The hydrophilicity vs. ion interaction selectivity plot revisited: The effect of mobile phase pH and buffer concentration on hydrophilic interaction liquid chromatography selectivity behavior

G Model

ARTICLE IN PRESS

CHROMA-357688; No. of Pages 8

Journal of Chromatography A, xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma

The hydrophilicity vs. ion interaction selectivity plot revisited: The effect of mobile phase pH and buffer concentration on hydrophilic interaction liquid chromatography selectivity behavior Chad D. Iverson, Xinyun Gu, Charles A. Lucy ∗ Department of Chemistry, University of Alberta, Gunning/Lemieux Chemistry Centre, Edmonton, Alberta T6G 2G2, Canada

a r t i c l e

i n f o

Article history: Received 17 March 2016 Received in revised form 16 June 2016 Accepted 18 June 2016 Available online xxx Keywords: Hydrophilic interaction liquid chromatography (HILIC) Electrostatic interactions Selectivity Silanol HPLC Stationary phase

a b s t r a c t This work systematically investigates the selectivity changes on many HILIC phases from w w pH 3.7–6.8, at 5 and 25 mM buffer concentrations. Hydrophilicity (kcytosine /kuracil ) vs. ion interaction (kBTMA /kuracil ) selectivity plots developed by Ibrahim et al. (J. Chromatogr. A 1260 (2012) 126–131) are used to investigate the effect of mobile phase changes on the selectivity of 18 HILIC columns from various classes. “Selectivity change plots” focus on the change in hydrophilicity and ion interaction that the columns exhibit upon changing mobile phase conditions. In general, the selectivity behavior of most HILIC columns is dominated by silanol activity. Minimal changes in selectivity are observed upon changing pH between w w pH 5 and 6.8. However, a reduction in ionic interaction is observed when the buffer concentration is increased at w w pH ≥ 5.0 due to ionic shielding. Reduction of the w w pH to < 5.0 results in decreasing cation exchange activity due to silanol protonation. Under all eluent conditions, the majority of phases show little change in their hydrophilicity. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Since Alpert [1] introduced the term hydrophilic interaction liquid chromatography (HILIC) in 1990, HILIC’s popularity has steadily increased. This popularity is due to HILIC’s ability to retain and resolve highly polar analytes that are difficult to separate by reversed phase chromatography, and to HILIC’s compatibility with mass spectrometry [2–6]. HILIC has been the subject of many reviews [2,4,5,7–11] and several studies covering a wide range of HILIC stationary phases and/or mobile phase conditions [12–16]. HILIC has found utility in the analysis of small molecules such as metabolites [2,17,18], pharmaceuticals [19–24] and food chemicals [25–32], as well as larger biomolecules such as glycans [33–35] and peptides/proteins [36,37]. Today, many types of HILIC phases are commercially available, including bare silica, amine, amide, diol, and zwitterionic phases [8,38]. Retention in HILIC is due to a combination of the analyte partitioning into a surface water layer that forms on the surface of polar particles in the presence of an acetonitrile (ACN)- (or other suitable aqueous miscible polar aprotic organic solvent) rich mobile

∗ Corresponding author. E-mail address: [email protected] (C.A. Lucy).

phase, and direct interactions, such as adsorption, with the stationary phase surface [7,39–41]. Secondary interactions such as dipole-dipole, hydrophilic interactions, hydrophobic interactions, and electrostatic interactions are responsible for the different selectivity classes of HILIC phases [12,15,42]. In 2011, Dinh et al. [15] characterized these interactions on 22 HILIC stationary phases using principal component analysis (PCA). Their approach successfully classified the behavior of the different phases, but was necessarily complex. Inspired by the two-dimensional RPLC selectivity plot developed by Neue and co-workers [43,44], Ibrahim et al. [45] developed several twodimensional plots to characterize the selectivity of HILIC phases based on the relative retention of a subset of the test probes studied by Dinh et al. [15]. The objective of these two-dimensional plots was to frame HILIC selectivity in a format that was visually easier to comprehend. One drawback of the selectivity plots of Dinh et al. [15] and Ibrahim et al. [45] are that they reflect HILIC selectivity under a single set of mobile phase conditions. Altering mobile phase conditions such as pH and buffer concentration can fine-tune HILIC selectivity by affecting water layer thickness, silanol activity and/or the ionization state of polar bonded groups [8–11,16,46–48]. Herein we investigate the effect of pH and buffer concentration on the selectivity behavior of many classes of HILIC phases.

http://dx.doi.org/10.1016/j.chroma.2016.06.061 0021-9673/© 2016 Elsevier B.V. All rights reserved.

Please cite this article in press as: C.D. Iverson, et al., The hydrophilicity vs. ion interaction selectivity plot revisited: The effect of mobile phase pH and buffer concentration on hydrophilic interaction liquid chromatography selectivity behavior, J. Chromatogr. A (2016), http://dx.doi.org/10.1016/j.chroma.2016.06.061

G Model CHROMA-357688; No. of Pages 8

ARTICLE IN PRESS

2.3. Tested columns and test probes

4.6 4.6 4.6 2.1 4.6 3.0 4.6 4.6 4.6 4.6 4.6 4.6 4.6 4.6 4.6 4.6 4.6 4.6 4.6 4.6 4.6 100 100 100 150 150 50 150 150 150 150 100 100 100 100 150 150 100 100 100 150 150 160 300 150 185 300 300 450 180 – 180 450 130 380 150 440 188 300 150 300 188 300

Brand Name

Zorbax HILIC Plus Chromolith Si Ascentis Express HILIC Xbridge HILIC Cosmosil HILIC Ultra Amino TSKgel NH2- 100 ZIC HILIC ZIC-pHILIC ZIC-cHILIC TSKgel Amide-80 AdvanceBio Glycan Mapping Fortis HILIC Diol Ascentis Express OH5 FRULIC-N PolyHYDROXYETHYL A Ultra IBD Ascentis Express F5 Ultra PFP PolySULFOETHYL A Acclaim HILIC 10

#

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17a 18a 19a 20 21

The premixed mobile phases consisted of 80:20 v/v ACN:water containing ammonium acetate (w w pH 6.8 or 5.0) or ammonium formate (w w pH 3.7). Analytes (20 ␮L partial loop fill injection) were separated under ambient temperature (23 ± 3 ◦ C) at 0.5 mL/min and detected at 254 nm. Prior to the first injection under each condition, each column was equilibrated with ≥45 column volumes of fresh eluent. The buffer strength quoted (5 or 25 mM) is that present after ACN addition. Buffers were prepared by titrating the appropriate amount of the above salts with HCl. The % ACN quoted in this work represents the volume of ACN relative to the total volume of the solvents including buffer and ACN. The pH for each aqueous component was measured prior to adding ACN (w w pH). Early studies suggested that the w w pH value is more representative of the surface aqueous layer in HILIC [42,48]. More recently, direct pH measurement of the buffered aqueous/organic mobile phase (after calibrating in aqueous buffers; w s pH) has been advocated [12,49,50]. As neither measure is truly an ideal descriptor of the mobile phase acidity, both the w s pH and w w pH are quoted.

