Fundamental Evaluation of the Stationary Phase Retention and Selectivity on the Ascentis® Express Biphenyl

By: Daniel Shollenberger, Hugh Cramer, Dave Bell, Reporter US Volume 34.3

Introduction

While C18 alkyl-based phases remain the gold standard for reversed-phase liquid chromatography, stationary phases that invoke interactions in addition to hydrophobic partitioning afford the method developer a viable option for improving resolution without sacrificing robustness or reproducibility. Phenyl type stationary phases are one such alternative that are thought to benefit from pi-pi interactions promoting charge transfer between solute and stationary phase via Lewis acid/base mechanisms.1 Recently, there has been a growing interest in the use of phases based on the biphenyl moiety, and the advantages it sometimes exhibits in the separation of compounds not well resolved by C18 or existing phenyl chemistry. There are studies that have evaluated typical commercially available phenyl phases,2 however; there is not as much data in the primary literature to characterize the molecular interactions on biphenyl type phases.

Experiments previously conducted to characterize the retention and selectivity on the Ascentis Express Biphenyl stationary phase yielded insight into potential molecular interactions. These evaluations were based on methods described by Euerby and Tanaka, among others, and rely on using a set of defined probes to determine the relative extent of particular molecular interactions.3,4 For example, the hydrogen bonding capacity for a given stationary phase is determined by comparing the retention and selectivity of phenol versus caffeine, where the difference in hydrogen bond donating versus accepting potential of these probes can discriminate the stationary phase potential for this particular dipolar interaction.

These studies led to the observation that hydrogen bonding capacity on the biphenyl phase was much higher compared to a C18 alkyl phase. Figure 1 and Table 1 describe the results of this test. There is an elution order reversal of the analytes comparing the two phases and the biphenyl chemistry shows a much greater retention of the caffeine probe. Horak et al. have described a similar result, giving the rationale that the extended pi system of the purine ring compared to phenol drives retention by pi-pi interactions.5 While the contributions of pi-pi type contributions are likely important for aromatic stationary phases, the purpose of the current study is to further investigate the hydrogen bonding capacity observation by studying a test mix of substituted benzene compounds with functional groups capable of accepting a hydrogen bonding proton, similar to the caffeine probe.

Overlay of Chromatograms for Caffeine and Phenol

Figure 1. An Overlay of Chromatograms for Caffeine and Phenol on Discovery® C18 (Red) and Ascentis Express Biphenyl (Blue)

 

Table 1. Absolute Retention and Selectivity for Caffeine and Phenol Comparing an Alkyl C18 and Biphenyl Stationary Phase

Parameter Discovery C18
Ascentis Express Biphenyl
k’ caffeine 0.94 5.28
k’ phenol 1.93 2.91
α H-Bond Capacity 0.49 1.82

 

In order to understand the additional contributions to retention conferred by the functional group on the substituted benzene compounds, toluene is used as a reference for purely hydrophobic interactions. Isoelutropic conditions were selected so that the retention of toluene could be normalized for both stationary phases and organic modifiers. The Ascentis Express Biphenyl stationary phase is compared against a standard phenyl type phase and an alkyl C18 phase to evaluate contributions to retention inherent to increasing the aromatic character of the stationary phase. The probes were chosen to include compounds that were all hydrogen bond acceptors based on the previous observation that the biphenyl phase has a high degree of hydrogen bonding capacity based on an analysis of caffeine retention and selectivity (Figure 2).

Substituted Benzene Compounds Used as Selectivity Probes

Figure 2. Substituted Benzene Compounds Used as Selectivity Probes on Ascentis Express Biphenyl

 

The results of the substituted aromatic comparison are shown in Figure 3. Changing the organic component in the mobile phase had a significant impact on selectivity. The differences in retention and selectivity are more pronounced on the phenyl phases, particularly the biphenyl stationary phase. The change in elution order between benzonitrile and acetophenone is observed on all phases, implying the selectivity difference is likely not an attribute of stationary phase aromaticity. In methanol mobile phase, solute retention increases with the increasing aromatic character of the stationary phase. While there are different substituent effects on the phenyl ring based on the functional group chemistry, there is not a clear trend between the electron density of the ring and interactions with the stationary phase. Compounds with both electron donating and electron withdrawing groups all seem to have the same increased retention in the methanolic mobile phase.

