Comparison of the Retention and Separation Selectivity of Aromatic Hydrocarbons with Polar Groups in RP-HPLC Systems with Diﬀerent Stationary Phases and Eluents

molecules

Article Comparison of the Retention and Separation Selectivity of Aromatic Hydrocarbons with Polar Groups in RP-HPLC Systems with Diﬀerent Stationary Phases and Eluents

Anna Klimek-Turek 1,* , Beata Misiołek 1,2 and Tadeusz H. Dzido 1 1 Department of Physical Chemistry, Medical University of Lublin, Chod´zki4a, 20-093 Lublin, Poland; [email protected] (B.M.); [email protected] (T.H.D.) 2 Department for Variations and Renewals of Medicinal Products, The Oﬃce for Registration of Medicinal Products, Medical Devices and Biocidal Products, Al. Jerozolimskie 181C, 02-222 Warsaw, Poland * Correspondence: [email protected]; Tel.: +48-81448-7206

Academic Editors: Magdalena Wojciak-Kosior and Bogusław Buszewski Received: 29 September 2020; Accepted: 28 October 2020; Published: 1 November 2020

Abstract: In this manuscript, the retention of aromatic hydrocarbons with polar groups has been compared for systems with various nonpolar columns of the types from C3 to C18 and different mobile phases composed of methanol, acetonitrile, or tetrahydrofuran as modifiers. The selectivity separation of the solutes in systems with different adsorbents, when one eluent modifier is swapped by another, has been explained, taking into account molecular interactions of the solutes with components of the stationary phase region (i.e., extracted modifier depending on the chain length of the stationary phase).

Keywords: stationary phases; molecular interactions; separation selectivity; HPLC

1. Introduction Reverse-phase high-performance liquid chromatography (RP-HPLC) is one of the most commonly used analytical methods applicable in the pharmaceutical [1–3] and cosmetic industries [4–6], food analysis [7–9], diagnostics [10,11], biomedicine [12], environmental protection [9], and many other fields [13]. The main advantage of HPLC as an analytical method is the possibility of using many options for changing the retention and separation selectivity [11,14,15], like selecting the polarity of the eluent [16], tailoring the mobile phase, and so forth [17–19]. However, some aspects of the chromatographic process, especially those related to retention and separation selectivity mechanisms, are still debatable [20–27]. It is widely accepted that retention could be caused by adsorption of the analytes onto the adsorbent surface [28] and/or partitioning of the analytes between the stationary and the mobile phases [29,30]. The properties of the stationary phase strongly depend on the qualitative and quantitative compositions of the mobile phase [31]. The components of the mobile phase may adsorb on the stationary phase, forming a layer or even many layers [32–35]. The type and amount of adsorbed molecules of organic modifier have a strong influence on retention and selectivity [36–38]. Therefore, the stationary phase cannot be considered only as hydrocarbon chains with a certain length but as a combination of four components: hydrocarbon chains, solvent molecules, water molecules, and residual silanols [39,40]. Therefore, it is very hard to interpret the retention changes and put forward a clear conclusion. However, it is worth noting that nonpolar stationary phases of the C18 type of similar carbon density in systems with various organic modifiers are characterized by constant composition of the hydrocarbon chains, unreacted silanol groups, and water (in the limited range of modifier concentration) [39]. It means that the modifier in such chromatographic system is responsible for

Molecules 2020, 25, 5070; doi:10.3390/molecules25215070 www.mdpi.com/journal/molecules Molecules 2020, 25, 5070 2 of 16 different stationary phase properties. In addition, the partition constant of aromatic hydrocarbons with different polar groups in gas–liquid systems, where liquid is composed of organic modifier (methanol (MeOH), acetonitrile (ACN), or tetrahydrofuran (THF)) in water, demonstrates a very good correlation [41]. Based on these circumstances, in our previous papers we proposed an approach in which separation selectivity can be explained by molecular interactions of solutes and the component of the stationary phase when one modifier is replaced by another [39,41]. The application of this approach to the explanation of separation selectivity changes was demonstrated, among others, for aliphatic hydrocarbons with a polar group [42] and aromatic hydrocarbons with different polar groups [43] in systems with C18 stationary phase type. The results presented in our previous papers showed a clear dependence of the changes in separation selectivity on the mobile phase modifier type [39]. They also allowed us to develop rules explaining these changes. However, these rules refer to those stationary phases that were investigated so far (i.e., C18 type as mentioned above). In view of the wide range of stationary phases, it is justified to check whether the above-described approach is also applicable to systems with other nonpolar adsorbents containing alkyl chains of different lengths, and also differing in the degree of coverage with these chains, the size of the pores, and so forth. It is well known that the total amount of adsorbed organic solvent depends on the length of hydrocarbon chains [44], the density of the column surface coverage [44], and the presence of free silanol groups [45]. When the surface coverage and number of carbons in ligands increase, the amount of the adsorbed solvent also increases, but it is worth noting that when the coverage density is very high, molecules of an organic modifier cannot penetrate the stationary phase region, and they adsorb only on the outer surface [34]. In turn, the endcapping of the stationary phase decreases unwanted influence of accessible residual silanols on the separation process [33]. Thus, a better understanding of these phenomena could lead to the improvement of the prediction of separation selectivity changes with respect to the modifier choice. The aim of the paper is to extend investigations on relative retention changes of aromatic hydrocarbons with polar groups in systems with hydro-organic eluent and 10 various stationary phases with different hydrocarbon lengths (from 3 to 18) and coverage density. Such research will allow us to conclude whether the proposed approach to explaining retention changes is applicable to systems with stationary phases other than C18 type. The approach could be very useful for the optimization of separation conditions of various substances, not only aromatic hydrocarbons with polar groups.

2. Results and Discussion Retention data for a set of 34 compounds were obtained for systems with different nonpolar stationary phases (C3, C8, and C18 types) and organic modifiers (MeOH, ACN, and THF) in water mobile phases. The chosen concentration of the modifiers ensures similar retention of benzene in all systems. Moreover, the concentration of the organic modifier in the eluent did not exceed 50%. In order to compare the obtained results, the retention of substances was correlated as log k1 versus log k2, where k1 and k2 are solute retention factors in systems 1 and 2, respectively (see Figures S1–S10). On the resulting plots, one could see that the solutes formed two separate lines, the first one for substances with proton-donor and proton-acceptor properties (AH > 0, BH > 0, where AH-hydrogen bond acidity, BH-hydrogen bond basicity) and the second one for compounds capable of interacting as an electron-pair donor (BH > 0). The obtained parameters of linear equations and values of correlation coefficient are gathered in Table1. Molecules 2020, 25, 5070 3 of 16

Table 1. The obtained parameters of linear equations and values of correlation coeﬃcient for investigated chromatographic systems.

SB C18 Rx C18 XDB C18 Extend C18 AH > 0, AH > 0, AH > 0, AH > 0, BH > 0 BH > 0 BH > 0 BH > 0 BH > 0 BH > 0 BH > 0 BH > 0

a = 0.6811 a = 0.6626 34% ACN (k1) a = 0.7504 a = 0.7114 34% ACN (k1) a = 0.7749 a = 0.6837 35% ACN (k1) a = 0.7819 a = 0.6867 35% ACN (k1) b = 0.3047 b = 0.0626 - b = 0.1614 b = 0.0307 - b = 0.1871 b = 0.0397 - b = 0.1734 b = 0.0534 44% MeOH (k2) R = 0.7946 R = 0.9541 45% MeOH (k2)R = 0.9 R = 0.9630 43% MeOH (k2)R = 0.9048 R = 0.9575 32% MeOH (k2)R = 0.9239 R = 0.9633

a = 0.5507 a = 0.7765 32% THF (k1) a = 0.5938 a = 0.7513 31%THF (k1) a = 0.6729 a = 0.7607 31% THF (k1) a = 0.6447 a = 0.7578 32% THF (k1) b = 0.3885 b = 0.4242 - b = 0.2975 b = 0.4005 - b = 0.2640 b = 0.4519 - b = 0.3460 b = 0.4494 35% ACN (k2) R = 0.5846 R = 0.8414 34% ACN (k2)R = 0.7220 R = 0.8296 34% ACN (k2)R=0.7876 R = 0.8413 35% ACN (k2)R = 0.6876 R = 0.8321

a = 0.491 a = 0.4765 32% THF (k1) a= 0.4060 a = 0.5040 31% THF (k1) a = 0.5007 a = 0.4783 31% THF (k1) a = 0.5239 a = 0.4904 32% THF (k1) b = 0.4602 b = 0.4964 - b = 0.4252 b = 0.4375 - b = 0.4063 b = 0.5099 - b = 0.44 b = 0.5077 44% MeOH (k2) R = 0.6080 R = 0.7434 45% MeOH (k2)R = 0.5921 R = 0.7534 45%MeOH (k2)R = 0.6879 R = 0.7407 45% MeOH (k2)R = 0.6603 R = 0.7554 SB C8 XDB C8 SB C3 AH > 0, AH > 0, AH > 0, BH > 0 BH > 0 BH > 0 BH > 0 BH > 0 BH > 0

a = 0.5695 a = 0.59 36% ACN (k1) a = 0.6426 a = 0.6109 33% ACN (k1) a = 0.6249 a = 0.2294 35% ACN (k1) b = 0.225 b = 0.0002 - b = 0.2653 b = 0.0606 - b = 0.0761 b = 0.1295 42% MeOH (k2) R = 0.7892 R = 0.8612 45% MeOH (k2)R = 0.834 R = 0.9557 25% MeOH (k2)R = 0.6987 R = 0.4205

a = 0.6870 a = 0.8771 32% THF (k1) a = 0.6699 a = 0.9109 33% THF (k1) a = 0.1515 a = 0.613 33% THF (k1) b = 0.289 b = 0.3277 - b = 0.3413 b = 0.3867 - b = 0.531 b = 0.439 35% ACN (k2) R = 0.6504 R = 0.8371 36% ACN R = 0.658 R = 0.7837 33% ACN (k2)R = 0.2593 R = 0.387

a = 0.3191 a = 0.4338 32% THF (k1) a = 0.4129 a = 0.5325 33% THF (k1) a = 0.1293 a = 0.2026 33% THF (k1) b = 0.5135 b = 0.3964 - b = 0.5321 b = 0.4558 - b = 0.4538 b = 0.4203 42% MeOH (k2) R = 0.4187 R = 0.6044 45% MeOH (k2)R = 0.5264 R = 0.7164 25% MeOH (k2)R = 0.2571 R = 0.3876 Molecules 2020, 25, 5070 4 of 16

