3435
J. Sep. Sci. 2014, 37, 3435–3442
Szymon Bocian Bogusław Buszewski Faculty of Chemistry, Department of Environmental Chemistry & Bioanalytics, Nicolaus Copernicus University, Torun, Poland Received July 17, 2014 Revised August 18, 2014 Accepted September 9, 2014
Research Article
Phenyl-bonded stationary phases—The influence of polar functional groups on retention and selectivity in reversed-phase liquid chromatography The chromatographic properties of four phenyl-bonded phases with different structures were studied. The columns used were packed with a stationary phase containing a phenyl ring attached to the silica surface using different types of linkage molecules. As a basic characteristic of the bonded phases, the hydrophobicity and silanol activity (polarity) were investigated. The presence of the polar amino and amide groups in the structure of the bonded ligand strongly influences the polarity of the bonded phase. Columns were compared according to methylene selectivity using a series of benzene homologues and according to their shape and size selectivity using polycyclic aromatic hydrocarbons. The measurements were done using methanol/water and acetonitrile/water mobile phases. The presented results show that the presence of polar functional groups in the ligand structure strongly influences the chromatographic properties of the bonded phase. Keywords: Liquid chromatography / Phenyl-bonded phases / Polar functional groups DOI 10.1002/jssc.201400764
1 Introduction Phenyl-bonded stationary phases are unique separation materials that are gaining popularity as packing materials in HPLC analyses [1–3]. Phenyl-bonded stationary phases have been available for many years [4, 5]. The idea of this type of material is to promote the – interaction of the aromatic rings with a solute as an additional force, which modifies the RP retention mechanism. The presence of phenyl-bonded ligands also changes the solvation process and offers different selectivity of the chromatographic system in comparison with octadecyl (C18 ) stationary phases [6–8]. This special selectivity will be observed when analyzed compounds possess aromatic functionalities in their structure, which enable – interactions. As a result of these interactions, solutes with -electrons will display a different retention behavior on phenyl columns than on alkyl RP phases, while solutes without -electrons or with sterically hindered -electron systems will be retained by the classical C8 /C18 RP-like mode [3]. The methylene and aromatic selectivity of phenyl-bonded stationary phases depends on the
Correspondence: Professor Bogusław Buszewski, Faculty of Chemistry, Department of Environmental Chemistry & Bioanalytics, Nicolaus Copernicus University, Gagarin 7 St., 87-100 Torun, Poland E-mail: [email protected] Fax: +48-56-611-4837
Abbreviations: CP MAS, cross-polarization magic-angle spinning; PAH, polycyclic aromatic hydrocarbon C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
spacer length, which binds the phenyl ring with the silica surface [9]. The surface coverage of bonded phenyl ligands is also an important parameter that influences the selectivity of the stationary phase [10]. Because of the manifold possibilities to create – active stationary phases, numerous types of – active RP phases were already designed and investigated. Early and moderately – active phases contained mono-aromatic [11] or polyaromatic [12] moieties and stationary phases employing heteroaromatic ring systems [13]. Other types of phenylbonded stationary phases are fluorinated phenyl phases. This material is especially recommended for solutes that possess some halogens in their structure, as many pesticides, for example [14]. Recently, a novel type of aromatic stationary phase was introduced in the market that contains biphenyl ligands and offers selectivity significantly different from alkyl stationary phases. Another type of bonded stationary phase that contains phenyl ligand is a phenyl-hydride material [15, 16]. This type of stationary phase may be applied in both RP and aqueous normal-phase chromatography. Nowadays, polar-endcapped and polar-embedded RP phases with incorporated different polar functional groups are in the center of interest [17]. Due to their ability to undergo hydrogen bonding and other polar interactions with water molecules, such materials are significantly more stable in water-rich mobile phases than the conventional hydrophobic RP adsorbents. Such materials may also be stable in purely aqueous mobile phases [17, 18]. Colour Online: See the article online to view Fig. 2 in colour. www.jss-journal.com
3436
J. Sep. Sci. 2014, 37, 3435–3442
S. Bocian and B. Buszewski
The goal of our work was to synthesize a series of phenyl-bonded stationary phases with different polar functional groups in their structures and compare their chromatographic properties. Typical propyl-phenyl adsorbents were compared with polar-endcapped and polar-embedded phenyl-bonded stationary phases. The influence of the polar group on the retention and selectivity was determined. Methylene selectivity using a series of alkyl benzene derivatives and shape/size selectivity obtained from polycyclic aromatic hydrocarbon (PAH) separations in MeOH/water and ACN/water was compared.
