Reversible Polymeric Gels and Related Systems - American Chemical

0 C ^c ο where the growth rate is. Rm. = M ( 3 2 f /3c 2 ) 2 / 8 K. (16) c ο. R(q) can be ... optionally be parallel with 1/e diameter =0..5mm or focu...
0 downloads 0 Views 2MB Size
Chapter 11 Gelation of Poly(γ-benzyl-α,L-glutamate) 1,3

1

2

Paul S.Russo ,PaulMagestro ,and Wilmer G. Miller 1

Macromolecular Studies Group, Department of Chemistry, Louisiana State University, Baton Rouge, L A 70803-1804 Department of Chemistry, University of Minnesota, Minneapolis, MN 55455

Downloaded by CALIFORNIA INST OF TECHNOLOGY on November 16, 2017 | http://pubs.acs.org Publication Date: September 14, 1987 | doi: 10.1021/bk-1987-0350.ch011

2

The reversible gelation of poly-γ-benzyl-α, L-glutamate i n toluene and Ν,Ν-dimethylformamide is studied by l i g h t scattering and o p t i c a l microscopy. The gels contain inhomogeneities with a c h a r a c t e r i s t i c size of about 1 µm. Under favorable circumstances, this structure can be visualized i n the wet gel by e p i ­ -illumination microscopy. When it can be v i s u a l i z e d , the structure has a "spongy" appearance, consistent with what one might expect of a system that has undergone spinodal decomposition. So far, however, there i s no concrete evidence to support this mechanism of phase separation. The complex d i f f u s i o n of the rodlike polymers during phase separation i s also discussed.

Rodlike polymers are an increasingly important class of macromolecules prized for their u t i l i t y i n making high-strength fibers.(1,2) In general, rodlike polymers are not thermoplastic; hence, manipulation must occur i n solution, and the solution behavior of this class of polymer i s increasingly studied. Attention has been lavished on the lyotropic l i q u i d c r y s t a l forming a b i l i t i e s of rodlike polymers, and i s o t r o p i c - l i q u i d c r y s t a l l i n e phase boundaries have been determined for several rodlike polymers dissolved i n good solvents.(3) In agreement with the Flory l a t t i c e theory(4), i t has been found that the single phase i s o t r o p i c region i s separated by a narrow biphasic zone from the single phase l i q u i d c r y s t a l l i n e regime. The behavior of rodlike polymers i n poor solvents has received comparatively l i t t l e attention. However, the phase boundaries of the rodlike polypeptide poly-y-benzyl-α, L-glutamate i n dimethylformamide (PBLG/DMF) have been determined over a temperature range spanning both poor and good solvent l i m i t s . ( 5 , 6 ) As expected from the Flory Correspondence should be addressed to this author. 0097-6156/87/0350-0152$08.25/0 © 1987 American Chemical Society

Russo; Reversible Polymeric Gels and Related Systems ACS Symposium Series; American Chemical Society: Washington, DC, 1987.

Downloaded by CALIFORNIA INST OF TECHNOLOGY on November 16, 2017 | http://pubs.acs.org Publication Date: September 14, 1987 | doi: 10.1021/bk-1987-0350.ch011

11.

RUSSOETAL.

153

Gelation of Poly(y-benzyl-a,L-glutamate)

theory, the biphasic region i n the poor solvent l i m i t was found to be very broad. However, the polymer enriched phase was not readily discernible from the f l u i d isotropic phase, as i n the narrow biphasic regime i n the good solvent l i m i t where coexisting phases may often be seen either by the naked eye or by an o p t i c a l microscope. Instead, the system became a gel. The nature of these thermoreversible gels depended on sample treatment. In the f i r s t place, a gel was not formed by simple addition of a poor solvent to the polymer i n either powdered or fibrous form. The gel occurred only when the solvation power of the solvent i n a ready-made polymer solution was suddenly reduced — for example, on lowering temperature. Freeze fracture electron microscopic v i s u a l i z a t i o n of the gel i n the poor solvent, toluene, revealed a bicontinuous, phase separated "beam and girder" structure i n which the polymer was located i n thick (=1000Â) interconnected f i b r i l s . ( 7 ) By X-ray analysis, the actual alignment of the rodlike polymers within the f i b r i l s was judged to be rather poor.(8) Together, these observations prompted the suggestion(7,9) that a dynamic mechanism of phase separation — namely, spinodal decomposition(10,11) — might dictate morphology i n these systems. In this paper, we present further studies of the gel structure and gelation process i n PBLG/DMF and PBLG/toluene, using l i g h t scattering and o p t i c a l microscopic techniques. Gelation of another rodlike polymer, poly-p-phenylene benzobisthiazole, i n 97% H2S0i+ also discussed at the symposium. However, this work has been ' published elsewhere(12). w

