Studies on the Selectivity of Porous Methacrylate ... - SAGE Journals

523

Studies on the Selectivity of Porous Methacrylate Polymers† B. Gawdzik* and M. Maciejewska Faculty of Chemistry, Maria Curie-Sk l/ odowska University, pl. Marii Curie/ Sklodowskiej 3, 20-031 Lublin, Poland.

(Received 3 December 2001; accepted 8 March 2002)

ABSTRACT: Four types of porous methacrylate polymers were synthesized as stationary phases for gas chromatography. The influence of the chemical structure of the monomers used in the synthesis on the selectivities of the resulting polymers was studied. Two procedures were applied to determine the selectivities of the copolymers: the selectivity triangle and the general selectivity. Porapak Q, the least polar commercially available porous polymer, was used as a reference phase.

INTRODUCTION Stationary phases for gas chromatography (GC) have been characterized in a number of ways. Thus, Rohrschneider (1966) classified stationary phases by their ability to retard polar solute probes. McReynolds (1970) substracted the dispersive contribution to the retention index of these probes to yield the difference, DI, and used the sum of DI values to describe the total polarity of the phases. In these methods, the selectivity of the phase was characterized by five constants represented by the differences in the Kovats retention indices of test substances (benzene, n-butanol, 2-pentanone, 1-nitropropane and pyridine) on the phase under study and on a reference column packed with squalane, both at the same temperature. According to Smith et al. (1978), graphitized thermal carbon black (GTCB) is a more suitable reference for classifying solid phases in chromatography than squalane, because with this phase none of the modified McReynolds’ constants is negative. Using this modification of McReynolds’ system, it is possible to obtain a good measure of the selectivity of polymeric packings. This is known as a general selectivity system. Castello and D’Amato (1987) proposed the use of Porapak Q (styrene–divinylbenzene-type copolymer) as a ‘standard polymeric phase’ in selectivity studies of polymeric phases, as it is the least polar porous polymer available commercially. In order to achieve a detailed classification of the possible interaction of polymeric phases with different functional groups of solute molecules, they used the following polarity reference substances: benzene, n-butanol, 2-pentanone, 1-nitropropane, pyridine, ethanol, 2-butanone and nitromethane. Of these, the first five represent the test substances recommended by McReynolds. To distinguish the contribution of the common retention mechanism (hydrogen-bond donor, hydrogen-bond acceptor and dipole interaction) to the selectivity of porous polymers, Hepp and Klee (1987) adopted the selectivity triangle developed earlier by Snyder (1974) for liquid chromatography. The retention indices of n-butanol, 1,4-dioxan and 1-nitropropane referenced to those obtained on GTCB may be used for determining the hydrogen-bond donor, hydrogen-bond acceptor and dipole characteristics of a porous polymer. As retention is governed by the total energy of interaction, the extent to which any selectivity is exhibited depends on the amount of polar interaction relative to nonpolar dispersive interaction. GTCB was also used as a reference phase in this method. First presented at the 6th Ukrainian–Polish Symposium on Interfacial Phenomena and their Technological Applications, Odessa, Ukraine, 9–13 September 2001. *Author to whom all correspondence should be addressed. †

