Desorption Behavior of (Terminal-Functionalized) Polystyrenes from

Using this method, the desorption behavior of preadsorbed polystyrene (PS) and polystyrene (PS-X) functionalized by a terminal iminium ion (X) from α...
0 downloads 0 Views 181KB Size
Langmuir 1996, 12, 2015-2018

2015

Desorption Behavior of (Terminal-Functionalized) Polystyrenes from Alumina Surfaces. Studied by the Continuous Elution Method Ming-Hua Chen and Kunio Furusawa* Department of Chemistry, University of Tsukuba, Tsukuba, Ibaraki 305, Japan Received December 20, 1994. In Final Form: January 2, 1996X To explore the desorption behavior of polymers from the solid surfaces, a new experimental technique was introduced in this research. We referred to it as the “continuous elution method”. Using this method, the desorption behavior of preadsorbed polystyrene (PS) and polystyrene (PS-X) functionalized by a terminal iminium ion (X) from R- and γ-alumina surfaces was investigated. The particles preadsorbed by PS or PS-X were packed into a small column and eluted by various solvents. Irreversibility of polymer adsorption can be observed clearly, the adsorption/desorption amount can be determined, and much new information on kinetics of desorption, adsorption energy, solvency effect, and etc. was also obtained. On R-alumina, it was found that only the X-group is the driven force for adsorption, and the adsorbed layer has a higher energetic state than that on γ-alumina. A complicated desorption kinetics of polymer desorbed from γ-alumina was observed, which has a relation to the porosity of adsorbent.

Introduction Interest in polymers at interfaces has been steadily growing during the last few decades. Polymers may be considered as tools to manipulate the interfaces. Such modified interfaces are very important in many industrial products and technologies. Adhesives, composites, dispersants, or flocculents are just a few examples. The mechanisms of polymer adsorption on solid surfaces are not yet completely clear, even though various theories have been proposed. Generally, the adsorption process of a polymer is considered to be dependent on the properties of the polymer, adsorbent surface, liquid medium, and other kinetic factors.1 Theoretical and experimental studies on polymer adsorption have been carried out extensively,2-5 but the desorption behavior has not been studied in detail, even though it is considered that polymer adsorption is generally an irreversible process. Recently, some papers concerned with polymer desorption using a displacer have been published.6,7 It is apparent from such studies on the desorption process of adsorbed polymers that new information on polymer adsorption, such as the structure of adsorbed polymer layer, adsorption strength, adsorption kinetics, etc., can be obtained. It is also reported that the terminal-functional group of a polymer plays an important role in the adsorption process.8-11 So, the manner in which the terminal-functional group affects the desorption behavior is a very interesting problem. * To whom correspondence should be addressed: tel, 81-29853-4464; fax, 81-298-53-6503; e-mail, [email protected]. tsukuba.ac.jp. X Abstract published in Advance ACS Abstracts, March 15, 1996. (1) Napper, D. H. Polymer Stabilization of Colloid Dispersions; Academic Press: London, 1983. (2) Scheutjens, J. M. H. M.; Fleer, G. J. J. Phys. Chem. 1979, 83, 1619; 1980, 84, 178. (3) de Gennes, P. G. Adv. Colloid Interface Sci. 1987, 27, 189. (4) Van de Ven, T. G. M. Adv. Colloid Interface Sci. 1994, 48, 121. (5) Kawaguchi, M.; Takahashi, A. Adv. Colloid Interface Sci. 1992, 37, 219. (6) Cohen, M. A.; Fleer, G. J.; Scheutjens, J. M. H. M. J. Colloid Interface Sci. 1984, 97, 515, 526. (7) Dodson, P. J.; Somasundaran, P. J. Colloid Interface Sci. 1984, 97, 481. (8) Kawaguchi, M.; Kawakabayashi, M.; Takahashi, A.; Nagata, N.; Yoshioka, A. Colloids Surf. 1990, 48, 363. (9) Singh, N.; Karin, A.; Bates, F. S.; Tirrell, M.; Furusawa, K. Macromolecules 1994, 27, 2586. (10) Frantz, P.; Leonhardt, D. C.; Granick, S. Macromolecules 1991, 24, 1868.