2 Materials and methods 2.1 Instruments The liquid chromatograph was a Shimadzu Prominence system (Tokyo, Japan) equipped with ternary gradient pump (LC20AD), diode array detector (SPD-M20A), an autosampler (SIL-20A), and a column thermostat (CTO-10AS VP). Data were collected using LabSolutions software (Tokyo, Japan). The degree of coverage of the surface by alkylsilyl ligands (␣RP ) was calculated on the basis of the carbon percentage determined on a Model 240 CHN analyzer (Perkin Elmer, Norwalk, USA). Solid-state NMR measurements were performed on a Bruker Avance III 700 MHz (Karlsruhe, Germany). The 13 C cross-polarization magic-angle spinning (CP MAS) NMR spectra were obtained with rotation frequency 8 kHz, pulse time 2 ms, acquisition time 0.01643 s, and relaxation time 6 s. All spectra were externally referenced with liquid tetramethylsilane and the chemical shifts (␦) were given in parts per million (ppm). Adsorbents were packed using laboratory-made apparatus equipped with Haskel packing pump (Burbank, CA, USA) into 125 × 4.6 mm id stainless steel columns using the slurry method. About 1.5 g of the modified silica was prepared as a slurry with 15 mL of chloroform and placed into the packing apparatus. Methanol was used as a packing pressurizing solvent during the filling process. Columns were packed under a constant pressure of 40 MPa.
Aldrich Chemie (Steinheim, Germany). The concentration of test compounds was in the range of 10–40 g/mL. The injection volumes were in the range 2–5 L.
2.4 Methods Silanol activity (SAG ) and the hydrophobicity (HG ) of the stationary phases were determined according to the method described by Galushko [19] and commonly used for stationary phase characterization [20]. To determine these factors, the analysis of aniline, phenol, benzene, and toluene retention was done using a mobile phase containing 60% methanol in water. The silanol activity (SAG ) and hydrophobicity (HG ) can be calculated as follows using solute retention factor k: SAG = 1 + 3 kaniline /kphenol − 1
(1)
HG = (ktoluene + kbenzene ) /2
(2)
Methylene selectivity tests were performed in MeOH/water and ACN/water conditions using benzene, toluene, ethylbenzene, propylbenzene, and butylbenzene [20–22]. log k = log  + n log ␣
(3)
where ␣ is the measure of methylene (liphophilic) selectivity, n is the number of repeat methylene units, and  describes the aromatic (phenyl) contribution to the retention [23]. In the shape and size selectivity tests, six PAHs were used: naphthalene, phenanthrene, anthracene, pyrene, chrysene, and benzo[a]pyrene. In this group, phenanthrene and anthracene as well as pyrene and chrysene, are pairs of isomers with the same number of aromatic rings. The retention of PAHs was performed in ACN/water mixtures as a mobile phase. In all methods, the flow rate was 1 mL/min. Measurement was carried at 298 K. Column dead volumes were measured using thiourea as a marker.
