temperature, polymer, atmosphere, phase behavior, small-angle X-ray scattering, micelles, water content, differential scanning calorimetry, fluorescence, S. Straus, D. H. Wiliams, C. Djerassi, critical micelle concentration, anionic surfactants, fluorescence probe, M. Malanga, Fluorescence Probes of Surfactant Association, polymer system, SDS micelles, American Chemical Society 1322 Turro, Interactions, L. A. Wall, probe, Mass Spectrometry, Organic Compounds, surfactant concentration, J. Jachowicz, visual appearance, complex phase, optical microscopy, water contamination, phase boundaries, system, polarizing optical microscopy, University of Minnesota, Department of Chemistry, R. Simha, aqueous solution, New York, J. H. Flynn, J. Polym, Sodium Dodecyl Sulfate, H. Budzikiewicz, Water-Soluble Polymers
Macromolecules 1984,17, 1321-1324
(12) H. Budzikiewicz, C. Djerassi, and D. H. Wiliams, `mass spectrometry
of Organic Compounds", Holden-Day, San Francisco
, Cambridge, London, and Amsterdam, 1967. (13) J. Jachowicz, M. Kryszewski,and P. Kowalski,J.Appl. Polym. Sci., 22, 2891 (1978). (14) L. A. Wall, S. Straus, J. H. Flynn, D. McIntyre, and R. Simha, J. Phys. Chem., 70, 53 (1966). (15) H. H. G. Jellinek, "Thermal Degradation of Vinyl Polymers", Academic Press, New York, 1955. (16) R. Simha, L. A. Wall, and P. J. Blatz, J. Polym. Sci., 5, 615 (1950); R. Simha and L. A. Wall, ibid., 6, 39 (1951).
(17) `Thermal Stability of Polymers", Vol. 2, R. T. Conley, Ed., Marcel Dekker, New York, 1970. (18) G. G. Cameron, Mukromol. Chem., 100, 255 (1967). (19) L. A. Wall, S. Straus, R. E. Florin, and L. J. Fetters, J. Res. Nutl. Bur. Stand. (US.)77,A, 157 (1973). (20) G. G. Cameron, J. M. Meyer, and I. T. McWalter, Mucromol- ecules, 11, 696 (1978). (21) I. C. McNeill and M. A. J. Mohammad, Eur. Polym. J., 8,975 (1972). (22) M.Kryszewski, J. Jachowicz, M. Malanga, and 0. Vogl, Poly- mer, 23 (2), 271 (1982).
Photoluminescence Probes for the Investigation of Interactions between Sodium Dodecyl Sulfate and Water-Soluble Polymers Nicholas J. Turro,* Bruce H.Baretz, and Ping-Lin Kuo Department of Chemistry, Columbia University
, New York, New York
10027. Received January 18, 1983
ABSTRACT Fluorescence probe techniques have been employed to monitor interactions between sodium dodecyl sulfate (SDS) and two water-soluble polymers, poly(N-vinylpyrrolidone) and poly(ethy1ene oxide). The resdta me analogous to those found by conventional methods
and lead to the conclusion that SDS micelles bind to the polymers. "he fluorescence probe method also provides information on the solubilization mechanism, since the site of solubilization of the probe can be inferred.
Introduction Polymer/Surfactant Interactions i n Aqueous Solution. Investigations of the interactions of polymers and surfactants in aqueous solution are of interest from the fundamental standpoint of obtaining an understanding of the structure and dynamics of polymer/surfactant associates and from the practical standpoint of employing fundamental understanding to assist in formulations for polymer/swfactant system that can be used in the process of enhanced oil recovery
.' The ability of surfactants to aggregateand form micelles2adds a particularly intriguing dimension to their interactions with polymers. An ex- tensive body of evidence has accumulated that supports the proposal that anionic surfactants such as sodium dodecyl sulfate (SDS) and water-soluble polymers such as poly(N-vinylpyrrolidone)or poly(ethy1eneoxide) (PVP and PEO, respectively) form associates between surfactant aggregates and polymer rather than surfactant monomers and p01ymer.~ The following picture has emerged to describe the polymer/surfactant interactions that occur when SDS is added to a dilute aqueous solution of water-soluble PVP or PE0.2-4In the absence of polymer, SDS forms welldefined micelles above the critical micelle concentration (cmc) of 8 X (termed xo). When SDS is added to a dilute (51%) solution of PVP (or PEO), there is no association between polymer and surfactant until a concentration x1 is reached. Above x1 (which is a lower concentration than the cmc, xo) the polymer /surfactant association process starta abruptly and then saturates abruptly at a concentration x2(which is a higher concentration than the cmc, xo). Thus, three well-defined regions are suggested by available information: region I, for which [SDS] ranges from 0 to x,; region II,for which [SDS]ranges from x1 to x2;and region 111, for which [SDS] is greater than xa. In region I, there is no significant polymer/surfactant interaction. In region 11,clusters of SDS molecules (termed premicelles)cooperativelyassociate with the polymers. In region 111,free micelles of SDS form and exist in equilibrium with the polymer/surfactant associates. These situations are summarized schematically in Figure 1.