Table 1 Characteristics of the stationary phases evaluated in this study.

2.4. Chromatographic conditions

Manufacturer

Support

Functionality

Table 1 lists the 21 columns evaluated in this study and their characteristics. Retention factors (k) for cytosine, uracil and BTMA were calculated as the average of three injections of the standards prepared in the buffered eluent. Toluene was used as the unretained dead time marker (t0 ) for all HILIC phases. The standard deviations of the retention ratio measurements shown in Fig. 1 are smaller than the size of the marker symbol (RSD’s typically <1%). All analytes studied herein do not display acid/base character with the pH range studied. Thus, lower capacity buffers could be used, albeit with addition of buffer to the injected sample and extensive column equilibration (Section 2.4). With weak acid and base analytes, additional selectivity changes will result from protonation of the analyte, in which case adequate buffering capacity is essential.

Column was studied but excluded from plots as it exhibited reversed phase behavior under the conditions studied.

Particle size (␮m)

All solutions were prepared with Nanopure water (Barnstead, Dubuque, IA, USA). Cytosine, uracil, Optima-grade ACN, and HPLCgrade ammonium formate were from Sigma Aldrich (St. Louis, MO, USA). HPLC-grade ammonium acetate was from Fisher Scientific (Fair Lawn, NJ, USA). HCl was from Caledon Laboratory Chemicals (Caledon, ON, Canada) Benzyltrimethylammonium chloride (BTMA) was from Acros Organics (Fair Lawn, NJ, USA).

a

Pore size (Å)

2.2. Chemicals and reagents

95 130 90 130 120 100 100 100 – 100 80 120 100 90 100 200 100 90 100 200 120

Surface area (m2 /g)

All experiments were performed on a Varian ProStar HPLC (Varian, Palo Alto, CA, USA) consisting of a 210 ProStar Pump and a ProStar 410 Autosampler fit with a 40 ␮L loop. This system was connected to a Knauer Smartline 2500 UV detector (Knauer-ASI, Franklin, MA, USA) with a 2 ␮L flow cell connected via fibre optic cables. The detector time constant was 0.1 s.

3.5 N/A 2.7 3.5 5 3 3 5 5 3 5 2.7 3 2.7 5 5 5 2.7 5 5 3

2.1. Apparatus

Underivatized Underivatized Underivatized Underivatized Triazole Aminopropyl Aminoalkyl Polymeric sulfoalkylbetaine zwitterionic Polymeric sulfoalkylbetaine zwitterionic Polymeric phosphorylcholine zwitterionic Amide (polymeric carbamoyl) Proprietary amide Alkyl diol Penta hydroxy High loaded propylcarbamate cyclofructan 6 Poly(2-hydroxyethyl aspartamide) Proprietary polar alkyl embedded Pentafluorophenylpropyl Pentafluorophenylpropyl Poly(2-sulfoethyl aspartamide) Proprietary neutral polar functionality

Column length (mm)

2. Experimental

Silica Silica monolith Fused core silica Silica (BEH) Silica Silica Silica Silica Porous polymer Silica Silica Poroshell silica Silica Fused core silica Silica Silica Silica Fused core silica Silica Silica Silica

Specifically, we reconstruct the hydrophilicity vs. ion interaction selectivity plot of Ibrahim et al. [45] under three pH values and two buffer concentrations. We then focus on the changes in selectivity caused by the new mobile phase conditions.

Agilent Merck Supelco Waters Nacalai Restek Tosoh Bioscience Merck Merck Merck Tosoh Bioscience Agilent Fortis Techonolgies Supelco AZYP LLC PolyLC Restek Supelco Restek PolyLC Thermo Scientific

Column diameter (mm)

C.D. Iverson et al. / J. Chromatogr. A xxx (2016) xxx–xxx

2

Please cite this article in press as: C.D. Iverson, et al., The hydrophilicity vs. ion interaction selectivity plot revisited: The effect of mobile phase pH and buffer concentration on hydrophilic interaction liquid chromatography selectivity behavior, J. Chromatogr. A (2016), http://dx.doi.org/10.1016/j.chroma.2016.06.061

G Model CHROMA-357688; No. of Pages 8

ARTICLE IN PRESS C.D. Iverson et al. / J. Chromatogr. A xxx (2016) xxx–xxx

3

Fig. 1. Hydrophilicity (kcytosine /kuracil ) vs. ion interaction (kBTMA /kuracil ) selectivity plots of HILIC phases acquired under different mobile phase pH’s and buffer concentrations. Symbols: bare silica ( ); amine ( ); zwitterionic ( ); amide ( ); hydroxylated ( ); and specialty phases (*). Eluents: (A) 5 mM ammonium acetate, w w pH 6.8, in 80% ACN, (B) 25 mM ammonium acetate, w w pH 6.8, in 80% ACN, (C) 5 mM ammonium acetate, w w pH 5.0, in 80% ACN, (D) 25 mM ammonium acetate, w w pH 5.0, in 80% ACN, (E) 5 mM ammonium formate, w w pH 3.7, in 80% ACN, and (F) 25 mM ammonium formate, w w pH 3.7, in 80% ACN; conditions: columns, see column numbers in Table 1 and as listed throughout the text; flow rate, 0.5 mL/min; test analytes 0.25–1.5 mM cytosine, uracil, and BTMA prepared in 80% ACN containing 5 mM of the appropriate buffer; UV detection at 254 nm with a 20 ␮L partial loop injection. Note the cytosine/uracil ratio for PolySULFOETHYL A (column 22) was offset in Fig. 1E to bring it on-scale. The true value was 7.10. See Table S1 for raw retention data of all columns.