524

B. Gawdzik and M. Maciejewska/Adsorption Science & Technology Vol. 20 No. 5 2002

The aim of the present work was to study the influence of the chemical structure of the monomers employed in the synthesis of porous polymers on their subsequent selectivities. In addition to nonpolar styrene–divinylbenzene (ST–DVB) porous copolymer, three other copolymers containing methacrylate functional groups were used. These were: ethylene glycol dimethacrylate– divinylbenzene (DMGE–DVB), 1,4-phenylene dimethacrylate–divinylbenzene (PDM–DVB) and bis(4-methacryloylmethylphenyl)sulphone–divinylbenzene (MMPS–DVB). All copolymers were synthesized in our laboratory. Two procedures were applied to determine the selectivity of these copolymers, i.e. the selectivity triangle and the general selectivity. For more complete evaluation of their selectivities, the elution orders of the test mixture separated on columns packed with the studied copolymers were determined. The mixture contained the following compounds: 2-propanol (hydrogen-bond donor), triethylamine (hydrogen-bond acceptor), 1,2-dichloroethane (weak dipole), acetonitrile (strong dipole) and octane (no polar interactions). EXPERIMENTAL Synthesis of porous copolymers Porous copolymers of DMGE–DVB, MMPS–DVB and ST–DVB were obtained by suspension polymerization from an equivalent mole fraction of monomers. Since the monomer PDM indicated limited reactivity, its concentration in the copolymer with divinylbenzene was decreased (see Table 1) (Koziol et al., submitted for publication). Copolymerization was carried out at 80ºC for 20 h with toluene and n-dodecane being used as pore-forming diluents in the mixture. a,a¢-Azobisisobutyronitrile was used as the initiator. The copolymers obtained were washed with distilled water, filtered off, dried and extracted with boiling acetone, toluene and methanol in a Soxhlet apparatus. Visual observation using a microscope showed that the resulting particles were perfectly spherical in shape with sizes in the range 0.05–0.30 mm. Subsequently, the polymer beads were separated on sieves. The chemical structures of the monomers used for the synthesis of the porous copolymers are presented in Figure 1. Measurements of specific surface areas of polymers The surface areas of the copolymers were measured via nitrogen adsorption at low temperatures, using an ASAP 2405 adsorption analyzer (Micrometrics Inc., Norcross, GA, USA). The specific TABLE 1. Properties of Copolymers Employed Copolymer

Polymer mole fraction

Specific surface area (m2/g)

Initial decomposition temperature (ºC)

DMGE–DVB PDM–DVB MMPS–DVB ST–DVB Porapak Q

0.50:0.50 0.25:0.75 0.50:0.50 0.50:0.50 unknown

315 655 153 163 660

260 280 220 260 200

Selectivity of Porous Methacrylate Polymers

525

Figure 1. Chemical structures of monomers used for the synthesis of porous copolymers: I, divinylbenzene; II, styrene; III, ethylene glycol dimethacrylate; IV, bis(4-methacryloylmethylphenyl)sulphone; V, 1,4-diphenylene dimethacrylate.

526


surface areas were calculated using the BET method assuming that the area of a single nitrogen / molecule is 0.162 nm2 (Oscik 1979). The measurements were preceded by activation of the samples at 200ºC for 2 h. Column preparation The columns were packed with porous polymers in the form of spherical beads with diameters in the range 0.08–0.12 mm. The Porapak Q of a similar sieve fraction employed was obtained from the Water Association, Milford, MA, USA. All the columns were conditioned in a stream of helium before use. They were then further conditioned by temperature programming from 60ºC to 250ºC (in the case of Porapak Q to 200ºC) at a ramp rate of 4ºC/min and then overnight isothermally at the final temperature. The initial decomposition temperatures of the copolymers were first determined from the course of the TG curves. These measurements were carried out on an MOM (Budapest, Hungary) derivatograph at a heating rate of 5ºC/min over the temperature range 20–500ºC in air. Chromatographic measurements Chromatographic measurements were carried out on a Dani GC 1000 gas chromatograph (Dani, Italy) equipped with a thermal conductivity detector, using stainless-steel columns (100 cm × 1.6 mm i.d.) and employing helium as the carrier gas at a flow rate of 50 ml/min. Measurements of the retention indices for the McReynolds substances (benzene, n-butanol, 2-pentanone, 1-nitropropane and pyridine) were carried out at 140ºC (Smith et al. 1978) and those of the selectivity parameters (n-butanol, 1-nitropropane and 1,4-dioxan) at 200ºC (Hepp and Klee 1987). The mixture of polar test solutes (acetonitrile, 2-propanol, triethylamine, 1,2-dichloroethane and octane) was also injected at 200ºC (Gawdzik 1990). The relationships between log VR and the carbon number of n-alkanes (C5–C12) were linear for all the porous copolymers at 140ºC and 200ºC, and hence the retention indices could be applied for the selectivity measurements. The retention time for the peak corresponding to air was considered to be the dead time. The samples were injected by means of a 1 ml syringe (SGE, North Melbourne, Australia). Each probe was injected separately. Selectivity calculations The selectivity parameters (xi) were calculated and plotted on the face of the selectivity triangle using the equation (Hepp and Klee 1987; Gawdzik and Matynia 1998): xi

=

DIi ––––––––––––– DIb + DIn + DId

(1)

where DIb, DIn and DId are the McReynolds constants for n-butanol, 1-nitropropane and 1,4-dioxan, respectively. The McReynolds constants (DIi) were calculated by substracting the average retention indices of the probe solute on GTCB from those on each of the porous copolymers. The values of the retention indices of n-butanol, 1-nitropropane and 1,4-dioxan on GTCB were those quoted by Hepp and Klee (1987).

