- Email: [email protected]
non-solvent gradient
Journal of Chromatography A, 1218 (2011) 7828–7831
Contents lists available at SciVerse ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Size exclusion chromatography-gradients, an alternative approach to polymer gradient chromatography: 2. Separation of poly(meth)acrylates using a size exclusion chromatography-solvent/non-solvent gradient Martin Schollenberger, Wolfgang Radke ∗ Deutsches Kunststoff-Institut (German Institute for Polymers), Schlossgartenstrasse, D-64289 Darmstadt, Germany
a r t i c l e
i n f o
Article history: Received 10 June 2011 Received in revised form 27 August 2011 Accepted 29 August 2011 Available online 6 September 2011 Keywords: Polymer separations SEC Gradients Polymer blends PMMA Polyacrylates
a b s t r a c t A gradient ranging from methanol to tetrahydrofuran (THF) was applied to a series of poly(methyl methacrylate) (PMMA) standards, using the recently developed concept of SEC-gradients. Contrasting to conventional gradients the samples eluted before the solvent, i.e. within the elution range typical for separations by SEC, however, the high molar mass PMMAs were retarded as compared to experiments on the same column using pure THF as the eluent. The molar mass dependence on retention volume showed a complex behaviour with a nearly molar mass independent elution for high molar masses. This molar mass dependence was explained in terms of solubility and size exclusion effects. The solubility based SEC-gradient was proven to be useful to separate PMMA and poly(n-butyl crylate) (PnBuA) from a poly(t-butyl crylate) (PtBuA) sample. These samples could be separated neither by SEC in THF, due to their very similar hydrodynamic volumes, nor by an SEC-gradient at adsorbing conditions, due to a too low selectivity. The example shows that SEC-gradients can be applied not only in adsorption/desorption mode, but also in precipitation/dissolution mode without risking blocking capillaries or breakthrough peaks. Thus, the new approach is a valuable alternative to conventional gradient chromatography. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Gradient chromatography has become a powerful tool for the characterization of chemically heterogeneous polymers. Polymer blends [1–4] can be separated into their components or copolymers [4–14] can be separated according to chemical composition allowing determination of the chemical heterogeneity. Due to the limited solubilities of polymers, it is often impossible to dissolve the samples in the starting eluent. This may result in so-called breakthrough peaks [15,16]. The breakthrough peaks are the consequence of the fact that the complete sample or parts of it are not adsorbed onto the stationary phase but migrate through the column within the solvent plug and elute without or nearly without retention. Berek used differences in the adsorptivity or solvent quality of solvent and eluent to establish an isocratic separation principle termed chromatography under limiting conditions [17–19]. However, the method is capable to separate only a rather low number of components differing distinctly in their adsorptivity. In addition even for a two component blend two solvents or solvent mixtures are required. One solvent in which both polymeric components
∗ Corresponding author. Tel.: +49 6151 162804; fax: +49 6151 292855. E-mail address: [email protected] (W. Radke). 0021-9673/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2011.08.090
are completely desorbed, while the second solvent is a desorbing liquid (desorli) for one and an adsorbing liquid (adsorli) for the other component. We have recently extended this approach to the application of gradients, resulting in a separation principle termed SEC-gradients [20]. Contrasting to conventional gradients, where the sample is injected at adsorbing conditions and the gradient is started with sample injection, in SEC-gradients the sample is dissolved in a strong eluent (desorli) and sample is injected at the end of the gradient, thus at desorbing conditions. Since the sample solvent and the eluent at the time of injection are strong eluents the sample components experience SEC conditions and will thus escape from the solvent band, due to its larger size as compared to the surrounding solvent, i.e. the polymer migrates through the column at a higher velocity than the surrounding eluent. Because the gradient was started before the actual sample injection, the macromolecules will experience continuously changing solvent compositions of decreasing eluent strength as they pass through the column. This will continue until the strength of the eluent surrounding the polymer is low enough to establish an adsorption threshold for the polymer. The sample will then migrate through the column with the same velocity as the solvent, as further surpassing the eluent is impossible due to the adsorbing eluent before the sample, while lagging behind is impossible due to the following SEC promoting eluent. Since the adsorption threshold depends on
M. Schollenberger, W. Radke / J. Chromatogr. A 1218 (2011) 7828–7831 Table 1 Characterization details of samples used in the investigation. Material
Mn (g/mol)
Mw (g/mol)
MP (g/mol)
PMMA 1 PMMA 2 PMMA 3 PMMA 4 PMMA 5 PMMA 6 PMMA 7 PnBuA PtBuA
1800 22,200 54,000 58,000 310,000 550,000 829,000 51,600 53,000
2000 23,800 55,900 60,000 322,400 570,000 865,000 61,200 55,000
2030 24,400 56,900 60,000 n.a. 570,000 903,000 64,600 55,000
n.a., not given by manufacturer.
