# In-Depth Performance Analysis and Comparison of Monolithic and Particulate Zwitterionic Hydrophilic Interaction Liquid Chromatography Polymer Columns

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

## 1. Introduction

_{i}the interstitial velocity; D

_{eff}, D

_{pz}and D

_{m}the effective, porous zone, and bulk molecular diffusion coefficients, respectively; k″ the zone retention factor; ε

_{e}the external porosity; α a geometrical constant; Sh

_{m}and Sh

_{pz}the Sherwood numbers relating to the mobile zone and the porous zone, respectively, and d a characteristic length. In Equation (1), the first term (${H}_{inhom}$) relates to band broadening originating from flow heterogeneities in the bed (traditionally referred to as eddy dispersion). The second term (B-term) represents the effective longitudinal diffusion. The third and fourth terms (C

_{m}-, and C

_{s}-terms) are the resistance to mass transfer in the mobile and stationary zones, respectively. The zone retention factor is calculated as:

_{R}the analyte retention time, L the column length and u

_{i}the interstitial velocity:

_{p}, while for monolithic columns, several measures can be used, such as the domain size (d

_{dom}), the throughpore size (d

_{tp}) and the skeleton size (d

_{skel}). Note that:

_{eff}) and intra-particle diffusion coefficients (D

_{pz}, where “pz” stands for “mesoPorous Zone”), which was attributed to very low surface diffusion rates, a strongly hindered diffusion in the polymer backbone, slow localized adsorption events, or a combination thereof [21].

## 2. Results and Discussion

#### 2.1. Geometrical Characterization

_{dom}) of the monoliths was determined from the skeleton size (d

_{skel}) and the throughpore size (d

_{tp}), according to Equation (4), by analyzing at least 100 of each for each monolith, as indicated in Figure 1. Note that for d

_{skel}, intermediate-sized, repeating units were considered, while for the determination of d

_{tp}, the largest voids were omitted (as can for example be observed on the left side of Figure 1a). The average d

_{skel}and d

_{tp}of the poly(SPE-co-EDMA) monolithic column were in this way estimated to be 0.7 µm and 0.8 µm, respectively, resulting in a domain size of 1.5 µm, while the skeleton size and throughpore size of the poly(SPE-co-MBA) monolithic column were determined to be 1.7 µm and 1.6 µm, respectively, resulting in a domain size of 3.3 µm, reflecting the smaller features of the poly(SPE-co-EDMA) monolithic column. These values are shown in Table 1, together with the standard deviations (SD) for at least 100 measurements. It should be noted that the SD are generally high, especially for the throughpore sizes (coefficient of variation ~50%), demonstrating the high heterogeneity of the throughpores.

_{e}for each column were determined in [21] via ISEC experiments, while total porosities ε

_{T}were determined from the elution time of toluene using tetrahydrofuran as the mobile phase. The porosity of the porous zone ε

_{pz}was calculated as:

_{e}-values of around 69%, in line with typical ε

_{e}-values observed for monolithic columns. The poly(SPE-co-MBA) monolithic column, however, had a smaller total porosity ${\epsilon}_{T}$, implying a significantly smaller internal porosity of ε

_{pz}=12.5% compared to the ε

_{pz}= 29.5% for the poly(SPE-co-EDMA) column.

_{e}-values for particle packed columns. The external porosity as measured via ISEC, however, was rather large (around 44%), especially considering the relatively large particle size distribution that was deduced from the SEM pictures. Whereas most random packings of spherical particles have external porosities of 36–40%, packings consisting of particles with a higher particle size distribution are known to yield smaller ε

_{e}-values since the smaller particles can position themselves between the larger particles and in this way fill up some of the ‘gaps’ in the interstitial volume. In fact, in [22] it was demonstrated that dense random packings of particles with a mean sphericity of 0.86 and a dimensionless standard deviation between 0.1 and 0.6, have external porosities ranging between 35% and 30%, respectively. For a dimensionless standard deviation of 0.23, as is the case for the ZIC-pHILIC particles considered here, ε

_{e}would be 34%. Assuming the ISEC measurements are inaccurate because of the large intra-particle voids visible in Figure 1d, we therefore considered a value of 34% for the external porosity ε

_{e}and a value of 40% for ε

_{pz}for the ZIC-pHILIC column in what follows.

#### 2.2. Evaluation of Column Performance

#### 2.2.1. Plate Height Curves

_{pz}= 40%), this flattening of the plate height curve at high velocities could be due to a convective (perfusion) flow through the particles.

_{pore}versus the superficial velocity u

_{s}, with ${u}_{S}={u}_{i}{\epsilon}_{e}$, was calculated using the correlation developed by Afeyan et al. [23]:

_{p}the particle permeability, K the column permeability, d

_{p}the particle size and d

_{pore}the mesopore size, here taken as equal to 0.10 µm, as deduced from the SEM pictures.