3. Results and discussion This work uses the hydrophilicity vs. ion interaction HILIC selectivity plot [45] to investigate the effect of pH and buffer concentration on the selectivity of various HILIC phases. The retention ratio of carefully chosen analytes are used to probe specific interactions [15,45,51]. By using the relative retention of probe pairs, column properties that affect absolute retention (e.g., column size, pore size, surface area, etc.) are factored out [15]. Cytosine and uracil are both hydrophilic (log P = −1.97 and −1.05, respectively, at pH 7), structurally similar (effectively negating other interactions), and both show significant but differing levels of HILIC retention. Recently, the retention ratio of these two compounds (kcytosine /kuracil ) has been ascribed as a convenient measure of the hydrophilicity of the HILIC phase [15].

Cytosine has a w w pKa of 4.6 [48]. The pKa of weak bases such as cytosine may be 1 or more pH units lower in 80% ACN than in water [52]. Studies of the UV absorbance spectra of cytosine (as per Ref. [53]) observed no changes in the ␭max over w w pH values of 3.7–6.8. Therefore it is concluded that cytosine remains uncharged under the mobile phase conditions utilized in this work [54]. The retention ratio of kBTMA /kuracil provides a measure of the electrostatic interactions of a phase [55,56]. Benzyltrimethylammonium (BTMA) is charged under all pH conditions and will undergo cation exchange interactions. Uracil remains uncharged under all chromatographically relevant pH conditions. Therefore uracil will not undergo cation exchange interactions but will experience polar interactions similar to BTMA. Further, the strong HILIC retention of BTMA on all HILIC phases allows the probe pair to also reflect the cationic nature of phases via electrostatic repulsion [57].

Please cite this article in press as: C.D. Iverson, et al., The hydrophilicity vs. ion interaction selectivity plot revisited: The effect of mobile phase pH and buffer concentration on hydrophilic interaction liquid chromatography selectivity behavior, J. Chromatogr. A (2016), http://dx.doi.org/10.1016/j.chroma.2016.06.061

G Model CHROMA-357688; No. of Pages 8

ARTICLE IN PRESS C.D. Iverson et al. / J. Chromatogr. A xxx (2016) xxx–xxx

4

Table S1 summarizes the retention for the three probes for all columns under the mobile phase conditions studied. Three columns (Ultra IBD phase (column no. 17), Ascentis Express F5 (18), and Ultra PFP (19)) were studied but displayed reversed phase behavior, as evidenced by weak retention of cytosine and uracil, and observable retention of toluene. Such reversed phase behavior for the fluorinated phases (18, 19) is consistent with the finding of less than a full monolayer water layer [40]. Similar hydration studies of the IBD phase have not been reported, albeit the manufacturer does advertise the IBD phase for use in HILIC separations. Based on their lack of HILIC behavior, columns 17–19 are excluded from the subsequent plots. Also note that while the selectivity behavior of the Acclaim HILIC 10 (column 21, black asterix) is shown in Fig. 1 and subsequent figures, it will not be discussed further as the proprietary nature of this phase precludes us from rationalizing its retention behavior. Fig. 1A shows the ion interaction (kBTMA /kuracil ) vs. hydrophilicity (kcytosine /kuracil ) using a 5 mM buffer at w w pH 6.8 (w s pH 7.7) in 80% ACN. As in Ref. [45], the columns cluster within groups (i.e., bare silica, amine, amide, hydroxylated, zwitterionic, and specialty phases). The high pH (w w pH 6.8) ensures deprotonation of silanols and the dilute 5 mM buffer ensures strong electrostatic interactions [15]. Fig. 1B–F shows selectivity plots for 5 and 25 mM total buffer concentrations at w w pH 6.8, 5.0, and 3.7 (w s pH 7.7, 7.2, and 5.5). Studies in this work were limited to w w pH < 7.0 since many of the phases investigated (Table 1) are bonded silica phases. Such phases are prone dissolution of the underlying silica at higher pH [46,58]. 3.1. Effect of pH To illustrate the effect of eluent on HILIC selectivity, Fig. 2 presents the change in the relative probe retention between two eluent conditions. For instance, Fig. 2A compares column selectivity using w w pH 5.0 (w s pH 7.2)/5.0 mM eluent relative to a w w pH 6.8 (w s pH 7.7)/5.0 mM eluent. The dotted horizontal and vertical lines indicate no change in hydrophilicity or ion interaction selectivity, respectively. The majority of the columns in Fig. 2A cluster around the (0, 0) point, indicating that the change in eluent w w pH has minimal effect on selectivity, consistent with Refs. [11,12]. The minimal change in selectivity observed upon decreasing w w pH from 6.8 to 5.0 is to be expected based on the small (7.7 vs. 7.2) difference in s w pH. Similar to other weak acids [52], addition of ACN causes up to a 1–2 pH unit increase in the pKa of the silanols relative to a purely aqueous system [59]. Hence, no significant change in silanol protonation would be expected over this pH range. At w w pH 3.7 (w s pH 5.5, Fig. 2B), a larger number of selectivity differences are evident. The ion-interaction selectivity of the bare silica phases (red circles, columns 1–4) is dominated by pHdependent silanol activity [11,15,16,42]. At w w pH 3.7 the silanols are largely protonated (silanol w w pKa = 4–7) [11,48]. Therefore, all bare silica phases investigated under this condition (1, 3, 4; column 2 was not studied at w w pH 3.7) displayed a moderate decrease in cation exchange activity at w w pH 3.7 relative to w w pH 6.8 (y < 0 in Fig. 2B). This is consistent with recent studies of pH on HILIC retention [16]. The amine columns (blue squares, 5–7) are protonated under all pH conditions used in this study (w w pH 6.8–3.7). Thus, BTMA experiences electrostatic repulsion [11,15]. At w w pH 5.0 and 6.8 this repulsion behavior is similar (kBTMA /kuracil ∼ 0 in Fig. 2A). At w w pH 3.7 the underlying exposed silanols are protonated (uncharged) and so repulsion of BTMA by amine phases should be enhanced (i.e., kBTMA /kuracil < 0 in Fig. 2B). Consistent with the above expectations, the Cosmosil (5) and Ultra Amine (6) showed increased anion exchange (lower kBTMA /kuracil ) at w w pH 3.7 in Fig. 2B (also compare Fig. 1E vs. A). The TSK-gel NH2-100 (7), however, maintained approximately the same level of anionic repulsion at w w pH 3.7 vs.