2.1. Comparison of Substance Retention and Selectivity in ACN and THF Systems with Different Adsorbents Selectivity differences between the systems with acetonitrile and tetrahydrofuran result from the different properties of these modifiers. The THF molecule is flatter and has greater volume and surface area than the ACN molecule. Tetrahydrofuran solvates and organizes the stationary phase area to a greater extent in comparison with ACN. It also exhibits greater proton-accepting properties than ACN (BH THF = 0.55, BH ACN = 0.31) [46] and a higher refractive index (nTHF = 1.405; nACN = 1.342); therefore, it is characterized by greater ability for dispersion interactions than acetonitrile. The dipole moment of ACN is 3.45D, while it is 1.75D for THF [47]. The correlation line for substances with proton-donor and proto-acceptor properties is above the line for solutes with an electron-donor group (Table1) for all columns investigated, which indicates their higher retention in a system with THF in comparison with ACN. Compounds with proton-donor properties have a greater tendency to interact with tetrahydrofuran, which has high proton-accepting property [46]. On the other hand, substances with abilities for interactions as a proton acceptor and dipolar interactions have higher retention in comparison with others in systems with acetonitrile. Acetonitrile is characterized by weak contribution to interaction as a proton acceptor and possesses high dipolar properties [48]. As can be seen in Table1, the values of slope of lines for substances with proton-donor and proton-acceptor properties are similar for all columns of C18 type and slightly higher for columns of type C8 (except column 300 SB 8). The lines for compounds capable of interacting as an electron-pair donor have slopes very similar to systems with all tested columns, except C3 columns. Based on retention data, one will notice that the substances with two OH groups (1,5-dihydroxynaphthalene,Molecules 2020, 25, x 1,6-dihydroxynaphthalene, 1,7-dihydroxynaphthalene) exhibit retention5 of 17 increase in comparison with the monofunctional solutes (phenol, o-cresol, p-cresol) in the THF system with tetrahydrofuran than with acetonitrile in the stationary phase region. The examples of relative to that of ACN. This effect is caused by stronger hydrogen bond interaction of the solutes with separation factor (selectivity), α, values for naphthalene relative to phenol are presented in Figure 1. tetrahydrofuran than with acetonitrile in the stationary phase region. The examples of separation factor (selectivity), α, values for naphthalene relative to phenol are presented in Figure1.

1,5-dihydroxynaphthalene/phenol 1,6-dihydroxynaphthalene/phenol  1,7-dihydroxynaphthalene/phenol

1.5

1.0

0.5 SB C3 SB T SB C8 SB T SB C3 SB A C3 SB C8 SB A C8 SB C18 SB T RX C18 T RX SB C18 SB A C18 RX C18 A C18 RX XDB C8 A XDB C8 T XDB C18 T XDB XDB C18 A C18 XDB 300 SB 300 C3SB T 300 C8SB T 300 SB A C3 300 SB A C8 Extent C18 T C18 Extent Extent C18 A C18 Extent 300 SB 300 C18SB T 300 SB A C18

FigureFigure 1. 1.Separation Separation factor,factor, αα,, valuesvalues ofof dihydroxynaphthalenesdihydroxynaphthalenes relative relative to to phenol phenol in in systems systems with with didifferentﬀerent nonpolar nonpolar stationary stationary phasesphases andand acetonitrileacetonitrile (A) in in water water and and tetrahydrofuran tetrahydrofuran (T) (T) in in water water (modiﬁer(modifier concentrations concentrations as as in in Table Table1 ).1).

ItIt is is worth worth underlining underlining that that this this eﬀ ecteffect is notis not noticeable noticeable in systems in systems with with the SBthe C3 SB stationary C3 stationary phase andphase all 300and SB all columns.300 SB columns. In such In cases, such the cases, separation the separation factor offactor the comparedof the compared compounds compounds has almost has almost identical values in the THF and ACN systems. As the length of the hydrocarbon chains decreases, less modifier takes part in their solvation, thus decreasing the intermolecular interactions of acetonitrile and tetrahydrofuran with the tested substances [49]. The influence of a hydroxyl functional group on substance retention can be observed when the retention of esters with a hydroxyl group (proton‐donor group) is compared with the retention of those without this group. In systems with acetonitrile, the selectivity of the latter towards the former is quite high (from 2.5 to 4.2 for the pair methyl phenylacetate/methyl 4‐hydroxybenzoate, from 1.9 to 4.5 for the pair methyl benzoate/methyl 4‐hydroxybenzoate, from 2.6 to 4.2 for the pair ethyl phenylacetate/ethyl 4‐hydroxybenzoate), and these values increase with increasing length of stationary phase chains (Figure 2). However, when tetrahydrofuran is used as a modifier, this value drops to approximately 1.5 for all columns. This proves the greater ability of THF to act as a proton acceptor in comparison with ACN [46].

Molecules 2020, 25, 5070 5 of 16 identical values in the THF and ACN systems. As the length of the hydrocarbon chains decreases, less modifier takes part in their solvation, thus decreasing the intermolecular interactions of acetonitrile and tetrahydrofuran with the tested substances [49]. The influence of a hydroxyl functional group on substance retention can be observed when the retention of esters with a hydroxyl group (proton-donor group) is compared with the retention of those without this group. In systems with acetonitrile, the selectivity of the latter towards the former is quite high (from 2.5 to 4.2 for the pair methyl phenylacetate/methyl 4-hydroxybenzoate, from 1.9 to 4.5 for the pair methyl benzoate/methyl 4-hydroxybenzoate, from 2.6 to 4.2 for the pair ethyl phenylacetate/ethyl 4-hydroxybenzoate), and these values increase with increasing length of stationary phase chains (Figure2). However, when tetrahydrofuran is used as a modifier, this value drops to approximately 1.5 for all columns. This proves the greater ability of THF to act as a proton acceptorMolecules in2020 comparison, 25, x with ACN [46]. 6 of 17

5 methyl phenylacetate/methyl 4 - hydroxybenzoate methyl benzoate/methyl 4 - hydroxybenzoate  ethyl phenylacetate/ethyl 4 - hydroxybenzoate

1 SB C3 T SB C8 T SB C3 A SB C8 A SB C18 T SB C18 A RX C18 T RX C18 A XDB C8 A C8 XDB XDB C8 T XDB C18 T XDBC18 A 300 SB C3 T 300 SB C8 T 300 SB C3 A 300 SB C8 A Extent C18Extent T Extent C18 A 300 SB C18 T 300 SB C18 A FigureFigure 2. 2.Separation Separation factor, factor,α α, values, values of of esters esters in in systems systems with with di ﬀdifferenterent nonpolar nonpolar stationary stationary phases phases and acetonitrileand acetonitrile (A) in (A) water in water and tetrahydrofuran and tetrahydrofuran (T) in (T) water in water (modiﬁer (modifier concentrations concentrations as inas Tablein Table1). 1).