_{e}for K and ε = ε

_{pz}for K

_{p}. The constant τ is related to the tortuosity of the packing, and has a typical value of 2. In this way, it can be calculated that the ratio u

_{pore}/u

_{s}is 0.003, or that 0.3% of the average velocity through the column passes through the mesopores of the particles. Although this seemingly only represents a small percentage of the average velocity, it is also important to consider the ‘time of transport’ through the particles via diffusion versus convection. This can be calculated as:

_{pore}in Equation (8) corresponds to an interstitial velocity of u

_{i}= 2.4 mm/s in the packed bed, the maximum u

_{i}measured during the plate height experiments. This is because the higher the velocity is, the more convection will become predominant. The calculations in Equations (8) and (9) show that the transport through the particles via diffusion is only four times more rapid than via convection at this highest velocity, indicating that transport via convection is significant in the ZIC-pHILIC particles. It is, however, not entirely clear whether the observed amount of convection is large enough to explain the observed flattening at the high-velocity end of the plate height curve. A possible explanation for this lower-than-expected contribution of the intra-particle convection could be due to the fact that a number of parameters in these calculations are based on estimations (ε

_{e}, ε

_{pz}, d

_{pore}), that could have resulted in a lower accuracy of the obtained results.

_{i}in Equation (1) is the C

_{s}-term (the fourth term), considering the Sh

_{m}-factor in the third term is velocity-dependent [25], it can be assumed that the plate heights obtained on the monoliths in the high-velocity range are C

_{s}-term dominated.

#### 2.2.2. Permeability Measurements

_{0}for the three evaluated columns. Note that for the construction of these curves, preference was given to the linear velocity u

_{0}over the interstitial velocity, as the total velocity through the column (including the zero-velocity inside the pores) determines the column permeability under actual separation conditions. The experimental pressure values were fitted to a linear equation, the obtained equations and goodness of fit (represented by R²-values) are also shown in Figure S1. In general, the pressures measured on all columns displayed a linear behavior with respect to the applied velocity, since all R² > 0.998. Looking closer at the values, it was, however, observed that the two monolithic columns displayed higher R²-values of 0.9997, whereas the R²-value obtained on the particulate column was slightly lower (0.998). When fitting the experimental pressure values obtained below 20 bar to a linear equation, a similar excellent R²-value of 0.9996 was obtained for the particle packed column. The values obtained above 20 bar deviated from this linearity in an upward manner. This could indicate that at pressures above 20 bar, the particles in the ZIC-pHILIC column become somewhat compressed and deformed, resulting in higher-than-expected pressures. The monolithic columns do not display this behavior, suggesting the monolithic structures are more mechanically stable at higher pressures, and hence inherently more suited as a chromatographic backbone in the case of polymeric stationary phases. Note that the maximum column pressure applied to all columns was below 100 bar. The maximum allowable backpressure specified by the manufacturer is 200 bar for the ZIC-pHILIC column.

_{v0}) values were subsequently calculated via Darcy’s law [26]:

_{v0}-values are shown as a function of u

_{0}in Figure 4. Interestingly, the K

_{v0}-values obtained for the two monolithic columns seem to stabilize above u

_{0}= 0.5 mm/s, whereas the values obtained for the poly(SPE-co-EDMA) column are higher for smaller u

_{0}-values, and those obtained on the poly(SPE-co-MBA) column are actually lower. It should be mentioned that the lowest K

_{v0}-values were obtained below the recommended operational range of the nanoLC flow selector (recommended range 50 nL–1000 nL/min, values measured here starting at 20 nL/min), which could have resulted in these deviating values. Note also that the standard deviations (denoted by the vertical error bars) obtained for the K

_{v0}-values at these low velocities are clearly larger. However, it is somewhat surprising that both monolithic columns show opposing trends in this low velocity range. For the ZIC-pHILIC column, a generally decreasing trend of K

_{v0}with increasing flow rate was observed. This reflects the higher-than-expected observed pressure values and could hence indicate a compression of the packed bed at higher velocities.

#### 2.2.3. Reduced Plate Height Curves

_{i}were subsequently constructed to further investigate the band broadening behavior observed in the different columns. Note that for the calculation of the reduced plate height h and the reduced interstitial velocity ν

_{i}a characteristic length d needs to be specified:

_{dom}) can for example be used. Figure S2 in the Supporting Information shows the reduced plate height curves that were obtained using the particle size for the particle packed column, and the domain size for the monolithic columns, as specified in Table 1.

