Dyer / Hudson / Williams Progress in Ion Exchange

Advances and Applications
1. Auflage 1997
ISBN: 978-1-84569-865-2
Verlag: Elsevier Science & Techn.
Format: EPUB
Kopierschutz: Adobe DRM (»Systemvoraussetzungen)

E-Book, Englisch, 510 Seiten

This volume is a record of a conference, which was the fourth in a series held at NWEI, in Wrexham. It brought together scientists with interests in the broadly based subject of ion exchange, with the aim to cover aspects of its application as well as advances in the theory of ion exchange.

Autoren/Hrsg.

Dyer, A.

Hudson, M J

Williams, Peter A.

Leseproben

POLYMER RESINS - SYNTHESIS AND STRUCTURE

D.C. Sherrington,     Department of Pure and Applied Chemistry, University of Strathclyde, 295 Cathedral Street, Glasgow G1 1XL

1 INTRODUCTION

Although styrene-divinylbenzene resins, and particularly the ion exchangers derived from them, have been readily and widely available for over thirty years, their further development, characterisation and exploitation continues unabated1 in an increasing number of fields. This has been stimulated in particular by the drive to produce cleaner processes and hence meet increasingly strict environmental demands. The present paper will describe work primarily from the author’s own laboratory and will cover results from the microscopic characterisation of conventional resins in the “wet” state, data from a state-of-the-art solid state 13C NMR study of chloromethylated resins, recent developments of reactive resins containing epoxide, thiirane and phenolic functions, data from work on improving the capacity of chelating ion exchange resins via the use of a functional comonomer, and a progress report on the development of resins with extremely high thermo-oxidative stability. Time limitation will prevent any details of applications being given, but an indication of areas of exploitation will be included as appropriate.

1.1 Macroporous Polystyrene/Divinylbenzene Resins

These spherical particulate materials typically 100-1000 µm in diameter are prepared by suspension polymerization methodologies.2 A great deal of work has been carried out to try and quantify the detailed morphology of these porous species, and to relate this to the conditions employed in polymerization. Seminal reviews have been published by Albright3 and Guyot4 and a generally accepted model is shown in Figure 1.

Figure 1 Schematic representation of the structure of a macroporous resin.

Despite this it has proved very difficult to relate the performance of a resin (e.g. as a hydrophobic sorbent5) to the conditions used in resin synthesis, and indeed to resin parameters such as surface area, pore size, pore volume, etc. Undoubtedly one of the problems has been that resin morphology is generally characterised using “dry” resins, whereas in practice resins are used in the “wet” state. As a result there has always been a suspicion that morphological changes might occur, even when a “rigid hydrophobic” sorbent is hydrated, and hence the “wet” performance correlation with “dry” parameters might be expected to break down. Manufacturers of resins have known for some time that solvent treatment can influence morphology even after polymerization is complete, and now this “rearrangement” of the porous structure has been quantified.6

We have recently used state-of-the-art election microscopic and image analysis techniques to evaluate resin ultrastructures in both the “dry” and “wet” state.7 Table 1 shows a matrix of styrene-divinylbenzene resins prepared with high levels of crosslinker and a “good” solvating porogen, toluene, in order to produce high surface area resins. Transmission electron micrographs (TEM) were obtained on unstained microtomed sections (70 nm) using contrast enhanced procedures on a Zeiss 902 electron microscope at 80 kV.

Table 1

Feed Composition and Surface Areas of Resin Sorbentsa)

PS55X

0.5T

0.5

103

1T

p-, m- 55

45

1

609

2T

2

655

3T

3

759

PS80X

0.5T

0.5

561

1T

p-, m- 80

20

1

587

2T

2

738

3T

3

870

PS100X

0.5T

0.5

530

1T

p-, m-

0

1

370

2T

100

2

450

3T

3

487

PS20X

1ET

P-, m- 20

16b)

lc)

68

a)From N2 sorption, BET method;

b)styrene 64 vol%;

c)2-ethylhexanol

Specimens were prepared in three ways: a) vacuum-dried and embedded; b) freeze-dried from the wet frozen state and embedded; and c) sectioned directly in the wet frozen state with no embedding. These samples are designated “dry”, “freeze-dried” and “frozen”, respectively (see reference 7 for more details). Images for pore structure analysis were collected at 20,000 magnification on a video camera and then subjected to non-sophisticated image analysis procedures. Figure 2 shows the pore profile parameters used to characterise quantitatively each pore examined. The pore profile cross-sectional area A, should correlate most closely with conventionally determined pore volume data; the pore profile perimeter parameter, p, with conventionally determined pore surface area; and the pore profile diameter, d, with conventional pore diameter.

Figure 2 Resin sorbent pore profile cross-sectional parameters: A = area; d = maximum diameter; w = maximum width; p = perimeter.

Figure 3 shows a histogram in which the sum of pore profile areas (µm2) (= pore volume) is correlated with resin composition data. From this it is clear that the pore volume, qualitatively seen to increase in the photomicrographs (see reference 7), does indeed do so as the volume of porogen is increased 0.5 ? 3 in the 55 and 80% crosslinked series. This agrees with earlier findings.8 However, the ~100% crosslinked series is quite anomalous with the pore volume falling as the level of porogen is increased. This is rather difficult to explain. Clearly in the 55X and 80X series the morphology seems to evolve according to the model detailed by Guyot4 i.e. crosslinked nuclei are formed at low conversion and interbonding occurs between these as polymerization continues. As the proportion of solvating porogen is increased interbonding is progressively reduced and the final sorbent possesses a larger total pore volume. A similar trend is seen in the average pore diameter and the morphology in the 55X and 80X series might be regarded as forming under thermodynamic control. The situation with the 100X series (Figure 3) is quite different, however, and with the lowest level of porogen (0.5 toluene) it seems that the very high level of divinylbenzene gives rise to the very rapid generation of a highly rigid, strained and dense matrix of crosslinked polymer chains even at low conversion, which quickly “locks-in” a well defined pore structure. Indeed, the situation probably corresponds quite closely to the case when a precipitating porogen is used, and the pore structure arises from kinetic rather than thermodynamic control. As the proportion of toluene porogen is increased the whole process of pore formation is probably increasingly delayed by more extensive solvation, and as a result a more uniform evolution of nuclei and interbonding occurs, with a closer adherence to the Guyot model.4 In many respects therefore the matrix formed with 100% p-divinylbenzene and volume ratio of toluene porogen of 0.5 has similarities with Davankov’s hypercrosslinked resins.9

Figure 3 Total pore profile areas for dry and freeze-dried resin samples.

Figures 3 also show the changes in pore volume determined for the “freeze-dried” resins and in most cases there is clear evidence for these “hydrophobic” resins swelling significantly when wet. The effect is largest for the species with the larger “dry” pore volumes, and in the extreme cases total pore volume increases by ~40%. Almost certainly therefore hydration allows considerable internal adjustment to the morphology, probably via plasticization of polymer chains and, in particular, the relief of steric strain. Again the effects are probably related closely to those seen on hydration of hypercrosslinked resins, when swelling is readily observed as a macroscopic change.9 With such changes...