a

g

Theoretical A short review of l i g h t scattering from biphasic systems w i l l be h e l p f u l . We s h a l l consider the scattering to arise from a two-phase medium with average p o l a r i z a b i l i t y α and l o c a l inhomogeneities n(r)=a(r)-a. Random Systems. The scattering from random two-phase systems was considered by Debye(13,14) and other accounts are available.(15,16) These formulations have been applied to gels of a f a i r l y s t i f f polymer by Goebel, Berry and Tanner.(17,18) If the system i s s p a t i a l l y isotropic then: oo I(q)« 4π / γ(Γ) K r d r (1) 2

slt

0

r

q

2

r

2 = /

Y(r)

(2)

q = J

sin γ (3) ο Here \ i s the i n vacuo laser wavelength, η the refractive index i n the scattering medium and Θ the scattering angle measured within the sample. The brackets indicate the average over a l l r i n the scattering volume. Thus, the zero angle scattered intensity i s Q

2

I(0)« w

(4a)

Q

where the "correlation volume", w , Q

i s given by

Russo; Reversible Polymeric Gels and Related Systems ACS Symposium Series; American Chemical Society: Washington, DC, 1987.

154

REVERSIBLE POLYMERIC GELS A N D R E L A T E D SYSTEMS

2

w

« 4ïï/ (r)r dr

Q

(4b)

Y

It i s convenient to expand the s i n (qr) term i n eq. 1 which, when compared to the usual expressions for isolated p a r t i c l e s , allows us to define a scattering radius of gyration: 2

2

2

2

= i / r ( r ) r d r / / ( r ) r d r G

Y

(5)

Y

2

As for independent particles can be obtained from the slope of a Guinier plot: In I vs. q . However, i s emphatically not to be considered as a p a r t i c l e radius i n the systems under consideration. Rather, i t i s a more generally defined parameter which characterizes the size of the o p t i c a l inhomogeneities. For a random two-phase system,

Downloaded by CALIFORNIA INST OF TECHNOLOGY on November 16, 2017 | http://pubs.acs.org Publication Date: September 14, 1987 | doi: 10.1021/bk-1987-0350.ch011

2

r

Y(r) = e "

/

2

a

(6)

The parameter a i s another measure of the extent of inhomogeneities, and i s simply related to ( c f . eq. 5 and 6). 2

2

= 6 a

2

(7)

G

1

2

2

Debye plots, I"" / vs. q , also provide a convenient measurement of a (slope/intercept) / Thus, the c o r r e l a t i o n volume for random twophase systems i s given by β

1

w

Q

2

3

2

= 2 a =

3/2

r

/(2/T)

a

(Y(r)-e- / )

(8)

Non-Random Systems. As pointed out by Cahn and H i l l i a r d ( 1 0 , l l ) , phase separation i n the thermodynamically unstable region may lead to a non-random morphology v i a spinodal decomposition. This model i s especially convenient for discussing the development of phase separating systems. In the linearized Cahn-Hilliard approach, the free energy of an inhomogeneous binary mixture i s taken as: 2

F = / dv[f(c)+K(Vc) ]

(9)

v

Here f ( c ) i s the free energy density of a hypothetical completely homogeneous system with volume f r a c t i o n c. It i s understood that c=c(r). For small fluctuations, one can expand f ( c ) about C , the bulk composition of the mixture. Keeping only terms to second order and invoking conservation of mass, one may obtain an expression for AF, the difference between the free energy of homogeneous and inhomogeneous mixtures : q

ΔΡ-/[

1/2 (

3 2 f

/

2 3 c

2

2

) ( c - c ) + K(Vc) ]dv

(10)

o

κ i s positive, representing the "surface" free energy at the boundary between emergent phases. Thus, i f ( 3 f / 3 c ) > 0 the solution i s stable to the small fluctuations applicable to eqn. 9 and phase separation by a random nucleation and growth mechanism can only be i n i t i a t e d by a f i n i t e , thermally driven f l u c t u a t i o n . The l i m i t of this metastability ( i . e . , the spinodal) occurs at ( 3 f / 3 c ) - 0 and the solution becomes unstable whenever ( 3 f / 3 c ) i s negative. The 2

2

2

2

2

2

Russo; Reversible Polymeric Gels and Related Systems ACS Symposium Series; American Chemical Society: Washington, DC, 1987.