2.5 Synthesis procedure 2.2 Materials As a support for the synthesis, the silica gel Kromasil 300 (Akzo Nobel, Bohus, Sweden) was used with particle size 5 m and pore diameter 300 Å. Two different mobile phase systems were used in the measurements: methanol/water and acetonitrile/water. Organic solvents (methanol and acetonitrile) were of high purity “for HPLC” isocratic grade from J. T. Baker (Deventer, The Netherlands). Water was purified using a Milli-Q system (Millipore, El Paso, TX, USA) in our laboratory. The standard test compounds—aniline, phenol, homologues of benzene—and PAHs were obtained from Sigma– C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Before the chemical modification reaction of bare silica gel, a sample of adsorbent was placed in a specially designed glass reactor protecting against the contact of the reagents with the external environment. Silica gel was treated at 180⬚C under vacuum (10–2 Pa) for 10 h to remove physically adsorbed water. Then, the temperature was decreased to 120⬚C and proper silane was added. The silanes, (3-phenylpropyl) dimethylchlorosilane, 3-phenoxypropyldimethylchlorosilane, and Nphenylaminopropyltrimethoxysilane, were used to obtain phenyl, phenoxy, and phenyl-amine stationary phases, respectively. Silica support surface modification with phenyl www.jss-journal.com
Liquid Chromatography
J. Sep. Sci. 2014, 37, 3435–3442
3437
density (␣RP ) on the silica surface. Calculation was performed using the Berendsen equation [25]: ␣RP I =
106 PC 1 (mol/m2 ) · ( ) 1200nC − PC M1 − nx SBET
(4)
where ␣RP I is the coverage density (mol/m2 ); PC , the percentage of carbon (%); nC , the number of carbon atoms in the ligand; M1 , molar mass of the ligand; nx , the number of functional groups in reactive group of the silane; and SBET is the specific surface area (m2 /g). Table 1 shows that three phases, phenyl, phenoxy, and phenyl-amide, have almost identical carbon load and as a result almost the same coverage density of phenyl ligands on the silica gel surface. Phenyl-amine stationary phase has a slightly lower carbon load and lower surface coverage. However, the difference is relatively low, about 5%. Thus, in this work we assume that the surface coverage is the same for all phases and the main parameter that differentiates the stationary phases are different polar groups (atoms) in the structure of bonded ligands. During the synthesis of phenyl-amide stationary phase, some of the amine ligands may become unreacted. According to elemental analysis, the coverage density of amine ligands was 3.59 mol/m2 and coverage of phenyl rings was 3.55 mol/m2 . It may be concluded that the differences in the coverage density are in the range of measurement error and there are almost any residual amines in the final stationary phase. Figure 1. phases.
13
C CP MAS NMR spectra of synthesized stationary
silanes was carried out in nonsolvent conditions as described in detail in Refs. [1, 24]. For the synthesis of phenyl-amide stationary phase, silica was first modified using ␥-aminopropyltrimethoxysilane. After 12 h, the reaction products were washed out with toluene, methanol, and hexane and dried. Next, the aminopropyl silica was placed in a glass reactor and heated up to 100⬚C. Further, the aminopropyl silica was modified using benzoyl chloride in toluene solution with an addition of triethylamine at 50⬚C for 12 h. The reaction products were washed out with toluene, methanol, and hexane and dried. Structures of synthesized phenyl-bonded stationary phases are illustrated in Fig. 1.
3.2 NMR spectroscopy The structures of the synthesized materials were confirmed using 13 C and CP MAS NMR spectroscopy. Figure 1 displays the obtained 13 C CP MAS NMR spectra. The carbon atoms of the bonded ligands for each observed signal are shown in the figure. The methyl group bonded to the silicon atom gives the signal at ␦ = 0 ppm and the carbon atom connected to silicon gives a signal indicated by peak at ␦ = 11 ppm. Carbon atoms from phenyl ring provide signals in the range of 111–128 ppm, excluding carbon atom bonded with the main chain whose position in the spectra varies depending on the polar functionalities in the chain. Carbonyl groups may be identified as a signal at ␦ = 170 ppm.
3.3 Polarity/hydrophobicity investigation
3 Results and discussion 3.1 Elemental analysis Synthesized adsorbents were a subject of elemental analysis. Table 1 presents the results of chemical modification of the silica gel surface, i.e. content of carbon, nitrogen, and hydrogen after each bonding reaction determined by elemental analysis. It allows the calculation of bonded ligands coverage C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
The retention of four standard compounds (aniline, phenol, benzene, and toluene) on the tested stationary phases was compared using mobile phase containing 60% methanol in water. Hydrophilic and silanol activity and stationary phase hydrophobicity were calculated according to Galushko test [19]. The results are presented in Table 2. These two parameters obtained from Galushko test differentiate phenyl-bonded stationary phases. The lowest polarity (including silanol activity) is observed on the phenyl-amide stationary phase and the www.jss-journal.com
3438
J. Sep. Sci. 2014, 37, 3435–3442
S. Bocian and B. Buszewski
Table 1. Physicochemical properties of stationary phases
Property
Phenyl
Phenoxy
Phenyl-amide
Phenyl-amine
Carbon percentage Hydrogen percentage Coverage density (mol/m2 )
4.856 0.921 3.57
4.738 0.936 3.50
4.501 0.837 3.55
3.731 0.955 3.36
Table 2. Silanol activity and hydrophobicity of synthesized phenyl-bonded stationary phases
Property
Phenyl
Phenoxy
Phenyl-amide
Phenyl-amine
Silanol activity Hydrophobicity
0.85 0.69
0.90 0.59
0.55 0.19
1.38 0.24
highest silanol activity exhibits phenyl-amine adsorbent. On the other hand, phenyl and phenoxy materials exhibit much higher hydrophobicity than the materials containing amine and amide groups. As a result, it may be expected that phenyl and phenoxy stationary phases should exhibit higher retention in RP conditions. However, the presence of functional groups capable of polar interaction should result in different selectivity.