Fig. 2. Difference plots showing the effect of changing the w w pH from 6.8 to (A) 5.0 and (B) 3.7 (while maintaining a 5 mM buffer concentration) on the hydrophilicity and ion interaction behavior of 19 HILIC columns. Symbols: bare silica ( ), amine ( ), zwitterionic ( ), amide ( ), hydroxylated ( ), and specialty phases (*). See Table 1 and the main text for column identities. All difference values were calculated relative to the data points obtained at pH 6.8 using a 5 mM buffer concentration. Note that in Fig. 2B the actual  kcytosine /kuracil for the PolySULFOETHYL A column was 4.17.

6.8 (Fig. 2B). Based on Ref. [60] the nature of the buffer (acetate vs. formate) does not affect retention of cationic species on amino columns. More likely, the differences in ion interaction behavior of column 7 vs. columns 5 and 6 are due to differences in their composition. Columns 5 and 6 are simply comprised of triazoleand aminopropyl-bonded silanes, respectively [61,62]. Column 7, however, is highly derivatized with a combination of primary, secondary, and tertiary amine groups [63]. Furthermore, unlike columns 5 and 6 [61,62], column 7 is endcapped with trimethylsilane [63]. Altogether this would lead to the TSK-gel NH2-100 (7) having much lower silanol activity (and therefore displaying less change in ion interaction behavior) than the Cosmosil (5) and Ultra Amine (6) phases. The hydrophilicity of columns 5 and 7 did not change when the w w pH was lowered to 3.7 (Fig. 2B). Column 6, conversely, displayed a small increase in kcytosine /kuracil . All zwitterionic phases (green triangles, 8–10) experienced a similar weak to moderate reduction in kBTMA /kuracil upon lowering w w pH to 3.7 (Fig. 2B). The reduction in cation exchange behavior of the ZIC-HILIC (8) and ZIC-cHILIC (10) phases at w w pH 3.7 may be ascribed to the protonation of residual silanols [47,48,64]. Similarly, a zwitterionic polymethacrylate monolith (similar polymer backbone to pHILIC [65]) showed less anionic character at w w pH 3.5 [66].

Please cite this article in press as: C.D. Iverson, et al., The hydrophilicity vs. ion interaction selectivity plot revisited: The effect of mobile phase pH and buffer concentration on hydrophilic interaction liquid chromatography selectivity behavior, J. Chromatogr. A (2016), http://dx.doi.org/10.1016/j.chroma.2016.06.061

G Model CHROMA-357688; No. of Pages 8

ARTICLE IN PRESS C.D. Iverson et al. / J. Chromatogr. A xxx (2016) xxx–xxx

5

Relative to other classes of HILIC phases investigated here, the zwitterionic phases (green triangles, 8–10) experienced the greatest change in hydrophilicity in Fig. 2B. Nevertheless, the overall increase in hydrophilicity for these three phases was relatively small (kcytosine /kuracil < 0.5). At w w pH 3.7 the amide phases (brown diamonds; 11, 12) displayed a comparable loss of cation exchange activity to the BEH silica (columns 11 and 12 vs. 4, Fig. 2B), ascribed to protonation of the underlying silanols [48,55]. The kcytosine /kuracil of these phases (11, 12) was not affected by reducing the w w pH to 3.7, consistent with the recent observations of McCalley [16]. It is noteworthy, however, that in expanded studies with buffering agents, such as trifluoroacetic acid, McCalley did observe unique salt-specific changes in selectivity. The selectivity behavior of the hydroxylated phases (pink open triangles, 13–16) in Fig. 2B were somewhat more column dependent than other classes, as observed in Ref. [55]. The Fortis Diol (13), Ascentis Express OH5 (14), and FRULIC-N (column 15; a commercialized analog of a column reported in Ref. [67]) phases displayed a comparable reduction in ion interaction activity to that of the TypeB bare silica phases (1, 3). The PolyHYDROXYETHYL A (16), on the other hand, experienced only a small reduction in cation exchange activity. In general, the PolyHYDROXYETHYL A (16) displayed relatively little shift in selectivity over all conditions examined. At w w pH 3.7 all hydroxylated phases had minimal (13, 16) or no change (14, 15) in hydrophilicity behavior vs. w w pH 6.8. The PolySULFOETHYL A (black asterisk, 20) behaves as a pseudozwitterionic phase (due to unbonded cationic taurine groups) with some strong cation exchange character (due to the anionic sulfonate groups) [15,55]. This column (20) remained near zero on the ionic interaction axis in Fig. 2B. Conversely, column 20 displayed almost a 2-fold increase in kcytosine /kuracil at w w pH 3.7 vs 6.8 (7.1 vs. 2.9, respectively; see Fig. 1A and E) due to a >2.5-fold increase in kcytosine (9.8 vs. 4.0). It has been previously noted that the zwitterionic character of column 20 gives it unique retention and selectivity characteristics [15,55]. 3.2. Effect of buffer concentration Fig. 3 illustrates the effect of increasing the buffer concentration while keeping the pH constant at w w pH 6.8 (Fig. 3A), 5.0 (Fig. 3B), and 3.7 (Fig. 3C). As noted above and in Refs. [39,47,48], more concentrated buffer mutes ionic interactions for most phases due to increased ionic shielding. At w w pH 6.8 (Fig. 3A) and 5.0 (Fig. 3B), all silica phases (red circles, 1–4) experienced a 50–100% decrease in kBTMA /kuracil arising from decreased ionic interactions. Additionally, one can distinguish the silanol activity of the phases based on the change in kBTMA /kuracil upon increasing buffer concentration. At w w pH 6.8/5 mM the silica phases (1–4) in Fig. 1A group together [15,45]. Upon increasing the buffer concentration, the silica phases respond differently (Fig. 3A). The Chromolith Si (2) exhibits strong silanol activity due to its use of a silica which contains a wide array of highly acidic and less acidic silanols; hence the strong cation exchange character in Fig. 1A [55]. Accordingly, Chromolith Si experiences the greatest shielding effects and loss of cation exchange activity upon increasing buffer concentration in Fig. 3A (kBTMA /kuracil 30.7 vs. 6.4 in 5 vs. 25 mM buffer at w w pH 6.8). At the other extreme, the blocked and less active silanols of the BEH silica (4) [48] display a much smaller decrease in cation exchange activity (Fig. 3A). The Type B silica phases (Zorbax HILIC Plus (1) and Ascentis Express HILIC (3)) displayed a median decrease in cation exchange activity. At w w pH 3.7, the cation exchange behavior of the silica columns (1–4) is weakened (Fig. 1E) due to protonation of the silanols (see Section 3.1). Hence, as shown in Fig. 3C, increasing the buffer concentration caused little change in the cation exchange behavior of

