Retention mechanisms of acidic and basic analytes on the Pentafluorophenyl stationary phase using fluorinated eluent additives

This work explores the effects of three selected fluoroalcohols - 1,1,1,3,3,3-hexafluoroisopropanol (HFIP), 1,1,1,3,3,3-hexafluorotert–butyl alcohol (HFTB) and hexafluoro-2,3-(trifluoromethyl)-2,3-butanediol (PP) as novel eluent additives and their effect on the retention of basic and acidic analytes, using a reversed phase (RP) column with a fluorophenyl (PFP) stationary phase.

Journal of Chromatography A 1666 (2022) 462850 Contents lists available at ScienceDirect Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma Tutorial Article Retention mechanisms of acidic and basic analytes on the Pentafluorophenyl stationary phase using fluorinated eluent additives Krit Lossmann a, Ruta Hecht a, Jaan Saame a, Agnes Heering a, Ivo Leito a, Karin Kipper a,b,c,∗ a University of Tartu, Institute of Chemistry, 14a Ravila Street, 50411 Tartu, Estonia Chalfont Centre for Epilepsy, Chesham Lane, Chalfont St Peter, Buckinghamshire, SL9 0RJ, United Kingdom c Department of Clinical and Experimental Epilepsy, Faculty of Brain Sciences, UCL Queen Square Institute of Neurology, Queen Square, London WC1N 3BG, University College London, United Kingdom b a r t i c l e i n f o Article history: Received 11 November 2021 Revised 20 January 2022 Accepted 23 January 2022 Available online 28 January 2022 Keywords: Pentafluorophenyl column HFTB, HFIP Perfluoropinacol LC-MS a b s t r a c t This work explores the effects of three selected fluoroalcohols - 1,1,1,3,3,3-hexafluoroisopropanol (HFIP), 1,1,1,3,3,3-hexafluorotert–butyl alcohol (HFTB) and hexafluoro-2,3-(trifluoromethyl)-2,3-butanediol (PP) as novel eluent additives and their effect on the retention of basic and acidic analytes, using a reversed phase (RP) column with a fluorophenyl (PFP) stationary phase In order to observe the changes in the model analytes’ retention, chromatograms were obtained at multiple (5.0; 6.0; 7.0; 8.5; 9.0 and 9.5) pH values depending on the eluent The retention observed with fluoroalcohols was compared with that of a conventional eluent additive - ammonium acetate When fluoroalcohols were used as eluent additives, a decrease in the retention factors (compared with ammonium acetate) was generally observed for strong acids The retention factors of strong bases were generally higher when using HFIP and HFTB as eluent additives The behaviour of weak bases and weak acids was more nuanced, potentially enabling interesting selectivity The extent of the effect regarding different fluoroalcohols also varied, with HFIP and HFTB having a more significant effect on the retention of analytes than PP The retention data were interpreted in terms of the hypothesis that four interactions are at play: (a) hydrophobic retention typical to RP; (b) π -π interactions between the analytes containing an aromatic ring and the aromatic rings on the stationary phase; (c) charge-charge or hydrogen bond interactions between the analytes and partially deprotonated fluoroalcohols adsorbed on the stationary phase and (d) a hydrogen bond or charge-charge interaction between the free silanol groups or their deprotonated forms on the stationary phase and the analytes (either neutral or ionic) Alternative selectivity obtained through fluoroalcohols on the PFP stationary phase was compared with the C18 and biphenyl stationary phases It was demonstrated that at the same eluent pH but with a different buffer system and/or different RP stationary phases, very different selectivity and retention order can be obtained © 2022 The Author(s) Published by Elsevier B.V This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/) Introduction The most common stationary phase in reversed phase (RP) liquid chromatography, also in the field of bioanalytical applications, continues to be C18 However, other non-polar stationary phases are gaining popularity due to the offered alternative selectivity One such stationary phase is pentafluorophenyl (PFP) [1] So far, RP, reversed phase; PFP, pentafluorophenyl; HFIP, 1,1,1,3,3,3-hexafluoro-2propanol; HFTB, 1,1,1,3,3,3-hexafluoro-tert-butyl alcohol; PP, hexafluoro-2,3(trifluoromethyl)-2,3-butanediol, perfluoropinacol; MS, mass-spectrometric/massspectrometry; SST, system suitability test; LC, liquid chromatography; HB, hydrogen bond; TOC, total organic carbon; PDA, photodiode array detector ∗ Corresponding author E-mail address: karin.kipper@ut.ee (K Kipper) PFP has demonstrated usual RP retention patterns for neutral and acidic analytes [2] However, changes in the retention of the basic analytes indicate additional interactions It has been speculated that these changes in analyte retention could be caused by ionic interactions between the free silanol groups on the stationary phase and the basic analytes [1,2] It is estimated that almost 95% of active pharmaceutical ingredients contain some ionisable functional group and that almost 75% of them are weakly basic [3] Analytes with acidic or basic properties