Pyrene and ll-(3-Hexyl-l-indolyl)undecylSulfate as Fluorescence Probes of Surfactant Association. Pyrene is a strongly hydrophobic probe with low solubility (ca. 3 X M) in water. The fluorescence spectrum of pyrene at low concentration (C1 X lo4 M) in Homogeneous Solutions
possesses considerable fine structure
whose relative peak intensities undergo significant perturbation upon going from polar to nonpolar solvent^.^ The ratio of the fluorescence intensity of the highest energy vibrational band (II)to the fluorescence intensity of the third highest energy vibrational band (IIIIh)as been shown to correlate with solvent polarity for a range of solvent structures. For e ~ a m p l ei,n~hydrocarbon solvents II/IIII is ca. 0.6, in ethanol 11/11is1c1a. 1.1,and in water 11/11is1 ca. 1.6. Thisdistinct dependence of fluorescencevibrational fine structure has been utilized to investigate the formation of SDS micelles.6 In the presence of SDS micelles, pyrene is preferentially solubilized in or near the interior hydrophobic region of the micelle. The value of 11/11is1c1a. 1.1 when pyrene is solubilized in SDS micelles (Figure 2). Since the value of II/IIn for pyrene is a convenient, readily measured quantity and since pyrene is poorly water soluble
and is preferentially solubilizedin the hydrophobic regions of aqueous systems that are microheterogeneous,it seemed apparent that polymer/surfactant associations could be investigated by employing the II/IIII parameters. The functionalized detergent ll-(3-hexyl-l-indoyl)undecyl sulfate (6-1n-11-) possesses a "built-in" fluorophore that allows this detergent to serve as a photoluminescence probe.7 It has been shown that the fluorescence maximum (A,) of 6-In-11- is sensitive to solvent polarity so that XF (in a manner similar to 11/1111) may be employed to probe the formation of hydrophobic associates in aqueous solution. Results Pyrene as a Fluorescence Probe for Polymer/Surfactant Interactions for the PVP/SDS and PEO/ S D S Systems. The variation of 11/1111 was measured for aqueous solutions containing fixed concentrations of
0024-9297/84/2217-1321$01.50/0 0 1984 American Chemical Society
1322 Turro, Baretz, and Kuo & polymer (oq) v- S D S i a q l
Macromolecules, Vol. 17, No. 7: 1984 t :/Im
Figure 1. Idealized schematic representation of three regions
of polymer/surfactant interactions as a function of surfactant
concentration. In region I the water-soluble probe, SDS mono-
mers, and probes are solubilized in the aqueous phase and are
not associated. In region 11, at a concentration x1 (below the cmc
of SDS), association of SDS with the polymer begins, and at a
concentration x 2 (above the cmc of SDS), the association of SDS
with the polymer is saturated. In region I11 micelles (which exist
in equilibrium with polymer/SDS associates) become the pre-dominant species
(c,, 3 r 1 0 7 M ,
-P E O ( M W 6 0 0 0 0 0 ~0 4 %
id5 Y *
Figure 2. Demonstration of the use of pyrene as a fluorescence probe for micelle formation by SDS in aqueous solution. The points on the line correspond to the ZI/ZIII values (see text for discussion) as a function of SDS concentration. The inserts are spectra of pyrene fluorescence
below the cmc (concentration indicated by heavy arrow) and above the cmc (concentration indicated by heavy arrow).