AnotherAnother exampleexample of selectivity changes changes demonstrates demonstrates the the 4‐nitrobenzaldehyde/4 4-nitrobenzaldehyde‐ / 4-cyanobenzaldehydecyanobenzaldehyde pair. pair. The The higher higher value value of of the the retention retention of of the the first first substance substance in in relation relation to to the the secondsecond one one in in systems systems with with THF THF comparedcompared with those with ACN ACN is is due due to to the the stronger stronger interaction interaction of of thethe tetrahydrofuran tetrahydrofuran molecule molecule with with the the NO NO22 group. On On the the other other hand, hand, the the CN CN group group is is characterized characterized byby strong strong dipolar dipolar properties properties and and capabilitycapability for dipolar interaction with with ACN ACN in in the the stationary stationary phase phase region.region. For For this reason,this reason, 4-cyanobenzaldehyde 4‐cyanobenzaldehyde exhibits greaterexhibits retention greater relative retention to 4-nitrobenzaldehyde relative to 4‐ innitrobenzaldehyde ACN systems. The in values ACN systems. of separation The values factor areof separation very close factor for systems are very with close all for stationary systems phases with exceptall stationary 300 SB C3, phases 300 SBexcept C8, 300 and SB 300 C3, SB 300 C18. SB C8, and 300 SB C18. TheThe presence presence of of a a nitro nitro group group inin thethe solute molecule enhances enhances its its retention retention relative relative to to a asubstance substance withoutwithout this this group group in in THF THF relative relative to to ACN ACN systems.systems. This phenomenon phenomenon can can be be observed observed in in such such pairs pairs of substances as 2‐methylo‐4‐nitrophenol/o‐cresol and 4‐nitrophenol/phenol. It is worth noting that of substances as 2-methylo-4-nitrophenol/o-cresol and 4-nitrophenol/phenol. It is worth noting that the the ratio of separation coefficients for selected substance pairs in the THF system in relation to the ratio of separation coefficients for selected substance pairs in the THF system in relation to the ACN ACN system has the highest values for systems with C18 columns except 300 SB C18 (1.4–1.7), system has the highest values for systems with C18 columns except 300 SB C18 (1.4–1.7), intermediate intermediate values for SB C8 and XDB C8 (1.2–1.5), and smallest values for SB C3 and 300 SB C3, 300 values for SB C8 and XDB C8 (1.2–1.5), and smallest values for SB C3 and 300 SB C3, 300 SB C8, and 300 SB C8, and 300 SB C18 (1.1–1.4). This means that as the length of the hydrocarbon chains increases, SB C18 (1.1–1.4). This means that as the length of the hydrocarbon chains increases, the retention of the retention of substances with an NO2 group in relation to a substance without this group increases substances with an NO group in relation to a substance without this group increases in a system with in a system with THF2 compared with ACN. The nitro group possesses two highly polarized bonds THF compared with ACN. The nitro group possesses two highly polarized bonds with an electron with an electron density deficiency on the nitrogen atom. The THF molecule has two CO bonds with density deficiency on the nitrogen atom. The THF molecule has two CO bonds with high electron high electron density on the O atom. For this reason, relatively strong quadrupole interactions between nitrobenzene and THF take place [39]. As the length of the hydrocarbon chains increases, more modifiers take part in their solvation, thus increasing the intermolecular interactions of acetonitrile and tetrahydrofuran with the tested substances [40]. These interactions are stronger the more NO2 groups the molecule has. Nitrobenzene derivatives with two nitro groups (i.e., 1,2‐dinitrobenzene, 1,4‐dinitrobenzene) exhibit higher retention than nitrobenzene but are much weaker than 1,3,5‐trinitrobenzene in the THF system in comparison with ACN. The exemplification of this phenomenon is the retention of nitrobenzene and 1,3,5‐trinitrobenzene in systems with SB C18, SB C8, and SB C3 stationary phases (Figure 3).

Molecules 2020, 25, 5070 6 of 16 density on the O atom. For this reason, relatively strong quadrupole interactions between nitrobenzene and THF take place [39]. As the length of the hydrocarbon chains increases, more modiﬁers take part in their solvation, thus increasing the intermolecular interactions of acetonitrile and tetrahydrofuran with the tested substances [40]. These interactions are stronger the more NO2 groups the molecule has. Nitrobenzene derivatives with two nitro groups (i.e., 1,2-dinitrobenzene, 1,4-dinitrobenzene) exhibit higher retention than nitrobenzene but are much weaker than 1,3,5-trinitrobenzene in the THF system in comparison with ACN. The exempliﬁcation of this phenomenon is the retention of nitrobenzene and 1,3,5-trinitrobenzene inMolecules systems 2020 with, 25, x SB C18, SB C8, and SB C3 stationary phases (Figure3). 7 of 17

log k nitrobenzene 1,3,5-trinitrobenzene

1.2

0.9

0.6 SB C8/THF SB C3/THF SB SB C8/ACN SB C3/ACN SB C18/THF SB SB C18/ACN SB

FigureFigure 3. 3.Log Log k ofk nitrobenzeneof nitrobenzene and 1,3,5-trinitrobenzeneand 1,3,5‐trinitrobenzene in systems in systems with di ﬀwitherent different nonpolar nonpolar stationary phasesstationary and phases tetrahydrofuran and tetrahydrofuran (THF) in water (THF) and in acetonitrilewater and acetonitrile (ACN) in water.(ACN) in water.

TheThe influence influence of of the the shape shape ofof separated substance molecules molecules on on their their retention retention and and the the selectivity selectivity ofof the the separation, separation, depending depending onon thethe mobilemobile phase modifier modifier used, used, is is also also significant. significant. In In systems systems with with tetrahydrofuran,tetrahydrofuran, compounds compounds with with a brancheda branched structure structure of moleculesof molecules (isophthalates) (isophthalates) show show much much lower retentionlower retention relative relative to substances to substances that possess that possess a planar a moleculeplanar molecule in comparison in comparison with acetonitrile with acetonitrile systems. Thissystems. effect isThis visible effect for theis visible following for substance the following pairs: dimethylsubstance isophthalate pairs: dimethyl/4-cyanobenzaldehyde, isophthalate/4‐ dimethylcyanobenzaldehyde, isophthalate /4-nitrobenzaldehyde,dimethyl isophthalate/4 diethyl‐nitrobenzaldehyde, terephthalate/4-cyanobenzaldehyde, diethyl terephthalate/4 and diethyl‐ terephthalatecyanobenzaldehyde,/4-nitrobenzaldehyde. and diethyl terephthalate/4 In systems with‐nitrobenzaldehyde. acetonitrile, the valuesIn systems of the with separation acetonitrile, factor the of thesevalues pairs of the of compounds separation factor are much of these higher pairs relative of compounds to the systems are much with higher tetrahydrofuran. relative to the An systems increase ofwith ordering tetrahydrofuran. of a surface An region increase of the of stationary ordering phaseof a surface is represented region of by the the stationary sequence ofphase organic is modifiers:represented MeOH, by the ACN, sequence THF of [49 organic]. This meansmodifiers: that MeOH, relative ACN, retention THF of [49]. the soluteThis means with athat more relative highly retention of the solute with a more highly branched molecule structure decreases relative to the one branched molecule structure decreases relative to the one with a less branched structure according with a less branched structure according to the order of the systems presented. In Figure 4, the relative to the order of the systems presented. In Figure4, the relative selectivity changes for dimethyl selectivity changes for dimethyl isophthalate/4‐cyanobenzaldehyde and dimethyl isophthalate/4‐ isophthalate/4-cyanobenzaldehyde and dimethyl isophthalate/4-nitrobenzaldehyde are presented. nitrobenzaldehyde are presented. One can see that differences between values of separation factor in One can see that differences between values of separation factor in THF and ACN systems rise with THF and ACN systems rise with chain length and coverage density increase of the stationary phase. chain length and coverage density increase of the stationary phase.

Molecules 2020, 25, 5070 7 of 16 Molecules 2020, 25, x 8 of 17

diethyl terephthalate/4-cyanobenzaldehyde  diethyl terephthalate/4-nitrobenzaldehyde

0 -- SB C3T SB SB C8T SB SB C3 A C3 SB SB C8 A C8 SB SB C18 T SB C18 A RX C18 T RX C18 A XDB C8 A XDB C8 XDB C8 T XDB C18 T XDB C18 A 300 SB300 C3 T SB300 C8 T 300 SB C3 A 300 SB C8 A Extent C18 T Extent Extent C18 A Extent 300 SB C18 T 300 SB C18 A SB 300

FigureFigure 4. 4.Separation Separation factor,factor, αα,, valuesvalues ofof dimethyldimethyl isophthalateisophthalate/4/4-cyanobenzaldehyde‐cyanobenzaldehyde and and dimethyl dimethyl isophthalateisophthalate/4/4-nitrobenzaldehyde‐nitrobenzaldehyde in systems with with different diﬀerent nonpolar nonpolar stationary stationary phases phases and and (T) (T) tetrahydrofurantetrahydrofuran in in water, water, and and (A) (A) acetonitrile acetonitrile inin waterwater (modiﬁer(modifier concentrations as as in in Table Table 11).).