may be partially or completely in a neutral or an ionic form, depending on the pH of the mobile phase If analytes are partially in an ionic form, already a small change in the pH can lead to a considerable change in the retention of analytes Analytes in a completely ionic form are polar and often elute too rapidly https://doi.org/10.1016/j.chroma.2022.462850 0021-9673/© 2022 The Author(s) Published by Elsevier B.V This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/) K Lossmann, R Hecht, J Saame et al Journal of Chromatography A 1666 (2022) 462850 Table Researched fluoroalcohols and their structures from the column [4] In the case of basic analytes, this means that mobile phases with basic pH might be preferred For such mobile phases, one would need acidic and/or basic buffer components with pKa values in the range of to 10 [4] Adding an organic solvent to the buffer solution can change the pKa values of the analytes and buffer components, thus also changing the real pH of the mobile phase In order to enable comparing pH values of different mobile phases, an unified pH scale has been established [5], which was also used in this work to estimate the change of the aqueous pH values after adding methanol Using mass-spectrometric (MS) detectors allows to determine a very small analyte quantity in a sample, but it also requires for all of the eluent components to be volatile and not suppress analyte ionisation [6] One such group of compounds, which can be used as eluent additives and buffer components, are fluoroalcohols They are volatile, provide buffering capacity in the basic pH range with ammonia as well as alternative selectivity [7,8] 1,1,1,3,3,3hexafluoro-2-propanol (HFIP) and 1,1,1,3,3,3-hexafluoro–tert–butyl alcohol (HFTB) have been shown to improve the separation and ionisation of oligonucleotides [9,10], antibiotics [11] and sedative drugs and their metabolites [7] However, 1,1,1,4,4,4-hexafluoro2,3-bis(trifluoromethyl)butane-2,3-diol (perfluoropinacol, PP) has, to the best of our knowledge, not been researched as an eluent additive outside our group [12] and it has been possible to achieve different retention from ammonium acetate when PP is used as an eluent additive Furthermore, using a buffer system of perfluoropinacol and ammonium ions provides a system with a wide and continuous buffering range in the pH range 5.0–11.5 The proposed additional retention mechanism of using novel fluorinated eluent additives is that the fluoroalcohols (which can partially deprotonate) are retained on the stationary phase (potentially forming a fluorous layer) and thus can form ion pairs with the analytes [8] Fluoroalcohols as eluent additives and PFP stationary phase have both shown potential as alternatives for more common additives and stationary phases, respectively, to enhance selectivity This work aims to broaden the selection of LC-MS compatible eluent additives which could be used for basic analytes In this work, the authors explored the effect of three fluorinated alcohols - HFIP, HFTB and perfluoropinacol - as novel eluent additives/buffer components on the retention of basic and acidic model analytes at different mobile phase pH values using a column with a PFP stationary phase To the best of authors’ knowledge, this is the first article where influence of fluoroalcohols as eluent additives is discussed on PFP stationary phase for small molecules The retention of the same analytes with a mobile phase based on ammonium acetate was used as a comparison The results were also compared with earlier results obtained when using fluorinated alcohols and columns with the C18 and biphenyl stationary phases Name pKa 1,1,1,4,4,4-hexafluoro-2,3bis(trifluoromethyl)butane-2,3-diol, perfluoropinacol 5.95; 10.43 [13] 1,1,1,3,3,3-hexafluoro-tert-butyl alcohol 9.6 [14] 1,1,1,3,3,3-hexafluoroisopropanol 9.3 [15] Structure 2-metoxypyridine, 2-methylpyridine, naphthylamine, urea, acetone, benzene, toluene and aniline were obtained from SigmaAldrich (Missouri, USA) N–hydroxy-6-bromobenzotriazole, N– hydroxy-6-trifuoromethylbenzotriazole and N–hydroxy-5–chloro4-methylbenzotriazole were kindly donated by Dr W Köning (Hoechst AG, Frankfurt, Germany) The analytes were divided into groups based on their acid-base behaviour: weak and strong acids and weak and strong bases (Table 2) The anaylte mixture for system suitability tests (SST) contained acetone, benzene and toluene 2.1 Preparation of the stock solution and working solution Stock solutions were prepared at a concentration of mg/mL in a water–methanol 20:80 (v:v) mixture All solutions were stored at −20 °C All working solutions were filtered using a 0.2 μm Sartorius Minisart RC syringe filter before transferring the solution to a vial The samples were stored in an autosampler for up to days at °C 2.2 Chromatographic conditions The analytical column Raptor Fluorophenyl 2.1 × 100 mm, particle size 2.7 μm, resistant in the pH range 2.0 to 8.0 [21] was kindly donated by Restek (Pennsylvania, USA) The column was equilibrated for 60 min, with the