_{min}= 6–9) are now much closer to those obtained on the poly(SPE-co-MBA) column (h

_{min}= 4–6), while those obtained on the poly(SPE-co-EDMA) column (h

_{min}= 9–18) are much higher. Note also the much steeper slope of the plate height curve in the high velocity range of the latter. This is entirely due to the much smaller characteristic lengths obtained for the poly(SPE-co-EDMA) column (Table 1), impacting the calculation of both the reduced plate height h and the reduced velocity ν

_{i}, as shown in Equations (12) and (13). However, as was already mentioned earlier, the standard deviations observed for these characteristic lengths were quite large, raising suspicions about the validity of using the domain size as the characteristic length for the monolithic columns. As an alternative, the permeability-based characteristic length proposed by Halasz (d

_{Halasz}) was therefore investigated next [27]:

_{Halasz}for the different columns, the permeability values K

_{v0}obtained at the highest measured pressure were used. This resulted in values of d

_{Halasz}= 5.2, 5.7 and 5.5 µm for the poly(SPE-co-EDMA), the poly(SPE-co-MBA) column and the ZIC-pHILIC column, respectively (Table 1). Note that these values are in very close agreement with each other despite the completely different structure of the packing. Figure 5 shows the reduced plate height curves obtained using d

_{Halasz}as the characteristic length. Unsurprisingly, given the close proximity of the d

_{Halasz}-values, the curves show similar trends as observed for their non-reduced counterparts in Figure 3. Minimum reduced plate heights observed for the monolithic columns are h

_{min}= 3–5 for the poly(SPE-co-EDMA) column, and h

_{min}= 2–4 for the poly(SPE-co-MBA) column, and hence slightly lower for the latter, in line with the more homogeneous structure of the poly(SPE-co-MBA) column (Figure 1b). Minimum plate heights for the ZIC-pHILIC column are h

_{min}= 5–7, and hence larger than for the monolithic column, while also the c-term is steeper and shows the same curvature/flattening as in Figure 3c. Despite the fact that these curves seem to present a more realistic view on the performance of the columns, and are more in line with one another, it should be mentioned that the square root of K

_{v0}in fact has no structural meaning, and can hence not be linked to the morphology, disorder or heterogeneity of the columns.

_{s}-term (at least for the monolithic columns), which decreases with increasing k″, as can be deduced from Equation (1), this is somewhat surprising as a systematic decrease in the c-term region with k″ would be expected. However, surprisingly, there does seem to be a correlation between the molecular weight (MW) of the compounds (Table S1 in the Supporting Information) and the order of the plate height curves, where the highest MW compounds generally have the steepest c-terms, and the low MW compounds the flattest c-terms. One exception to this behavior is the compound with a k″ = 9.69 on the poly(SPE-co-MBA) column.

_{s}-term dominated, at least for the monolithic columns, Equation (1) demonstrates the importance of intra-particle diffusion (D

_{pz}) in the c

_{s}-term. D

_{pz}can be written as [28]:

_{s}D

_{s}represents the stationary phase diffusion, γ

_{mp}D

_{m}the mesopore diffusion, and k

_{0}″ the zone retention factor k″ for k′= 0. Note that the phase retention factor k’ and the zone retention factor k″ are related as follows:

_{0}″ is a structural feature of the column packing, depending only on the interstitial porosity ε

_{e}and the total porosity ε

_{T}.

_{pz}/D

_{m}and hence D

_{pz}are very low for the columns evaluated in this work, and this was attributed to a very low amount of surface diffusion, a very strong and localized adsorption mechanism and/or strongly hindered diffusion in the polymer matrix. Under these circumstances, it can be assumed that the contribution of the stationary phase diffusion to the overall intra-particle diffusion is very low, or in other words:

_{part}= 10 [17].

_{mp}:

_{mp}essentially depends on structural column characteristics (d, ε

_{e}and ε

_{T}), C

_{s}, D

_{m}and k″. Even though the actual value of d is not entirely clear for the monolithic columns, this value is constant per column. In other words, when evaluating a single column, the exact value of d does not matter that much, as long as the same value is consistently used for that column. Therefore, d

_{dom}was taken to calculate γ

_{mp}for the monolithic columns. C

_{s}-values were taken as equal to the C-term values obtained by fitting the plate height curves to Equation (10), as shown in Table 2, considering the C-term region was C

_{s}-dominated. Figure 6 shows the obtained calculated values of γ

_{mp}plotted as a function of the MW of the compounds for the two monolithic columns. Interestingly, a rough trend can be observed where γ

_{mp}seems to decrease as the MW increases, with the exception of the compound with a MW = 268 g/mol (k″ = 9.69) on the poly(SPE-co-MBA) column. Although this is highly speculative, and more data are required to confirm this trend, these observations seem to suggest that compounds experience more obstruction against free movement in the mesoporous space of the monolithic polymer matrix, as their MW increases. Since it was impossible to obtain the C

_{s}-term for the ZIC-pHILIC column, as its fitted C-term was also impacted by the observed flattening of the curve, the same calculations were not made for the ZIC-pHILIC column.