Fig. 3. Difference plots showing the effect of increasing the buffer concentration from 5 to 25 mM at w w pH (A) 6.8, (B) 5.0, and (C) 3.7 on the hydrophilicity and ion interaction behavior of 19 HILIC columns. Symbols: bare silica ( ), amine ( ), zwitterionic ( ), amide ( ), hydroxylated ( ), and specialty phases (*). See Table 1 and the main text for column identities. All difference values were calculated relative to the data points obtained at the same pH using a 5 mM buffer concentration. Note that in Fig. 3C the actual  kcytosine /kuracil for the PolySULFOETHYL A column was −4.28.

the Ascentis Express HILIC (3) and BEH silica (4). The Zorbax HILIC Plus (1), however, did show a comparable loss of cation exchange activity in Fig. 3C to that observed in Fig. 3B. This suggests that even though both columns 1 and 3 utilize Type B silica, the free silanols in column 1 are more acidic than column 3. The Chromolith Si (2) was not investigated at w w pH 3.7/25 mM.

Please cite this article in press as: C.D. Iverson, et al., The hydrophilicity vs. ion interaction selectivity plot revisited: The effect of mobile phase pH and buffer concentration on hydrophilic interaction liquid chromatography selectivity behavior, J. Chromatogr. A (2016), http://dx.doi.org/10.1016/j.chroma.2016.06.061

G Model CHROMA-357688; No. of Pages 8 6

ARTICLE IN PRESS C.D. Iverson et al. / J. Chromatogr. A xxx (2016) xxx–xxx

Overall, the hydrophilicity behavior of the silicas was complex in response to increased buffer concentration. At w w pH 6.8 and 5.0 (Fig. 3A and B), all silica phases (1–4) experienced a decrease in kcytosine /kuracil , arising from a decrease in cytosine retention and increase in uracil retention (Table S1). At w w pH 3.7, the increased buffer caused a smaller decrease in the kcytosine /kuracil for Ascentis Express (Fig. 3C) than observed at w w pH 6.8 or w w pH 5.0 (Fig. 3A and B). The Zorbax HILIC Plus (1), conversely, had a small increase in kcytosine /kuracil in Fig. 3C, while the BEH silica was unaffected by the increased buffer concentration at w w pH 3.7. For the amine columns (blue squares, 5–7), the ion interaction behavior was more complex. At w w pH 6.8 (Fig. 3A) ionic shielding of the cationic amino groups due to increased buffer concentration reduced electrostatic repulsion of BTMA, leading to a 100–150% increase in kBTMA /kuracil , consistent with Ref. [15]. Similar changes in ion interaction behavior were observed at w w pH 5.0 (Fig. 3B), albeit the TSKgel NH2 -100 (7) was unaffected by increasing the buffer concentration at this pH. At w w pH 3.7/25 mM (Fig. 3C), the Cosmosil HILIC (5) and Ultra Amine (6) phases experienced a significantly greater reduction in cationic repulsion (as evidenced by increased kBTMA under this condition; see Table S1) relative to Fig. 3A and B. The TSKgel NH2 -100 (7) in Fig. 3C showed comparable behavior to Fig. 3B. All zwitterionic phases (green triangles, 8–10) experienced some loss of cation exchange activity in Fig. 3A, mainly ascribed to shielding of multi-point ionic interactions by the buffer salt, as noted in Refs. [15,64]. Furthermore, although the surface functionalities significantly shield the underlying backbone of the ZIC-HILIC phase (8) [7,64], interactions with the silica backbone are still apparent. Namely, more significant reductions in cation exchange interactions (resulting from a 50–110% reduction in BTMA retention) were observed for the ZIC-HILIC phase (8) as compared to the ZIC-pHILIC (9). Interestingly, the ZIC-cHILIC (10) remained nearly unaffected (0, 0 in Fig. 3A) when the buffer was increased at w w pH 6.8. At w w pH 5.0 (Fig. 3B), the differences in ion interaction between the zwitterionic phases (green triangles) collapse such that the columns 8–10 all display similar loss of cation exchange activity (i.e.,  kBTMA /kuracil is near equal for columns 8–10). At w w pH 3.7, the three zwitterionic phases once again show some differences in behavior. The ZIC-pHILIC (9) maintained a comparable loss of ion interaction in Fig. 3C relative to Fig. 3B, while the ZIC-HILIC (8) and ZIC-cHILIC (10) were minimally affected. Focusing on hydrophilicity, all zwitterionic phases (8–10) displayed minimal changes in kcytosine /kuracil when the buffer concentration was increased (Fig. 3A–C). The amide phases (brown diamonds, 11 and 12) at w w pH 6.8 and 5.0 (Fig. 3A and B) experienced a small reduction in kBTMA /kuracil when the buffer concentration was increased (Fig. 3A–C), attributed to ionic shielding of residual silanols. This effect was small as the bonded groups or polymeric layer block most silanols [15,55]. At w w pH 3.7, the TSKgel Amide-80 (11) showed increased cation exchange activity (arising from increased retention of BTMA) while the Agilent Glycan Mapping (12) column was unaffected. The hydroxylated phases (pink open triangles, 13–16) behaved similarly to the amide phases (11, 12) in Fig. 3A and B. At w w pH 3.7 (Fig. 3C), the Fortis HILIC diol (13) and PolyHYDROXYETHYL A (16) columns experienced a small loss of cation exchange activity, while the Ascentis Express OH5 (14) and FRULIC-N (15) phases displayed a small increase. Given that these hydroxylated phases (14–16) show multiple interactions beyond ion exchange [15], it is probable that the observed ion interaction behavior of the hydroxylated phases (14–16) is due to other interactions not studied in this work. The hydrophilicity of the amide (11, 12) and hydroxylated phases (13–16) was generally unaffected by increased buffer con-