buffer used in the following experiments The elution was isocratic at 25% MeOH with the flow rate 0.5 mL/min for analytes and 0.4 mL/min for system suitability test (SST) The column was thermostated at 40 °C The injection volume was 10 μL The results used for comparison were obtained from earlier works by Veigure et al [12,22] The analysis was performed with the Shimadzu Nexera XR LC20AD HPLC system, PDA detector SPD-M20A and Shimadzu LCMS2020 MS detector The detector choice depended on the analyte (see Table S1) The UV/Vis detector was set to record between 190 nm and 700 nm, for chromatograms the extracted wavelength was (254 ± 2) nm The reference wavelength was 600 nm with a bandwidth of 50 nm The MS was operating in a positive and negative scanning mode in the m/z range of 60–250 The chromatograms were processed using the Shimadzu LabSolutions version 5.75 SP2 The eluent consisted of a buffer and methanol 75:25 (v:v) Eluent additive concentrations in the buffer solutions were mM for all additives Aqueous buffer solutions were prepared at pH 5.0, 6.0, 7.0, 8.5, 9.0, 9.5 for PP and ammonium acetate, 8.5, 9.0, 9.5 for HFIP and 7.0, 8.5, 9.0, 9.5 for HFTB Due to the pKa values of HFTB and HFIP, these compounds lack buffering ability at pH values lower than Therefore, experiments with HFTB and HFIP were not conducted at lower pH values Ammonium hydroxide was used in most cases to modify the buffer pH, except for ammonium ac- Material and methods Eluent additives: ammonium acetate, 25% ammonium hydroxide, acetic acid, 1,1,1,4,4,4-hexafluoro-2,3bis(trifluoromethyl)butane-2,3-diol (perfluoropinacol, PP, Table 1), 1,1,1,3,3,3-hexafluoro–tert–butyl alcohol (HFTB) were LC-MS grade and obtained from Sigma-Aldrich (Missouri, USA), 1,1,1,3,3,3hexafluoroisopropanol (HFIP) was obtained from Fluka (Buchs, Switzerland) LC-MS grade MeOH used in the mobile phases was obtained from Sigma Aldrich (Missouri, USA), water was purified in-house using a Millipore Advantage A10 system (18.2 M cm at 25 °C and a total organic carbon (TOC) value 2–3 ppb) from Millipore (Bedford, USA) Analytes: 4-nitrobenzoic acid, 2,3,4,5,6-pentafluorophenol, 2,4-dichlorophenol, phenol, p-cresol, hydroquinone, isopropylphenol, 3-fluorophenol, -hyrdoxyacetophenone, diisopropylamine, cyclohexylamine, pyrrolidine, piperidine, 2,6-dimethylpyridine, K Lossmann, R Hecht, J Saame et al Journal of Chromatography A 1666 (2022) 462850 etate at pH = 5.0, where acetic acid was taken as the pH modifier instead The pH values measured in the aqueous buffer before adding the organic phase are denoted as w w pH [23] further in the text - if making a distinction was necessary from the pH of the whole mobile phase (where the organic component has been added) This is because the addition of the organic phase changes the solvated proton activity and thus also the final mobile phase’s pH For expressing the pH of the whole mobile phase, we have employed H2 O the absolute pH scale (further denoted as pHabs ) values [24] The w aqueous buffer w pH values were measured with an Elmetron EPP1 combination electrode connected to the Evikon pH metre E6115, which was calibrated daily with standard aqueous buffers at pH H2 O values 4.01, 7.00 and 10.00 The mobile phase pHabs measurements were conducted using the modified version of the 2020 differential potentiometric method described by Heering et al [5] Measurements were done with a Metrohm 713 pH metre (with a Pt wire) and a Keysight B2987A Electrometer (without a Pt wire) No pre-soaking was done Data was collected for 60 at 10 s intervals and points from 30 to 60 were used for the analysis Measurements were done only at one polarity The w w pH H2 O and pHabs values of the used mobile phases are listed in Table S2 in the Supplementary data For some of these mobile phases the H2 O pHabs values were reported by us already previously [12], using a simplified measurement method The values reported in this work have been obtained using the more accurate differential potentiometric method and they differ from the earlier values by up to 0.5 pH units, demonstrating the low accuracy of the previous simH2 O plified method The pHabs values from this work should be preferred Table Model analytes, their pKa and pKaH values (all values come from the iBonD database, unless indicated otherwise) and structures pKaH value corresponds to the pKa of the protonated base Analyte pKa Structure Strong acids 4-nitrobenzoic acid 3.4 2,3,4,5,6-pentafluorophenol 5.5 2,4-dichlorophenol 7.9 N–hydroxy-6-bromobenzotriazole 3.97 [16] N–hydroxy-6-trifluoromethyl benzotriazole 3.60 [13] N–hydroxy-5–chloro-4-methyl benzotriazole 4.09 [16] Weak acids phenol 9.99 p-cresol 10.28 hydroquinone 11.40; 11.65 [17] 2-isopropylphenol 10.53 3-fluorophenol 9.24 Results and discussion 3.1 Column stability -hydroxyacetophenone The recommended w w pH range for the used PFPcolumn was 2–8 A broader pH range was tested in this work due to the chemical properties of fluoroalcohols and to gain a better understanding of their effect Using various aqueous