#### 2.2.4. Kinetic Plot Analysis

_{max}the maximum pressure the column can withstand and η the mobile phase viscosity.

_{0}versus H is converted into a measure of the time that is required to obtain a certain plate count when operating the column at the maximum pressure, and at the corresponding velocity u

_{0}. In this way, each datapoint of t

_{0}versus N is in fact obtained in a different column length, where low values of u

_{0}will typically be obtained in long column lengths and hence result in high N-values, while high u

_{0}-values will typically be obtained in short column lengths and hence result in lower N-values.

_{max}= 200 bar and are shown in Figure 7. The compound with k″~2 was the same for all columns (uracil) and was purposely chosen to avoid any bias on the column performance that might be due to MW effects. The plots in Figure 7 show that the two monolithic columns outperform the ZIC-pHILIC column over the entire relevant range of plate counts between 10

^{3}and 8 × 10

^{5}plates. This is entirely attributed to the higher efficiency of the monolithic columns, as the permeability values of the monoliths and the particle-packed column are relatively similar.

_{v0}-values and slightly better efficiency of the former. The ZIC-pHILIC becomes more performant than the two monolithic columns for very high plate counts (≥10

^{6}plates) only. This is due to the slightly lower B-term values that were obtained for the ZIC-pHILIC column, as can be deduced from Table 2. As the high N-range of the kinetic plot is typically obtained at very low velocities, where the B-term is dominant, this lower B-term results in the better performance of the ZIC-pHILIC column in this region.

## 3. Materials and Methods

#### 3.1. Reagents and Materials

#### 3.2. Instrumentation

#### 3.3. Samples and Mobile Phases

_{2}O (50:50, v:v). These stock solutions were subsequently diluted in pure ACN to a final concentration of 50 ppm for each compound. Mobile phases were prepared by mixing ACN in different ratios with an ammonium acetate solution (adjusted to pH = 6.0 with glacial acetic acid) as shown in Table S1 in the Supporting Information [21]. The ammonium acetate concentration in Table S1 represents the total concentration in the mobile phase. Molecular diffusion coefficients (D

_{m}) for each compound in their respective mobile phases were determined via Taylor-Aris experiments, as detailed in [21]. The obtained D

_{m}-values are also given in Table S1.

#### 3.4. Plate Height Measurements

_{R}were obtained from the first moments of the peaks, while peak widths were determined at 4.4% of the peak height. All measured data were corrected for the extra-column contribution (ECC). For the Agilent 1290 UHPLC system, the ECC was experimentally determined using a zero-dead volume union instead of the column. For the Dionex Ultimate 3000 RSLC nano system, the ECC was calculated from the geometrical volume of the tubing, the injection volume and the volume of the flow cell.

## 4. Conclusions

_{e}= 0.34 was considered for further calculations, based on literature data for packings with similar particle size distributions.

_{s}-term dominated, revealed a correlation between the obstruction factor in the mesoporous zone and the molecular weight of the compounds. Although speculative at this instance, this suggests that a compound experiences more obstruction against free movement in the mesopores of these polymer monoliths as its MW increases.

## Supplementary Materials

_{i}obtained by using the domain size (d

_{dom}) as the characteristic length for the monolithic columns, and the particle size (d

_{p}) for the particle packed column.