527


The general selectivity (SI = x + y + z) was calculated using the reported retention data of GTCB for benzene (x), n-butanol (y) and 2-pentanone (z) (Kalashnikova et al. 1976). Retention indices were calculated from the equation given by Kovats (1958): Ix

=

100 log(t¢R,x/t¢R,z)/log(t¢R,z+1/t¢R,z)

+

(2)

100z

where t¢R,x denotes the reduced retention time of the substance x, t¢R,z is the reduced retention time of the homologous alkane with the nearest shortest retention time and t¢R,z+1 is the reduced retention time of the next higher homologue eluted after homologue z, where z denotes the number of carbon atoms in the n-alkane molecule. RESULTS AND DISCUSSION From the data given in Table 1, it appears that copolymers obtained from an equivalent mole fraction of monomers had rather low specific surface areas in comparison with those of Porapak Q. Copolymer PDM–DVB containing a high concentration of DVB in its chemical structure behaved exceptionally. Its specific surface area was similar to that of Porapak Q but its thermal resistance was significantly higher. Table 2 lists the retention indices, modified McReynolds constants and the general selectivities (SI) for the studied copolymers. The general selectivity index SI, defined as a measure of the polarity of the copolymer, indicates that all copolymers synthesized by us had a polar character. Even the ST–DVB copolymer was insignificantly more polar than Porapak Q (ST–DVB-type). The most polar was the MMPS–DVB copolymer which contained ester and sulphonyl functional groups. Copolymers containing only methacrylate groups (DMGE–DVB and PDM–DVB) had similar polarities. It was noteworthy that the PDM–DVB copolymer was less polar than DMGE– DVB; however, the low concentration of the PDM monomer in the chemical structure of the copolymer PDM–DVB (Table 1) was responsible for this phenomenon. It should be stressed that every polar molecule is capable of several polar interactions and that retention on any particular stationary phase depends on the total interaction. It is possible to evaluate the relative interactions between any particular stationary phase and the polar probes provided that they have identifiable polar characteristics which predominate (Abraham et al. 1999; Santiuste 2000). The values of SDII describing the relative polar contributions to the retention and selectivity

TABLE 2. Kovats Retention Indices for McReynolds’ Test Substances and General Selectivities (SI) for Porous Copolymers (at 140ºC)

Copolymer

DMGE–DVB PDM–DVB MMPS–DVB ST–DVB Porapak Q

Kovats retention indices Benzene n-Butanol 2-Pentanone

1-Nitropropane Pyridine

647 623 740 635 617

757 723 860 710 654

698 663 752 629 607

721 699 812 660 651

750 718 845 708 660

x

y

z

SI = x + y + z

73 49 166 61 43

209 174 263 140 118

156 134 248 95 86

438 357 677 296 247

528


TABLE 3. Kovats Retention Indices for n-Butanol, 1-Nitropropane and 1,4-Dioxan and Porous Copolymer Selectivity Parameters (at 200ºC)