polymer a n d variable concentrations of SDS. T h e results for PVP/SDS systems are shown in Figure 3 and the results for t h e PEO/SDS systems are shown in Figure 4. For t h e PVP/SDS system t h e 11/11v1s.1[SDS] profiles (Figure 3) look like "textbook" examples of regions I, 11, and III.3b T h e first break (transition from region I t o region 11) in t h e value of 11/1111occurs at x1 zz 2 X M. Literature valuess of x1for PVP/SDS systems fall in t h e range (1-2) X M SDS. T h e second break (transition from region I1 t o region 111)occurs at x2 5 X M a n d is independent of t h e molecular weight
of PVP for 0.170 concentration of polymer. For solutions containing 1% PVP break at x2 is not sharp, b u t t h e value of x 2 is clearly larger than that for the more dilute polymer system
. The value of 11/111a1t high concentration (>0.1 M) of SDS approaches t h a t for SDS micelles in t h e absence of poly- mer. For t h e PEO/SDS system t h e 11/11v1s.1[SDS] profile (Figure 4)shows only one s h a r p break for PEO concentrations varying from 0.4% t o 0.02%. For 0.4% and 0.2% PEO, t h e first break allows assignment of t h e value of x1
Macromolecules, Vol. 17, No. 7, 1984
SDS/Water-Soluble Polymer Interactions 1323
I 0.004 [Co`*]
-0.006 0.008 ,M
Figure 5. Stern-Volmer quenching of 6-In-11-(0.9 X lo-*M) * in SDS (25 X M)in the presence of hydrophilic polymers: (1)no polymer, slope = (9.1 0.5)X 102 M-l;(2)1%(w/v) PEO, *slope = (11.4f 1.7) X 10 M-l;(3)1% (w/v) PVP,slope = (5.9 1.4) X 10 M-l.All error limits are to 95% confidence limit.
Table I Quenching of 6-In-11-Fluorescence by Co(I1)
Kgv, M-' 109~ps, k,, M-'s - ~
92 X 1O'O
6.4 x 109
12 x 109
fluorescencein dilute aqueous solutions of SDS (25 X M) is quenched by Co(I1) and follows Stern-Volmer kinetia`* (Figure 5 and Table I). However, a Stern-Volmer constant of 59 M-l is found for the PVP/SDS system and 114 M-'is found for the PEO/SDS system. From the 6-111-11- lifetimes of 9.2 X lo4 s (PVP/SDS system) and of 9.5 X 10-9s (PEO/SDS system) in the absence of Co(II), quenching constants of 6.4 X lo9 M-ls-l and 12.0 X log M-l, respectively, are evaluated. The rate constant for quenching of 6-In-11-fluorescence by Co(I1) in pure water has been determined' to be 5 X lo-' M. For the more dilute (0.02%) PEO/SDS system, a fairly sharp break is seen, and a plateau region occurs between ca. 7 X and 6 X M SDS. The value of II/IIIoIf ca. 1.25 in the plateau region is close to the value of I I / I ~ = of ca.1.15 for SDS micelles. This in itself explainsthe lack of sharpness for the break from region I1 to region I11for the PEO/SDS, compared to the PVP/SDS. In the latter case the plateau region corresponds to 11/11o1f 1ca. 1.6, a value that is considerablydifferent from that for the probe in SDS micelles. In addition to serving as a probe of polymer/surfactant interactions, the measurement of II/IIn provides information on the time-average location of the pyrene probe in the polymer/surfactant systems. In the case of PVP/SDS, the pyrene probe experiences in region I1 an environment whose polarity is intermediate between that of water and SDS micelles (II/Im N 1.6). For comparison, the value of II/Imis ca. 1.6 for the following homogeneous
solvents: ethylene glycol (1.6), formic acid (1.6), acetone (1.5). On the other hand, in the case of PEO/SDS the pyrene probe experiences in region I1 an environment whose polarity is very close to that of SDS micelles (1.1). For comparison, the value of 11/11is1c1a. 1.1for the following solvents: benzene (l.l), benzyl alcohol (1.2), 1chlorobutane (1.1). Thus, in general, the pyrene probe in polymer/SDS associates experiences a more hydrophilic environment than that experienced in SDS micelles. This conclusion is consistent with smaller SDS micelles (bound to the polymer) with too loose aggregation of the SDS along the polymer strand. We favor the former model, which has independent support from NMR measurements and from thermodynamic calculations. For smaller micelles, water penetration is expected to be greater so that the probe "sees"more of the polar palisade layer of the micelle and the water associated with the palisade layer. Recent results that overlap with and complement our studies come to a similar conclusion.1° Indeed, evaluation of the "aggregationnumber" of the polymer-bound micelles