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Acknowledgments

## Conflicts of Interest

## References

- Bayatloo, M.R.; Tabani, H.; Nojavan, S.; Alexovič, M.; Ozkan, S.A. Liquid-Phase Microextraction Approaches for Preconcentration and Analysis of Chiral Compounds: A Review on Current Advances. Crit. Rev. Anal. Chem.
**2022**, 1–15. [Google Scholar] - Schiesel, S.; Lämmerhofer, M. Multitarget quantitative metabolic profiling of hydrophilic metabolites in fermentation broths of β -lactam antibiotics production by HILIC–ESI–MS/MS. Anal. Bioanal. Chem.
**2010**, 396, 1655–1679. [Google Scholar] [CrossRef] [PubMed] - Nelis, M.; Decraecker, L.; Boeckxstaens, G.; Augustijns, P.; Cabooter, D. Development of a HILIC-MS/MS method for the quantification of histamine and its main metabolites in human urine samples. Talanta
**2020**, 220, 121328. [Google Scholar] [CrossRef] [PubMed] - Díez, C.; Guillarme, D.; Spörri, A.S.; Cognard, E.; Ortelli, D.; Edder, P.; Rudaz, S. Aminoglycoside analysis in food of animal origin with a zwitterionic stationary phase and liquid chromatography–tandem mass spectrometry. Anal. Chim. Acta
**2015**, 882, 127–139. [Google Scholar] [CrossRef] [PubMed] - Kahsay, G.; Song, H.; Van Schepdael, A.; Cabooter, D.; Adams, E. Hydrophilic interaction chromatography (HILIC) in the analysis of antibiotics. J. Pharm. Biomed. Anal.
**2014**, 87, 142–154. [Google Scholar] [CrossRef] - Ikegami, T.; Horie, K.; Saad, N. Highly efficient analysis of underivatized carbohydrates using monolithic-silica-based capillary hydrophilic interaction (HILIC) HPLC. Anal. Bioanal. Chem.
**2008**, 391, 2533–2542. [Google Scholar] [CrossRef] - Spicer, V.; Krokhin, O.V. Peptide retention time prediction in hydrophilic interaction liquid chromatography. Comparison of separation selectivity between bare silica and bonded stationary phases. J. Chromatogr. A
**2018**, 1534, 75–84. [Google Scholar] [CrossRef] - Gilar, M.; Jaworski, A. Retention behavior of peptides in hydrophilic-interaction chromatography. J. Chromatogr. A
**2011**, 1218, 8890–8896. [Google Scholar] [CrossRef] - Periat, A.; Fekete, S.; Cusumano, A.; Veuthey, J.; Beck, A.; Lauber, M.; Guillarme, D. Potential of hydrophilic interaction chromatography for the analytical characterization of protein biopharmaceuticals. J. Chromatogr. A
**2016**, 1448, 81–92. [Google Scholar] [CrossRef] [Green Version] - Alpert, A.J. Hydrophilic-interaction chromatography for the separation of peptides, nucleic acids and other polar compounds. J. Chromatogr. A
**1990**, 499, 177–196. [Google Scholar] [CrossRef] - McCalley, D.V.; Neue, U.D. Estimation of the extent of the water-rich layer associated with the silica surface in hydrophilic interaction chromatography. J. Chromatogr. A
**2008**, 1192, 225–229. [Google Scholar] [CrossRef] [PubMed] - Buszewski, B.; Noga, S. Hydrophilic interaction liquid chromatography (HILIC)-a powerful separation technique. Anal. Bioanal. Chem.
**2012**, 402, 231–247. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Jandera, P. Stationary and mobile phases in hydrophilic interaction chromatography: A review. Anal. Chim. Acta
**2011**, 692, 1–25. [Google Scholar] [CrossRef] [PubMed] - Heaton, J.C.; Mccalley, D.V. Comparison of the kinetic performance and retentivity of sub-2 µm core–shell, hybrid and conventional bare silica phases in hydrophilic interaction chromatography. J. Chromatogr. A
**2014**, 1371, 106–116. [Google Scholar] [CrossRef] [PubMed] - Chauve, B.; Guillarme, D.; Cléon, P.; Veuthey, J.L. Evaluation of various HILIC materials for the fast separation of polar compounds. J. Sep. Sci.
**2010**, 33, 752–764. [Google Scholar] [CrossRef] - Jiang, Z.; William, N.; Liu, Z. Preparation and application of hydrophilic monolithic columns. J. Chromatogr. A
**2011**, 1218, 2350–2361. [Google Scholar] [CrossRef] - Desmet, G.; Broeckhoven, K. Equivalence of the different Cm- and Cs-term expressions used in liquid chromatography and a geometrical model uniting them. Anal. Chem.
**2008**, 80, 8076–8088. [Google Scholar] [CrossRef] - Giddings, C. Dynamics of Chromatography: Principles and Theory; CRC Press: New York, NY, USA, 1965. [Google Scholar]
- Lapidus, L.; Amundson, N.R. Mathematics of Adsorption in Beds. VI. The Effect of Longitudinal Diffusion in Ion Exchange and Chromatographic Columns. J. Phys. Chem.