phases with a w w pH above the recommended range (the w w pH varied from 8.5 to 9.5) caused the stationary phase to degrade rapidly, as was to be expected The influence of the column condition and mobile phase pH on the column was monitored using the SST solution (Figure S1) All analytes in the SST mixture eluted at progressively shorter times over the use of the column at high w w pH values with the most extreme change being toluene whose retention time changed from 3.5 to 2.1 Increased tailing was also observed with degraded stationary phases Measured retention factors were confirmed with unused PFP column after the SST solution had demonstrated the loss of retention The column was in use for 65 h at pH = 8.5 and 23 h at pH = 9.0 before system pressure reached the set limit and H2 O the column became blocked In addition, pHabs measurements re- 10.28 [18] Strong bases diisopropylamine 11.05 [19] cyclohexylamine 10.49 [19] pyrrolidine 11.27 piperidine 11.22 Weak bases 2,6-dimethylpyridine 6.72 H O 2-methoxypyridine 3.28 2-methylpyridine 5.96 1-naphthylamine 3.92 [20] aniline 4.62 vealed that the pHabs values for the eluents were higher than the w pH of the aqueous buffer component of the mobile phase (Table w H2 O S2 in Supplementary data) The pHabs values of the mobile phases w with a w pH in the range of 8.5 to 9.5 ranged from 8.17 to 10.17 Therefore, no measurements were performed with HFIP at pH 9.5, and with HFTB only acidic analytes were tested Additional chromatograms can be seen in Supplementary data 3.2 Trends observed The main interactions that can influence retention when using fluoroalcohols as eluent additives with a PFP column are the following: (a) hydrophobic retention typical to RP, (b) π -π interac3 K Lossmann, R Hecht, J Saame et al Journal of Chromatography A 1666 (2022) 462850 Fig Retention factors of weak bases using different eluent additives at different pH values Error bars represent standard deviations tions between the analytes containing an aromatic ring and aromatic rings on the stationary phase, (c) charge-charge or hydrogen bond interactions between the analytes and partially deprotonated fluoroalcohols adsorbed on the stationary phase and (d) a hydrogen bond and/or charge-charge interaction between the free silanol groups or their deprotonated forms on the stationary phase and the analytes The first three are expected to markedly affect retention time, while the main effect of the silanol groups is in the peak asymmetry, and their effect on retention time is probably smaller The average pKa values of silica’s silanol groups in water have been estimated to be at around [25] Thus, at first sight, in most of our mobile phases silanols are predominantly ionised However, there are three factors that additionally influence ionisation of SiOH groups: (1) adding organic solvent to aqueous buffer increases the pHabs , (2) adding organic solvent to the aqueous buffer increases the pKa of silanol groups and (3) the protonated form of silanol can be additionally stabilised by hydrogen bond (HB) formation with a base or protonated form of a base can be stabilised by charge-charge interaction and HB formation with a deprotonated silanol These three factors put together mean that HB interaction between silanol groups and bases are possible also if the pH value of the aqueous buffer is above To a large extent, the same considerations apply also to the fluoroalcohols erate, but the free silanol groups on the stationary phase are now to a large extent ionised and have limited ability to form hydrogen bonds with analytes With PP up to w w pH 7.0 the analytes were partially ionised, as with ammonium acetate As they became more neutral, the retention of analytes increased In addition, the partially ionised PP on the stationary phase could have ion-ion interactions with the ionised analytes Therefore, the retention factors in this pH region were higher when using PP rather than when using ammonium acetate At higher pH values, the retention mechanisms were largely the same as with ammonium acetate PP was strongly ionised in the mobile phase and did not attach to the stationary phase to the same extent Using HFTB and HFIP enabled the analytes similar retention mechanisms, thus creating similar retention trends as with PP 3.4 Strong bases (pKaH 9.6–11.3) Strong bases with pKaH 9.6–11.3 were to a large extent in a cationic (protonated) form throughout the researched pH range They demonstrated an increase in the retention factors as the pH increased (Fig 2) both when PP and ammonium acetate were used This is due to a gradual decrease in the extent of protonation of the analytes – they become less polar when the mobile phase pH increases Throughout the used w w pH range there is a possibility of interacting with the free silanol groups: either HB between neutral base and SiOH or charge-charge interaction between the protonated base and SiO– None of the selected analytes in the strong bases group had an aromatic ring and thus π -π interactions did not occur With ammonium acetate