11.

155

Gelation of Poly(y-benzyl-a,L-glutamate)

RUSSOETAL.

dynamics of phase separation i s developed by computing the r e l a t i v e flux of volume elements associated with the two components, J-MV(p^-y2), where M represents the r e l a t i v e mobility of the molecules i n the presence of a chemical gradient i n a binary system, and applying continuity to get the d i f f u s i o n equation: 2 3 C

/

at

= M(

3 f

/.2

2

)

d e c

4

V c - 2MKV C v o

(11)

u m e

Appropriate units for J and M are ( — ^ —\ area - sec'

a

Downloaded by CALIFORNIA INST OF TECHNOLOGY on November 16, 2017 | http://pubs.acs.org Publication Date: September 14, 1987 | doi: 10.1021/bk-1987-0350.ch011

v

n

(

j /length J units. energy

respectively. Since the system i s assumed to be at rest o v e r a l l , J1--J2» The c o e f f i c i e n t of the V c term i s associated with the i n t e r d i f f u s i o n c o e f f i c i e n t for the 2 components i n the nonequilibrium system: 2

2

3f D - M( V d

(12) Ο Thus, this d i f f u s i o n c o e f f i c i e n t changes sign at the spinodal. The general solution to eq. 11 i s given by: c-c

Q

f t

dC

2) C

- liexp (R(q)t)}{A(Ocos(qr)+B(Osin(qr)}

(13)

1 2 R(q) = -M (

3 f

/

2)

3 c

c

q ο

2

4

-2M20°, a FICA model 42000 l i g h t scattering goniometer was equipped with s o l i d state electronics, a 2 mW HeNe laser and c e l l holders to accept flamesealable 13 mm diameter c y l i n d r i c a l c e l l s or 1-2 mm pathlength rectangular spectroscopy c e l l s . The gaussian incident beam could optionally be p a r a l l e l with 1/e diameter =0..5mm or focused at the sample to -0.2 mm. With toluene or s i l i c o n e o i l as the i s o r e f r a c t i v e vat material, r e l i a b l e background subtraction was possible down to 0=20° ( c y l i n d r i c a l c e l l s ) or 0-5° (spectroscopy c e l l s ) . Scattering measurements at low angles 0.5°510 nm). For fluorescence observation PBLG was end-labelled with fluorescein isothiocyanate (FITC) i n tetrahydrofuran. Unreacted FITC was removed by repeatedly: dissolving the labelled PBLG i n DMF; p r e c i p i t a t i n g i t into f i l t e r e d , deionized water; recovery by ultracentrifuge; washing with copious amounts of acetone; and vacuum drying at less than 60°C. A control on unlabelled PBLG was performed. Dynamic l i g h t scattering confirmed i d e n t i c a l solution properties for labelled and unlabelled PBLG i n DMF.

Russo; Reversible Polymeric Gels and Related Systems ACS Symposium Series; American Chemical Society: Washington, DC, 1987.

158

REVERSIBLE POLYMERIC GELS AND RELATED SYSTEMS

Results High Angle Light Scattering Photometry. A t y p i c a l set of high angle scattering envelopes collected during gelation (kinetic envelopes) i s shown i n Figure 1. No scattering maximum i s observed. The i n t e n s i f i c a t i o n at various angles i s shown i n Figure 2, and scattering does seem to increase approximately exponentially over a limited time range. We estimated R(q) from the " l i n e a r " region of the semi-log plots and R(q)/q vs. q^ i s plotted i n Figure 3. Figures 1-3 do not resemble c l a s s i c a l spinodal decomposition. Various times after quench are defined according to the precentange of t o t a l i n t e n s i f i c a t i o n achieved: at tgo the i n t e n s i f i c a t i o n i s 90% complete. Figure 4 shows that the delay time prior to gelation, as well as the time required after the process begins, depends strongly on quench temperature. Experience shows that, at a given temperature, the rate of gelation increases with concentration, but this observation awaits quantitation.

