ADSORPTION OF POLYSTYRENE-POLY (METHYL METHACRYLATE) MIXTURES
3783
The Adsorption of Polystyrene-Poly (methyl methacrylate) Mixtures at
I
Solid-Liquid Interface'
by Curt Thies The National Cash Register Company, Dayton, Ohio 46409
(Received March 17, 1966)
The adsorption of polystyrene (PS) and poly(methy1 methacrylate) (PMMA) mixtures on finely divided silica from dilute trichloroethylene solutions at 25" has been studied in order to establish the competitive adsorption behavior of these polymers. It has been shown that under equilibrium conditions, reached within 2-4 hr, PS is not adsorbed unless the available PMMA is unable to saturate the adsorbent surface. If PS is first equilibrated on the silica surface and excess PMMA is added later in a second step, complete PS displacement occurs within 2-4 hr. Within limits of the experimental technique used, the rate and extent of displacement are essentially independent of the time the PS is equilibrated on the surface and the PS surface coverage. Polymer-polymer incompatibility was found to have little effect on the adsorbance and/or adsorbed structure of either polymer. Infrared bound fraction data establish that simultaneous adsorption of PS and PMhlA on silica induces relatively small changes in the average number of segments of either polymer that are attached to the adsorbing surface. However, these results may be uniquely applicable to dynamic systems involving small adsorbent particles such as employed in this study.
Introduction The absorption of polymers at solid-liquid interfaces has been studied extensively and is the subject of Past studies have led to the several recent conclusion that polymers adsorbed at such interfaces have a looped or coiled structure in which only a fraction of their segments are attached directly to the interface. Nany such structures are possible, ranging from ones which yield a relatively flat and compressed adsorbed polymer layer to ones which give adsorbed layers highly extended away from the interface. Thus, efforts have been made to measure directly the structure Of particular interest is an of adsorbed infrared method first reported by Fontana and Thomas.s This utilizes the fact that various molecular groups possess a characteristic infrared band which appears at a certain frequency when the group is unadsorbed and then shifts slightly to a new frequency when the group is bound or attached to an interface. If only a fraction of a polymer's segments are attached, this fraction can be determined by resolving the infrared peaks due to the adsorbed and unadsorbed segments. From these
peaks, the concentrations of free and bound segments can be determined, and the infrared bound fraction, p , can be calculated. Hence, p represents the average fraction of groups that are bound directly to the interface. By combining bound fraction data with isotherm data, considerable insight into polymer adsorption phenomena can be gained. The purpose of this study is to utilize these tools to explore the competitive ad(1) Portions of this paper were presented a t the 149th National Meeting of the American Chemical Society, Polymer Division, Detroit, Mich., April 1965, and the 13th Canadian High Polymer Forum, Ottawa, Canada, Sept 1965. (2) R. E. Hughes and C. A. von Frankenberg, Ann. Rev. Phya. Chem., 14, 290 (1963). (3) F. Rowland, R. Bulas, E. Rothstein, and F. R. Eirich, I d . Eng. Chem., 57, No. 9, 46 (1965). (4) F. R. Eirich, Consig. Naz. Ric. (Rome), 1963. ( 5 ) R. R. Stromberg, D. J. Tutas, and E. Passaglia, J . Phys. Chem., 69, 3955 (1965). (6) R. R. Stromberg, E. Passaglia, and D. J. Tutas, J . Res. Natl. Bur. Std., 67A, 431 (1963). (7) M. Gottlieb, J . Phys. Chem., 64, 427 (1960). (8) B. J. Fontana and J. R. Thomas, ibid., 65,480 (1961).