in region 11via a dynamic fluorescenceprobe method leads to values 3040% smaller (25-45) than those for SDS micelles in the absence of polymer (60-70). From the results for quenching of 6-In-11- emission by Co(II), it is concluded that although water may penetrate the interior of polymer-bound micelles, metal cations may not. Thus, probes solubilized by polymer-bound micelles are protected from quenching by cations, compared to the probe in free SDS micelles. For example,the rate constant for quenching of 6-In-11- fluorescence in region I1 decreases from 92 X 1O1O M-' s-l (a "pseudo"-second-order rate constant) for SDS micelles to 12 X lo9 M-' s-l for the PEO/SDS system to 6.4 X lo9M-l s-l for the PVP/SDS system. A detailed interpretation of these data is not warranted at this time because of potential special effects in expelling cations, which might be possible for PVP (cationictautomer)l' or for stabilizingcations (crown ether type coordination).12 Conclusion The results of our investigation demonstrate that fluorescence probe methods may be used to investigate interactions in polymer/surfactant systems in a manner quite analogous to conventional methods such as measurements of surface tension, conductivity, and viscosity. Our results are consistent with the existence of SDS micelles bound to PVP and PEO polymers. The size of these polymer-bound micelles is probably smaller than that of SDS micelles in the absence of polymer. The average environmental polarity experienced by the pyrene probe is in the order H 2 0 > region I1 (PVP/SDS) > region I1 (PEO/SDS) 2 SDS micelles. In the case of the 6-In-11- - probe, from the values of XF its average environment is in the order H 2 0 > region I1 (PEO/SDS) SDS micelles > region I1 (PVP/SDS). In summary, in addition to providing a convenient method for measuring polymer/ surfactant interactions,the fluorescenceprobe method also provides information on the mechanism of solubilization by polymer/surfactant aggregates, since the site of solubilization can be inferred. Acknowledgment. We thank the Army Research Office for generous support of this work. References and Notes (1) Taber, J. J. Pure Appl. Chem. 1980,52, 1323. (2) Robb, I. D., Surf. Sci. Ser. 1981,11, 109. (3) (3Cabane, B. J.Phys. Chem. 1977,81,1639.(b) Robb, I. D. In Surf. Sc'i. Ser. 1981,11.
Macromolecules 1984, 17, 1324-1331
(4) Nagarajan, R. Polym. Prepr., Am. Chem. SOCD. iu. Polym.
(5) (a) Dong, D. C.; Winnik, M. A. Photochem. Photobiol. 1982,
35717' (b) Nakajima9A' " papers.
lg7" 'If 429 and
(6) Kalyanasundaram, K.; Thomas, J. K. J. Am. Chem. SOC1. 977,
(7) (a)Turro, N. J.; Tanimoto, Y.; Gabor, G. Photochem. Photo-
biol. 1980,31,527. (b) Schore, N. E.; Turro, N. J. J . Am.
Chem. SOC1. 975,97,2488.
(8) Arai, H.; Murata, M.; Shinoda, K. J . Colloid Interface Scz. 1971,37,223. (9) Shirakama, K. Colloid Polym. Sci. 1974,252,978. (10) Zana, R. Polym. Prepr., Am. Chem. SOCD.,iu. Polym. Chem. 1982,23,41. (11) Takagishi, T.;Kuroki, N. J . Polym. Sci., Polym. Chem. Ed. 1973,11,1889. (12) Arkhipovich, G. N.;Dubronski, s.A,; Kazanski, K. s.;Ptitsina, K. S.; Shupik, A. N. Eur. Polym. J . 1982,18,569.
On the Nature of the Poly(?-benzyl glutamate)-Dimethylformamide "Complex Phase" Paul S. Russot and Wilmer G. Miller* Department of Chemistry, University of Minnesota, Minneapolis, Minnesota 55455. Received May 17, 1983
ABSTRACT: The system poly(y-benzyl L-glutamate)/dimethylformamide(PBLG/DMF) has been studied by small-angle X-ray scattering, polarizing optical microscopy, differential scanning calorimetry, and visual observation. The effects of a nonsolvent, water, have been assessed. We find that a small amount of water, which can be easily absorbed from the atmosphere under normal ambient conditions, even when samples are stored in capped containers, seriously alters the visual appearance, phase behavior, and morphology of the system. The various morphologies available in PBLG/DMF/H,O are classified and discussed in terms of a pseudobinary, biphasic system, in which a polymer-rich ordered state coexists with a polymer-poor phase, which may be ordered or disordered depending on water content and temperature. We conclude that the *complexphase" reported for PBLG/DMF is a result of water contamination and is related to the phase behavior of the PBLG/DMF/H20 ternary system.