2.2.2.2. Comparison Comparison of of Substance Substance Retention Retention and and Selectivity Selectivity in in MeOH MeOH and and THF THF Systems Systems with with Di Differentfferent Adsorbents Adsorbents Methanol is characterized by high proton-acceptor properties (BH MeOH = 0.62) [46]. Methanol can alsoMethanol be a proton is characterized donor (AH MeOH by high= 0.93).proton Moreover,‐acceptor properties the MeOH (B moleculeH MeOH is = much 0.62) [46]. smaller Methanol than the THFcan moleculealso be a proton and is lessdonor adsorbed (AH MeOH on/in = 0.93). the stationary Moreover, phase the MeOH and therefore molecule organizes is much smaller the stationary than phasethe THF region molecule to a lesser and extent is less than adsorbed THF [49 on/in]. These the characteristics stationary phase are the and reason therefore for the organizes differences the in thestationary relative retentionphase region of the to testeda lesser substances extent than in THF the discussed[49]. These systems. characteristics are the reason for the differencesSimilarly, in asthe in relative the case retention of ACN–THF of the tested systems, substances the retention in the discussed of substances systems. has been correlated as logSimilarly, k1 (obtained as in in the systems case of with ACN–THF THF) versus systems, log the k2 (obtainedretention of in substances systems with has MeOH). been correlated It can be observedas log k1 that,(obtained as in in the systems ACN systemwith THF) versus versus the log THF k2 system,(obtained the in solutessystems form with twoMeOH). separate It can lines, be oneobserved for substances that, as in with the ACN proton-donor system versus and proton-acceptor the THF system, properties the solutes and form the two other separate for compounds lines, one for substances with proton‐donor and proton‐acceptor properties and the other for compounds not not capable of interacting as a proton donor (AH = 0) (see Figures S11–S20). In the case of systems withcapable the C18 of interacting adsorbent, as the a correlationproton donor line (A forH = substances 0) (see Figures with S11–S20). proton-donor In the groups case of is systems slightly with above thethe line C18 for adsorbent, compounds the correlation without these line for groups. substances The explanation with proton for‐donor this groups fact is relatedis slightly to above the greater the sorptionline for ofcompounds THF in/on without the stationary these phasegroups. compared The explanation with that for of MeOH.this fact is related to the greater sorptionThis of eff THFect is in/on also the present stationary when phase comparing compared systems with that (for of allMeOH. tested stationary phases) with This effect is also present when comparing systems (for all tested stationary phases) with tetrahydrofuran and acetonitrile, but the distances between the correlation lines are greater relative to tetrahydrofuran and acetonitrile, but the distances between the correlation lines are greater relative the MeOH system versus the THF system. to the MeOH system versus the THF system. Moreover, regarding the THF system versus the MeOH system, this difference is practically absent Moreover, regarding the THF system versus the MeOH system, this difference is practically for C8 (80 Å) phases, or the line position is reversed for C3, C8, and C18 phases with a pore size of absent for C8 (80 Å) phases, or the line position is reversed for C3, C8, and C18 phases with a pore 300 Å. This phenomenon is probably due to solvent properties, which have proton-accepting properties size of 300 Å. This phenomenon is probably due to solvent properties, which have proton‐accepting and can form hydrogen bonds with the analytes [46]. The longer are the hydrocarbon chains of the properties and can form hydrogen bonds with the analytes [46]. The longer are the hydrocarbon stationary phases and the greater is their specific surface area (adsorbents with a larger pore diameter chains of the stationary phases and the greater is their specific surface area (adsorbents with a larger havepore a diameter lower specific have a surface lower area),specific the surface more solventarea), the can more be adsorbed solvent can in/ beon adsorbed its surface in/on [40]. its Shortening surface the[40]. length Shortening of the stationary the length phase of the ligands stationary reduces phase the ligands amount reduces of THF the in amount this region. of THF The in lower this region. amount ofThe tetrahydrofuran lower amount in thisof tetrahydrofuran zone reduces the likelihoodin this zone of hydrogenreduces the bond likelihood interactions of withhydrogen proton-donor bond substances.interactions However, with proton MeOH,‐donor contrary substances. to ACN However, and THF,MeOH, may contrary be strongly to ACN adsorbed and THF, on themay silica be surfacestrongly through adsorbed hydrogen on the silica bonding surface with through residual hydrogen accessible bonding silanols with [40], residual apart from accessible the solvation silanols of hydrocarbon[40], apart from chains. the Therefore,solvation of methanol hydrocarbon sorption chains. in the Therefore, area of the methanol stationary sorption phase in does the notarea decrease of the stationary phase does not decrease as much as in case of THF with the shortening of the chain length.

Molecules 2020, 25, x 9 of 17

Moreover, the shorter are adsorbent hydrocarbon chains, the easier is the access of MeOH molecules to silanol groups [40]. It can be observed that nitrophenols show a higher retention relative to phenols in systems with Moleculestetrahydrofuran2020, 25, 5070 compared with that with methanol. Examples of these relative retention changes8 for of 16 selected pairs of compounds are presented in Figure 5. This effect is visible for all C8 and C18 type columns. as muchWhen as in the case C3 of type THF or with 300 theSB C18 shortening and 300 of SB the C8 chain stationary length. phases Moreover, are used, the shorter the relationship are adsorbent is hydrocarbonreversed and chains, the values the easier of the is retention the access of of nitrophenol MeOH molecules relative toto silanolphenol groups are higher [40]. in the MeOH systemIt can compared be observed with that THF. nitrophenols This is probably show adue higher to the retention reduction relative in the to phenolsamount inof systemsTHF in the with tetrahydrofuranstationary phase compared area. The adsorbed with that amount with methanol. of organic Examplesmodifiers strongly of these depends relative retentionon the stationary changes forphase selected chain pairs length of and compounds surface area. are presentedThe shorter in chain Figure length5. This (or esmallerﬀect is surface visible area) for all leads C8 and to less C18 typeamount columns. of THF in the stationary phase zone [40].

2-methyl 4-nitrophenol/o-cresole 4-nitrophenol/phenol 

0 -- SB C3 T C3 SB SB C8 T C8 SB SB C3 M C3 SB SB C8 M C8 SB SB C18 T RX C18 T SB C18SB M RX C18 M XDB C8 M XDBC8 XDB C8 TC8 XDB XDB C18 T XDB C18 M C18 XDB 300 SB C3 T 300 SB C8 T Extent C18 T Extent 300 SB C3 M 300 SB C8 M Extent C18 M Extent 300 SB C18 T SB 300 300 SB C18 M A FigureFigure 5. 5.Separation Separation factor,factor, αα,, valuesvalues of hydrocarbons with with different different functional functional groups groups in in systems systems withwith di differentfferent nonpolar nonpolar stationary stationary phases phases andand methanol (M) in in water water and and tetrahydrofuran tetrahydrofuran (T) (T) in in water water (modifier(modifier concentrations concentrations as as in in Table Table1 ).1).

WhenMethanol the C3 has type the orability 300 SBto solvate C18 and silanol 300 SB groups, C8 stationary which are phases “more are accessible” used, the to relationship those with is reversedshorter carbon and the chains values [50]. of Therefore, the retention the ofconcentration nitrophenol of relative MeOH toin phenolthe area are of the higher stationary in the phase MeOH systemdoes not compared decrease with much, THF. and This therefore, is probably it is duepossible to the to reduction have a greater in the participation amount of THF of intermolecular in the stationary phaseinteractions area. The of adsorbedthis modifier amount with of nitrophenols organic modifiers relative strongly to monophenols depends on in thea system stationary with phaseMeOH chain in lengthcomparison and surface with THF. area. The shorter chain length (or smaller surface area) leads to less amount of THF in theAdditionally, stationary phase the zoneinfluence [40]. of the number of ‐NO2 groups in a solute molecule on separation selectivityMethanol changes, has the depending ability to on solvate whether silanol THF groups, or MeOH which was areused “more as the accessible” mobile phase to thosemodifier, with shortercan be carbonobserved. chains In systems [50]. Therefore, with methanol, the concentration the order of of elution MeOH is in consistent the area ofwith the the stationary solvophobic phase doestheory not [51]: decrease as more much, polar and groups therefore, (in this it case is possible ‐NO2) are to havepresent, a greater a solute participation is less retained of intermolecularin the system interactionswith a hydrophobic of this modifier stationary with phase. nitrophenols In the system relative with to monophenols THF, 1,3,5‐trinitrobenzene in a system with has MeOHa much in higher retention relative to 1,2‐dinitrobenzene or nitrophenols. The reason for this are the electrostatic comparison with THF. interactions of THF molecules with nitro groups (quadrupole interactions), as described above. Additionally, the influence of the number of -NO2 groups in a solute molecule on separation Another example of changes in separation selectivity with the replacement of one modifier by selectivity changes, depending on whether THF or MeOH was used as the mobile phase modifier, another occurs when comparing the ester retention with and without the additional hydroxyl group. can be observed. In systems with methanol, the order of elution is consistent with the solvophobic Esters that possess a hydroxyl group exhibit greater retention relative to esters without such group theory [51]: as more polar groups (in this case -NO ) are present, a solute is less retained in the in systems with THF compared with systems with MeOH2 (Figure 6). system with a hydrophobic stationary phase. In the system with THF, 1,3,5-trinitrobenzene has a much A similar effect, however, more pronounced, is observed for the correlation of THF versus ACN. higher retention relative to 1,2-dinitrobenzene or nitrophenols. The reason for this are the electrostatic This effect is present for systems with all tested stationary phases except C3, while for systems with interactions of THF molecules with nitro groups (quadrupole interactions), as described above. Another example of changes in separation selectivity with the replacement of one modifier by another occurs when comparing the ester retention with and without the additional hydroxyl group. Esters that possess a hydroxyl group exhibit greater retention relative to esters without such group in systems with THF compared with systems with MeOH (Figure6). Molecules 2020, 25, x 10 of 17

Moleculesall C8 and2020 ,30025, 5070SB C18 stationary phases, this effect is less visible. This phenomenon can be explained9 of 16 in a similar way as that for the nitrophenol/phenol selectivity changes discussed above.

Molecules 2020, 25, x 10 of 17

all C8 and 300 SB C18 stationary1 methyl phases, 4-hydroxybenzoate/methyl this effect is less phenylacetate visible. This phenomenon can be explained in a similar way as that for the ethyl nitrophenol/phenol 4-hydroxybenzoate/ethyl selectivity phenylacetate changes discussed above.

