makromolek. Chem. 37: 105, 1974