This effect of organic modifier has been observed on phenyl phases in previous reported studies.5 In the present study, this result is confirmed on phenyl hexyl phase, though it is shown to be of greater significance on the biphenyl column. The accepted explanation for this observation is that pi-pi interactions are shielded in an acetonitrile reversed phase environment because of the potential for charge transfer between the pi electrons of the solvent and stationary phase. While pi-pi molecular interactions are likely to occur, it is unclear whether the energetics of this interaction or the formation and proximity of a reversed phase interface driven by solvation of the stationary phase is the driving force for the differences observed. The explanation is likely more complex; a combination of molecular contributions and solvation dependent on organic mobile phase component that is not yet understood.

To interpret the chromatographic data for the substituted benzene compounds, a method was adapted based on the Hammet linear free energy approach.6 Using toluene as a reference compound, the hydrophobic contributions to selectivity were subtracted so that correlations could be made between remaining molecular contributions to retention from the stationary phase. Linear regressions are then used to correlate retention to previously calculated solute parameters in the determination of linear solvation energy relationships.7 Molecular descriptors were chosen that are most applicable to interactions for phenyl stationary phases based on previous characterizations of column selectivity, namely polarity, polarizability, hydrogen bond basicity, and steric contributions (shape selectivity).

Overlay of Substituted Benzene Compounds

Figure 3. Overlay of Substituted Benzene Compounds Comparing Selectivity on Different Stationary Phases and the Effect of Mobile Phase Organic Modifier

 

The analysis of the data in Figure 4 shows no correlation between hydrogen bond basicity of the analytes and retention on the biphenyl stationary phase, in contradiction to our previous observation in the Euerby column characterization. The most meaningful correlation was to dipole moment, with slightly less significant contributions from shape selectivity and polarizability. The polar nature of the aromatic ring, which has net negative charge above and below the plane of the ring because of the p-orbital conjugation appears to be the major driving force in retention. Whether this is an indication of pi-pi charge transfer or polarity dictating a difference in the stationary solvation remains unclear. The data clearly demonstrates that parameters like shape selectivity, polarizability, and polarity of the analyte should be parameters than can be exploited for successful method development. These factors will differentiate analytes and are the major contributors to selectivity that will ultimately promote resolution of critical pairs in an analysis. Given the small sample set of probes and the potentially limited span of molecular interactions for this group of analytes, a more in depth LSER study is likely needed, and does not yet exist for biphenyl type stationary phases in the primary literature.

:Correlations

Figure 4. Correlations Between Non-hydrophobic Contributions to Retention and Solute Parameters

Conclusions

The Ascentis® Express Biphenyl stationary phase demonstrates an enhancement of many of the effects associated with phenyl type stationary phases. Significant differences in selectivity based on choice of organic modifier distinguish it from C18 alkyl type phases. Factors of polarity, shape selectivity, and polarizability are the driving forces behind retention and selectivity, though the relative contributions of these interactions are still unknown. The differences compared to alkyl and even existing phenyl chemistries make this phase a suitable tool for method development in reversed-phase where existing approaches lack selectivity.

Materials

     

 References

  1. Brindle, R.R.; Albert, K.; J. Chrom. A.; 757, 1997, 3.
  2. Marchand, D.H.; Coes, K.; Dolan, J.W.; Snyder, L.R.; Henry, R.A.; Kallury, K.M.R.; Waite, S.; Carr, P.; J. Chrom. A.; 1062, 2005, 65-78.
  3. Euerby, m.R.; Peterson, P.; Campbell, W.; Roe, W.; J. Chrom. A.; 1154, 2007, 138-151.
  4. Tanaka, N.; Okuda, Y.; iwaguchi, K.; Araki, m.; J. Chrom. A.; 239, 1982, 761.
  5. Horak, J.; Maier, N.M.; Linder, W.; J. Chrom. A; 1045, 2004, 43-58.
  6. Hammet, L.P, Physical organic Chemistry, McGraw-Hill, New York, London, 1940 (Chapter 7).
  7. Snyder, L.R.; Dolan, J.W.; Carr, P.W.; J. Chrom. A; 1060, 2004, 77-116.

 

Related Links