0743-7463/96/2412-2015$12.00/0

Table 1. The Properties of Polymer Samples sample

M h w × 10-3

M h w/M hn

PS0.9 PS5.7 PS40 PS200 PS498 PS1800 PSX5.7 PSX40 PSX200

0.9 5.7 40 200 498 1800 5.7 40 200

∼1.05

chemical structure

∼1.05

In this work, we introduce a new experimental technique to explore the desorption behavior of polymers from solid surfaces. We call this the “continuous elution method”. By this method, we studied desorption behaviors of polystyrene and terminal-functionalized polystyrene from an alumina surface in order to investigate the adsorption/ desorption mechanism and the adsorption/desorption characteristics of the terminal-functionalized polymer. These experiments also show what the new method can do. Experimental Section Materials. The monodisperse polystyrenes (PS) and their terminal-functionalized ions (PS-X) were supplied by Nihon Zeon Co., Ltd., Japan. PS-X was synthesized by the anionic polymerization technique and one end of the chain terminates with an iminium cation.12-13 The characterization data of the polymer samples are shown in Table 1. Two kinds of alumina powders, R-Al2O3 and γ-Al2O3, were used as adsorbents. These were supplied by Sumitomo Chemical Co., Ltd., Japan, and were used without any additional treatment. Some properties of these adsorbents are listed in Table 2. Both types alumina particles were made from aluminum oxide gel. The first was prepared by heating at a temperature lower than 400 °C and was dubbed γ-Al2O3. It has a porous nature and highly specific surface areas. γ-Al2O3 was considered to be covered with hydroxyl groups and has both Lewis acid and base (11) Kumacheva, E.; Klein, J.; Pincus, P.; Fetters, L. J. Colloids Surf. 1994, 89, 283. (12) Furusawa, K.; Ogawa, T.; Itabashi, T.; Kitahara, S.; Watanabe, H. Kobunshi Ronbunshu 1992, 49, 915. (13) Furusawa, K.; Ogawa, T.; Itabashi, T.; Miyahara, T.; Kitahara, S.; Kawanaka, T. Colloid Polym. Sci. 1994, 272, 1514.

© 1996 American Chemical Society

2016 Langmuir, Vol. 12, No. 8, 1996

Chen and Furusawa

Figure 1. Schematic diagram of continuous elution method: E, eluent; P, pump; H, heater (controlled at 35 °C); C, column; D, detector (UV); W, waste; R, recorder. Table 2. Characteristics of Alumina Powders sample

particle size (µm)

surface area (m2/g)

mean pore diameter (nm)

R-alumina (R-Al2O3) γ-alumina (γ-Al2O3)

50 50

5.7 273

167 17

properties. The second was prepared from aluminum oxide gel by drying at about 1200 °C and was dubbed R-Al2O3. The surface hydroxyl groups of R-Al2O3 were decreased by dehydrolysis and the surface has only Lewis base properties. Adsorption and Desorption Experiments. Each polymer was adsorbed on the alumina powders by the usual depletion technique in cyclohexane solution.12,13 Alumina powder and a cyclohexane solution of polymer were added in a glass tube (10 mL in volume) and the tube was agitated slowly in a water bath maintained at 35 °C (the θ condition of polystyrene in cyclohexane). After some adsorption time (0-120 h), the adsorbent particles covered by the polymer layers were packed into a Teflon tube (20 cm, 0.8 mm i.d.) by a slurry method. In this process, some quantity of polymer solution was also transferred into the Teflon column. The schematic diagram of the “continuous elution method” is shown in Figure 1. After setting the column into the elution system, an eluent was pumped through the column, where a micropump with a stable flow rate (∼6 µL/min) was used. The concentration of polymer components in the liquid that flowed out from the column was analyzed by a UV detector (λ ) 260 nm) and recorded as an absorbance-time curve. Selection of Eluent and Elution Procedure. As a first step (step I), elution with pure solvent (cyclohexane) was conducted at 35 °C, because the column contained, both preadsorbed polymers on adsorbent and the free polymer in solution. Only the free polymers in the column were drained out by flowing cyclohexane for a few hours. As a second step (step II), a desorption process of preadsorbed polymer was conducted using a mixed solvent of cyclohexane and chloroform (good solvent). From the peak area of this elution curve in step II, the desorbed amounts of preadsorbed polymer were calculated using a calibration curve. As an eluent in step II, a mixed solvent of φcf ) 0.25 was usually employed. Here φcf is the volume fraction of chloroform in the mixture, i.e., φcf ) chloroform/(chloroform + cyclohexane). In some cases, step II was further divided into substeps, i.e., step IIa, IIb, etc., where the φcf was increased step by step, for detail examination of the effect of solvent power on desorption behavior. As the third step (step III), the elution was conducted by using pure chloroform (φcf ) 1.0) for checking whether the desorption was complete or not. After all these processes, the column was dried at 70 °C and the amounts of adsorbent were weighed to calculate the amount of desorbed polymer.