Downloaded by CALIFORNIA INST OF TECHNOLOGY on November 16, 2017 | http://pubs.acs.org Publication Date: September 14, 1987 | doi: 10.1021/bk-1987-0350.ch011

2

Low Angle Light Scattering Photometry: Static Envelopes. We s h a l l f i r s t consider the f i n a l scattering envelopes obtained after gelation i s complete. The next subsection w i l l be devoted to the k i n e t i c evolution of the scattering envelopes. 2

_ 1

2

PBLG/toluene: Typical Guinier (In I vs. q ) and Debye ( I / 2 vs q ) plots are shown i n Figure 5 for 0.91 wt % PBLG-138,000/toluene gels made by quenching from 70°C to different temperatures. Although i n t e n s i t i e s are i n arbitrary units, v a r i a t i o n i n experimental parameters has been taken into account so the plots may be compared d i r e c t l y . The plots were frequently linear (Figure 5a) but sometimes displayed nonlinear behavior (Figure 5b) or a maximum (Figure 5c). Between runs, the sample was "renewed" by heating at 100°C (well above the melting transition) with slow tumbling for at least 15 minutes. Samples that were not renewed but merely heated at 70°C and re-gelled generally gave reduced scattering intensity, as may be seen by comparing 5-a and 5-d. The adherence to linear Debye or Guinier plots was capricious, to say the least. Nevertheless, i n order to examine the general trends, we performed least squares analysis on the linear regions of the p l o t s . The results are l i s t e d i n Table I. We see that the inhomogeneities i n the system have a size of a few microns and that the extrapolated forward scattered intensity can vary by a factor of ~ 1000. It should be remarked that i n 1-mm pathlength c e l l s none of these gels was v i s u a l l y opaque. The more weakly scattering gels had a v i s u a l c l a r i t y indistinguishable from pure toluene and the most strongly scattering samples were very s l i g h t l y turbid i n appearance. The Guinier intercept i s plotted against temperature i n Figure 6. The relationship /a -6 (eq. 7) i s not well obeyed although this i s probably due to experimental uncertainty i n estimating the intercept of the Debye plots. Studies at other concentrations and temperatures have not yet indicated a clear trend. Q u a l i t a t i v e l y , the results are similar: linear and nonlinear plots are found; i l l - d e f i n e d peaks are occasionally seen; and the c h a r a c t e r i s t i c size of the inhomogeneities i s generally a few microns. Some of these results are l i s t e d at the bottom of Table I. 2

2

Russo; Reversible Polymeric Gels and Related Systems ACS Symposium Series; American Chemical Society: Washington, DC, 1987.

Downloaded by CALIFORNIA INST OF TECHNOLOGY on November 16, 2017 | http://pubs.acs.org Publication Date: September 14, 1987 | doi: 10.1021/bk-1987-0350.ch011

11.

Gelation of Poly (y-benzyl-a,L-glutamate)

RUSSOETAL.

30

50

70

θ (DEGREES) Fie- 1 Scattering at Θ > 20° for gelation of PBLG-138,000/toluene (c » 7.3xl0~ 7.3xlO~ gm/ml). gm/ml). I n i t i a l temperature: 115°C; f i n a l temp: 10°C. Times after quench are indicated by symbol. Note the absence of a maximum. (115°C obtainable because sample i s sealed)· nA

Russo; Reversible Polymeric Gels and Related Systems ACS Symposium Series; American Chemical Society: Washington, DC, 1987.

159

REVERSIBLE POLYMERIC GELS AND RELATED SYSTEMS

5000 >-

< cr

H m

j r

1000 ^

500j

^ —0-15° x y

°

^-O

Ο 25°

/L^O

s*

2 0

rr--0

°

0-30°

cr

ζ

/ 100

LU

50

ι IJIH^

CO




°-90°

I

Downloaded by CALIFORNIA INST OF TECHNOLOGY on November 16, 2017 | http://pubs.acs.org Publication Date: September 14, 1987 | doi: 10.1021/bk-1987-0350.ch011