1 methyl 4-hydroxybenzoate/methyl phenylacetate  ethyl 4-hydroxybenzoate/ethyl phenylacetate -- SB C3 T C3 SB SB C8 T C8 SB SB C3 M SB C8 M SB C18 SB T RX C18 T SB C18 M RX C18 M XDB C8 M XDB C8 T XDB C8 XDB C18 T C18 XDB XDB C18 M 300 SB C3 T 300 SB C8 T Extent C18 T Extent 300 SB C3 M 300 SB C8 M Extent C18 M 300 SB C18 T 300 SB C18 M

FigureFigure 6. 6.Separation Separation factor, factor,α α, values, values of of esters esters in in systems systems with with di ffdifferenterent nonpolar nonpolar-- stationary stationary phases phases and methanoland methanol (M) in (M) water in water and tetrahydrofuran and tetrahydrofuran (T) in (T) water in water (modifier (modifier concentrations concentrations as in as Table in Table1). 1). SB C3 T C3 SB SB C8 T C8 SB SB C3 M SB C8 M SB C18 SB T RX C18 T SB C18 M RX C18 M XDB C8 M XDB C8 T XDB C8 XDB C18 T C18 XDB XDB C18 M 300 SB C3 T 300 SB C8 T Extent C18 T Extent 300 SB C3 M 300 SB C8 M Extent C18 M 300 SB C18 T AAnother similar einterestingffect, however, effect more of separation pronounced,300 SB C18 M selectivity is observed changes for the on correlationthe mobile of phase THF modifier versus ACN. is

Thisobserved effect isfor present the forsubstance systems pairs with alldimethyl tested stationary isophthalate/4 phases‐cyanobenzaldehyde except C3, while for systemsand dimethyl with all Figure 6. Separation factor, α, values of esters in systems with different nonpolar stationary phases C8isophthalate/4 and 300 SB C18‐nitrobenzaldehyde. stationary phases, As this can e ffbeect seen is less in visible.Figure 7, This separation phenomenon factor can values be explained are much in and methanol (M) in water and tetrahydrofuran (T) in water (modifier concentrations as in Table 1). ahigher similar in way the assystems that for with the MeOH nitrophenol relative/phenol to THF. selectivity A similar changes effect was discussed also observed above. for THF versus ACN systems, but the differences were much smaller. AnotherAnother interestinginteresting eeffectffect of separation separation selectivity selectivity changes changes on on the the mobile mobile phase phase modifier modifier is isobserved observed for for the the substance substance pairs pairs dimethyl dimethyl isophthalate/4 isophthalate/4-cyanobenzaldehyde‐cyanobenzaldehyde and and dimethyl dimethyl isophthalateisophthalate/4/4-nitrobenzaldehyde.‐nitrobenzaldehyde. As As can can be be seen seen in Figurein Figure7, separation 7, separation factor factor values values are much are much higher inhigher the systems in the systems with MeOH with MeOH relative dimethyl relativeto isophtalate/4-cyanobenzaldehyde THF. to A THF. similar A similar effect waseffect also was observed also observed for THF for THF versus versus ACN dimethyl isophtalate/4-nitrobenzaldehyde systems,ACN systems, but the but diff theerences differences were much were smaller.much smaller. 14

dimethyl isophtalate/4-cyanobenzaldehyde  dimethyl isophtalate/4-nitrobenzaldehyde 7 14

0 7 -- SB C3 T SB C8 T SB C3 M SB C8 M SB C18 T RX C18 T SB C18 M C18 SB RX C18 M XDB C8 M XDB C8 T XDBC8 XDB C18 T XDB C18 M XDBC18 300 SB C3 T 300 SB C8 T Extent C18 T 300 SB C3 SB M 300 C8 SB M 300

0 Extent C18 M 300 SB C18 T 300 SB C18 M Figure 7. Separation factor, α, values of hydrocarbons with different functional groups in systems -- with different nonpolar stationary phases and methanol (M) in water and tetrahydrofuran (T) in water SB C3 T SB C8 T SB C3 M SB C8 M SB C18 T RX C18 T SB C18 M C18 SB RX C18 M XDB C8 M (modifier concentrations as in Table 1). T XDBC8 XDB C18 T XDB C18 M XDBC18 300 SB C3 T 300 SB C8 T Extent C18 T 300 SB C3 SB M 300 C8 SB M 300 Extent C18 M 300 SB C18 T 300 SB C18 M FigureFigure 7. 7.Separation Separation factor,factor, αα,, valuesvalues of hydrocarbons with with different diﬀerent functional functional groups groups in in systems systems

withwith di differentﬀerent nonpolar nonpolar stationary stationary phases phases andand methanol (M) in in water water and and tetrahydrofuran tetrahydrofuran (T) (T) in in water water (modiﬁer(modifier concentrations concentrations as as in in Table Table1 ).1).

Molecules 2020, 25, 5070 10 of 16

It is worth underlining that the ability of a modiﬁer to increase the ordering of the hydrocarbon stationary phase rises in the following order: MeOH, ACN, THF [49]. For this reason, in the system with MeOH, the stationary phase is more accessible for entropic penetration by branched molecules relative to the system with THF. A stationary phase in the system with THF is more ordered than in that with ACN and, especially, the one containing MeOH [49]. Molecules with a branched structure (e.g., dimethyl isophthalate) exhibit lower entropic penetration of this stationary phase region relative to molecules of planar structure (4-cyanobenzaldehyde, 4-nitrobenzaldehyde).

2.3. Comparison of Substance Retention and Selectivity in MeOH and ACN Systems with Different Adsorbents Adsorption of acetonitrile molecules on chemically bonded phases is caused by their interaction with bonded ligands and dipole–dipole interactions, while their interactions with residual silanols by hydrogen bonding are weak. Methanol may be adsorbed near the silica surface by hydrogen bonding and dipole–dipole interaction with residual accessible silanols [40]. Therefore, the low-coverage-density phase is better solvated by methanol molecules than by acetonitrile [52]. The C18 stationary phase in the system with ACN is more ordered than with MeOH [53]. However, changes in separation selectivity between these modifier systems regarding the shape of the molecules are much smaller than between the systems of these modifiers relative to THF. The solutes, as in the ACN system versus the THF system, form two separate lines, one for substances with groups of proton-donor and electron-pair-donor (proton-acceptor) characters and the second for compounds with groups of electron-pair donors. However, contrary to the compared ACN and THF systems, the correlation line for analytes that can act only as a proton acceptor are placed above the first line, which means that these compounds show higher retention in relation to other substances in systems with acetonitrile (see Figures S21–S30). This is due to the stronger intermolecular interaction of ACN with these substances in the stationary phase region compared with MeOH [46]. On the other hand, substances that possess groups with both proton-donor and proton-acceptor properties show relatively stronger interactions with methanol. The influence of the number of –NO2 groups on separation selectivity changes, depending on the organic modifier used, can be observed. In systems with methanol, especially in the case of C18 and C8 columns, the order of elution is consistent with the solvophobic theory [50]: the larger the number of -NO2 groups a molecule has, the less it is retained. However, in a system with ACN, all nitro derivatives exhibit similar retention. This phenomenon is probably caused by stronger dipole–dipole interactions of the discussed compounds with ACN relative to the MeOH system. An increase in the relative retention of phenols with an ester group in relation to phenols in systems with methanol compared with systems with acetonitrile stands as another example of changes in separation selectivity in respect of the shape of the molecules (Figure8). Phenols with an ester group (methyl, ethyl, and propyl 4-hydroxybenzoate) have a more branched molecular shape than monophenols (phenol, o-cresol), and their entropic penetration of the stationary phase region in a system with ACN is more restricted than with MeOH. Methanol orders organic ligands to a lesser extent; therefore, in systems with this modifier, molecules with branched shapes have a greater possibility of entropic penetration, and their relative retention is higher [49]. An interesting effect can be seen when comparing the correlation plots for ACN versus MeOH and THF versus MeOH. There is a decrease in the slope of the correlation lines with the shortening of the hydrocarbon chain length and with an increase in the pore diameter. This phenomenon is especially visible for the systems with columns of SB type. For the THF system versus MeOH system, the slope of the correlation lines is smaller than for ACN versus MeOH, and for both, it is much less than one. This reduction in the slope means that the range of retention changes of the tested substances in the THF and ACN systems is smaller compared with the MeOH system. It is probably caused by the reduction of intermolecular interactions of acetonitrile and tetrahydrofuran with the substances in Molecules 2020, 25, 5070 11 of 16 the stationary phase region (the shorter hydrocarbon chains, then the less modifier in the stationary phase region) [40]. In the case of systems with methanol, the reduction of its content in the area of the stationary phase is not essential because, as already mentioned, apart from solvation of carbon chains, itMolecules can also 2020 form, 25, hydrogenx bonds with residual silanol groups [34]. 12 of 17

ethyl 4-hydroxybenzoate/phenol methyl 4-hydroxybenzoate/phenol 15 propyl 4-hydroxybenzoate/p-cresole ethyl 4-hydroxybenzoate/p-cresole 

0 -- SB C3A SB SB C8A SB SB C3M SB SB C8M SB SB C18 A RX C18 A SB C18 M RX C18 M XDB C8 M C8 XDB XDB C8 A XDB C18 A XDB C18 M 300 SB300 C3 A SB300 C8 A 300 SB300 C3 M SB300 C8 M 300 SB C18 A 300 SB C18 M ExAenA C18 A ExAenA C18 M FigureFigure 8. 8.Separation Separation factor, factor,α α, values, values of of hydrocarbons hydrocarbons with with di ffdifferenterent functional functional groups groups in systems in systems with diwithfferent different nonpolar nonpolar stationary stationary phases phases and methanol and methanol (M) in water(M) in and water acetonitrile and acetonitrile (A) in water (A) in (modifier water concentrations(modifier concentrations as in Table 1as). in Table 1).