Results and Discussion Elution Behavior and Adsorption State from r-Al2O3. Figure 2A shows some typical elution curves of PS-X from the R-Al2O3 surface. After the process of step I, the polymer remaining in the column was the preadsorbed polymer. So the peak that appears at step II is attributed to the desorption of the polymer. At step III, which used a better solvent than that at step II, no peak

Figure 2. Typical elution curves: abscissa, elution time (corresponds to elution volume); ordinate, absorbance (UV, corresponds to concentration of polymer), arbitrary unit. φcf is volume fraction of chloroform in eluent. φcf ) chloroform/ (chloroform + cyclohexane) (v/v).

appeared, meaning that all the preadsorbed polymer had desorbed at step II. From the form and area of the peak at step II, some desorption kinetics behavior and (pre)adsorbed amount can be obtained, these will be discussed later. As mentioned above, Figure 2A clearly shows that the adsorbed polymers PS-X cannot be desorbed at step I, even if they were displaced by a pure cyclohexane for a long time (at lab. scales). This means that preadsorbed PS-X from the cyclohexane solution cannot be desorbed under the same solvent conditions at least at the macroscopic level; i.e., in the present adsorption conditions, an irreversible adsorption of polymers has taken place. Similar results are reported in previous publications.3,14,15 This irreversibility of PS-X is also confirmed by another method; i.e., after adsorption of PS-X under a definite solution condition, the supernatant solution was displaced several times by a pure solvent (cyclohexane). Then, the polymer-coated R-Al2O3 powders were packed into the column by the slurry method. In this case, no peak was detected in step I as showing in Figure 2A. However, in step II, the same peak area as found in the original step II can be detected. Therefore the washing process will not influence the amount of adsorption even in the column inside. However, the washing process in column inside as the step I, is more convenient and can easily detect whether the washing of free polymer is complete by using an UV detector. The amounts of desorption (ΓD) for the samples of PS and PS-X with different molecular weight were calculated from the peak areas (step II), considering the elapsed time of adsorption. These results are shown in Figure 3. If we regard the amounts of desorption (ΓD) to be equal to the amounts of adsorption at the respective elapsed time, Figure 3 can be viewed as the usual adsorption rate curves. It is evident from Figure 3 that when R-Al2O3 powders are used as the adsorbent, the amount of adsorbed polymer is significantly enhanced by the existence of the endfunctional group on the polymer chain. In fact there is nearly zero adsorption in the PS-R-Al2O3 system (this is why we only present the curve of PS-X-R-Al2O3 in Figure 2A). This result indicates that the end-functional group (iminium ion) on the polymer chain end plays a dominating role in the adsorption of PS-X on the R-Al2O3 surface. (14) Terashima, H. J. Colloid Interface Sci. 1988, 125, 444. (15) Dijt, J. C.; Cohen Stuart, M. A.; Fleer, G. J. Macromolecules 1992, 25, 5416.

Desorption Behavior

Figure 3. Desorbed amounts of PS and PS-X from R-Al2O3 at various adsorption times: b, PS-X200; 2, PS-X40; 9, PS-X5.7; O, PS200; 4, PS40; 0, PS5.7.