as the eluent additive, the analytes’ ionisation decreased with the increase in the pH, causing stronger retention, especially at the w w pH above 7.0 (Table S4) In the case of PP, the above considerations hold as well Additionally, partially ionised PP can interact with ionised analytes (ion-pairing effect), which in turn causes a stronger retention of analytes than with ammonium acetate Otherwise, the retention mechanisms are similar to those of ammonium acetate When HFIP or HFTB were used as mobile phase components, the retention mechanism was expected to be similar to that of PP However, the analytes’ retention factors were significantly higher with HFIP or HFTB than either with ammonium acetate or PP This implies that the ion pair effect might be more prominent with HFIP and HFTB than with PP indicating strong influence of additives 3.3 Weak bases (pKaH 3.3–7.0) The weak bases with pKaH 3.3–7.0 are only partially ionised in the used pH range and are mostly in a neutral form in the upper part of the used pH range When using ammonium acetate as an eluent, the retention factors increased in the w w pH range 5.0– 7.0 and with w w pH 8.5 began to decrease (Fig 1) In the case of PP, such a trend was observed only in the case of 1-naphthylamine The retention factors of other analytes decreased with increasing the pH of the eluent The full data set can be seen in Table S5 and chromatograms in Figure S3 With ammonium acetate as the eluent additive, up to w w pH 7.0 the analytes were partially ionised An increase in the eluent pH influences analytes to become more neutral and increases retention, as is common with RP LC In addition, π -π interactions and hydrogen bonding between the neutral analytes and free silanol groups on the stationary phase are possible Starting from w w pH 8.5, the analytes are essentially in their neutral forms and the proportion of the neutral form no longer increases with an increase in the pH The hydrophobic retention and π -π interactions still op4 K Lossmann, R Hecht, J Saame et al Journal of Chromatography A 1666 (2022) 462850 Fig Retention factors of strong bases using different eluent additives at different pH values Error bars represent standard deviations Fig Retention factors of weak acids using different eluent additives at different pH values Error bars represent standard deviations 3.5 Weak acids (pKa 9.3–11.4) With PP as the eluent additive, there is a more visible increase in the retention factors with the increase in the w w pH up to 7.0 (Fig 3) A further increase in the mobile phase pH leads to a decrease in the retention factors (Table S3) Using HFTB and HFIP, the analytes have retention trends similar to PP Most likely, the initial increase in retention is due to an increase in the proportion of the ionised form of PP, which may bind the (still almost fully neutral) analytes via hydrogen bonds As in the case of ammonium acetate, the main reason for the decrease in retention above pH 7.0 is that the analytes are increasingly in an anionic form With all mobile phases the interactions between anionic analytes and free silanol groups as well as π -π interactions are weakened due to the high hydrophilicity of the anions and increasing deprotonation of the silanol groups Anionic analytes are repelled from deprotonated silanol groups Weakly acidic analytes with pKa 9.3–11.4 are mostly in a neutral form throughout the used pH range Thus, they were expected to have similar retention factors at every mobile phase pH researched However, a more nuanced pattern of retention factor changes was observed with the changing of the pH, as seen in Fig The full data set can be seen in Table S3 and chromatograms can be seen in Figure S5 Using ammonium acetate as the eluent additive, the analytes are mostly in a neutral form up to w w pH 8.5 and their retention is essentially constant The π -π interactions are possible, depending on the analyte’s structure At higher pH values, the proportion of the ionised (anionic) form of the analytes begins to increase and thus the retention decreases K Lossmann, R Hecht, J Saame et al Journal of Chromatography A 1666 (2022) 462850 Fig Retention factors of strong acids using different eluent additives at different pH values Due to 4-nitrobenzoic acid’s very weak retention, this analyte is not shown in the graph A – 2,3,4,5,6-pentafluorophenol, B – N–hydroxy-6-bromobenzotriazole, C – N–hydroxy-6-trifluoromethylbenzotriazole, D – N–hydroxy-5–chloro-4methylbenzotriazole Error bars represent standard deviations Fig Comparison of the logarithms to the base 10 of the retention factors of pyrrolidine (strong base), 2,6-dimethylpyridine (weak base) and phenol (weak acid) with different stationary and mobile phases 3.6 Strong acids (pKa 3.4–7.9) All analytes except 2,4-dichlorophenol, which had the highest pKa value of the strong acids and thus was deprotonated to a smaller degree than the others, eluted close to dead time using HFIP and HFTB as eluent additives A decrease in the retention factors was observed with an increase in