Volume 70,Number 1% December 1966
CURTTHIES
3784
sorption behavior of polystyrene (PS)-poly(methy1 methacrylate) (PMMA) mixtures and determine whether polymer incompatibility effects alter the adsorbance or adsorbed structure of either polymer when they are simultaneously adsorbed on silica. Equilibrium and reversibility of the adsorption processes occurring in such systems also were investigated. Experimental Section Materials. A series of PMMA and PS polymers was prepared by azobisisobutyronitrile initiated polymerization of freshly distilled monomers at 55" in sealed glass ampoules. Except for PS-1, which was carried to approximately 90% conversion, all polymerizations were carried to about 20% conversion. Each polymer was repeatedly reprecipitated from benzene into methanol and then freeze-dried from benzene. Characterization data are given in Table I. Number average molecular weights were determined in toluene a t 37" using a Mechrolab Model 502 membrane osmometer (F and RI 1\lechrolab Division, Hewlett-Packard Corp., Mountain View, Calif.). Intrinsic viscosity measurements were made in benzene at 30 f 0.05" using a Cannon-Ubbelohde viscometer. Viscosity averwere calculated from age molecular weights the equation [ V I = KM", where h] represents the intrinsic viscosity, M the molecular weight, and K and a appropriate constants. The K and a values used for PS were those of Ewart and Tingeye while values reported by Stanley'* were used for PMMA.
(nn)
(ivy)
Table I: Characterization of Polymers Used for Adsorption Studies
Sample
PMMA-I PMMA-5
PS-1 PS-2
Polymerization method
Bulk Solution" Bulk Solution"
anx
10-5
8.22 3.20 1.05 0.36
a, x 10-6 19.0 3.68 2.46 0.39
Toluene/monomer ratio was 1: 1 by volume.
The adsorbent for all experiments was Cab-0-Si1 M-5 (Cabot Corporation, Boston, Mass.). This is a finely divided silica powder having a manufacturer's specified surface area (nitrogen adsorption) of 175-200 m2/g. It was heated at 110" for 48-72 hr, cooled, and stored in a desiccator until used. Most experiments were run using a single lot of Cab-0-Si1 in order to avoid lot-tolot variations in adsorption behavior. Only the rate of adsorption plot for PS-1 shown in Figure 2 was obtained using a second lot. The Journal of Physical Chemistry
The solvent for all experiments was Chromatoquality trichloroethylene (C2CHCI,) (Matheson Coleman and Bell, Nonvood, Ohio). It was used as received. Adsorption Isotherms. Adsorption isotherms for the individual polymers were constructed at 25 f 1" by agitating 25-ml sealed glass ampoules containing a known weight of adsorbent, w (grams), and 15 ml of polymer solution of initial concentration, co (grams/100 ml), on an Eberbach laboratory shaker at 170 1.5-in. strokes/min (Eberbach, Ann Arbor, Mich.). The shaker was enclosed in order to prevent photoinitiated decomposition of C2HCl, during prolonged agitation periods. All ampoules were flushed with nitrogen before sealing. After known agitation periods, the suspension was centrifuged, and the concentration of unadsorbed polymer remaining in the supernatant, ca (grams/100 ml), was determined by infrared analysis. The characteristic band of PS at 697 cm-' and the band for PMMA a t 1720 cm-' were utilized for these analyses. Polymer adsorbances are reported as weight of polymer adsorbed per unit weight of adsorbent, s/?n (milligrams per gram). The absence of surface-active impurities was confirmed by varying the ratio of adsorbent to volume of solution used in carrying out the adsorption experiment. l1 Adsorption isotherms for PS-PRIMA mixtures were obtained in a manner similar to that used for the individual polymers. I n this case, initial concentrations of both polymers were fixed while the amount of adsorbent added was varied. Mixture isotherms are plotted as values of cs for both polymers us. to. Since neither PS or PMMA has infrared bands which overlap the other's characteristic infrared band, the concentrations of both in binary mixtures were readily determined by infrared analysis. PS Displacement. PS was adsorbed from CzHCls on silica and equilibrated for a known time. Excess PMMA then was added to the system and the ea values for both polymers were determined by infrared analysis after known displacement periods. In all cases, the PS was added to the silica first as 10 ml of solution and 5 ml of PMMA solution was added later. &'lost runs were made as outlined for construction of adsorption isotherms. When PS was equilibrated on the surface for 3 min before PMMA addition, the samples were agitated vigorously by hand rather than on the mechani(9) R. H. Ewart and H. c. Tingey in "Styrene, Its Polymers, CGpolymers and Derivatives," R. H. Boundy and R. F. Boyer, Ed., Reinhold Publishing Co., New York, N. Y., 1952, p 334. (10) E. L. Stanley in "Analytical Chemistry of Polymers," Part I, G. M. Kline, Ed., Interscience Publishers, Inc., New York, N. Y., 1959, p 14. (11) R. Perkel and R. Ullman, J. Polymer Sci., 54, 127 (1961).