Figure 4. Profiles of elution curve (step II in Figure 2) affected with molecular weight of polymers in γ-Al2O3-PS adsorption systems (preadsorption time, 100-120 h; Rg, radii of gyration of PS in cyclohexane at 35 °C, the mean pore diameter of γ-Al2O3 is 17 nm): abscissa, elution time; ordinate, absorbance (UV), arbitrary unit.

Furthermore, it is seen from the figure that the desorbed amount (ΓD) is not significantly influenced by the molecular weight of the polymer. This reveals that the adsorption layer has a monolayer structure where the polymer coils are packed closely with each other. Elution Behavior and Adsorption State from γ-Al2O3. As can be seen from Figure 1B, the elution curves of polymer (both PS and PS-X) desorbed from γ-Al2O3 show usually a complicated form, consisting of a narrow initial peak followed by a broad plateau. Presumably, this differs with the case of R-Al2O3, due to the porous nature of γ-Al2O3. The pore diameter of γ-Al2O3 is ∼17 nm and roughly has the same order of the diameter of the PS200 coil (∼27 nm) in cyclohexane at 35 °C. To clarify the effect of pore size of γ-Al2O3 on the elution profile of polymer, systematic experiments using the various PS samples with different molecular weights have been conducted. As seen from Figure 4, the profiles of elution curves in step II do depend strongly on the molecular weight of PS, showing the strong effect of porosity geometry of γ-Al2O3; i.e., when polymer size is appreciably smaller than the pore size (Figure 4a,b), one single peak without the following plateau appeared. When the polymer size is roughly equal to the pore size of

Langmuir, Vol. 12, No. 8, 1996 2017

Figure 5. Desorbed amounts of PS and PS-X from γ-Al2O3 at various adsorption times: b, PS-X200; 2, PS-X40; 9, PS-X5.7; O, PS200; 4, PS40; 0, PS5.7.

adsorbent (Figure 4c,d), the curves show an initial narrow peak followed by a plateau similar to that seen in Figure 1B. Further, when the polymer size becomes appreciably larger than the pore diameter (Figure 4e,f), the elution curve has a single broad peak without an initial narrow peak. All these results indicate (it seems more likely) that the different elution profile in γ-Al2O3-PS series can be related to the size ratio of the diameter of PS coils to the pore diameter of adsorbent particles. The geometric effect between them plays an important role on the elution process of adsorbed polymer. The reader is reminded that our method differs from GPC, in which the polymer is added to a prepacked column, rather than preadsorbed. Figure 4 suggests the existence of a specific γ-Al2O3-polymer interaction, whose strength might be influenced strongly from the geometrical effect between the pore size of adsorbent and the radius of polymer coil in solvent. Some polymers having radii of gyration roughly equal or larger than the pore size of γ-Al2O3 will be introduced with some degree into the empty pores by their strong entropic reason. It is assumed that the polymer chains introduced into the narrow pore have taken an elongated and flat conformation on the pore surface. Therefore, the elution curve in such a system becomes a complicated profile and the longer elution times will be necessary for these polymer molecules. The detailed effect of porosity of adsorbent was investigated by using a well-defined porosity adsorbent (controlled pore glass beads, CPG-10) and was submitted separately.16 The desorption amounts from γ-Al2O3, which were calculated from the peak area (step II) of respective elution curves, are plotted in Figure 5. It is found that the amount of adsorption/desorption (ΓD) at the long-time saturation level increased gradually with the increasing molecule weight of the polymer. The existence of the end-functional group hardly influences the amount of adsorption, indicating that the polymer adsorption progresses by the usual loop-train model. Driving Force for PS and PS-X Adsorption on Alumina. It is known that as a driving force for polymer adsorption, Lewis acid-base (electron acceptor-donor) interaction plays an important role. This concept was introduced originally by Fowkes,17 and our present data on polymer adsorption/desorption (in Figure 3 and Figure 5) have been analyzed using this concept. There are Lewis base sites (B) on the R-Al2O3 surface, and on the γ-Al2O3 surface there are both the Lewis acid (16) Chen, M.-H.; Furusawa, K. Submitted to colloids surf. (17) Fowkes, F. M. J. Adhes. Sci. Technol. 1987, 1, 7.