the chemical structure of the polymer, a separation of chemically different macromolecules can be realized. The application of a small pore size column is beneficial as it allows reducing exclusion effects for the polymer, resulting in a purely adsorption based separation for high molar mass polymers. The use of a SEC column with pore sizes comparable to the size of the polymer molecules would result in a high molar mass fraction being capable to reach the adsorption threshold, while the slowly travelling low molar mass fraction cannot catch up with the adsorption/precipitation threshold. Therefore the elution volume of the lower molar masses is determined solely by the hydrodynamic volume. This complicates method development and reduces the potential for separations based on chemical composition. In addition, the separation range for the proposed separation mechanism in general is restricted by the analytes’ SEC elution volume on one hand, and the solvent elution volume on the other hand. Thus, the use of a small pore size column allows obtaining the maximum separation efficiency. The concept has been shown to be useful for the separation of a blend of poly(methyl methacrylate) (PMMA) and polystyrene (PS) of similar hydrodynamic size in a gradient ranging from chloroform to THF. These liquids are good solvents for both components, but differ in their eluent strength for PMMA and PS. Thus, the performed separation was based on adsorptivity, rather than on solubility. However, the concept of SEC-gradients should also be applicable to gradients of different solubilities for the polymer components, resulting in separation by precipitation/dissolution. This idea will be explored in the present manuscript. It should be noticed that the approach using a solubility threshold is not new, but has been proposed long ago by Porath under the name “zone precipitation” [21]. However, to our knowledge, the concept of zone precipitation was not further evaluated probably due to inappropriate instrumentation at that time. 2. Experimental 2.1. Samples and solvents Poly(methyl methacrylate) (PMMA), poly(t-butyl acrylate) (PtBuA) and poly(n-butyl acrylate) (PnBuA) standards were purchased from PSS Polymer Standards Service GmbH (Mainz, Germany). The details of the standards as given by the manufacturer are given in Table 1. Methanol (MeOH) and chloroform (CHCl3 ) were delivered by VWR (Darmstadt, Germany) and used as received. Technical tetrahydrofuran (THF, BASF Ludwigshafen, Germany) was dried over calcium hydride and distilled. 2.2. HPLC The HPLC measurements were performed using an Agilent Series 1100 (Agilent Technologies, Santa Clara, USA) chromatography system, consisting of a Degasser (G1379A), a quaternary pump
7829
(G1311A), an autosampler (G1313A) and a column thermostat (G1316A) operated at 35 ◦ C. For detection an Evaporative Light Scattering Detector (ELSD, gas flow 1.0 L/min, nebulizer temperature 40 ◦ C, evaporator temperature 80 ◦ C), ELS 1000 (Polymer Laboratories, Church Stretton, England) was used. Data acquisition and evaluation were performed using PSS WINGPC Unity Software (PSS Polymer Standards Service GmbH, Mainz, Germany). Separations were carried out using a Proteema SEC-column (3 m particles, 300 m length and 8 mm ID), produced by PSS Polymer Standards Service GmbH (Mainz, Germany) at a flow rate of 1 mL/min. Samples were dissolved at concentrations of approximately 1–1.5 g/L in the respective solvent. The injection volume was typically 20 L. Since the HPLC system starts the gradient with sample injection, the gradient was started by a blank run and sample introduction was performed by a separate injection 6.5 min later. It should be noted that the Agilent system returns to the initial conditions with the start of the sample preparation step. If the time for sample preparation is large as compared to the flow time between mixer and injector, the sample is injected at initial conditions and not into the conditions at the end of the gradient. Therefore sample injection time was optimized using the “Overlap Injection Cycle” option. Cloud points were determined using a DAWN DSP light scattering detector (Wyatt Technology, Santa Barbara, USA) in batch mode using scintillation vials. For the measurements the samples were dissolved in THF at concentrations of 2–4 mg/mL. After filtering through 0.45 m syringe filter, the scattering intensity at 90◦ was monitored at room temperature. Several known aliquots of filtered methanol were added and the scattering intensities were determined. The cloud points were determined as onset of the scattering intensity. For PnBA only two experiments were carried out, due to the restricted sample amount.
3. Results and discussion 3.1. Experimental results Recently we have shown that the application of a linear gradient ranging from chloroform to THF and injection into the end of the gradient result in a delayed elution as compared to isocratic elution in pure THF [20]. This was attributed to the fact that the polymer molecules move faster than the solvent and reach the adsorption threshold within the column. Due to differences in the adsorption thresholds it was also possible to separate a mixture of PS and PMMA using the same gradient. The separation of a mixture of different poly(meth)acrylates presents a more challenging problem than the separation of PS and PMMA. In order to prove whether the application of the same gradient also results in delayed retention for poly(n-BuA) and poly(t-BuA) samples the samples were first run in pure THF and the elution times were compared with the ones in a 5 min SEC-gradient ranging from CHCl3 to THF. A comparison of the chromatograms of the selected PMMA, poly(n-BuA) and poly(t-BuA) standards is shown in Fig. 1. The SEC chromatograms obtained by isocratic runs in pure THF show very similar retention volumes in the range 6.0–6.5 mL. Thus, the three chemically different polymers cannot be separated by SEC, due to their very similar hydrodynamic sizes. In the SEC-gradient all samples reveal higher retention volumes as compared to the experiment in pure THF (Fig. 1). However, the samples elute before the dead volume of the column, which is approximately 12 mL. The elution before the pure solvent indicates that the polymers elute at elution volumes typical for SEC, i.e. the samples are excluded from the pore of the stationary phase and migrate faster than the surrounding eluent. The later elution as compared to the isocratic experiment proves that all three samples
7830
M. Schollenberger, W. Radke / J. Chromatogr. A 1218 (2011) 7828–7831
Fig. 1. Comparison of chromatograms of PMMA 3, poly(n-BuA) and poly(t-BuA) for isocratic elution in pure THF (SEC) and for 5 min linear SEC-gradient ranging from CHCl3 to THF (SEC-gradient).