3785
ADSORPTION OF POLYSTYRENE-POLY (METHYL METHACRYLATE) MIXTURES
cal shaker. Displacement times < 0.3 hr could not be examined, since this much time was required to separate the adsorbent from the supernatant. Infrared Bound Fractions. The average fraction of segments on an adsorbed polymer chain which are bound to the surface ( p ) was calculated for each polymer from a shift in its characteristic infrared band which occurs upon adsorption. The bands used were (a) the carbonyl stretching frequency of PMMA which shifts from 1720 to 1703 cm-’, and (b) the out-of-plane C-C ring vibration of PS which shifts from 697 to 701 cm-’. The procedures*12involved adsorbing the polymer(s) on silica and forming a uniform suspension. This was transferred to a demountable fixed-thickness cell which was placed in the sample beam of the infrared spectrophotometer (Perkin-Elmer 337). A variable cell containing a known concentration of unadsorbed polymer was placed in the reference beam and adjusted by trial and error until the concentration of free polymer segments was exactly compensated. The fraction of bound segments, p , was calculated from the known concentration of adsorbed polymer, ca, and the measured concentration of bound segments, Cb. Stable suspensions necessary for determining infrared p values were formed using a ratio of 0.15 g of silica to 5 ml of CZHC12. In order to minimize errors arising from nonuniform suspensions, each reported bound fraction value represents an average of two to four infrared scans on the same sample. The measurement error associated with these repeated scans was estimated by calculating the standard tlevirttion, u, as suggested by Wilson.la Based on nine samples (32 observations), u = *0.025 for PS-1; based on ten samples (24 observations), u = 10.013 for PMRIA-5. These values of u reflect scatter in p associated with transferring uniform, stable suspensions into the infrared cell and recording the spectra.
Results and Discussion Equilibrium adsorption isotherms for each PS and PMMA polymer studied are shown in Figure 1. Included are representative infrared bound fraction values for PS-3 and PMMA-5 obtained from samples agitated >24 hr. The isotherms are typical for polymers, and have a steep rise in x/m at low values of cs followed by a long plateau region in which x/m is almost independent of cs. The values of p for PMMA-5 vary somewhat with surface coverage, particularly at x/m = 65 mg/g, where p appears to increase rather sharply from approximately 0.29 to 0.35. The high values of p reported here are similar to those observed previously when two bulk polymerized PMMA polymers were adsorbed on silica from chloroform.12 This implies that PMMA is adsorbed from both solvents by silica
k
50
i
0.28
0’23
I00
,
5:f!, 0.04
o.;2
0.08
o,;b
,
4; .o
0.20
EQUILIBRIUM C O N C E N T R A T I O N , c s , g i l 0 0 ml.
Figure 1. Equilibrium adsorption isotherms on silica from CzHCla a t 25”. The numbers on the graph are the values of p , the bound fraction, at that point: 0, PMMA-5;
0,PS-1; A, PMMA-1;
0, PS-2.
to form a highly compressed adsorbed polymer layer. In contrast to PMMA-5, bound fraction values for PS-1 were both lower and more variable. The trend toward lower values of p at higher surface coverages indicates that a more extended adsorbed structure is formed as the surface fills with adsorbed PS-1, and is consistent l2 with previous Rate of adsorption curves given in Figure 2 show that PMMA-5 and PS-1 reach their equilibrium adsorbance values within 1.5 and 0.5 hr, respectively, when suspensions are agitated on the mechanical shaker. PS-1 samples agitated by vigorous hand shaking reached essentially equilibrium adsorbances in 3 min, demonstrating that the rate of adsorption increases with agitation as La Mer and Healy reported.I4 The question of the extent to which the bound fraction data represent equilibrium values was examined by determining p as a function of agitation time. Representative data for PS-1 and PMMA-5 mixtures are shown in Table I1 together with data obtained for a (12) C. Thies, P. Peyser, and R. Ullman, Proceedings of the 4th International Congress on Surface Activity, Brussels, in press. (13) E. B. Wilson, Jr., “An Introduction to Scientific Research,” McGraw-Hill Book Co., New York, N. Y., 1952, p 245. (14) V. K. La Mer and T. W. Healy, Rev. Pure AppE. Chem. (Australia), 13, 112 (1963).