centration across all pH’s tested. The one exception was the Agilent Glycan Mapping (12) phase which, at w w pH 3.7/25 mM showed notably increased kcytosine /kuracil in Fig. 3C. Lastly, as observed in Ref. [15], the PolySULFOETHYL A (20) showed significantly reduced cation exchange activity with increased buffer concentration at all tested pH’s (Fig. 3A–C) due to shielding of the electrostatic interaction between the anionic sulfonate groups and BTMA. Accordingly, BTMA experienced a > 2fold reduction in retention at both w w pH 6.8 and 5.0 when a 25 mM buffer was used (Table S1). Further, at w w pH 3.7/25 mM the PolySULFOETHYL A exhibited a sharp decrease in kcytosine /kuracil , arising from a nearly 2-fold decrease in the retention of cytosine relative to w w pH 3.0/5 mM buffer (k = 9.8 vs. 5.5). 4. Conclusions We have investigated and plotted the hydrophilicity vs. ion interaction selectivity behavior of 18 HILIC columns. Plots of the changes in selectivity between different pH values and buffer concentrations pinpoint the effect of mobile phase conditions on column interactions. From these difference plots several trends emerge. At w w pH ≥ 5, only minor changes in selectivity are observed. Increasing buffer concentration at w w pH ≥ 5 resulted in a general muting of ionic interactions due to ionic shielding by the buffer salt. Lowering the w w pH below 5 caused large changes in the ionic interaction selectivity (especially in phases that utilize ion exchange as part of their retention mechanism) due to protonation of the silanols. Under all pH and buffer conditions investigated, the hydrophilic selectivity of most HILIC phases was not significantly affected. In closing, the reader is reminded that in order to limit complications in interpreting selectivity behaviors, this study focused on the use of a w w pH range (3.7–6.8) which kept the probe analytes in an uncharged state. The buffers were also carefully selected to ensure that additional salt-specific selectivity behaviors [16] would not unduly influence our investigation. Acknowledgements This work was supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) and the University of Alberta. The columns were the generous gift of their manufacturers. The authors are grateful to David Bell (Supelco, Belefonte, PA, USA), Xiaoli Wang (Agilent Technologies, Santa Clara, CA, USA) and Frances Carroll (Restek Technologies, Belefonte, PA, USA) for helpful discussions. CDI gratefully acknowledges his Queen Elizabeth II Fellowship. XG is grateful for funding from the University of Alberta’s International Student Work Study Program (ISWSP). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.chroma.2016.06. 061. References [1] A.J. Alpert, Hydrophilic-interaction chromatography for the separation of peptides, nucleic acids and other polar compounds, J. Chromatogr. A 499 (1990) 177–196. [2] D.-Q. Tang, L. Zou, X.-X. Yin, C.N. Ong, HILIC-MS for metabolomics: an attractive and complementary approach to RPLC-MS, Mass Spectrom. Rev. (2014), http://dx.doi.org/10.1002/mas.21445. [3] R.E. Majors, Current trends in HPLC column usage, LC-GC North Am. 30 (2012) 20–34. [4] B. Buszewski, S. Noga, Hydrophilic interaction liquid chromatography (HILIC)-a powerful separation technique, Anal. Bioanal. Chem. 402 (2012) 231–247.

Please cite this article in press as: C.D. Iverson, et al., The hydrophilicity vs. ion interaction selectivity plot revisited: The effect of mobile phase pH and buffer concentration on hydrophilic interaction liquid chromatography selectivity behavior, J. Chromatogr. A (2016), http://dx.doi.org/10.1016/j.chroma.2016.06.061