Fig. 2. Chromatograms of PMMA (MP = 829, 500, 310, 58, 54, 22, 1.8 kg/mol) standards in a 5 min gradient ranging from methanol to THF. Injection 6.5 min after start of gradient. The dotted line represents the eluent composition at the detector at the time of elution.
reach eluent conditions that result in adsorption onto the stationary phase. As a consequence, the polymer cannot escape further from the injection band. Despite the change in retention volumes when going from isocratic elution in THF to the SEC-gradient, a separation could not be achieved, since all samples elute at very similar elution volumes (6.7–7 mL) in the SEC-gradient. One obvious way in trying to increase the separation efficiency would be changing gradient slope. However, we have decided to vary the solvents, in order to prove at the same time, whether SEC-gradients can be applied also for gradients utilizing differences in the solubility of the polymers. Therefore, chloroform, which is a good solvent but a weak eluent for all samples was replaced by methanol, which is a non-solvent for all samples. All other chromatographic parameters were kept identical. Since methanol should act as displacer on the polar column used, i.e. it should promote desorption, retardation of the macromolecules must be due to solubility effects and not based on changes in adsorptivity. First it was tested, whether the retention will be altered when starting the gradient with a non-solvent instead of an adsorption promoting liquid. For this purpose, a series of PMMA-standards with molar masses ranging form 1.8 kg/mol to 829 kg/mol was dissolved in THF and injected into the end of a 5 min gradient ranging from pure methanol to pure THF. Fig. 2 shows the chromatograms obtained. All samples elute in the range 6.7–10 mL, i.e. in the volume range expected for elution under SEC conditions.
Fig. 3. Dependence of retention volume on molar mass of PMMA standards for isocratic elution in pure THF (䊉) and for 5 min gradient ranging from MeOH to 100 THF ().
The samples having a molar mass above 10 kg/mol elute later than under isocratic conditions in pure THF, as expected. However, the elution volume depends on molar mass in a non-trivial fashion. This becomes clearer in Fig. 3, which compares the calibration curves of the PMMA standards in pure THF and in the SEC-gradient. With the exception of the lowest molar mass sample, the retention volumes in the SEC-gradient are significantly higher than those in pure THF. However, the molar mass dependence of the retention volume in the SEC-gradient is quite complicated. The retention volume of the lowest molar mass is nearly independent of whether a SEC-gradient is applied or elution occurs in pure THF. This indicates that the sample is not sufficiently excluded from the pores of the stationary phase to experience the change of the solvent quality, due to the increasing methanol content. The samples of higher molar mass reach the solubility threshold within the column and are therefore retarded in the gradient. However, since the fraction of non-solvent required for precipitation increases with decreasing molar mass, the sample of molar mass 22,000 g/mol reaches its precipitation threshold at a higher methanol fraction, i.e. at a lower elution volume. With increasing molar mass the amount of methanol required to precipitate the sample onto the stationary phase decreases and consequently the retention time increases with increasing molar mass, until a nearly molar mass independent elution is observed for molar masses exceeding approximately 100 kg/mol. It should be noted that the influence of molar mass molar on elution volume for the PMMAs is more pronounced for the non-solvent/solvent gradient (methanol/THF) than for the adsorption/desorption gradient (CHCl3 /THF) investigated in the first paper of this series. While in the latter gradient the difference in elution volumes for the samples 22.2 kg/mol and 829 kg/mol amounts to 0.16 mL, the corresponding elution volume difference reads 1 mL in the present investigation. It remains to be investigated whether this is a general feature of non-solvent/solvent gradients. Sample recovery was not determined explicitly. Due to the proposed mechanism it is expected that the samples are never fully adsorbed/precipitated, but are at the verge on adsorption/precipitation, minimizing the danger of limited sample recovery. In addition, the gradient ended always at 100% THF which has been shown (Fig. 1) to elute all samples well before the solvent peak, i.e. in SEC mode. For the present investigation, a column with rather small particles was used, which raises the question of potential chain degradation for the highest molar mass samples. However, it has been shown on polystyrenes that for the molar mass range investigated, chain deformation rather than chain rupture occurs under similar experimental conditions [22,23].