**1952**, 56, 984–988. [Google Scholar] [CrossRef] - Horvath, C.; Lin, H.-J. Movement and band spreading liquid chromatography. J. Chromatogr. A
**1976**, 126, 401–420. [Google Scholar] [CrossRef] - Li, H.; Desmet, G.; Jiang, Z.; Cabooter, D. On the occurrence of very low intra-particle diffusion rates in zwitterionic hydrophilic interaction liquid chromatography polymer columns. J. Chromatogr. A
**2022**, 1683, 463531. [Google Scholar] [CrossRef] - Sohn, H.Y.; Moreland, C. The effect of particle size distribution on packing density. Can. J. Chem. Eng.
**1968**, 46, 162–167. [Google Scholar] [CrossRef] - Afeyan, N.B.; Gordon, N.; Mazsaroff, I.; Varady, L.; Fulton, S.P.; Yang, Y.B.; Regnier, F.E. Flow-through particles for the high-performance liquid chromatographic separation of biomolecules: Perfusion chromatography. J. Chromatogr. A
**1990**, 519, 1–29. [Google Scholar] [CrossRef] - Desmet, G.; Song, H.; Makey, D.; Stoll, D.R.; Cabooter, D. Experimental investigation of the retention factor dependency of eddy dispersion in packed bed columns and relation to knox’s empirical model parameters. J. Chromatogr. A
**2020**, 1626, 461339. [Google Scholar] [CrossRef] - Deridder, S.; Desmet, G. New insights in the velocity dependency of the external mass transfer coefficient in 2D and 3D porous media for liquid chromatography. J. Chromatogr. A
**2012**, 1227, 194–202. [Google Scholar] [CrossRef] [PubMed] - Bear, J. Dynamics of Fluids in Porous Media; Dover Publications: New York, NY, USA, 1988. [Google Scholar]
- Halasz, I.; Görlitz, G. Optimal Parameters in High Speed Liquid Chromatography (HPLC). Angew. Chem. Int. Ed.
**1982**, 21, 50–61. [Google Scholar] [CrossRef] - Desmet, G.; Broeckhoven, K.; De Smet, J.; Deridder, S.; Baron, G.V.; Gzil, P. Errors involved in the existing B-term expressions for the longitudinal diffusion in fully porous chromatographic media Part I: Computational data in ordered pillar arrays and effective medium theory. J. Chromatogr. A
**2008**, 1188, 171–188. [Google Scholar] [CrossRef] - De Villiers, A.; Lestremau, F.; Szucs, R.; Gélébart, S.; David, F.; Sandra, P. Evaluation of ultra performance liquid chromatography. Part I. Possibilities and limitations. J. Chromatogr. A
**2006**, 1127, 60–69. [Google Scholar] [CrossRef] - Hsieh, S.; Jorgenson, J.W. Preparation and evaluation of slurry-packed liquid chromatography microcolumns with inner diameters from 12 to 33 μm. Anal. Chem.
**1996**, 68, 1212–1217. [Google Scholar] [CrossRef] - Desmet, G.; Clicq, D.; Gzil, P. Geometry-independent plate height representation methods for the direct comparison of the kinetic performance of LC supports with a different size or morphology. Anal. Chem.
**2005**, 77, 4058–4070. [Google Scholar] [CrossRef] - Clicq, D.; Heinisch, S.; Rocca, J.L.; Cabooter, D.; Gzil, P.; Desmet, G. Use of the kinetic plot method to analyze commercial high-temperature liquid chromatography systems II. Practically constrained performance comparison. J. Chromatogr. A
**2007**, 1146, 193–201. [Google Scholar] [CrossRef] - Eeltink, S.; Decrop, W.M.C.; Steiner, F.; Ursem, M.; Cabooter, D.; Desmet, G.; Kok, W.T. Use of kinetic plots for the optimization of the separation time in ultra-high-pressure LC. J. Sep. Sci.
**2010**, 33, 2629–2635. [Google Scholar] [CrossRef] [PubMed] - Jiang, Z.; Smith, N.W.; Ferguson, P.D.; Taylor, M.R. Hydrophilic interaction chromatography using methacrylate-based monolithic capillary column for the separation of polar analytes. Anal. Chem.
**2007**, 79, 1243–1250. [Google Scholar] [CrossRef] [PubMed] - Pena-Pereira, F.; Wojnowski, W.; Tobiszewski, M. AGREE—Analytical GREEnness Metric Approach and Software. Anal. Chem.
**2020**, 92, 10076–10082. [Google Scholar] [CrossRef] [PubMed]