Volume 70, Number 12 December 1966
CURTTHIES
3786
-
1:1 FhChlA-I FS-2 MIXTURE
1.0
I.5
1.5
'L3
2.0
l B ( E . hrr.
Figure 2. Rate of polymer adsorption on silica from CzHCls a t 25" when suspensions are agitated on a mechanical shaker: 0, PMMA-5; 0,PS-1.
series of experiments in which PS-1 was adsorbed alone. The mixture results indicate that values of p for both polymers scatter in a random manner when samples are agitated 24 to 192 hr. As a rule, similar data were obtained when these polymers were adsorbed alone and agitated up to 400 hr. The only significant time-dependent effect on p observed throughout this study was that for the series of PS-1 samples included in Table 11. Major alterations in z/m or c, did not accompany this change. The explanation may lie in the relatively high value of cs (0.175-0.211 g/100 ml) here, relative to that of most systems examined. As the equilibrium concentration of unadsorbed polyTable 11: Effect of Agitation Time on Infrared Bound Fractions Agitation time, hr
(z/~)Ps, mg/g
(Z/~)PMMA,
mg/g
PPMMA
24.8 98.5 169.8
PS-1-PMMA-5 38.0 0.29 0.29 38.0 0.34 38.0
45.7 160.3
36.0 36.0
0.34 0.30
102.6 98.9
47.5 162.5
36.0 36.0
0.25 0.32
94.0 191.8
63.6 65.6
0.37 0.37
18.0 44.7 166.0 184,9
... ...
... ...
...
...
(dP8,
g/100 ml
Mixture" 98.3 0.029 93.4 0.047 87.8 0.052
PPS
0.19 0.17 0.14
0.101 0.107
0.15 0.15
45.0 45.0
0 0
0.16 0.21
91.2 92.5
0.108 0.101
0.18 0.19
118.6 125.3 123.2 121.1
0.211 0.176 0.175 0.181
0.24 0.21 0.13 0.13
Ps-1
...
...
For PMMA-5, cB = 0 in all cases.
The Journal of Physiml Chmistru
0.08
0.16
0.12
0.24
WEIGHT O F A D S O R B E h l
i.4:
u, e.
Figure 3. Equilibrium adsorption isotherms for 1: 1 (by weight) PMMA-PS mixtures. The solid and dashed curves represent PMMA and PS isotherms calculated assuming complete PMMA adsorption occurs before PS is adsorbed. Experimental points shown represent,: 0 , PMMA-5; 0,PS-1; m, PMMA-1: 0 , PS-2.
mer increases, it is possible that one may encounter relatively slow adsorption-displacement phenomena involving various molecular species which differ grossly in size. More extensive studies involving systems with high values of cS are needed. Equilibrium adsorption isotherms for 1: 1 (by weight) PS-PMMA mixtures are presented in Figures 3 and 4. The solid curves shown represent the calculated relationship between the cs values for PSIRIA and weight of adsorbent, while the dashed lines represent the same relationship for PS. These calculations involved two basic assumptions. (1) PR/IR!IA is completely adsorbed before any PS adsorbs and PS has no influence on the adsorbance of PMMA. (2) PS adsorbs once PMMA adsorption is complete and the presence of PMMA on the surface has no effect on the adsorbance of PS. Since the plateau region of each of the isotherms in Figure 1 extends to low values of cs, the relation between ce and w for each polymer was calculated using its appropriate adsorbance at surface saturation, (z/")*. As shown by the experimental points, PS adsorption does not occur until complete removal of PbIhIA from solution. The adsorbance of PMMA is unaffected by the presence of PS in the system, except at low values
ADSORPTION OF POLYSTYRENE-POLY (METHYL METHACRYLATE) MIXTURES
3787
0
I
L
3
4 TME, -_
WEIGHT OF ADSORBENT. w, P.