M. Schollenberger, W. Radke / J. Chromatogr. A 1218 (2011) 7828–7831
7831
for PtBuA, PnBuA and PMMA at the time of elution were estimated to be 1%, 27% and 23%. Thus, PMMA and PnBuA elute earlier than expected from cloud point measurements. However, since additional adsorptive interaction would result in longer retention times, it can be concluded that it is possible to use SEC-gradients under conditions of precipitation/dissolution in order to change the selectivity. 4. Conclusions
Fig. 4. Normalized chromatograms of PMMA 3, poly(n-BuA), poly(t-BuA) standards and a blend of PMMA and PtBuA in a SEC-gradient of 5 min duration ranging from pure methanol to pure THF. Injection time: 6.5 min after start of gradient.
Having shown that the elution volume can be altered by non-solvent/solvent gradients similar to the adsorption/desorption experiments, it was investigated whether a separation of the chemically different poly(meth)acrylates can be obtained. Therefore the samples of Fig. 1 were investigated in the same 5 min SEC-gradient ranging from methanol to THF. A comparison of the chromatograms of the individual samples as well as of a mixture of PMMA and the PtBuA obtained using this gradient is given in Fig. 4. As can be seen, the PMMA and the PnBuA elute at very similar elution volumes and cannot be separated under these conditions. However, the PtBA, which could not be separated by either SEC or SEC-gradient running from CHCl3 to THF elutes significantly earlier, i.e. at higher levels of MeOH and is separated from both other components. In order to investigate whether the separation is based on precipitation/dissolution, cloud points were determined. While PtBuA could not be precipitated by the addition of MeOH, the PMMA and PnBuA samples precipitated at MeOH contents of 65% and 70–80% MeOH, respectively. Thus, based on cloud points, only weak retention of PtBA, i.e. elution close to the SEC elution volume, is expected, while PMMA and PnBuA are expected to elute significantly later, as compared to SEC behaviour. This is observed experimentally indeed. However, knowing the retention times, the gradient slope, the dwell, pore and void volume of the column, the MeOH contents
It has been shown that the concept of SEC-gradients can be successfully applied to gradients based on polymer solubility. The advantage of the method is that the samples can be dissolved in good solvents without producing break through peaks. Since the sample is also injected into a good solvent no problems are expected due to the blocking of capillaries when applying solvent/nonsolvent gradients. Thus, the new approach provides a valuable alternative to conventional gradients. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23]
L.-C. Heinz, H. Pasch, Polymer 46 (2005) 12040. M.A. Bashir, W. Radke, J. Chromatogr. A 1163 (2007) 86. F. Fitzpatrick, R. Edam, P. Schoenmakers, J. Chromatogr. A 988 (2003) 53. T. Chang, J. Polym. Sci. B: Polym. Phys. 43 (2005) 1591. S. Park, I. Park, T. Chang, C.Y. Ryu, J. Am. Chem. Soc. 126 (2004) 8906. N. Fandrich, J. Falkenhagen, S.M. Weidner, B. Staal, A.F. Thuenemann, A. Laschewsky, Macromol. Chem. Phys. 211 (2010) 1678. B. Trathnigg, M.I. Malik, N. Pircher, S. Hayden, J. Sep. Sci. 33 (2010) 2052. S.M. Graef, A.J.P. van Zyl, R.D. Sanderson, B. Klumperman, H. Pasch, J. Appl. Polym. Sci. 88 (2003) 2530. P.J.C.H. Cools, F. Maesen, B. Klumperman, A.M. Van Herk, A.L. German, J. Chromatogr. A 736 (1996) 125. I. Krämer, W. Hiller, H. Pasch, Macromol. Chem. Phys. 201 (2000) 1662. D. Braun, I. Kramer, H. Pasch, Macromol. Chem. Phys. 201 (2000) 1048. A. Albrecht, R. Brüll, T. Macko, H. Pasch, Macromolecules (2007). Y. Brun, J. Liq. Chromatogr. Relat. Technol. 22 (1999) 3067. A.M. Striegel, J. Chromatogr. A 971 (2002) 151. X. Jiang, A. van der Horst, P.J. Schoenmakers, J. Chromatogr. A 982 (2002) 55. G. Glöckner, Gradient HPLC of Copolymers and Chromatographic Crossfractionation, Springer, Berlin, 1991. D. Berek, Macromolecules 31 (1998) 8517. D. Berek, Macromol. Symp. 231 (2006) 134. D. Berek, D. Hunkeler, J. Liq. Chromatogr. Relat. Technol. 22 (19) (1999) 2867. M. Schollenberger, W. Radke, Polymer 52 (2011) 3259. J. Porath, Nature 196 (1962) 47. E. Uliyanchenko, S.v.d. Wal, P.J. Schoenmakers, J. Chromatogr. A (2011), doi:10.1016/j.chroma.2011.08.014. Y. Liu, W. Radke, H. Pasch, Macromolecules 39 (2006) 2004.