G Model CHROMA-357688; No. of Pages 8

ARTICLE IN PRESS C.D. Iverson et al. / J. Chromatogr. A xxx (2016) xxx–xxx

[5] L. Nováková, L. Havlíková, H. Vlˇcková, Hydrophilic interaction chromatography of polar and ionizable compounds by UHPLC, TrAC Trends Anal. Chem. 63 (2014) 55–64. [6] P. Hemström, T. Jonsson, P. Appelblad, W. Jiang, HILIC after the hype: a separation technology here to stay, Chromatogr. Today (2011) 4–8. [7] P. Hemström, K. Irgum, Hydrophilic interaction chromatography, J. Sep. Sci. 29 (2006) 1784–1821. [8] P. Jandera, Stationary and mobile phases in hydrophilic interaction chromatography: a review, Anal. Chim. Acta 692 (2011) 1–25. [9] Y. Guo, Recent progress in fundamental understanding of hydrophilic interaction chromatography (HILIC), Analyst 140 (2015) 6452–6466. [10] Y. Guo, Retention and selectivity of polar stationary phases for hydrophilic interaction chromatography, in: P.G. Wang, W. He (Eds.), Hydrophilic Interaction Liquid Chromatography and Advanced Applications, CRC Press, Boca Raton, FL, 2011, pp. 401–426. [11] Y. Guo, S. Gaiki, Retention and selectivity of stationary phases for hydrophilic interaction chromatography, J. Chromatogr. A 1218 (2011) 5920–5938. [12] G. Schuster, W. Lindner, Additional investigations into the retention mechanism of hydrophilic interaction liquid chromatography by linear solvation energy relationships, J. Chromatogr. A 1301 (2013) 98–110. [13] A. Periat, B. Debrus, S. Rudaz, D. Guillarme, Screening of the most relevant parameters for method development in ultra-high performance hydrophilic interaction chromatography, J. Chromatogr. A 1282 (2013) 72–83. [14] J. Wang, Z. Guo, A. Shen, L. Yu, Y. Xiao, X. Xue, et al., Hydrophilic-subtraction model for the characterization and comparison of hydrophilic interaction liquid chromatography columns, J. Chromatogr. A 1398 (2015) 29–46. [15] N.P. Dinh, T. Jonsson, K. Irgum, Probing the interaction mode in hydrophilic interaction chromatography, J. Chromatogr. A 1218 (2011) 5880–5891. [16] D.V. McCalley, Study of retention and peak shape in hydrophilic interaction chromatography over a wide pH range, J. Chromatogr. A 1411 (2015) 41–49. [17] R. Zhang, D.G. Watson, L. Wang, G.D. Westrop, G.H. Coombs, T. Zhang, Evaluation of mobile phase characteristics on three zwitterionic columns in hydrophilic interaction liquid chromatography mode for liquid chromatography-high resolution mass spectrometry based untargeted metabolite profiling of Leishmania parasites, J. Chromatogr. A 1362 (2014) 168–179. [18] H.G. Gika, G.A. Theodoridis, U. Vrhovsek, F. Mattivi, Quantitative profiling of polar primary metabolites using hydrophilic interaction ultrahigh performance liquid chromatography-tandem mass spectrometry, J. Chromatogr. A 1259 (2012) 121–127. [19] M. Isokawa, T. Kanamori, T. Funatsu, M. Tsunoda, Recent advances in hydrophilic interaction chromatography for quantitative analysis of endogenous and pharmaceutical compounds in plasma samples, Bioanalysis 6 (2014) 2421–2439. [20] G. Kahsay, H. Song, A. Van Schepdael, D. Cabooter, E. Adams, Hydrophilic interaction chromatography (HILIC) in the analysis of antibiotics, J. Pharm. Biomed. Anal. 87 (2014) 142–154. [21] T. Mercier, F. Tissot, C. Gardiol, N. Corti, S. Wehrli, M. Guidi, et al., High-throughput hydrophilic interaction chromatography coupled to tandem mass spectrometry for the optimized quantification of the anti-gram-negatives antibiotic colistin A/B and its pro-drug colistimethate, J. Chromatogr. A 1369 (2014) 52–63. [22] Z. Zhou, X. Wu, Q. Wei, Y. Liu, P. Liu, A. Ma, et al., Development and validation of a hydrophilic interaction liquid chromatography-tandem mass spectrometry method for the simultaneous determination of five first-line antituberculosis drugs in plasma, Anal. Bioanal. Chem. 405 (2013) 6323–6335. [23] T. Nemoto, X.-P. Lee, T. Kumazawa, C. Hasegawa, M. Fujishiro, A. Marumo, et al., High-throughput determination of nonsteroidal anti-inflammatory drugs in human plasma by HILIC-MS/MS, J. Pharm. Biomed. Anal. 88 (2014) 71–80. [24] R.-H. Zhu, H.-D. Li, H.-L. Cai, Z.-P. Jiang, P. Xu, L.-B. Dai, et al., Validated HILIC-MS/MS assay for determination of vindesine in human plasma: application to a population pharmacokinetic study, J. Pharm. Biomed. Anal. 96 (2014) 31–36. [25] S. Guo, J. Duan, D. Qian, Y. Tang, Y. Qian, D. Wu, et al., Rapid determination of amino acids in fruits of ziziphus jujuba by hydrophilic interaction ultra-high-performance liquid chromatography coupled with triple-quadrupole mass spectrometry, J. Agric. Food Chem. 61 (2013) 2709–2719. [26] A.E. Karatapanis, Y.C. Fiamegos, C.D. Stalikas, HILIC separation and quantitation of water-soluble vitamins using diol column, J. Sep. Sci. 32 (2009) 909–917. [27] V. Spackova, J. Pazourek, Identification of carbohydrate isomers in flavonoid glycosides after hydrolysis by hydrophilic interaction chromatography, Chem. Pap. 67 (2013) 357–364. [28] Y. Xiong, Y.-Y. Zhao, S. Goruk, K. Oilund, C.J. Field, R.L. Jacobs, et al., Validation of an LC–MS/MS method for the quantification of choline-related compounds and phospholipids in foods and tissues, J. Chromatogr. B 911 (2012) 170–179. [29] J. Bernal, A.M. Ares, J. Pol, S.K. Wiedmer, Hydrophilic interaction liquid chromatography in food analysis, J. Chromatogr. A 1218 (2011) 7438–7452. [30] A.L.N. van Nuijs, I. Tarcomnicu, A. Covaci, Application of hydrophilic interaction chromatography for the analysis of polar contaminants in food and environmental samples, J. Chromatogr. A 1218 (2011) 5964–5974. [31] H. Schlichtherle-Cerny, M. Affolter, C. Cerny, Hydrophilic interaction liquid chromatography coupled to electrospray mass spectrometry of small polar compounds in food analysis, Anal. Chem. 75 (2003) 2349–2354.