5 “20
40
60
80
100
hrn.
Figure 4. Equilibrium adsorption isotherms for a 1:1 (by weight) PMMA-PS mixture. The solid and dashed curves represent PMMA and PS isotherms calculated assuming that complete PhlMA adsorption occurs before PS is adsorbed. Experimental points shown represent: 0, PMMA-5; 0, PS-2.
Figure 5. Rate of adsorption of 1: 1 (by weight) PS-PMMA mixtures on silica from C2HC1, a t 25’. Curves I and I1 designate cases where PS is not adsorbed a t equilibrium; curve I11 is a case where it is. Experimental points shown represent: 0, A, 0, PS-1 ca values; W, A, 0 , PMMA-5 cs values.
of cs for PMMA. These deviations are well beyond the small error introduced in the calculated isotherms when ( ~ / mfor ) ~PMMA was assumed to extend to cs = 0. Once PMMA adsorption is complete, PS adsorption begins immediately. The measured values of ce for PS are somewhat greater than the calculated ones, but in all cases, the adsorbance of PS is nearly independent of PRIMA surface coverage. These results demonstrate that at equilibrium PMMA is preferentially adsorbed on silica from CiHC13 containing PS-PMMA mixtures. This reflects PRIMA’S stronger interaction with the silica surface arising from its more polar character and greater hydrogen bonding ability. Related to this finding is Fontana’s15 observation that the ester segments in an alkyl methacrylate-polyglycol methacrylate copolymer tend to be excluded from the silica surface while the polyglycol ether segments are preferentially adsorbed. Rate of adsorption curves for several 1: 1 (by weight) PS-PMMA mixtures are plotted in Figure 5. These show that equilibrium adsorbances in such systems are reached quickly. This is true whether PS is (curve 111) or is not (curves I and 11) adsorbed at equilibrium. Values of cs for PAIRSA decrease uniformly with agitation time until their equilibrium levels are reached. I n contrast, the first experimental values for PS on curves 11 and I11 suggest that ca for PS may increase with agitation time. Such a trend implies that some PS is adsorbed initially when PS-PMMA mixtures are added to silica and then is completely displaced from the interface by excess PMMA in the system. This would be
reasonable, since the rate data of Figure 2 indicate that PS-1 is adsorbed more rapidly than PMMA-5. However, more complete rate curves are needed to confirm this observation. They were not obtained since the procedure utilized to isolate the supernatant polymer solution from the silica required approximately 0.3 hr. The displacement of PS by PMMA was examined more fully by first adsorbing varying amounts of PS-1 on silica from C2HC1, and then adding excess PMRIA to the system. Values of cs for both polymers were measured as a function of time. The amount of PMMA added was calculated to give an equilibrium PMMA cs value of 0.128 g/100 ml, if complete PS displacement occurred. Results are given in Figure 6. The dashed lines represent the unadsorbed PS concentrations, cs, if PS displacement is complete. The (x/m)O values shown on the graph are the PS-1 adsorbances before addition of PMMA-5. The experimental points represent cs values measured as a function of displacement time. The solid and open experimental points are cS values obtained for samples equilibrated on the silica surface 3 min and 22 hr, respectively, before excess PMMA was added. The 3-min samples were shaken by hand, but control experiments established that they essentially reached their equilibrium adsorbances within this time. The data of Figure 6 establish that PMMA rapidly and completely displaces PS from silica under all conditions examined. The extent and rate of displacement (15) B. J. Fontana,
J. Phys. Chem., 67, 2360 (1963).
Volume 70,Number 1.2 December 1966
CURTTHIES
3788
0.40
0.25
i
*-++---.-
.--e---
c.05
0
----
(ximy.
121.5 mgip.
'xim:i 34'syg'
__
95
4 p
t '
bi
"
'
"
"
"
,,i
'
DISPIACEMENT TIME, Lrs.