the pH for the analytes mostly in an anionic form (analytes with pKa < 7.9), as can be seen in Fig This can be explained with the shift of the acid’s dissociation equilibrium from the neutral acid HA to the anionic (deprotonated) A– form, which results in an increase in the analyte’s polarity and decrease in its retention The full data set can be seen in Table S2 and chromatograms can be seen in Figure S4 When using an ammonium acetate buffer with the w w pH values up to 7.0, there is a possibility for π -π interactions between the analytes and the stationary phase Although at w w pH values up to 7.0 an appreciable share of silanol groups are in neutral form, hydrogen bonds between silanols and anions are unlikely to form due to the high hydrophilicity of the anions In addition, the deprotonated silanols will repel the ionic analytes At higher w w pH values (8.5–9.5), the analytes are essentially fully ionised and retention is weak With PP as the eluent additive, a similar trend was observed as with ammonium acetate (see the paragraph above) In the w w pH range 5.0–7.0, the analytes’ retention was weaker than with ammonium acetate because partially anionic PP can be present on the stationary phase and repel anionic analytes At higher w w pH values (8.5–9.5), the analytes are almost fully ionised and retention is weak Comparison of the PFP stationary phase with the C18 and biphenyl stationary phases When fluoroalcohols were used as eluent additives with the C18 and biphenyl columns, the retention patterns displayed both similar and different trends in comparison to the PFP column We examined the situation on the basis of three exemplary analytes: pyrrolidine (strong base), 2,6-dimethylpyridine (weak base) and phenol (weak acid), see Fig We left out strong acids as their retention was weak overall and clear comparisons could not be made Fig reveals that very large differences in the retention factors were achievable for the same analyte using the same mobile phase composition with the three different RP stationary phases The difference in the order of tens of times is not rare with basic analytes (the largest is around 60 times) Short retention times of basic analytes on C18 column could be improved by changing the stationary phase The strong bases included in this work and in the works by Veigure et al [12,22] did not contain aromatic rings and therefore could not have π -π interactions, which makes the differences in the retention mechanisms between biphenyl, PFP and C18 K Lossmann, R Hecht, J Saame et al Journal of Chromatography A 1666 (2022) 462850 stationary phases smaller With phenol, the difference was around times These differences led to the observed highly differentiated extents of separation As an example, with HFTB at w w pH 8.5 as seen on Fig 5, the selectivity factor (the ratio of retention factors) between the strongest and weakest retained analyte (2,6dimethylpyridine and phenol) was around in the case of C18, while with the PFP column it was over 50 (pyrrolidine being the strongest retained and phenol being the weakest retained analyte) As a generalisation, Fig clearly shows that on average the best differentiation of retention with these analytes was obtained with the PFP phase When using the same stationary phase and the same w w pH, while changing the nature of the buffer system, the difference in retention can also be striking (see Figure S2) Although the PFP on average ensured the highest selectivity factors between the compounds, it was the remaining two stationary phases that displayed the most interesting retention order behaviour In the case of the biphenyl phase, just at w w pH 8.5 the four different buffer systems led to three (!) different retention orders and the selectivity factor between the strongest and weakest retained analyte ranged from 2.2 to 6.5 While PP has shown promising influence in our earlier study on C18 column, on PFP column HFIP and HFTB showed much stronger influence In Fig 5, the analyte that stands out is pyrrolidine – representative of strong bases The figure shows that PFP column with fluorinated eluent additives is the best stationary and mobile phase combination for pyrrolidine and similar analytes Using common mobile and stationary phases may not provide good enough retention or separation for strongly basic analytes Even when using a common eluent additive, the difference in retention factors is remarkable between different stationary phases: using ammonium acetate pyrrolidine has retention factors well below on C18 column, while the retention factor is well over 10 on the PFP stationary phase, showing an over 60 times increase in retention Similar trends can be seen for the fluorinated eluent additives as well While ammonium acetate is possibly the most common eluent additive in LC-MS, it generally leads to lower retention factors compared to results obtained with fluoroalcohols as eluent additives Thus, when using ammonium acetate, it would be necessary to use higher pH to be able to separate strong bases Higher pH