Contents lists available at SciVerse ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Size exclusion chromatography-gradients, an alternative approach to polymer gradient chromatography: 2. Separation of poly(meth)acrylates using a size exclusion chromatography-solvent/non-solvent gradient Martin Schollenberger, Wolfgang Radke ∗ Deutsches Kunststoff-Institut (German Institute for Polymers), Schlossgartenstrasse, D-64289 Darmstadt, Germany
a r t i c l e
i n f o
Article history: Received 10 June 2011 Received in revised form 27 August 2011 Accepted 29 August 2011 Available online 6 September 2011 Keywords: Polymer separations SEC Gradients Polymer blends PMMA Polyacrylates
a b s t r a c t A gradient ranging from methanol to tetrahydrofuran (THF) was applied to a series of poly(methyl methacrylate) (PMMA) standards, using the recently developed concept of SEC-gradients. Contrasting to conventional gradients the samples eluted before the solvent, i.e. within the elution range typical for separations by SEC, however, the high molar mass PMMAs were retarded as compared to experiments on the same column using pure THF as the eluent. The molar mass dependence on retention volume showed a complex behaviour with a nearly molar mass independent elution for high molar masses. This molar mass dependence was explained in terms of solubility and size exclusion effects. The solubility based SEC-gradient was proven to be useful to separate PMMA and poly(n-butyl crylate) (PnBuA) from a poly(t-butyl crylate) (PtBuA) sample. These samples could be separated neither by SEC in THF, due to their very similar hydrodynamic volumes, nor by an SEC-gradient at adsorbing conditions, due to a too low selectivity. The example shows that SEC-gradients can be applied not only in adsorption/desorption mode, but also in precipitation/dissolution mode without risking blocking capillaries or breakthrough peaks. Thus, the new approach is a valuable alternative to conventional gradient chromatography. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Gradient chromatography has become a powerful tool for the characterization of chemically heterogeneous polymers. Polymer blends [1–4] can be separated into their components or copolymers [4–14] can be separated according to chemical composition allowing determination of the chemical heterogeneity. Due to the limited solubilities of polymers, it is often impossible to dissolve the samples in the starting eluent. This may result in so-called breakthrough peaks [15,16]. The breakthrough peaks are the consequence of the fact that the complete sample or parts of it are not adsorbed onto the stationary phase but migrate through the column within the solvent plug and elute without or nearly without retention. Berek used differences in the adsorptivity or solvent quality of solvent and eluent to establish an isocratic separation principle termed chromatography under limiting conditions [17–19]. However, the method is capable to separate only a rather low number of components differing distinctly in their adsorptivity. In addition even for a two component blend two solvents or solvent mixtures are required. One solvent in which both polymeric components
∗ Corresponding author. Tel.: +49 6151 162804; fax: +49 6151 292855. E-mail address: [email protected] (W. Radke). 0021-9673/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2011.08.090
are completely desorbed, while the second solvent is a desorbing liquid (desorli) for one and an adsorbing liquid (adsorli) for the other component. We have recently extended this approach to the application of gradients, resulting in a separation principle termed SEC-gradients [20]. Contrasting to conventional gradients, where the sample is injected at adsorbing conditions and the gradient is started with sample injection, in SEC-gradients the sample is dissolved in a strong eluent (desorli) and sample is injected at the end of the gradient, thus at desorbing conditions. Since the sample solvent and the eluent at the time of injection are strong eluents the sample components experience SEC conditions and will thus escape from the solvent band, due to its larger size as compared to the surrounding solvent, i.e. the polymer migrates through the column at a higher velocity than the surrounding eluent. Because the gradient was started before the actual sample injection, the macromolecules will experience continuously changing solvent compositions of decreasing eluent strength as they pass through the column. This will continue until the strength of the eluent surrounding the polymer is low enough to establish an adsorption threshold for the polymer. The sample will then migrate through the column with the same velocity as the solvent, as further surpassing the eluent is impossible due to the adsorbing eluent before the sample, while lagging behind is impossible due to the following SEC promoting eluent. Since the adsorption threshold depends on
M. Schollenberger, W. Radke / J. Chromatogr. A 1218 (2011) 7828–7831 Table 1 Characterization details of samples used in the investigation. Material
Mn (g/mol)
Mw (g/mol)
MP (g/mol)
PMMA 1 PMMA 2 PMMA 3 PMMA 4 PMMA 5 PMMA 6 PMMA 7 PnBuA PtBuA
1800 22,200 54,000 58,000 310,000 550,000 829,000 51,600 53,000
2000 23,800 55,900 60,000 322,400 570,000 865,000 61,200 55,000
2030 24,400 56,900 60,000 n.a. 570,000 903,000 64,600 55,000
n.a., not given by manufacturer.