Figure 6. Rate of PS displacement from silica at 25" by excess PMMA. Dashed curves represent unadsorbed PS-1 concentrations, cat calculated for complete PS displacement. (z/m)Ovalues on the graph are PS-1 absorbances before displacement. Experimental points shown represent: A, D, 0, X, PS-1 equilibrated on the surface 3 min before displacement; A, 0, PS-1 equilibrated on the surface 22 hr before displacement.
are essentially independent of PS surface coverage or length of time the PS was equilibrated on the surface before being displaced. Marked changes in the conformation of an adsorbed polymer molecule can occur as a function of these parameters. The fact that no major effects are observed demonstrates that any changes which do occur have little effect upon the displacement process, within limits of the experimental technique used. A trend toward longer displacement times at lower PS surface coverages suggests that PMMA has more difficulty displacing PS under these conditions. This is consistent with the increased values of p for PS-1 at lower surface coverages (see Figure l), since a greater number of polymer-surface attachments must be ruptured in order to effect displacement. However, the concentration changes involved are so small that more complete displacement rate data are required to confirm this trend. The complete displacement of PS by PMMA observed throughout this study demonstrates the reversibility of the PS adsorption process and is consistent with previous observations that preferential adsorption of more surface-active molecules can effect polymer l 9 have stressed displacement. l7 Eirich4 and that polymer adsorption is reversible under suitable conditions, showing that no irreversible polymer-surface bonds are formed. The comparatively rapid rate of PS displacement shows that PMMA molecules have the ability to rupture quickly and irreversibly all PSsurface bonds. This establishes that multiple surface The Journal of P h y s h l Chemistry
attachments characteristic of adsorbed polymers do not necessarily limit the extent of polymer displacement or cause it to be extremely slow, especially when molecular species which have a stronger affinity for the adsorbing surface are present. Significantly, PMMA rapidly displaces PS from silica even though PS and PIfMA are incompatible polymers. Dobry and Boyer-Kawenoki established that they are incompatible in while several qualitative experiments in this study established that they also are incompatible in C2CHC13. Except at low concentrations, these incompatible polymers form two liquid phases with the solute in each phase consisting of a preponderance of one polymer component and a small proportion of the other. Such incompatibility arises because polymers characteristically have small, but positive, heats of mixing accompanied by negligible entropies of mixing.*l All solutions used for the mixture adsorption experiments had initial PS and PIIMA concentrations well below the critical point of phase separation and hence were homogeneous. However, the concentration of PS segments in an adsorbed layer could be much greater than that in the original bulk solution, depending on the thickness of this layer and the distribution of segments in it. The concentration could be so great that most PMMA molecules would be excluded from the adsorbed layer due to incompatibility with the "PS-rich" environment. The fact that PhIRIA rapidly and completely displaces PS from silica in all cases examined establishes that PMhlA readily penetrates the adsorbed PS layer. Polymerpolymer incompatibility effects do not seem to affect the displacement process. Additional evidence that polymer-polymer incompatibility effects are small is provided by the data in Figures 3 and 4, which show that PS's adsorbance is nearly independent of the amount of PhlMA adsorbed on the surface. Interactions between PS and PJlhlA responsible for incompatibility are not sufficiently strong to exclude PS from the surface totally or markedly reduce its adsorbance. Another measure of the interactions between PS and PMMA is gained by determining values of p for both polymers when simultaneously adsorbed on silica. Such data are presented in Table 111, where most of the (16) J. Koral, R. Ullman, and F. R. Eirich, J . Phys. Chem., 62, 541 (1958). (17) S. Ellerstein and R. Ullman, J . Polymer Sci., 55, 123 (1961). (18) A. Silberberg, J. Phys. Chem., 66, 1872 (1962). (19) R. R. Stromberg, W. H. Grant, and E. Passaglia, J. Res. Nail. Bur. Std., 68A, 391 (1964). (20) A. Dobry and F. Boyer-Kawenoki, J . Polymer Sci., 2 , 90 (1947). (21) P. J. Flory, "Principles of Polymer Chemistry," Cornel1 University Press, Ithaca, N. Y., 1953, Chapter XIII.
