Adsorption from Good Solvents

In the theory of Hoeve based on the loop-train model, at is correlated with the excluded volume derived from Eq. (B-34) by 

The theory of Jones and Richmond based on the self-consistent mean field theory predicts Eq. (B-122), which may be rewritten 

The values of (a, - l)/n are plotted against Mpt ) (1 - 0/T) in Fig. 18. The predicted linear relation is not obeyed and the measured values of (a, - l)/n are far smaller than the theoretical ones. Thus, we find that the excluded volume effect formulated by Jones and Richmond is entirely irrelevant. 

The smaller expansion factor predicted by the theory of Hoeve originates from neglecting the tail portions of the adsorbed polymer chain, while the larger expansion factor predicted by Jones and Richmond is due to their failure of correctly evaluating the elastic free energy, as has been pointed out by Kawaguchi and Takahashi74.  

Killmann and Wiegand77) investigated the adsorption of poly(ethyiene glycol) and poly(vinyl pyrrolidone) on a chromium plate from water and methanol solutions. The value of tnns for poly(ethylene glycol) (Mw = 40 x 103) in water came dose to the root-mean-square end-to-end distance of the sample in bulk solution, whereas t for polyethylene glycol) (M = 6.13 x 103) in water and that for poly(vinyl pyrrolidone) (Mw = 38 X 103) in water or methanol were much larger than those in bulk solutions. No comparison between theory and experiment was made. 

---

The theoretical description of excluded volume effects on the adsorption from good solvents is still unsatisfactory. The scaling theory for polymer adsorption has not yet been subject to experimental tests. 

For polymer adsorption from the poor solvents, when the macromolecule volume is small, the number of macromolecules adsorbed on the unit surface area is greater than the adsorption from good solvents. A change of the solvent from poor to good leads to an increase in the volume of adsorbed macromolecules and partial desorption of macromolecules occurs. 

The study of the system PDMS-silica-cyclohexane provides the first direct evidence for the de Gennes description of polymer adsorption from good solvent. A difficulty of this system is, however, that the contrast between PDMS and silica is relatively small. 

De Gennes24,54) investigated the adsorption of a flexible polymer on a flat surface from good solvents using the scaling theory. Three different regimes, i.e. dilute, semidilute, and plateau regions, were considered. 

The basic ideas of the scaling theory for homogeneous polymer solutions have also been used to set up a framework for a theoiy describing the adsorption of polymers from good solvents. The aim Is to derive power laws for q> z), valid in certain regimes in most cases numerical coefficients are ignored. So far, the model has only been formulated for weak adsorption, l.e., small x - We shall treat some more details of this model in the following subsection (5.4c). 

One may wonder whether adsorption from selective solvents leads to more dense brushes. As was pointed out above, this implies that the anchoring blocks attract each other so that they are prone to form self-assembled structures. One therefore has to bear in mind that not only does adsorption take place from a micellar solution, but also that there is a good chance that these micelles, once they are on the surface, remain micelles so that inhomogeneous adsorption layers are produced. Examples of adsorption studies of diblock copolymers from selective solvents are those by Marra and Hair [46] (polystyrene-poly(ethylene oxide) from heptane/toluene on mica). 

It was found that PE can be adsorbed on specific zeolites from some polar nonsolvents as well as from good solvents, such as decalin or TCB which are typically used for SEC of polyolefins. Full or partial adsorption of PE and isotactic PP on different column packings were found see a summary in Table 4 [143]. The pores of some zeolites, e.g., SH-300, correspond to the diameter of PE in its linear conformation but are smaller than the corresponding diameter of PP. Thus, PE but not PP may be adsorbed into the zeolite SH-300 from a specific solvent, see Table 4. The adsorption of PP requires a zeolite with larger pore diameter, e.g., zeolite CBV-700. NMR measurements confirmed the penetration of PE into the... 

Figure 6 gives a comparison between experimental and theoretical results for adsorption of homodisperse polystyrene from 0 solvent and from good solvent. The agreement between theory and experiment is quite good. 

Usually," " more polymer is adsorbed from poorer solvents. Variation in solvent-substrate interactions can complicate this simple finding, which was predicted theoretically by Scheutjens and Fleer. In good solvents, repulsion between segments e.g. in tails) of adsorbed polymers will be greater than in poor solvents, leading to lower adsorption in the former. The lower adsorption in good solvents has been wrongly attributed to less-extended conformations. 

The adsorption transition also shows up in the behavior of the chain linear dimension. Fig. 6(a) shows the mean-square gyration radii parallel, i gl, and perpendicular, to the adsorbing plate. While these components do not differ from each other for e for e > ej i g strongly increases whereas Rh decreases. In the first case (non-adsorbed chain) oc oc N as a dilute solution in a good solvent in the bulk. For adsorbed chains R /N 0 for oo because the thickness is finite it is controlled by the distance e- e from the adsorption threshold, but does not diverge as N oo. The adsorbed chain follows in fact a... 

As an example of composite core/shell submicron particles, we made colloidal spheres with a polystyrene core and a silica shell. The polar vapors preferentially affect the silica shell of the composite nanospheres by sorbing into the mesoscale pores of the shell surface. This vapor sorption follows two mechanisms physical adsorption and capillary condensation of condensable vapors17. Similar vapor adsorption mechanisms have been observed in porous silicon20 and colloidal crystal films fabricated from silica submicron particles32, however, with lack of selectivity in vapor response. The nonpolar vapors preferentially affect the properties of the polystyrene core. Sorption of vapors of good solvents for a glassy polymer leads to the increase in polymer free volume and polymer plasticization32. 

In the range of 15-30% acetonitrile, the different components of the mixture were separated according to size, in a size-exclusion mode, with very little nonspecific adsorption and good peak symmetry. It was believed that the addition of TEA to the solvent contributed to these desirable effects by neutralizing residual silanol groups. Present supports should not suffer from the presence of such uncapped silanol groups. 

Several reviews (e.g., Takahashi and Kawaguchi, 1982 Cohen Stuart et al, 1986) discuss the extensive experimental data and the predictions of the SF model for the variation of polymer molecular weight. In general theta solvent, but is less in a good solvent, and appears to be limited at high r. Proper selection of parameters in the model produces excellent agreement with experimental data (Fig. 13) for both [Pg.173]

The same procedure has been applied to the data shown in Fig. 3.40 (solid line 3) though Eq. (3.87) strictly holds only for neutral chains in good solvent [248], It this case all points that cannot be accounted for by the DLVO-theory have been included. The resulting h = 186 A is larger than the thickness h of the adsorption layer. It may be speculated that this ensues from the diffuse double layer interactions omitted in this treatment. 

Figure 5.26. Adsorption isotherms for various molecular weights. Diagram (a) shows theoretical data from SF theory the chain length N for each curve is indicated. Solid curves apply to a -solvent [x = 0.5) and dashed curves to a good solvent ( t 0). Parameters = 1> hexagonal...

---