the chemical structure of the polymer, a separation of chemically different macromolecules can be realized. The application of a small pore size column is beneficial as it allows reducing exclusion effects for the polymer, resulting in a purely adsorption based separation for high molar mass polymers. The use of a SEC column with pore sizes comparable to the size of the polymer molecules would result in a high molar mass fraction being capable to reach the adsorption threshold, while the slowly travelling low molar mass fraction cannot catch up with the adsorption/precipitation threshold. Therefore the elution volume of the lower molar masses is determined solely by the hydrodynamic volume. This complicates method development and reduces the potential for separations based on chemical composition. In addition, the separation range for the proposed separation mechanism in general is restricted by the analytes’ SEC elution volume on one hand, and the solvent elution volume on the other hand. Thus, the use of a small pore size column allows obtaining the maximum separation efficiency. The concept has been shown to be useful for the separation of a blend of poly(methyl methacrylate) (PMMA) and polystyrene (PS) of similar hydrodynamic size in a gradient ranging from chloroform to THF. These liquids are good solvents for both components, but differ in their eluent strength for PMMA and PS. Thus, the performed separation was based on adsorptivity, rather than on solubility. However, the concept of SEC-gradients should also be applicable to gradients of different solubilities for the polymer components, resulting in separation by precipitation/dissolution. This idea will be explored in the present manuscript. It should be noticed that the approach using a solubility threshold is not new, but has been proposed long ago by Porath under the name “zone precipitation” [21]. However, to our knowledge, the concept of zone precipitation was not further evaluated probably due to inappropriate instrumentation at that time. 2. Experimental 2.1. Samples and solvents Poly(methyl methacrylate) (PMMA), poly(t-butyl acrylate) (PtBuA) and poly(n-butyl acrylate) (PnBuA) standards were purchased from PSS Polymer Standards Service GmbH (Mainz, Germany). The details of the standards as given by the manufacturer are given in Table 1. Methanol (MeOH) and chloroform (CHCl3 ) were delivered by VWR (Darmstadt, Germany) and used as received. Technical tetrahydrofuran (THF, BASF Ludwigshafen, Germany) was dried over calcium hydride and distilled. 2.2. HPLC The HPLC measurements were performed using an Agilent Series 1100 (Agilent Technologies, Santa Clara, USA) chromatography system, consisting of a Degasser (G1379A), a quaternary pump
7829
(G1311A), an autosampler (G1313A) and a column thermostat (G1316A) operated at 35 ◦ C. For detection an Evaporative Light Scattering Detector (ELSD, gas flow 1.0 L/min, nebulizer temperature 40 ◦ C, evaporator temperature 80 ◦ C), ELS 1000 (Polymer Laboratories, Church Stretton, England) was used. Data acquisition and evaluation were performed using PSS WINGPC Unity Software (PSS Polymer Standards Service GmbH, Mainz, Germany). Separations were carried out using a Proteema SEC-column (3 m particles, 300 m length and 8 mm ID), produced by PSS Polymer Standards Service GmbH (Mainz, Germany) at a flow rate of 1 mL/min. Samples were dissolved at concentrations of approximately 1–1.5 g/L in the respective solvent. The injection volume was typically 20 L. Since the HPLC system starts the gradient with sample injection, the gradient was started by a blank run and sample introduction was performed by a separate injection 6.5 min later. It should be noted that the Agilent system returns to the initial conditions with the start of the sample preparation step. If the time for sample preparation is large as compared to the flow time between mixer and injector, the sample is injected at initial conditions and not into the conditions at the end of the gradient. Therefore sample injection time was optimized using the “Overlap Injection Cycle” option. Cloud points were determined using a DAWN DSP light scattering detector (Wyatt Technology, Santa Barbara, USA) in batch mode using scintillation vials. For the measurements the samples were dissolved in THF at concentrations of 2–4 mg/mL. After filtering through 0.45 m syringe filter, the scattering intensity at 90◦ was monitored at room temperature. Several known aliquots of filtered methanol were added and the scattering intensities were determined. The cloud points were determined as onset of the scattering intensity. For PnBA only two experiments were carried out, due to the restricted sample amount.
3. Results and discussion 3.1. Experimental results Recently we have shown that the application of a linear gradient ranging from chloroform to THF and injection into the end of the gradient result in a delayed elution as compared to isocratic elution in pure THF [20]. This was attributed to the fact that the polymer molecules move faster than the solvent and reach the adsorption threshold within the column. Due to differences in the adsorption thresholds it was also possible to separate a mixture of PS and PMMA using the same gradient. The separation of a mixture of different poly(meth)acrylates presents a more challenging problem than the separation of PS and PMMA. In order to prove whether the application of the same gradient also results in delayed retention for poly(n-BuA) and poly(t-BuA) samples the samples were first run in pure THF and the elution times were compared with the ones in a 5 min SEC-gradient ranging from CHCl3 to THF. A comparison of the chromatograms of the selected PMMA, poly(n-BuA) and poly(t-BuA) standards is shown in Fig. 1. The SEC chromatograms obtained by isocratic runs in pure THF show very similar retention volumes in the range 6.0–6.5 mL. Thus, the three chemically different polymers cannot be separated by SEC, due to their very similar hydrodynamic sizes. In the SEC-gradient all samples reveal higher retention volumes as compared to the experiment in pure THF (Fig. 1). However, the samples elute before the dead volume of the column, which is approximately 12 mL. The elution before the pure solvent indicates that the polymers elute at elution volumes typical for SEC, i.e. the samples are excluded from the pore of the stationary phase and migrate faster than the surrounding eluent. The later elution as compared to the isocratic experiment proves that all three samples
7830
M. Schollenberger, W. Radke / J. Chromatogr. A 1218 (2011) 7828–7831
Fig. 1. Comparison of chromatograms of PMMA 3, poly(n-BuA) and poly(t-BuA) for isocratic elution in pure THF (SEC) and for 5 min linear SEC-gradient ranging from CHCl3 to THF (SEC-gradient).