Poly(y-benzyl L-glutamate) (PBLG) was the first syn-
thetic polymer to exhibit cholesteric liquid crystalline
A cholesteric phase is known to
exist in a number of s o l ~ e n t s . ~ H- lo~wever, there is dis-
agreement over the morphological state in dimethylform-
amide (DMF). In two independent X-ray investigations
on the PBLG/DMF system, a "complex phase" was re-
ported."J2 This state, which occurred below ca. 40 "C, was
described as an opaque gel." Watanabe et al. noted an opaque gel at T < 60 O C in 20 w t % PBLG/DMF.8 Over
the years, opaque whitish states have occasionally been
seen in this laboratory. However, the vast majority of
PBLG/DMF samples have been clear, with the cholesteric
thumbprint pattern clearly evident in the polarizing optical
micro~cope.'-~Many flame-sealed ~ a m p l e s , ~us~edJ ~to
determine the phase boundaries in this system, have re-
mained clear for a decade. The occasional whitish samples
have been routinely discarded in the belief that a non-
solvent, such as water, caused the whitish appearance.
In this paper, we report the findings of a reexamination
of both clear and whitish samples. Using small-angle X-ray
scattering (SAXS),polarizing optical microscopy (POM),
differential scanning calorimetry (DSC), and visual ob-
servations, we have determined the effect of a small
amount of water. For comparison, the effect of a non-
solventon the phase behavior has also been calculated. We
conclude that a small amount of water drastically changes
the appearance and morphology of the system in a fashion
consistent with its phase behavior.
Materials and Methods PBLG of 310 000 and 130 000 daltons (M.,,) was obtained from New England Nuclear and Miles Yeda, respectively, and desig-
Present address: Department of Chemistry, Louisiana State University, Baton Rouge, LA 70803.
nated as PBLG-310000 and PBLG-130000. On the basis of the polymerization mechanism and results from other studies,15' M,/M, is considered to be 1.2-1.3. PBLG was vacuum-dried to constant weight at temperatures not exceeding 60 "C prior to use. Reagent grade DMF was dried over 4-A molecular sieves and distilled under reduced pressure a t ca. 50 "C. Solutions were prepared by weight in a dry atmosphere or as rapidly as possible in the open atmosphere. Solutions were homogenized in capped vials containing magnetic stirbars placed on a stirring hotplate in a dry nitrogen atmosphere. The temperature required for dissolution increased with water content. If necessary, the temperature was raised to as much as 80-100 "C for a short time to ensure homogeneous samples. For conversion to volume fraction (up)t,he specific volume of PBLG was taken as 0.791 cm3/g and the density of DMF as 0.944 g/cm3. All X-ray studies were performed in 1-mmglass capillaries with 0.01-mm-thick walls (Charles Supper Co.). The sample was transferred into the capillary and spun to the bottom in a lowspeed centrifuge, both operations taking place under a dry nitrogen atmosphere. The capillaries were removed from the dry atmosphere and immediately flame sealed. A good seal was ensured by dunking the capillary tip several times into molten beeswax. The capillaries were then weighed over a period of several days to check for leaks. The delicate capillaries were stored inside tubular holders glued to microscope slides held in a slide box. Temperature equilibration was achieved by immersing a watertight slide box into a temperature-controlled bath (*0.005 "C). The samples were transported to the X-ray camera in a capped Dewar containing some water from the bath. The sample was mounted in a brass cell holder of original design. The holder had a volume of 3 in.3,through which thermostated water circulated, and was designed to fit into the standard slots of the Warhus pinhole-collimated camera used throughout these studies. Kel-F insulators prevented heat transfer to the rest of the camera, and the water inlet lines were sealed to permit vacuum operation for reduced air scatter. Since facilities for measurement of the temperature at the brass block during the measurement were not available, the temperature reported is that at the water bath, which was connected to the cell holder by well-insulated lines. All measurements were performed under vacuum. The camera had several standard film positions. The distance from the sample
0024-9297/84/2217-1324$01.50/0 0 1984 American Chemical Society
NJ Turro, BH Baretz, PL Kuo