7

[32] M.A. Koupparis, N.C. Megoulas, A.M. Gremilogianni, HILIC for food environmental, and other applications, Chem. Anal. 177 (2013) 239–264. [33] M. Melmer, T. Stangler, A. Premstaller, W. Lindner, Comparison of hydrophilic-interaction, reversed-phase and porous graphitic carbon chromatography for glycan analysis, J. Chromatogr. A 1218 (2011) 118–123. [34] B. Adamczyk, W.B. Struwe, A. Ercan, P.A. Nigrovic, P.M. Rudd, Characterization of fibrinogen glycosylation and its importance for serum/plasma N-glycome analysis, J. Proteome Res. 12 (2013) 444–454. [35] S. Tao, Y. Huang, B.E. Boyes, R. Orlando, Liquid chromatography-selected reaction monitoring (LC-SRM) approach for the separation and quantitation of sialylated N-glycans linkage isomers, Anal. Chem. 86 (2014) 10584–10590. [36] T. Tetaz, S. Detzner, A. Friedlein, B. Molitor, J.-L. Mary, Hydrophilic interaction chromatography of intact, soluble proteins, J. Chromatogr. A 1218 (2011) 5892–5896. [37] O. Hernandez-Hernandez, J.E. Quintanilla-Lopez, R. Lebron-Aguilar, M.L. Sanz, F.J. Moreno, Characterization of post-translationally modified peptides by hydrophilic interaction and reverse phase liquid chromatography coupled to quadrupole-time-of-flight mass spectrometry, J. Chromatogr. A 1428 (2016) 202–211. [38] M.E.A. Ibrahim, C.A. Lucy, Stationary phases for HILIC., in: B.A. Olsen, B.W. Pack (Eds.), Hydrophilic Interact. Chromatogr. A Guid. Pract., John Wiley & Sons, Hoboken, NJ, 2013, pp. 43–85. [39] N.P. Dinh, T. Jonsson, K. Irgum, Water uptake on polar stationary phases under conditions for hydrophilic interaction chromatography and its relation to solute retention, J. Chromatogr. A 1320 (2013) 33–47. [40] J. Soukup, P. Jandera, Adsorption of water from aqueous acetonitrile on silica-based stationary phases in HILIC (aqueous normal-phase) liquid chromatography, J. Chromatogr. A 1374 (2014) 102–111. [41] F. Gritti, A. Höltzel, U. Tallarek, G. Guiochon, The relative importance of the adsorption and partitioning mechanisms in hydrophilic interaction liquid chromatography, J. Chromatogr. A 1376 (2015) 112–125. [42] D.V. McCalley, Study of the selectivity, retention mechanisms and performance of alternative silica-based stationary phases for separation of ionised solutes in hydrophilic interaction chromatography, J. Chromatogr. A 1217 (2010) 3408–3417. [43] U.D. Neue, Stationary phase characterization and method development, J. Sep. Sci. 30 (2007) 1611–1627. [44] U.D. Neue, K. VanTran, P.C. Iraneta, B.A. Alden, Characterization of HPLC packings, J. Sep. Sci. 26 (2003) 174–186. [45] M.E.A. Ibrahim, Y. Liu, C.A. Lucy, A simple graphical representation of selectivity in hydrophilic interaction liquid chromatography, J. Chromatogr. A 1260 (2012) 126–131. [46] J. Nawrocki, The silanol group and its role in liquid chromatography, J. Chromatogr. A 779 (1997) 29–71. [47] G. Greco, S. Grosse, T. Letzel, Study of the retention behavior in zwitterionic hydrophilic interaction chromatography of isomeric hydroxy- and aminobenzoic acids, J. Chromatogr. A 1235 (2012) 60–67. [48] A. Kumar, J.C. Heaton, D.V. McCalley, Practical investigation of the factors that affect the selectivity in hydrophilic interaction chromatography, J. Chromatogr. A 1276 (2013) 33–46. [49] J.C. Heaton, D.V. McCalley, Comparison of the kinetic performance and retentivity of sub-2 ␮m core–shell, hybrid and conventional bare silica phases in hydrophilic interaction chromatography, J. Chromatogr. A 1371 (2014) 106–116. [50] J.C. Heaton, J.J. Russell, T. Underwood, R. Boughtflower, D.V. McCalley, Comparison of peak shape in hydrophilic interaction chromatography using acidic salt buffers and simple acid solutions, J. Chromatogr. A 1347 (2014) 39–48. [51] T. Ikegami, K. Tomomatsu, H. Takubo, K. Horie, N. Tanaka, Separation efficiencies in hydrophilic interaction chromatography, J. Chromatogr. A 1184 (2008) 474–503. [52] X. Subirats, M. Rosés, E. Bosch, On the effect of organic solvent composition on the pH of buffered HPLC mobile phases and the pKa of analytes-a review, Sep. Purif. Rev. 36 (2007) 231–255. [53] B.E. Billinghurst, S.A. Oladepo, G.R. Loppnow, pH-Dependent UV resonance raman spectra of cytosine and uracil, J. Phys. Chem. B 113 (2009) 7392–7397. [54] C.D. Iverson, Preparation and Characterization of Stationary Phases for Hydrophilic Interaction Liquid Chromatography and Reversed Phase Liquid Chromatography, University of Alberta, 2016. [55] Y. Kawachi, T. Ikegami, H. Takubo, Y. Ikegami, M. Miyamoto, N. Tanaka, Chromatographic characterization of hydrophilic interaction liquid chromatography stationary phases: hydrophilicity, charge effects, structural selectivity, and separation efficiency, J. Chromatogr. A 1218 (2011) 5903–5919. [56] M.F. Wahab, M.E.A. Ibrahim, C.A. Lucy, Carboxylate modified porous graphitic carbon: a new class of hydrophilic interaction liquid chromatography phases, Anal. Chem. 85 (2013) 5684–5691. [57] A.J. Alpert, Electrostatic repulsion hydrophilic interaction chromatography for isocratic separation of charged solutes and selective isolation of phosphopeptides, Anal. Chem. 80 (2008) 62–76. [58] J.E. O’Gara, B.A. Alden, C.A. Gendreau, P.C. Iraneta, T.H. Walter, Dependence of cyano bonded phase hydrolytic stability on ligand structure and solution pH, J. Chromatogr. A 893 (2000) 245–251. [59] C. Schwer, E. Kenndler, Electrophoresis in fused-silica capillaries: the influence of organic solvents on the electroosmotic velocity and the.zeta. potential, Anal. Chem. 63 (1991) 1801–1807.

Please cite this article in press as: C.D. Iverson, et al., The hydrophilicity vs. ion interaction selectivity plot revisited: The effect of mobile phase pH and buffer concentration on hydrophilic interaction liquid chromatography selectivity behavior, J. Chromatogr. A (2016), http://dx.doi.org/10.1016/j.chroma.2016.06.061

G Model CHROMA-357688; No. of Pages 8 8

ARTICLE IN PRESS C.D. Iverson et al. / J. Chromatogr. A xxx (2016) xxx–xxx

[60] Y. Guo, S. Gaiki, Retention behavior of small polar compounds on polar stationary phases in hydrophilic interaction chromatography, J. Chromatogr. A 1074 (2005) 71–80. [61] Nacali, Cosmosil HILIC, 2016 http://www.nacalai.co.jp/global/cosmosil/ column/08.html. [62] Restek Corporation, Ultra Amine, 2016 http://www.restek.com/catalog/view/ 2138. [63] Tosoh Bioscience, TSKgel NH2-100, 2016 http://www.separations.us. tosohbioscience.com/Products/HPLCColumns/NormalHydrophilic/ TSKgel+NH2-100.htm. [64] R.-I. Chirita, C. West, S. Zubrzycki, A.-L. Finaru, C. Elfakir, Investigations on the chromatographic behaviour of zwitterionic stationary phases used in hydrophilic interaction chromatography, J. Chromatogr. A 1218 (2011) 5939–5963.

[65] W. Jiang, K. Irgum, Covalently bonded polymeric zwitterionic stationary phase for simultaneous separation of inorganic cations and anions, Anal. Chem. 71 (1999) 333–344. [66] X. Wang, X. Lin, Z. Xie, J.P. Giesy, Preparation and evaluation of a neutral methacrylate-based monolithic column for hydrophilic interaction stationary phase by pressurized capillary electrochromatography, J. Chromatogr. A 1216 (2009) 4611–4617. [67] H. Qiu, L. Loukotková, P. Sun, E. Tesaˇrová, Z. Bosáková, D.W. Armstrong, Cyclofructan 6 based stationary phases for hydrophilic interaction liquid chromatography, J. Chromatogr. A 1218 (2011) 270–279.

Please cite this article in press as: C.D. Iverson, et al., The hydrophilicity vs. ion interaction selectivity plot revisited: The effect of mobile phase pH and buffer concentration on hydrophilic interaction liquid chromatography selectivity behavior, J. Chromatogr. A (2016), http://dx.doi.org/10.1016/j.chroma.2016.06.061