**Figure 1.**Scanning electron microscopy (SEM) pictures of the evaluated column materials. (

**a**) poly(SPE-co-EDMA) monolithic stationary phase, (

**b**) poly(SPE-co-MBA) monolithic stationary phase, (

**c**) ZIC-pHILIC particles and (

**d**) close-up of particle surface. Red lines in (

**a**,

**b**) indicate the size of d

_{glob}and yellow lines the size of d

_{tp}. Red lines in (

**d**) indicate the ‘skeleton’ sizes of the polymer structure, yellow lines indicate the mesopore sizes.

**Figure 2.**Chromatograms obtained on the (

**a**) poly(SPE-co-EDMA) monolithic column, flow rate = 0.0006 mL/min; (

**b**) poly(SPE-co-MBA) monolithic column, flow rate = 0.0006 mL/min; (

**c**) ZIC-pHILIC column, flow rate = 0.1 mL/min. Mobile phase compositions are shown in Table S1; column temperature: room temperature; peak annotation: (1) uracil, (2) adenosine, (3) thiourea, (4) uridine, (5) inosine, (6) hypoxanthine. The t

_{0}-marker was toluene.

**Figure 3.**Plate height curves of H versus u

_{i}obtained on the (

**a**) poly(SPE-co-EDMA) monolithic column, (

**b**) poly(SPE-co-MBA) monolithic column and (

**c**) ZIC-pHILIC column. k″ = 1.9–2.2 (), k″ = 2.9–3.7 (), k″ = 4.7–6.4 (), k″ = 6.4–7.3 (), k″ = 9.7–10.5 (). Mobile-phase conditions are given in Table S1.

**Figure 4.**Curves of permeability (K

_{v0}) as a function of the linear velocity (u

_{0}) for the columns evaluated in this work: () poly(SPE-co-EDMA) monolithic column, () poly(SPE-co-MBA) monolithic column, () ZIC-pHILIC column. The error bars are the standard deviations obtained from three replicate measurements.

**Figure 5.**Reduced plate height curves of h versus ν

_{i}obtained by using d

_{Halasz}(Equation (14)) as the characteristic length for the (

**a**) poly(SPE-co-EDMA) monolithic stationary phase, (

**b**) poly(SPE-co-MBA) monolithic stationary phase and (

**c**) ZIC-pHILIC column. k″= 1.9–2.2 (), k″= 2.9–3.7 (), k″= 4.7–6.4 (), k″= 6.4–7.3 (), k″= 9.7–10.5 (). Mobile-phase conditions as in Table S1.

**Figure 6.**Plots of γ

_{mp}versus MW for the (

**a**) poly(SPE-co-EDMA) monolithic column and (

**b**) poly(SPE-co-MBA) monolithic column.

**Figure 7.**Kinetic plots of time (t

_{0}) versus plate count (N) for the materials evaluated in this work: () poly(SPE-co-EDMA) monolithic column, () poly(SPE-co-MBA) monolithic column, () ZIC-pHILIC column. Component = uracil, having k″ ≅ 2 on all three materials.

**Table 1.**Structural characteristics (d

_{glob}, d

_{tp}, d

_{dom}and d

_{Halasz}) and column porosities of the columns evaluated in this work. The reported sizes show the average values of at least 100 independent measurements and their standard deviations.

Column | ${\mathbf{d}}_{\mathbf{s}\mathbf{k}\mathbf{e}\mathbf{l}}$ (µm) | ${\mathbf{d}}_{\mathbf{t}\mathbf{p}}$ (µm) | ${\mathbf{d}}_{\mathbf{p}}$ or ${\mathbf{d}}_{\mathbf{d}\mathbf{o}\mathbf{m}}$ (µm) | d_{Halasz} (µm) | ${\mathbf{\epsilon}}_{\mathbf{e}}$ | ${\mathbf{\epsilon}}_{\mathbf{T}}$ | ${\mathbf{\epsilon}}_{\mathbf{i}}$ |
---|---|---|---|---|---|---|---|

poly(SPE-co-EDMA) | 0.7 ± 0.1 | 0.8 ± 0.4 | 1.5 ± 0.4 | 5.2 | 0.6923 | 0.7830 | 0.2948 |

poly(SPE-co-MBA) | 1.7 ± 0.3 | 1.6 ± 0.7 | 3.3 ± 0.8 | 5.7 | 0.6995 | 0.7370 | 0.1248 |

ZIC-pHILIC | / | / | 4.7 ± 1.1 | 5.5 | 0.4398 | 0.6040 | 0.2931 |

**Table 2.**A-, B- and C-term values obtained by fitting the experimental plate height data to Equation (10), either fitting all terms freely (BFIT) or fixing the B-term value to the value obtained by peak parking in [21], and subsequently fitting A and B (BPP).

Column | Compound | k″ | H_{min} (mm) | Equation (10) B_{FIT} | Equation (10) B_{PP} | ||||
---|---|---|---|---|---|---|---|---|---|

A (mm^{1/2}/s^{1/2}) | B (mm²/s) | C (s) | A (mm^{1/2}/s^{1/2}) | B (mm²/s) | C (s) | ||||

SPE-co-EDMA | Uracil | 1.92 | 2.17 × 10^{−2} | 1.38 × 10^{−2} | 3.35 × 10^{−3} | 1.22 × 10^{−2} | 1.25 × 10^{−2} | 3.48 × 10^{−3} | 1.30 × 10^{−2} |

Uracil | 3.08 | 2.76 × 10^{−2} | 2.29 × 10^{−2} | 3.80 × 10^{−3} | 1.31 × 10^{−2} | 2.89 × 10^{−2} | 3.27 × 10^{−3} | 8.74 × 10^{−3} | |