ADSORPT~ON OF POLYSTYRENE-POLY (METHYL METHACRYLATE) MIXTURES
reported values of p represent averages of two to four determinations. The average errors given are mean deviations of p frcm the reported mean values. Comparisons of the values of p shown here with those reported in Figure 1 for cases where the individual polymers are adsorbed alone establish the following points. Table I11 : Infrared Bound Fraction Values, p , for PS-PNNA Mixtures” 7-PMMA-5--
Adsorbance,
anee. mg/gb
PS-1
,-
Adsorb-
P
es,
w / g b
g/100 ml
P
0 0.040 0.11
0.19f0.025 0.18 f 0.020 0.14 f 0.010
36.0 37.5 37.0
0.29f0.035 0.29 3: 0.023 0.32 4: 0.017
36.0 73.0 81.5
66.8 65.3 64.6 67.5
0.373ir0.025 0.38 3: 0,010 0.374ir0.0050 0.40 3: 0.0050
45.6 52.3 60.0 49.9
0 0.040 0.11 0.18
0.16 f 0.025 0.18 f 0.025 0.19~0.025 0.12 f 0.0050
104.0
0.37
29.9
0.237
0.13
156.3
0.33
4.0
0.433
...
a For PMMA-5, cB = 0 in all cases. of silica in system.
Based on total weight
1. Average values of p for PMRlA-5 are relatively unaffected by the simultaneous adsorption of PS-1 and PhI;1IA-5. Only at a PAIMA-5 adsorbance of approximately 65 mg/g does the presence of PS-1 cause an obvious change from p = 0.30-0.32 to 0.37-0.40. Variations in PS-1 concentration a t constant PM1C’IA-5 adsorbance have no significant effect on the value of p for PMAIA-5 over the range of PS-1 concentrations examined. 2 . Average values of p for PS-1 vary from 0.12 to 0.19 when it is simultaneously adsorbed with PMMA-5. For those mixtures having a PS-1 cs value >0, this range of p values approaches that observed when PS-1 is adsorbed alone and having cs > 0. Values of p for two mixtures where cs = 0 are somewhat below those observed when equivalent amounts of PS-1 are adsorbed alone. The relatively small changes in values of p for both polymers when simultaneously adsorbed on silica are undoubtedly due in part to the increased amounts of polymer adsorbed. For a given PS or PMMA adsorbance, the total weight of adsorbed polymer is greater in the mixture system than when either polymer is adsorbed alone. Since p for both polymers varies with surface coverage as seen in Figure 1, one would anticipate slight changes in the measured values of p .
3789
I n general, it appears that PS cannot compete with PR/IMA for available surface sites and must occupy only those adsorption sites that the PlSMA molecules are unable to fill. However, the PS does not undergo marked structural rearrangements in order to fill these sites. The PS and PMMA molecules occupy similar numbers of sites when adsorbed simultaneously as when adsorbed individually. In conclusion, it should be noted that this study was carried out under dynamic conditions and involved a very finely divided silica as adsorbent. Hence, the results obtained may be uniquely applicable to such systems. The adsorbent, Cab-0-Si1 M-5, is reported to have an average particle size range of 0.015-0.020 p and is formed by a pyrogenic process.22 Fully dispersed particles of this size approach the root-meansquare end-to-end dimensions of random polymer coils in solution. Assuming the silica actually was well dispersed to give this particle size range, the possibility exists that one or perhaps a few adsorbed polymer molecules effectively saturate the surface of each small adsorbent particle. Bridging of several particles might occur as La Mer and Healy have disc~ssed,’~ but still comparatively few molecules would be required to saturate the available surface area on a given particle or aggregate of particles. Alternatively, the PS and PMRZA could be fractionated in such a manner that certain adsorbent particles adsorb only PS molecules and others only PRIMA molecules. In both cases, a pronounced effect on the adsorbance and/or adsorbed structure of either polymer arising from polymer-polymer incompatibility might not be observed, since each polymer molecule is being adsorbed in an environment virtually unchanged from that encountered in systems where it is the only polymeric adsorbate present. Unfortunately, it is not known whether Cab-0-Si1 ever is completely dispersed into independent particles of 0.015-0.020-p diameter, since electron photomicrographs consistently yield aggregates of 50-100 individual particlesz2which conceivably could be chemically fused together. I n any case, additional studies involving adsorbents which are known to be essentially infinite planar surfaces should give more insight into this question. Both the small adsorbent particle size and dynamic nature of these experiments must be recognized when attempts are made to compare the results of this
(22) Technical Bulletin Describing Cab-0-Si1 M-5, Cabot Corporation, Boston, Mass. (23) See ref 2 for a recent review of these theories. More recent contributions include: C. A. J. Hoeve, E. A. DiMarzio, and P. Peyser, J . Chem. Phys., 42, 2558 (1965); R. J. Roe, ibid., 43, 1591 (1965); E. A. DiMarzio and F. L. McCrackin, ibid 43, 539 (1965); C . A. J. Hoeve, ibid., 44, 1505 (1966).