2018 Langmuir, Vol. 12, No. 8, 1996

Figure 6. Schematic picture showing the Lewis acid-base interaction between polymer and adsorbent, and the structure model of adsorbed polymer layers: a/B, interaction between terminal-functional group and basic site of surface; A/nb, interaction between acidic sites and phenyl group of polymer.

and base sites (A+B). Meanwhile, the phenyl group in PS and PS-X and the iminium cation in PS-X are the base and acid sites, respectively. Therefore, PS has n base sites (nb) and PS-X has n base sites and one acid site (nb+a) in a single polymer molecule, where n means the number of repeating units in a single polymer. Figure 6 shows the acid-base interaction models schematically. The binding interaction operates predominantly between X in PS-X and base sites on R-Al2O3 (a/B) and between phenyl groups in PS (or PS-X) and acid sites on γ-Al2O3 surfaces (A/nb). On the basis of these considerations, we can understand that the R-Al2O3 cannot adsorb PS but can adsorb PS-X through the anchored action of the endfunctional group (X) on the solid surface. On the γ-Al2O3 surfaces, however, both the PS and PS-X molecules can be adsorbed by the interaction between the phenyl groups included in the respective molecule and the solid surface. Here it should be noted that cyclohexane used as the solvent does not compete as an adsorption component. Effect of Solvency of Eluent. In above studies, we selected the eluent φcf ) 0.25 as a good solvent, which is good enough to desorb the preadsorbed polymers completely as shown in Figure 1. If the solvent power of the eluent is strengthened gradually from φcf ) 0 to φcf ) 1.0 step by step to elute the same column, the elution curves obtained will give us other information on the adsorption energy of polymer layer. Figure 7 shows the accumulated fraction of desorbed polymer (Σf) at each step plotted against the component fraction φcf of eluent employed at the respective step. It is well-known that, for polystyrene polymer, the solvation power of eluent of the present binary mixture solvent (chloroform + cyclohexane) increases linearly with increasing the φcf value. Figure 7 shows firstly that the Σf increases about linearly with increasing φcf, the solvent power of eluent used. It indicates that in a polymer adsorption/desorption process, adsorbed amount is inversely proportion to the solvent power of the solvent used.

Chen and Furusawa

Figure 7. Relation between the fraction of accumulated desorbed amounts (Σf) and the fraction of chloroform in mixed eluent (φcf): 9, γ-Al2O3/PS-X200; 0, γ-Al2O3/PS200; 2, R-Al2O3/ PS-X200.

Secondly, Figure 7 shows that the slope (dotted line) of Σf vs φcf plot in the PS-X-R-Al2O3 system is smaller than that in PS- and PS-X-γ-Al2O3 systems. In order to get complete desorption, the solvent power of eluent needs to be about φcf ) 0.25 and 0.12, respectively. These indicate that the polymers adsorbed with different energetic states; i.e., the PS-X on R-Al2O3 is adsorbed more strongly than the PS (X) on γ-Al2O3, due to the strong acid-base interaction between the end group of PS-X molecule and the surface of R-Al2O3. Figure 7 also shows that on the γ-Al2O3 surface, the end-functional group has no clear effect on the plot of Σf vs. φcf. It agrees with the result of Figure 5. Conclusion The “continuous elution method” introduced in this research is valuable for studying the desorption behavior of preadsorbed polymer on the surface of solid particles and can supply some new information. According to this new method, the irreversibility of polymer adsorption can be clearly observed, the adsorption/desorption amount can be determined, and new information of kinetics of desorption, adsorption energy, and solvency effect, etc. can also be obtained. The desorption behavior of preadsorbed polystyrene (PS) and polystyrene (PS-X) functionalized by a terminal iminium ion (X) from R- and γ-alumina surface was investigated. It was found that, on R-alumina, only the X group is the driving force for adsorption, and the adsorbed layer has a higher energetic state than that on γ-alumina. A complicated desorption kinetics of polymer was observed from γ-alumina, which has a relation to the porosity of adsorbent. The solvency of medium has a strong effect on an adsorption/desorption behavior. All of these supply us with a new direction of more systematic studies. LA941024K