Fig. 2. Chromatograms of PMMA (MP = 829, 500, 310, 58, 54, 22, 1.8 kg/mol) standards in a 5 min gradient ranging from methanol to THF. Injection 6.5 min after start of gradient. The dotted line represents the eluent composition at the detector at the time of elution.
reach eluent conditions that result in adsorption onto the stationary phase. As a consequence, the polymer cannot escape further from the injection band. Despite the change in retention volumes when going from isocratic elution in THF to the SEC-gradient, a separation could not be achieved, since all samples elute at very similar elution volumes (6.7–7 mL) in the SEC-gradient. One obvious way in trying to increase the separation efficiency would be changing gradient slope. However, we have decided to vary the solvents, in order to prove at the same time, whether SEC-gradients can be applied also for gradients utilizing differences in the solubility of the polymers. Therefore, chloroform, which is a good solvent but a weak eluent for all samples was replaced by methanol, which is a non-solvent for all samples. All other chromatographic parameters were kept identical. Since methanol should act as displacer on the polar column used, i.e. it should promote desorption, retardation of the macromolecules must be due to solubility effects and not based on changes in adsorptivity. First it was tested, whether the retention will be altered when starting the gradient with a non-solvent instead of an adsorption promoting liquid. For this purpose, a series of PMMA-standards with molar masses ranging form 1.8 kg/mol to 829 kg/mol was dissolved in THF and injected into the end of a 5 min gradient ranging from pure methanol to pure THF. Fig. 2 shows the chromatograms obtained. All samples elute in the range 6.7–10 mL, i.e. in the volume range expected for elution under SEC conditions.
Fig. 3. Dependence of retention volume on molar mass of PMMA standards for isocratic elution in pure THF (䊉) and for 5 min gradient ranging from MeOH to 100 THF ().
The samples having a molar mass above 10 kg/mol elute later than under isocratic conditions in pure THF, as expected. However, the elution volume depends on molar mass in a non-trivial fashion. This becomes clearer in Fig. 3, which compares the calibration curves of the PMMA standards in pure THF and in the SEC-gradient. With the exception of the lowest molar mass sample, the retention volumes in the SEC-gradient are significantly higher than those in pure THF. However, the molar mass dependence of the retention volume in the SEC-gradient is quite complicated. The retention volume of the lowest molar mass is nearly independent of whether a SEC-gradient is applied or elution occurs in pure THF. This indicates that the sample is not sufficiently excluded from the pores of the stationary phase to experience the change of the solvent quality, due to the increasing methanol content. The samples of higher molar mass reach the solubility threshold within the column and are therefore retarded in the gradient. However, since the fraction of non-solvent required for precipitation increases with decreasing molar mass, the sample of molar mass 22,000 g/mol reaches its precipitation threshold at a higher methanol fraction, i.e. at a lower elution volume. With increasing molar mass the amount of methanol required to precipitate the sample onto the stationary phase decreases and consequently the retention time increases with increasing molar mass, until a nearly molar mass independent elution is observed for molar masses exceeding approximately 100 kg/mol. It should be noted that the influence of molar mass molar on elution volume for the PMMAs is more pronounced for the non-solvent/solvent gradient (methanol/THF) than for the adsorption/desorption gradient (CHCl3 /THF) investigated in the first paper of this series. While in the latter gradient the difference in elution volumes for the samples 22.2 kg/mol and 829 kg/mol amounts to 0.16 mL, the corresponding elution volume difference reads 1 mL in the present investigation. It remains to be investigated whether this is a general feature of non-solvent/solvent gradients. Sample recovery was not determined explicitly. Due to the proposed mechanism it is expected that the samples are never fully adsorbed/precipitated, but are at the verge on adsorption/precipitation, minimizing the danger of limited sample recovery. In addition, the gradient ended always at 100% THF which has been shown (Fig. 1) to elute all samples well before the solvent peak, i.e. in SEC mode. For the present investigation, a column with rather small particles was used, which raises the question of potential chain degradation for the highest molar mass samples. However, it has been shown on polystyrenes that for the molar mass range investigated, chain deformation rather than chain rupture occurs under similar experimental conditions [22,23].