3.4 RP retention and selectivity All synthesized chemically bonded stationary phases were tested in RP-LC conditions. As a test sample, the mixture of alkyl benzene derivatives was used. The retention factors of alkyl benzene derivatives in 50% of organic solvent in the mobile phase, both MeOH and ACN, are shown in Fig. 2. Phenyl and phenoxy stationary phases exhibit significantly higher retention of hydrophobic compounds as compared with phenylamine and phenyl-amide materials. This observation is in agreement with hydrophobicity of synthesized materials (Table 2). More hydrophobic materials provide higher retention. The highest retention in MeOH environment provides a phenyl stationary phase that does not possess any polar group in the structure. The presence of polar groups reduces the retention of hydrophobic compounds. In ACN environment, the highest retention is observed on phenoxy column, which is less hydrophobic than the phenyl column. Additionally, the differences of retention between columns are not so significant as in the case of MeOH. These may be caused by differences in the solvation process of phenyl stationary phases with polar functional groups. Probably, in MeOH environment, polar phenyl phases exhibit lower retention of hydrophobic compounds as compared with mobile phases containing ACN, as a result of MeOH interaction with polar functionalities and relatively stronger elution strength of MeOH. It may be explained that methanol shows higher specific hydrogen bonding interactions compared to acetonitrile. C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 2. Retention factor of alkyl benzene derivatives in 50% MeOH in water (A) and 50% ACN in water (B).
Alkyl benzene derivatives were also used for methylene selectivity test. In binary mobile phases containing various volume fractions, , of organic modifier in water, linear dependencies of log k on were found according to Eq. (3) [20–22], where a = a0 + a1 n
(5)
m = m0 + m1 n
(6)
The increments in the parameters a and m per methylene group, a0 , a1 , m0 , and m1 were determined using multilinear regression; n is the number of repeat methylene units in alkyl benzene derivatives homologues series. Measurements were performed using MeOH and ACN as an organic modifier. The results are listed in Table 3A and B. The parameter a1 characterizes the methylene selectivity and a0 the phenyl selectivity. Higher methylene selectivity was obtained in MeOH/water mobile phases. The phenyl stationary phase exhibits the highest methylene selectivity, which may be attributed to the highest hydrophobicity of this material. The lowest values were obtained for phenyl-amine www.jss-journal.com
Liquid Chromatography
J. Sep. Sci. 2014, 37, 3435–3442
3439
Table 3A. The parameters a0 , a1 , m0 , m1 , and SD and determination coefficient of Eqs. (5) and (6) in the homologous series of nalkylbenzenes (C0–C4) in MeOH/water mobile phase
MeOH
Phenyl
Phenoxy
Phenyl-amide
Phenyl-amine
a0 a1 R2 m0 m1 R2
1.328 ± 0.036 0.439 ± 0.015 0.9967 2.684 ± 0.045 0.405 ± 0.018 0.9938
1.295 ± 0.017 0.399 ± 0.007 0.9991 2.733 ± 0.028 0.359 ± 0.011 0.9970
0.338 ± 0.024 0.342 ± 0.010 0.9975 1.813 ± 0.033 0.389 ± 0.014 0.9964
0.260 ± 0.038 0.350 ± 0.016 0.9940 1.486 ± 0.086 0.358 ± 0.035 0.9722
Table 3B. The parameters a0 , a1 , m0 , m1 , and SD and determination coefficient of Eqs. (5) and (6) in the homologous series of nalkylbenzenes (C0–C4) in ACN/water mobile phase
ACN
Phenyl
Phenoxy
Phenyl-amide
Phenyl-amine
a0 R2 a1 m0 m1 R2
0.656 ± 0.015 0.9990 0.324 ± 0.006 2.308 ± 0.028 0.372 ± 0.0112 0.9971
1.082 ± 0.008 0.9995 0.280 ± 0.004 2.650 ± 0.015 0.267 ± 0.006 0.9985
0.369 ± 0.015 0.9990 0.331 ± 0.006 1.689 ± 0.035 0.470 ± 0.014 0.9972
0.719 ± 0.014 0.9972 0.331 ± 0.006 2.279 ± 0.034 0.449 ± 0.014 0.9992