V o l u m e 70, N u m b e r 12
December I966
L. BENJAMIN
3790
study with theoretical treatments based on infinite adsorbing planar surfaces and static systems.23 Acknowledgment. The author gratefully acknowl-
edges Mr. J. A. Herbig's encouragement of this work as well as many stimulating discussions with Dr. Hans F. Huber.
Partial Molal Volume Changes during Micellization and Solution of Nonionic Surfactants and Perfluorocarboxylates Using a Magnetic Density Balance
by L. Benjamin Miami Valley Labm-atm-ks, the Procter & Gamble Co., Cincinnati, Ohw
(Received March 24, 1966)
A simple magnetic density balance is described for obtaining partial molal volume data at The data show that the standard volume change of micellization per mole, AVO,, is always positive and becomes increasingly so the longer the alkyl chain length of the dimethylalkylamine oxides (DC,AO). For these compounds AVO, approaches zero at -Ca chain length below which micelles do not form. It is inferred that a part of the alkyl chain near the head group retains its hydration in the micellar state. Solution of the fluorinated molecules studied is attended by more contraction than with their hydrogen counterparts and this leads to larger positive AVO, values.
25" for various nonionic surfactants and perfhorocarboxylates.
Aqueous solutions of compounds partially or totally hydrophobic in character often exhibit unusual thermodynamic properties associated with ordering of water molecules around the solute. Thus the unfavorable positive free energy of solution of, for example, hydrocarbons has favorable enthalpy contributions (hydrogen bond formation) but overriding negative entropy contributions from the resulting water structure. Although such unusual properties had been recognized for some time previously,' the classification of various solutes as structure makers and structure breakers in aqueous solution was first generalized by Frank and Evans. More recently, considerable interest has developed in entropic contributions arising from the breakdown of such water structure during hydrophobic bonding-the nonspecific interaction accompanying the transfer of hydrophobic groups from an aqueous to a less aqueous en~ironment.~-'~Volume The Journal of Physical Chemistry
changes associated with this process, all positive in nature, have been discussed.3*18-18 (1) J. A. V. Butler, Trans. Faraday sot., 33, 229 (1937) (2) H. 9. Frank and M. W. Evans, J . Clam. Phys., 13, 507 (1945). (3) W. Kaurmann, Advan. Protein Chem., 14, 1 (1959). (4) I. M. Klotr and S. W. Luborsky, J . Am. C h a . soc., 81, 5119 (1959). (5) H. A. Scheraga, G. Nbmethy, and I. Z. Steinberg, J . Biol. Chem., 237,2506 (1962). (6) G. NQmethy and H . A. Scheraga, J . Phys. Chem., 66, 1773 (1962). (7) E. D. Goddard, C. A. J. Hoeve, and G. C. Benson, ibid., 61, 593 (1957). (8) H. Schneider, G. C. Kresheck, and H. A. Scheraga, {bid., 69, 1310 (1965). (9) W. P. Jencks, Federation Proc., 24, Suppl. 15, S-50 (1965). (IO) C. Tanford, J . Am. Chem. soc., 84, 4240 (1962). (11) L. Benjamin, J. Phys. Chem., 68, 3575 (1964). (12) G. C. Kresheck and L. Benjamin, ibid., 68, 2476 (1964).