3. MaterialsAn interesting and Methods effect can be seen when comparing the correlation plots for ACN versus MeOH andSolutes THF versus (Table MeOH.2, Table S1)There were is a obtained decrease from in the different slope of sources. the correlation Benzene (lines99.9%), with toluene the shortening ( 99.8%), ≥ ≥ phenolof the (hydrocarbon99%), o-cresol chain ( 99%), lengthp-cresol and with ( 99%), an increase 2-naphthol in the ( pore99%), diameter. acetophenone ( 99%), nitrobenzene This≥ phenomenon≥ is especially visible≥ for the systems≥ with columns of SB ≥type. For the THF ( 99%), 4-cyanobenzaldehyde ( 98%), benzonitrile ( 99%), 1,5-dihydroxynaphthalene ( 97%), ≥system versus MeOH system, the≥ slope of the correlation ≥lines is smaller than for ACN versus MeOH,≥ 1,6-dihydroxynaphthalene ( 99%), 2-cyanophenol ( 99%), dimethyl-4,40-diphenyl dicarboxylate ( 99%), and for both, it is much less≥ than one. ≥ ≥ ethyl 4-hydroxybenzoate ( 99%), methyl 4-hydroxybenzoate ( 99%), propyl 4-hydroxybenzoate ( 99%), This reduction in the≥ slope means that the range of retention≥ changes of the tested substances≥ in dimethyl isophthalate ( 99%), 4-nitrobenzaldehyde ( 98%), 4-nitrobenzyl alcohol ( 99%), 2-nitrophenol the THF and ACN systems≥ is smaller compared with≥ the MeOH system. It is probably≥ caused by the ( 98%), 3-nitrophenol ( 99%), 2-nitro-4-chlorophenol ( 97%), 2-methylo-4-nitrophenol ( 97%), ≥reduction of intermolecular≥ interactions of acetonitrile and≥ tetrahydrofuran with the substances≥ in 4-nitrophenol ( 99%), 1-chloro-2,4-dinitrobenzene ( 97%), 1,2-dinitrobenzene ( 99%), 1,4-dinitrobenzene the stationary≥ phase region (the shorter hydrocarbon≥ chains, then the less modifier≥ in the stationary ( 98%), methyl phenylacetate ( 98%), ethyl phenylacetate ( 98%), and 1,3,5-trinitrobenzene (analytical ≥phase region) [40]. In the case ≥of systems with methanol, the≥ reduction of its content in the area of the standard) were supplied by Sigma-Aldrich (St. Louis, MO, USA). 1,7-Dihydroxynaphthalene ( 97%) stationary phase is not essential because, as already mentioned, apart from solvation of carbon chains,≥ and diethyl terephthalate ( 95%) were supplied by Alfa Aesar (Heysham, Lancashire, UK), and methyl it can also form hydrogen≥ bonds with residual silanol groups [34]. benzoate (analytical standard) was supplied by Fluka (Buchs, Switzerland). All solvents (methanol, acetonitrile,3. Materials and and tetrahydrofuran), Methods HPLC grade, were purchased from Merck (Darmstadt, Germany). Water was bidistilled. Eluents contained 0.1% acetic acid (analytical grade, Merck). Solutes (Table 2, Table S1) were obtained from different sources. Benzene (≥99.9%), toluene(≥99.8%), phenol (≥99%), o‐cresol (≥99%), p‐cresol (≥99%), 2‐naphthol (≥99%), acetophenone (≥99%), nitrobenzene (≥99%), 4‐cyanobenzaldehyde (≥98%), benzonitrile (≥99%), 1,5‐ dihydroxynaphthalene (≥97%), 1,6‐dihydroxynaphthalene (≥99%), 2‐cyanophenol (≥99%), dimethyl‐ 4,4′‐diphenyl dicarboxylate (≥99%), ethyl 4‐hydroxybenzoate (≥99%), methyl 4‐hydroxybenzoate (≥99%), propyl 4‐hydroxybenzoate (≥99%), dimethyl isophthalate (≥99%), 4‐nitrobenzaldehyde (≥98%), 4‐nitrobenzyl alcohol (≥99%), 2‐nitrophenol (≥98%), 3‐nitrophenol (≥99%), 2‐nitro‐4‐ chlorophenol (≥97%), 2‐methylo‐4‐nitrophenol (≥97%), 4‐nitrophenol (≥99%), 1‐chloro‐2,4‐ dinitrobenzene (≥97%), 1,2‐dinitrobenzene (≥99%), 1,4‐dinitrobenzene (≥98%), methyl phenylacetate (≥98%), ethyl phenylacetate (≥98%), and 1,3,5‐trinitrobenzene (analytical standard) were supplied by Sigma‐Aldrich (St. Louis, MO, USA). 1,7‐Dihydroxynaphthalene (≥97%) and diethyl terephthalate (≥95%) were supplied by Alfa Aesar (Heysham, Lancashire, UK), and methyl benzoate (analytical standard) was supplied by Fluka (Buchs, Switzerland). All solvents (methanol, acetonitrile, and tetrahydrofuran), HPLC grade, were purchased from Merck (Darmstadt, Germany). Water was bidistilled. Eluents contained 0.1% acetic acid (analytical grade, Merck).

Molecules 2020, 25, 5070 12 of 16

Table 2. List of investigated solutes and their descriptors [54].

A , Hydrogen B , Hydrogen A , Hydrogen B , Hydrogen Substance H H Substance H H Bond Acidity * Bond Basicity * Bond Acidity * Bond Basicity * 1. Benzene 0 0.14 18. 3-Nitrophenol 0.79 0.23 2. Toluene 0 0.14 19. 4-Nitrophenol 0.82 0.26 3. Phenol 0.6 0.3 20. 2-Methyl-4-nitrophenol 0.78 0.25 4. 2-Cresol 0 0.14 21. Methyl 4-hydroxybenzoate 0.69 0.45 5. 4-Cresol 0.52 0.3 22. Ethyl 4-hydroxybenzoate 0.69 0.45 6. 2-Naphthol 0.57 0.31 23. Propyl 4-hydroxybenzoate 0.69 0.45 7. Methyl phenylacetate 0.61 0.4 24. 4-Nitrobenzyl alcohol 0.44 0.62 8. Ethyl phenylacetate 0 0.58 25. 1,2-Dinitrobenzene 0 0.38 9. Methyl benzoate 0 0.57 26. 1,4-Dinitrobenzene 0 0.46 10. Acetophenone 0 0.46 27. 1-Chloro-2,4-dinitrobenzene 0 0.42 11. Nitrobenzene 0 0.28 28. 4-Nitrobenzaldehyde 0 0.44 12. Benzonitrile 0 0.33 29. 4-Cyanobenzaldehyde 0 0.33 13. 1,5-Dihydroxynaphthalene 0.93 0.58 30. Dimethyl isophthalate 0 0.67 14. 1,6-Dihydroxynaphthalene 0.98 0.57 31. Diethyl terephthalate 32. Dimethyl 15. 1,7-Dihydroxynaphthalene 1.02 0.54 0 0.91 4,4-diphenylcarboxylate 16. 2-Cyanophenol 0.78 0.34 33. 2-Nitro-4-chlorophenol 0.1 0.3 17. 2-Nitrophenol 0.05 0.37 34. 1,3,5-Trinitrobenzene 0 0.6 * Experimentally determined, from various sources.

Measurements of retention were performed with the 1290 Agilent Inﬁnity LC System (Santa Clara, CA, USA) equipped with a DAD detector. The C3, C8, and C18 columns were used in the study. The properties of the stationary phases are listed in Table3. The columns were thermostated at 20 ◦C. The ﬂow rate was set at 0.5 mL/min, the injection volume was 10 µL, and the eluate was monitored at 254 nm. All experiments were performed in triplicate.

Table 3. Characterization of used columns.

Dimensions Particle Pore Surface Area Carbon Coverage Density No. Column Endcapped (mm) Diameter (µm) Diameter (Å) (m2/g) Load (%) (µmol/m2) 1 Zorbax SB C18 3 150 5 80 180 no 10 2.98 × 2 Zorbax Rx C18 3 150 5 80 180 no 12 2.98 × Zorbax Eclipse 3 3 150 5 80 180 yes 10 4 XDB C18 × Zorbax Extend 4 3 150 5 80 180 yes 12.5 3.8 C18 × 5 Zorbax SB C8 3 150 5 80 180 no 5.5 2.4 × Zorbax Eclipse 6 3 150 5 80 180 yes 7.6 3.8 XDB C8 × 7 ProntoSIL C4 3 125 5 120 300 yes 5 - × 8 Zorbax SB C3 3 150 5 80 180 no 4 - × Zorbax 300SB 9 3 150 5 300 45 no 2.8 - C18 × Zorbax 300SB 10 3 150 5 300 45 no 1.5 - C8 × Zorbax 300SB 11 3 150 5 300 45 no 1.1 - C3 ×

4. Conclusions Retention correlation and comparison of aromatic hydrocarbons with a polar group in methanol, acetonitrile, and tetrahydrofuran systems demonstrate differences in separation selectivity between these systems. Moreover these differences are closely related to the stationary phase used. The results presented in the paper confirm our previous presumption that the explanation of separation selectivity changes in reversed-phase systems, when one modifier is replaced by another in the mobile phase, can be performed, taking into account molecular interactions of the solutes with the stationary phase Molecules 2020, 25, 5070 13 of 16 components, especially mobile phase modifier (organic solvent) extracted into the stationary phase. Moreover, these results point out that the previously described approach can also be applied to systems with other nonpolar adsorbents of the C3 and C8 types, keeping in mind that the adsorbed amount of organic modifiers strongly depends on the chain length of the stationary phase. Additionally, one has to take into account that the organic modifiers interact with free silanol groups at different extents.