values, however, are detrimental for any silica-based column and thus, not advisable The greatest difference obtained in retention factors between ammonium acetate and fluoroalcohols was more than 13 times (for pyrrolidine), remaining in the range of 3–10 times on average for other model analytes In conclusion, very different selectivity is possible with the same organic phase content and w w pH when using a different type of buffer and a different RP phase, however PFP with combined influence of fluoroalcohols has demonstrated a great potential of use with pyrrolidine-like analytes charge-charge interactions between protonated analytes and deprotonated silanols) and charge-charge interactions between the analytes and partially deprotonated fluoroalcohols which have attached to the stationary phase The effect of different fluoroalcohols was also varied, but HFIP and HFTB had a more noticeable effect on the retention of analytes than PP The comparison of the used mobile and stationary phases showed that different selectivity was possible by changing the mobile buffer or stationary phase Using the PFP column gave on average the best selectivity compared to the C18 or biphenyl columns Fluoroalcohols significantly improved the weak retention of basic analytes For weakly retained basic analytes this offers the possibilities of improvement of retention by either changing the stationary phase to PFP, changing the eluent additive to fluoroalcohols or both, without increasing mobile phase pH to highly basic Authors understand that fluoroalcohols might not outcompete tried and trusted eluent additives like ammonium acetate –as the routinely used buffer (as few as they are for LC-MS methods) components usually outcompete fluoroalcohols in price and performance combined competition, as well as for most aspects when acidic analytes are to be researched and analysed However, authors find this research very valuable nonetheless, for the cases, when the results obtained with the routine eluent additives are sub-optimal or not fit for the purpose at all – especially for basic analytes (as is the most common case when analysing active pharmaceutical ingredients) Both broadening the selection of LCMS compatible eluent additives and giving explanation for most probable interactions and thus elution patterns to expect, the authors hope to save time and effort for any bioanalytical chemist analysing medical products Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper CRediT authorship contribution statement Krit Lossmann: Investigation, Formal analysis, Writing – original draft, Visualization Ruta Hecht: Investigation, Writing – original draft, Visualization, Supervision Jaan Saame: Investigation, Formal analysis Agnes Heering: Investigation, Formal analysis Ivo Leito: Conceptualization, Resources, Writing – review & editing, Supervision Karin Kipper: Conceptualization, Methodology, Writing – review & editing, Supervision Acknowledgments The current research was supported by Restek Corporation through the Academic Support Program (RASP) This research was funded from the EMPIR programme (project 17FUN09 “UnipHied”, www.uniphied.eu) co-financed by the Participating States and from the European Union’s Horizon 2020 research and innovation programme, by the Estonian Research Council grants (PRG690) and by the EU through the European Regional Development Fund under the project TK141 “Advanced materials and high-technology devices for energy recuperation systems” (2014–2020.4.01.15–0011) This work was carried out using the instrumentation at the Estonian center of Analytical Chemistry (www.akki.ee) Conclusions When using fluoroalcohols as eluent additives, the retention factors of strong acids were generally lower, and the retention factors of strong bases were generally higher than when using ammonium acetate as the eluent additive The retention factors obtained for weak bases and weak acids were more comparable in the case of both ammonium acetate and fluoroalcohols Retention mechanisms with the PFP column using fluoroalcohols as the eluent additives appeared to be due to a combination of four interactions These interactions are RP hydrophobic retention mechanism, π -π interactions between aromatic analytes and the aromatic ring on the stationary phase, hydrogen bond interactions between the analytes and free silanols on the stationary phase (or Supplementary materials Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.chroma.2022.462850 K Lossmann, R Hecht, J Saame et al Journal of Chromatography A 1666 (2022) 462850 References [12] R Veigure, K Lossmann, M Hecht, E Parman, R Born, I Leito, K Herodes, K Kipper, Retention of acidic and basic analytes in reversed phase column using fluorinated and novel eluent additives for liquid chromatographytandem mass spectrometry, J Chromatogr A 1613 (2020) 460667, doi:10.1016/ j.chroma.2019.460667 [13] E Parman, L Toom, S Selberg, I Leito, Determination of pKa values of fluorocompounds in water using 19F NMR, J Phys Org Chem (2018) e3940 [14] R Filler, R.M Schure, Highly