M. Schollenberger, W. Radke / J. Chromatogr. A 1218 (2011) 7828–7831
7831
for PtBuA, PnBuA and PMMA at the time of elution were estimated to be 1%, 27% and 23%. Thus, PMMA and PnBuA elute earlier than expected from cloud point measurements. However, since additional adsorptive interaction would result in longer retention times, it can be concluded that it is possible to use SEC-gradients under conditions of precipitation/dissolution in order to change the selectivity. 4. Conclusions
Fig. 4. Normalized chromatograms of PMMA 3, poly(n-BuA), poly(t-BuA) standards and a blend of PMMA and PtBuA in a SEC-gradient of 5 min duration ranging from pure methanol to pure THF. Injection time: 6.5 min after start of gradient.
Having shown that the elution volume can be altered by non-solvent/solvent gradients similar to the adsorption/desorption experiments, it was investigated whether a separation of the chemically different poly(meth)acrylates can be obtained. Therefore the samples of Fig. 1 were investigated in the same 5 min SEC-gradient ranging from methanol to THF. A comparison of the chromatograms of the individual samples as well as of a mixture of PMMA and the PtBuA obtained using this gradient is given in Fig. 4. As can be seen, the PMMA and the PnBuA elute at very similar elution volumes and cannot be separated under these conditions. However, the PtBA, which could not be separated by either SEC or SEC-gradient running from CHCl3 to THF elutes significantly earlier, i.e. at higher levels of MeOH and is separated from both other components. In order to investigate whether the separation is based on precipitation/dissolution, cloud points were determined. While PtBuA could not be precipitated by the addition of MeOH, the PMMA and PnBuA samples precipitated at MeOH contents of 65% and 70–80% MeOH, respectively. Thus, based on cloud points, only weak retention of PtBA, i.e. elution close to the SEC elution volume, is expected, while PMMA and PnBuA are expected to elute significantly later, as compared to SEC behaviour. This is observed experimentally indeed. However, knowing the retention times, the gradient slope, the dwell, pore and void volume of the column, the MeOH contents
It has been shown that the concept of SEC-gradients can be successfully applied to gradients based on polymer solubility. The advantage of the method is that the samples can be dissolved in good solvents without producing break through peaks. Since the sample is also injected into a good solvent no problems are expected due to the blocking of capillaries when applying solvent/nonsolvent gradients. Thus, the new approach provides a valuable alternative to conventional gradients. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23]
L.-C. Heinz, H. Pasch, Polymer 46 (2005) 12040. M.A. Bashir, W. Radke, J. Chromatogr. A 1163 (2007) 86. F. Fitzpatrick, R. Edam, P. Schoenmakers, J. Chromatogr. A 988 (2003) 53. T. Chang, J. Polym. Sci. B: Polym. Phys. 43 (2005) 1591. S. Park, I. Park, T. Chang, C.Y. Ryu, J. Am. Chem. Soc. 126 (2004) 8906. N. Fandrich, J. Falkenhagen, S.M. Weidner, B. Staal, A.F. Thuenemann, A. Laschewsky, Macromol. Chem. Phys. 211 (2010) 1678. B. Trathnigg, M.I. Malik, N. Pircher, S. Hayden, J. Sep. Sci. 33 (2010) 2052. S.M. Graef, A.J.P. van Zyl, R.D. Sanderson, B. Klumperman, H. Pasch, J. Appl. Polym. Sci. 88 (2003) 2530. P.J.C.H. Cools, F. Maesen, B. Klumperman, A.M. Van Herk, A.L. German, J. Chromatogr. A 736 (1996) 125. I. Krämer, W. Hiller, H. Pasch, Macromol. Chem. Phys. 201 (2000) 1662. D. Braun, I. Kramer, H. Pasch, Macromol. Chem. Phys. 201 (2000) 1048. A. Albrecht, R. Brüll, T. Macko, H. Pasch, Macromolecules (2007). Y. Brun, J. Liq. Chromatogr. Relat. Technol. 22 (1999) 3067. A.M. Striegel, J. Chromatogr. A 971 (2002) 151. X. Jiang, A. van der Horst, P.J. Schoenmakers, J. Chromatogr. A 982 (2002) 55. G. Glöckner, Gradient HPLC of Copolymers and Chromatographic Crossfractionation, Springer, Berlin, 1991. D. Berek, Macromolecules 31 (1998) 8517. D. Berek, Macromol. Symp. 231 (2006) 134. D. Berek, D. Hunkeler, J. Liq. Chromatogr. Relat. Technol. 22 (19) (1999) 2867. M. Schollenberger, W. Radke, Polymer 52 (2011) 3259. J. Porath, Nature 196 (1962) 47. E. Uliyanchenko, S.v.d. Wal, P.J. Schoenmakers, J. Chromatogr. A (2011), doi:10.1016/j.chroma.2011.08.014. Y. Liu, W. Radke, H. Pasch, Macromolecules 39 (2006) 2004.