Supplementary Materials: The following are available online, Figure S1. Log k in 32% v/v THF plotted against log k in 35% v/v ACN. SB C18 stationary phase. Solute numbers as in Table2. Figure S2. Log k in 32% v/v THF plotted against log k in 34% v/v ACN. Rx C18 stationary phase. Solute numbers as in Table2. Figure S3. Log k in 31% v/v THF plotted against log k in 34% v/v ACN. Eclipse XDB C18 stationary phase. Solute numbers as in Table2. Figure S4. Log k in 31% v/v THF plotted against log k in 35% v/v ACN. Extend C18 stationary phase. Solute numbers as in Table2. Figure S5. Log k in 33% v/v THF plotted against log k in 35% v/v ACN. SB C8 stationary phase. Solute numbers as in Table2. Figure S6. Log k in 32% v/v THF plotted against log k in 36% v/v ACN. Eclipse XDB C8 stationary phase. Solute numbers as in Table2. Figure S7. Log k in 33% v/v THF plotted against log k in 33% v/v ACN. SB C3 stationary phase. Solute numbers as in Table2. Figure S8. Log k in 35% v/v THF plotted against log k in 34% v/v ACN. 300 SB C18 stationary phase. Solute numbers as in Table2. Figure S9. Log k in 35% v/v THF plotted against log k in 36% v/v ACN. 300 SB C8 stationary phase. Solute numbers as in Table2. Figure S10. Log k in 34% v/v THF plotted against log k in 32% v/v ACN. 300 SB C3 stationary phase. Solute numbers as in Table2. Figure S11. Log k in 32% v/v THF plotted against log k in 44% v/v MeOH. SB C18 stationary phase. Solute numbers as in Table2. Figure S12. Log k in 32% v/v THF plotted against log k in 45% v/v MeOH. Rx C18 stationary phase. Solute numbers as in Table2. Figure S13. Log k in 31% v/v THF plotted against log k in 43% v/v MeOH. Eclipse XDB C18 stationary phase. Solute numbers as in Table2. Figure S14. Log k in 31% v/v THF plotted against log k in 45% v/v MeOH. Extend C18 stationary phase. Solute numbers as in Table2. Figure S15. Log k in 33% v/v THF plotted against log k in 42% v/v MeOH. SB C8 stationary phase. Solute numbers as in Table2. Figure S16. Log k in 32% v/v THF plotted against log k in 45% v/v MeOH. XDB C8 stationary phase. Solute numbers as in Table2. Figure S17. Log k in 33% v/v THF plotted against log k in 25% v/v MeOH. SB C3 stationary phase. Solute numbers as in Table2. Figure S18. Log k in 35% v/v THF plotted against log k in 44% v/v MeOH. 300 SB C18 stationary phase. Solute numbers as in Table2. Figure S19. Log k in 35% v/v THF plotted against log k in 41% v/v MeOH. 300 SB C8 stationary phase. Solute numbers as in Table2. Figure S20. Log k in 33% v/v ACN plotted against log k in 26% v/v MeOH. 300 SB C3 stationary phase. Solute numbers as in Table2. Figure S21. Log k in 35% v/v ACN plotted against log k in 44% v/v MeOH. SB C18 stationary phase. Solute numbers as in Table2. Figure S22. Log k in 34% v/v ACN plotted against log k in 45% v/v MeOH. Rx C18 stationary phase. Solute numbers as in Table2. Figure S23. Log k in 34% v/v ACN plotted against log k in 43% v/v MeOH. Eclipse XDB C18 stationary phase. Solute numbers as in Table2. Figure S24. Log k in 35% v/v ACN plotted against log k in 45% v/v MeOH. Extend C18 stationary phase. Solute numbers as in Table2. Figure S25. Log k in 35% v/v ACN plotted against log k in 42% v/v MeOH. SB C8 stationary phase. Solute numbers as in Table2. Figure S26. Log k in 36% v/v ACN plotted against log k in 45% v/v MeOH. Eclipse XDB C8 stationary phase. Solute numbers as in Table2. Figure S27. Log k in 33% v/v ACN plotted against log k in 25% v/v MeOH. SB C3 stationary phase. Solute numbers as in Table2. Figure S28. Log k in 34% v/v ACN plotted against log k in 44% v/v MeOH. 300 SB C18 stationary phase. Solute numbers as in Table2. Figure S29. Log k in 36% v/v ACN plotted against log k in 41% v/v MeOH. 300 SB C8 stationary phase. Solute numbers as in Table2. Figure S30. Log k in 32% v/v ACN plotted against log k in 26% v/v MeOH. 300 SB C3 stationary phase. Solute numbers as in Table2. Table S1. Structural formulas of substances used in experiments. Author Contributions: Conceptualization, A.K.-T. and T.H.D.; methodology, A.K.-T. and T.H.D.; investigation, B.M. and A.K.-T.; data curation, A.K.-T. and B.M.; writing—original draft preparation, A.K.-T.; writing—review and editing, T.H.D.; visualization, A.K.-T.; supervision, A.K.-T. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Youdim, K.A.; Saunders, K.C. A review of LC-MS techniques and high-throughput approaches used to investigate drug metabolism by cytochrome P450s. J. Chromatogr. B 2010, 878, 1326–1336. [CrossRef] [PubMed] 2. Lindon, J.C.; Nicholson, J.K.; Wilson, I.D. Directly coupled HPLC-NMR and HPLC-NMR-MS in pharmaceutical research and development. J. Chromatogr. B 2000, 748, 233–258. [CrossRef] 3. Zhou, L.Z. Handbook of Pharmaceutical Analysis by HPLC; Elsevier: Amsterdam, The Netherlands, 2005; ISBN 9780120885473. Molecules 2020, 25, 5070 14 of 16

4. Gao, W.; Legido-Quigley, C. Fast and sensitive high performance liquid chromatography analysis of cosmetic creams for hydroquinone, phenol and six preservatives. J. Chromatogr. A 2011, 1218, 4307–4311. [CrossRef] [PubMed] 5. Zgoła-Grze´skowiak,A.; Werner, J.; Jeszka-Skowron, M.; Czarczyńska-Go´slińska,B. Determination of parabens in cosmetic products using high performance liquid chromatography with fluorescence detection. Anal. Methods 2016, 8, 3903–3909. [CrossRef] 6. Zhang, Q.; Lian, M.; Liu, L.; Cui, H. High-performance liquid chromatographic assay of parabens in wash-off cosmetic products and foods using chemiluminescence detection. Anal. Chim. Acta 2005, 537, 31–39. [CrossRef] 7. Stachniuk, A.; Szmagara, A.; Czeczko, R.; Fornal, E. LC-MS/MS determination of pesticide residues in fruits and vegetables. J. Environ. Sci. Health. Part B 2017, 52, 446–457. [CrossRef] 8. Hefni, M.; McEntyre, C.; Lever, M.; Slow, S. Validation of HPLC-UV methods for the quantification of betaine in foods by comparison with LC-MS. Food Anal. Methods 2016, 9, 292–299. [CrossRef] 9. Núñez, O.; Gallart-Ayala, H.; Martins, C.P.B.; Lucci, P. New trends in fast liquid chromatography for food and environmental analysis. J. Chromatogr. A 2012, 1228, 298–323. [CrossRef] 10. Joutovsky, A.; Hadzi-Nesic, J.; Nardi, M.A. HPLC retention time as a diagnostic tool for hemoglobin variants and hemoglobinopathies: A study of 60 000 samples in a clinical diagnostic laboratory. Clin. Chem. 2004, 50, 1436–1447. [CrossRef] 11. Wright, M.; Hepburn, S. Positive Impacts of HPLC Innovations on Clinical Diagnostic Analysis. LC GC Europe 2016, 29, 34–38. 12. Citti, C.; Braghiroli, D.; Vandelli, M.A.; Cannazza, G. Pharmaceutical and biomedical analysis of cannabinoids: A critical review. J. Pharm. Biomed. Anal. 2018, 147, 565–579. [CrossRef][PubMed] 13. Zhao, J.; Liu, F.; Wang, G.; Cao, T.; Guo, Z.; Zhang, Y. High performance liquid chromatography determination of formaldehyde in engine exhaust with unsymmetrical dimethylhydrazine as a new derivatization agent. Anal. Methods 2015, 7, 309–312. [CrossRef] 14. Siouffi, A.M. High performance liquid chromatography. In Handbook of Food Science, Technology, and Engineering-4 Volume Set; CRC Press: Boca Raton, FL, USA, 2005; ISBN 9781466507876. 15. Snyder, L.R. Changing reversed-phase high performance liquid chromatography selectivity Which variables should be tried first? J. Chromatogr. B Biomed. Sci. Appl. 1997, 689, 105–115. [CrossRef] 16. Tan, L.C.; Carr, P.W. Study of retention in reversed-phase liquid chromatography using linear solvation energy relationships II. The mobile phase. J. Chromatogr. A 1998, 752, 1–18. [CrossRef] 17. Rosés, M.; Subirats, X.; Bosch, E. Retention models for ionizable compounds in reversed-phase liquid chromatography. Effect of variation of mobile phase composition and temperature. J. Chromatogr. A 2009, 1216, 1756–1775. [CrossRef] 18. Ramis-Ramos, G.; García-Álvarez-Coque, M.C. Solvent Selection in Liquid Chromatography. In Liquid Chromatography: Fundamentals and Instrumentation; Elsevier: Amsterdam, The Netherlands, 2013; ISBN 9780124158078. 19. Tanaka, N.; Goodell, H.; Karger, B.L. The role of organic modifiers on polar group selectivity in reversed-phase liquid chromatography. J. Chromatogr. A 1978, 158, 233–248. [CrossRef] 20. Colin, H.; Guiochon, G. Introduction to reversed-phase high-performance liquid chromatography. J. Chromatogr. A 1977, 141, 289–312. [CrossRef] 21. Snyder, L.R.; Dolan, J.W.; Carr, P.W. The hydrophobic-subtraction model of reversed-phase column selectivity. J. Chromatogr. A 2004, 1060, 77–116. [CrossRef] 22. Vailaya, A.; Horváth, C. Retention in reversed-phase chromatography: Partition or adsorption? J. Chromatogr. A 1998, 829, 1–27. [CrossRef] 23. Žuvela, P.; Skoczylas, M.; Liu, J.J.; B ˛aczek,T.; Kaliszan, R.; Wong, M.W.; Buszewski, B. Column Characterization and Selection Systems in Reversed-Phase High-Performance Liquid Chromatography. Chem. Rev. 2019, 119, 3674–3729. [CrossRef] 24. Jaroniec, M. Partition and displacement models in reversed-phase liquid chromatography with mixed eluents. J. Chromatogr. A 1993, 656, 37–50. [CrossRef] 25. Sander, L.C.; Wise, S.A. Shape selectivity in reversed-phase liquid chromatography for the separation of planar and non-planar solutes. J. Chromatogr. A 1993, 656, 335–351. [CrossRef] Molecules 2020, 25, 5070 15 of 16