Copolymer

Kovats retention indices

DIb

DIn

DId

SDIi

xacc

xdip

xdon

191 174 289 164 127

251 235 380 211 168

215 203 317 206 164

657 612 986 581 459

0.291 0.284 0.293 0.282 0.276

0.382 0.384 0.385 0.363 0.366

0.327 0.332 0.321 0.355 0.358

n-Butanol 1-Nitropropane 1,4-Dioxan DMGE–DVB PDM–DVB MMPS–DVB ST–DVB Porapak Q

677 660 775 650 636

757 741 886 717 683

699 687 801 690 648

parameters (xi) for n-butanol, 1,4-dioxan and 1-nitropropane are summarized in Table 3. In this study, n-butanol (hydrogen-bond donor) was used to measure the hydrogen-bond acceptor characteristics, 1,4-dioxan (hydrogen-bond acceptor) to measure the hydrogen-bond donor characteristics and 1-nitropropane to measure the dipole characteristics of the polymer. In comparison to Porapak Q and the ST–DVB copolymer, all others studied had stronger hydrogen-bond acceptor properties. This was undoubtedly associated with the presence of lone electron pairs on the oxygen and sulphur atoms in the DMGE, PDM and MMPS monomers. Values of selectivity parameters xi indicate that all the polar polymers interacted with molecules with large dipole moments in a similar fashion. The location of these copolymers on the selectivity triangle is shown in Figure 2. The elution order of the test mixture proposed by Hepp and Klee (1987) is presented in Figure 3. The test mixture was composed of a proton donor (2-propanol), proton acceptor (triethylamine), a weak dipole (1,2-dichloroethane), a strong dipole (acetonitrile) and octane which has no polar functionality. The retention of these substances on GTCB was due almost entirely to dispersive interaction and for this reason they eluted in the order of their boiling points and molar volumes (Table 4). All the porous methacrylate copolymers are located in the middle of the selectivity triangle. In comparison with non-polar ST–DVB stationary phases, all polar methacrylate copolymers exhibited stronger hydrogen-acceptor properties. 2-Propanol (hydrogen-bond donor) was retained on the latter in a similar strength to acetonitrile (dipole). The porous PDM–DVB copolymer exhibited a slightly better affinity towards hydrogen-bond acceptors. Unfortunately, on this phase, 2-propanol (representing hydrogen-bond donors) co-eluted with acetonitrile (dipole). It should be noted that the triethylamine peak was more symmetrical on all the methacrylate copolymers than that obtained on less polar porous ST–DVB copolymers. TABLE 4. Properties of Test Mixture Probesa Solute

Boiling point (ºC)

Molar volume (ml/mol)

Dipole moment (Debye)

Acetonitrile 2-Propanol 1,2-Dichloroethane Triethylamine Octane

81.6 82.4 83.5 89.3 125.7

52.6 76.5 79.0 139.1 162.6

3.44 1.66 1.86 0.66 0.00

Taken from Klee et al. (1983).

a


529

Figure 2. Selectivity triangle showing relative selectivities of the porous polymers: DMGE–DVB ( ); PDM–DVB ( ); MMPS–DVB ( s ); ST–DVB ( ); Porapak Q (F ).

The results presented here indicate that the chemical structure of the monomers used in the synthesis of the polymeric stationary phases had an influence on their selectivities, but that the role of functional groups present in their structures was predominant. REFERENCES Abraham, M.H., Poole, C.F. and Poole, S.K. (1999) J. Chromatogr. A 842, 79. Castello, G. and D’Amato, G. (1987) Chromatographia 23, 839. Gawdzik, B. (1990) J. Chromatogr. 503, 41. Gawdzik, B. and Matynia, T. (1998) Adsorption 4, 251. Hepp, M.A. and Klee, M.S. (1987) J. Chromatogr. 404, 145. Kalashnikova, E.V., Kiselev, A.V., Poshkus, D.P. and Shcherbakova, K.D. (1976) J. Chromatogr. 119, 233. Klee, M.S., Kaiser, M.A. and Laughlin, K.B. (1983) J. Chromatogr. 279, 681. Kovats, E. (1958) Helv. Chim. Acta 41, 1915. / Koziol/, A., Grabska, A.K., Maciejewska, M., Rudz, W. and Gawdzik, B. J. Polym. Sci., submitted for publication. McReynolds, O. (1970) J. Chromatogr. Sci. 8, 685. / O scik, J. (1979) Adsorption, PWN, Warsaw, Poland, p. 77. Rohrschneider, L. (1966) J. Chromatogr. 22, 6. Santiuste, J.M. (2000) Chromatographia 52, 225. Smith, J.R.L., Tameesh, A.H.H. and Waddington, D.J. (1978) J. Chromatogr. 151, 21. Snyder, L.R. (1974) J. Chromatogr. 92, 223.

Figure 3. Separations of test mixtures as obtained on copolymers: (A) DMGE–DVB; (B) PDM–DVB; (C) MMPS–DVB; (D) ST–DVB. Peaks: 1 = acetonitrile; 2 = 2-propanol; 3 = 1,2-dichloroethane; 4 = triethylamine; 5 = octane.

530 B. Gawdzik and M. Maciejewska/Adsorption Science & Technology Vol. 20 No. 5 2002