Thiourea | 4.96 | 1.37 × 10^{−2} | 6.90 × 10^{−3} | 3.83 × 10^{−3} | 3.71 × 10^{−3} | 7.59 × 10^{−3} | 3.70 × 10^{−3} | 3.21 × 10^{−3} | |

Thiourea | 7.00 | 1.53 × 10^{−2} | 8.43 × 10^{−3} | 4.11 × 10^{−3} | 4.11 × 10^{−3} | 9.53 × 10^{−3} | 3.91 × 10^{−3} | 3.42 × 10^{−3} | |

Hypoxanthine | 10.29 | 2.47 × 10^{−2} | 1.46 × 10^{−2} | 3.53 × 10^{−3} | 2.18 × 10^{−2} | 1.39 × 10^{−2} | 3.64 × 10^{−3} | 2.23 × 10^{−2} | |

SPE-co-MBA | Uracil | 1.86 | 1.70 × 10^{−2} | 1.78 × 10^{−2} | 2.08 × 10^{−3} | 1.38 × 10^{−3} | 1.56 × 10^{−2} | 2.31 × 10^{−3} | 2.80 × 10^{−3} |

Adenosine | 2.93 | 2.12 × 10^{−2} | 2.22 × 10^{−2} | 1.53 × 10^{−3} | 1.96 × 10^{−2} | 1.96 × 10^{−2} | 1.80 × 10^{−3} | 2.11 × 10^{−2} | |

Thiourea | 4.73 | 1.22 × 10^{−2} | 8.80 × 10^{−3} | 2.66 × 10^{−3} | 1.63 × 10^{−3} | 6.25 × 10^{−3} | 3.05 × 10^{−3} | 3.20 × 10^{−3} | |

Uridine | 6.37 | 1.48 × 10^{−2} | 1.25 × 10^{−2} | 1.47 × 10^{−3} | 1.21 × 10^{−2} | 1.21 × 10^{−2} | 1.72 × 10^{−3} | 1.20 × 10^{−2} | |

Inosine | 9.69 | 1.36 × 10^{−2} | 1.42 × 10^{−2} | 1.30 × 10^{−3} | 5.98 × 10^{−3} | 1.41 × 10^{−2} | 1.25 × 10^{−3} | 6.04 × 10^{−3} | |

ZIC-pHILIC | Uracil | 2.18 | 2.78 × 10^{−2} | 3.84 × 10^{−2} | 2.07 × 10^{−3} | 0.00 | 3.92 × 10^{−2} | 1.84 × 10^{−3} | 0.00 |

Adenosine | 3.70 | 4.11 × 10^{−2} | 8.70 × 10^{−2} | 1.37 × 10^{−3} | 9.12 × 10^{−3} | 8.31 × 10^{−2} | 1.50 × 10^{−3} | 1.26 × 10^{−2} | |

Uridine | 6.39 | 2.87 × 10^{−2} | 4.98 × 10^{−2} | 1.29 × 10^{−3} | 5.71 × 10^{−3} | 4.42 × 10^{−2} | 1.61 × 10^{−3} | 1.05 × 10^{−2} | |

Uridine | 7.31 | 3.26 × 10^{−2} | 5.45 × 10^{−2} | 1.50 × 10^{−3} | 9.31 × 10^{−3} | 5.84 × 10^{−2} | 1.25 × 10^{−3} | 5.97 × 10^{−3} | |

Uridine | 10.54 | 3.14 × 10^{−2} | 4.71 × 10^{−2} | 1.62 × 10^{−3} | 1.31 × 10^{−2} | 5.33 × 10^{−2} | 1.33 × 10^{−3} | 7.41 × 10^{−3} |

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |

© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Li, H.; Jiang, Z.; Desmet, G.; Cabooter, D.
In-Depth Performance Analysis and Comparison of Monolithic and Particulate Zwitterionic Hydrophilic Interaction Liquid Chromatography Polymer Columns. *Molecules* **2023**, *28*, 2902.
https://doi.org/10.3390/molecules28072902

**AMA Style**

Li H, Jiang Z, Desmet G, Cabooter D.
In-Depth Performance Analysis and Comparison of Monolithic and Particulate Zwitterionic Hydrophilic Interaction Liquid Chromatography Polymer Columns. *Molecules*. 2023; 28(7):2902.
https://doi.org/10.3390/molecules28072902

**Chicago/Turabian Style**

Li, Haibin, Zhengjin Jiang, Gert Desmet, and Deirdre Cabooter.
2023. "In-Depth Performance Analysis and Comparison of Monolithic and Particulate Zwitterionic Hydrophilic Interaction Liquid Chromatography Polymer Columns" *Molecules* 28, no. 7: 2902.
https://doi.org/10.3390/molecules28072902