26. Tanaka, N.; Kimata, K.; Hosoya, K.; Miyanishi, H.; Araki, T. Stationary phase effects in reversed-phase liquid chromatography. J. Chromatogr. A 1993, 656, 265–287. [CrossRef] 27. Wheeler, J.F.; Beck, T.L.; Klatte, S.J.; Cole, L.A.; Dorsey, J.G. Phase transitions of reversed-phase stationary phases. J. Chromatogr. A 1993, 656, 317–333. [CrossRef] 28. Horváth, C.; Melander, W.; Molnár, I. Solvophobic interactions in liquid chromatography with nonpolar stationary phases. J. Chromatogr. A 1976, 125, 129–156. [CrossRef] 29. Dill, K.A. The mechanism of solute retention in reversed-phase liquid chromatography. J. Phys. Chem. 1987, 91, 1980–1988. [CrossRef] 30. Dorsey, J.G.; Dill, K.A. The Molecular Mechanism of Retention in Reversed-Phase Liquid Chromatography. Chem. Rev. 1989, 89, 331–346. [CrossRef] 31. Nasuto, R.; Kwietniewski, L.; Ró´zyło,J.K. Relationship between stationary and mobile phase composition and its influence on retention factors of aromatic hydrocarbons in reversed-phase high-performance liquid chromatography. J. Chromatogr. A 1997, 762, 27–33. [CrossRef] 32. Bocian, S. Solvation processes in liquid chromatography: The importance and measurements. J. Liq. Chromatogr. Relat. Technol. 2016, 39, 731–738. [CrossRef] 33. Gritti, F.; Kazakevich, Y.V.; Guiochon, G. Effect of the surface coverage of endcapped C18-silica on the excess adsorption isotherms of commonly used organic solvents from water in reversed phase liquid chromatography. J. Chromatogr. A 2007, 1169, 111–124. [CrossRef][PubMed] 34. Bocian, S.; Vajda, P.; Felinger, A.; Buszewski, B. Excess adsorption of commonly used organic solvents from water on nonend-capped C18-bonded phases in reversed-phase liquid chromatography. Anal. Chem. 2009, 81, 6334–6346. [CrossRef] 35. Gritti, F.; Guiochon, G. Thermodynamics of adsorption of binary aqueous organic liquid mixtures on a RPLC adsorbent. J. Chromatogr. A 2007, 1155, 85–99. [CrossRef] 36. Poole, C.F.; Poole, S.K. Column selectivity from the perspective of the solvation parameter model. J. Chromatogr. A 2002, 965, 263–299. [CrossRef] 37. Rafferty, J.L.; Siepmann, J.I.; Schure, M.R. Mobile phase effects in reversed-phase liquid chromatography: A comparison of acetonitrile/water and methanol/water solvents as studied by molecular simulation. J. Chromatogr. A 2011, 1218, 2203–2213. [CrossRef] 38. Zapala, W. Influence of mobile phase composition on retention factors in different HPLC systems with chemically bonded stationary phases. J. Chromatogr. Sci. 2003, 41, 289–294. [CrossRef][PubMed] 39. Dzido, T.H.; Kossowski, T.E.; Matosiuk, D. Comparison of retention of aromatic hydrocarbons with polar groups in binary reversed-phase high-performance liquid chromatography systems. J. Chromatogr. A 2002, 947, 167–183. [CrossRef] 40. Bocian, S.; Felinger, A.; Buszewski, B. Comparison of solvent adsorption on chemically bonded stationary phases in RP-LC. Chromatographia 2008, 68, 19–26. [CrossRef] 41. Dzido, T.H. Modifier influence on selectivity of reversed-phase hplc systems. J. Liq. Chrom. Relat. Technol. 2000, 23, 2773–2788. [CrossRef] 42. Klimek-Turek, A.; Misiołek, B.; Dzido, T.H. Comparison of the retention of aliphatic hydrocarbons with polar groups in RP-HPLC systems with different modifiers of the binary eluent. Chromatographia 2013, 76, 939–947. [CrossRef] 43. Klimek-Turek, A.; Jaglínska, K.; Imbierowicz, M.; Dzido, T.H. Solvent front position extraction with semi-automatic device as a powerful sample preparation procedure prior to quantitative instrumental analysis. Molecules 2019, 24, 1358. [CrossRef] 44. Moldoveanu, S.; David, V. Estimation of the phase ratio in reversed-phase high-performance liquid chromatography. J. Chromatogr. A 2015, 1381, 194–201. [CrossRef][PubMed] 45. Marchetti, N.; Guzzinati, R.; Catani, M.; Massi, A.; Pasti, L.; Cavazzini, A. New insights into perfluorinated adsorbents for analytical and bioanalytical applications. Anal. Bioanal. Chem. 2015, 407, 17–21. [CrossRef] [PubMed] 46. De Juan, A.; Fonrodona, G.; Casassas, E. Solvent classification based on solvatochromic parameters: A comparison with the Snyder approach. Trac Trends Anal. Chem. 1997, 16, 52–62. [CrossRef] 47. Karapetian, Y.A. Physical Chemistry Properties of Electrolytes in Non-Aqueous Solutions; Khimia: Moscow, Russia, 1989. Molecules 2020, 25, 5070 16 of 16

48. Hałka-Grysińska,A.; Skop, K.; Klimek-Turek, A.; Gorzkowska, M.; Dzido, T.H. Thin-layer chromatogram development with a moving pipette delivering the mobile phase onto the surface of the adsorbent layer. J. Chromatogr. A 2018, 9, 91–99. [CrossRef] 49. Cole, S.R.; Dorsey, J.G. Effect of stationary phase solvation on shape selectivity in reversed-phase high-performance liquid chromatography. J. Chromatogr. A 1993, 635, 177–186. [CrossRef] 50. Zhang, L.; Rafferty, J.L.; Siepmann, J.I.; Chen, B.; Schure, M.R. Chain conformation and solvent partitioning in reversed-phase liquid chromatography: Monte Carlo simulations for various water/methanol concentrations. J. Chromatogr. A 2006, 1126, 219–231. [CrossRef] 51. Heron, S.; Tchapla, A. Description of retention mechanism by solvophobic theory. Influence of organic modifiers on the retention behaviour of homologous series in reversed-phase liquid chromatography. J. Chromatogr. A 1991, 556, 219–234. [CrossRef] 52. Bocian, S.; Soukup, J.; Jandera, P.; Buszewski, B. Thermodynamics Study of Solvent Adsorption on Octadecyl-Modified Silica. Chromatographia 2014, 78, 21–30. [CrossRef] 53. Claustrat, B.; Brun, J.; Chazot, G. The basic physiology and pathophysiology of melatonin. Sleep Med. Rev. 2005, 9, 11–24. [CrossRef] 54. Abraham, M.H.; Rosés, M. Hydrogen bonding. 38. Effect of solute structure and mobile phase composition on reversed-phase high-performance liquid chromatographic capacity factors. J. Phys. Org. Chem. 1994, 7, 672–684. [CrossRef]

Sample Availability: Samples of the compounds are not available from the authors.

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional aﬃliations.

© 2020 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 (http://creativecommons.org/licenses/by/4.0/).