acidic perhalogenated alcohols A new synthesis of perfluoro-tert-butyl alcohol, J Org Chem 32 (1967) 1217–1219, doi:10.1021/ jo01279a081 [15] W.J Middleton, R.V Lindsey, Hydrogen bonding in Fluoro alcohols, J Am Chem Soc 86 (1964) 4948–4952, doi:10.1021/ja01076a041 [16] I Koppel, J Koppel, I Leito, V Pihl, L Grehn, U Ragnarsson, The acidity of substituted 1-Hydroxybenzotriazoles in water and Dimethyl sulfoxide, J Chem Res (1993) 446–447 [17] C Li, M.Z Hoffman, One-electron redox potentials of phenols in aqueous solution, J Phys Chem B 103 (1999) 6653–6656, doi:10.1021/jp983819w [18] J Stradins, B Hasanli, Anodic voltammetry of phenol and benzenethiol derivatives.: part Influence of pH on electro-oxidation potentials of substituted phenols and evaluation of pKa from anodic voltammetry data, J Electroanal Chem 353 (1993) 57–69, doi:10.1016/0022- 0728(93)80286- Q [19] S Zhang, A reliable and efficient first principles-based method for predicting pKa values organic bases, J Comput Chem 33 (2012) 2469–2482, doi:10 1002/jcc.23068 [20] N.F Hall, M.R Sprinkle, Relations between the structure and strength of certain organic bases in aqueous solution, J Am Chem Soc 54 (1932) 3469–3485 [21] Restek, Stationary Phase: fluoroPhenyl, (2018) https://www.restek.com/pdfs/ GNBR2368B-UNV.pdf (accessed 11 December 2021) [22] R Hecht (formerly Veigure), Novel Eluent Additives for LC-MS Based Bioanalytical Methods, University of Tartu, 2020 [23] M Rosés, Determination of the pH of binary mobile phases for reversed-phase liquid chromatography, J Chromatogr A 1037 (2004) 283–298, doi:10.1016/j chroma.2003.12.063 [24] A Suu, L Jalukse, J Liigand, A Kruve, D Himmel, I Krossing, M Rosés, I Leito, Unified pH values of liquid chromatography mobile phases, Anal Chem 87 (2015) 2623–2630, doi:10.1021/ac504692m [25] J Nawrocki, The silanol group and its role in liquid chromatography, J Chromatogr A 779 (1997) 29–71, doi:10.1016/S0 021-9673(97)0 0479-2 [1] M.R Euerby, A.P McKeown, P Petersson, Chromatographic classification and comparison of commercially available perfluorinated stationary phases for reversed-phase liquid chromatography using principal component analysis, J Sep Sci 26 (2003) 295–306, doi:10.1002/jssc.200390035 [2] D.S Bell, A.D Jones, Solute attributes and molecular interactions contributing to “U-shape” retention on a fluorinated high-performance liquid chromatography stationary phase, J Chromatogr A 1073 (2005) 99–109, doi:10.1016/j chroma.2004.08.163 [3] J Wells, Pharmaceutical Preformulation: the physicochemical properties of drug substances, 1998 [4] J.W Dolan, Back to basics: the role of pH in retention and selectivity, LC GC N Am 35 (2017) 22–28 [5] A Heering, D Stoica, F Camões, B Anes, D Nagy, Z Nagyné Szilágyi, R Quendera, L Ribeiro, F Bastkowski, R Born, J Nerut, J Saame, S Lainela, L Liv, E Uysal, M Roziková, M Vicˇ arová, A Snedden, L Deleebeeck, V Radtke, I Krossing, I Leito, Symmetric potentiometric cells for the measurement of unified pH values, Symmetry (Basel) 12 (2020), doi:10.3390/sym12071150 [6] A Tan, J.C Fanaras, Use of high-pH (basic/alkaline) mobile phases for LC–MS or LC–MS/MS bioanalysis, Biomed Chromatogr 33 (2019) e4409, doi:10.1002/ bmc.4409 [7] R Veigure, R Aro, T Metsvaht, J.F Standing, I Lutsar, K Herodes, K Kipper, A highly sensitive method for the simultaneous UHPLC–MS/MS analysis of clonidine, morphine, midazolam and their metabolites in blood plasma using HFIP as the eluent additive, J Chromatogr B 1052 (2017) 150–157, doi:10.1016/j.jchromb.2017.03.007 [8] K Kipper, K Herodes, I Leito, Fluoroalcohols as novel buffer components for basic buffer solutions for liquid chromatography electrospray ionization mass spectrometry: retention mechanisms, J Chromatogr A 1218 (2011) 8175–8180 [9] B Basiri, H van Hattum, W.D van Dongen, M.M Murph, M.G Bartlett, The role of fluorinated alcohols as mobile phase modifiers for LC-MS analysis of oligonucleotides, J Am Soc Mass Spectrom 28 (2017) 190–199 [10] L Gong, R Liu, Y Ruan, Z Liu, The role of fluoroalcohols as counter anion for ion-pair reversed-phase liquid chromatography/high resolution electrospray ionization mass spectrometry analysis of oligonucleotides, Rapid Commun Mass Spectrom 33 (2019) [11] K Kipper, K Herodes, I Leito, L Nei, Two fluoroalcohols as components of basic buffers for liquid chromatography electrospray ionization mass spectrometric determination of antibiotic residues, Analyst 136 (2011) 4587–4594 ... ammonium acetate As they became more neutral, the retention of analytes increased In addition, the partially ionised PP on the stationary phase could have ion-ion interactions with the ionised analytes. .. adsorbed on the stationary phase and (d) a hydrogen bond and/ or charge-charge interaction between the free silanol groups or their deprotonated forms on the stationary phase and the analytes The first... that on average the best differentiation of retention with these analytes was obtained with the PFP phase When using the same stationary phase and the same w w pH, while changing the nature of the