and phenyl-amide stationary phases. In acetonitrile environment, the methylene selectivity of phenyl, phenyl-amide, and phenyl-amine stationary phases was almost identical and for phenoxy column it was a little bit lower. Phenyl and phenoxy columns exhibit significantly higher phenyl selectivity in MeOH solutions compared with phenylamide and phenyl-amine stationary phases. It may be connected with the structure of bonded ligands and significantly lower hydrophobicity. The presence of nitrogen atoms changes the conformation of the main chain. In the ACN environment, the phenyl selectivity is reduced. It is observed especially for phenyl stationary phase that has lower selectivity than phenyl-amine stationary phase. Such a change may be caused by solvation process, which is different in MeOH and ACN solution. Parameter m1 represents the influence of solvent of the methylene selectivity and m0 describes the influence of organic solvent on the phenyl selectivity. Although in MeOH solution, the parameter m1 is comparable for all tested stationary phases, in ACN environment the significant differences are observed. Methylene selectivity of phenyl-amide and phenyl-amine is more influenced by organic solvent concentration than phenyl and phenoxy columns. On the other hand, in MeOH solution, the concentration of organic solvent influences more strongly phenyl and phenoxy stationary phases, whereas the m0 value obtained for phenyl-amine stationary phase is relatively low. Synthesized stationary phases were also compared according to their size and shape selectivity. For such tests, the mixture of PAHs was applied. The separations of PAH isomers may be a good parameter for the comparison of stationary phase selectivity. The chromatograms of PAHs sepa C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
ration in mobile phase containing 40% of ACN in water are shown in Fig. 3. One can observe, that phenyl and phenoxy stationary phases offer about two times higher retention as a result of higher hydrophobicity. Significant hydrophobicity of these two phases causes the better selectivity that results in the separation of anthracene and phenanthrene, which was not possible on less hydrophobic phenyl-amine and phenylamide stationary phases. Detailed parameters of PAHs separation, such as retention factor, resolution, and selectivity are listed in Table 4. As it was mentioned above, higher hydrophobicity causes higher retention that results in better resolution of the PAHs separations. The lowest resolution provides phenyl-amide stationary phase that exhibits the lowest hydrophobicity. For a better investigation of stationary phase selectivity, a critical pair of isomers—anthracene/phenanthrene— was evaluated in the test mixture. Figure 4 shows that all compared stationary phases offer a good selectivity of this pair. However, it has to be emphasized that this selectivity is obtained in different concentrations of ACN in the mobile phase. For the mobile phase that contains 50% of ACN in water, only phenyl and phenoxy phases provide selectivity higher than 2. To obtain the separation of anthracene and phenanthrene to the baseline, it is necessary to decrease the ACN concentration to 40%. Similar selectivity may be obtained on phenyl-amine stationary phase, when concentration of ACN mobile phase is decreased to 45%. On the phenyl-amide stationary phase, the separation of anthracene and phenanthrene was possible with mobile phases containing 40% and less ACN. The decreasing concentration of organic solvent for the separation of anthracene and phenanthrene results from the lower
www.jss-journal.com
3440
S. Bocian and B. Buszewski
J. Sep. Sci. 2014, 37, 3435–3442
Figure 3. Separation of PAHs on different phenyl-bonded stationary phases (A–D); mobile phase 40% ACN/60% H2 O; and on phenyl-amide in 30% ACN (E) and on phenyl-amine at 35% ACN in the mobile phase (F); compounds in order of appearance: benzene, naphthalene, phenanthrene, anthracene, pyrene, chrysene, benzo[a]pyrene.
hydrophobicity of phenyl-amine and phenyl-amide stationary phases. However, all four phases exhibit good isomer selectivity in different mobile phases. To obtain a satisfactory separation, it is necessary to decrease the concentration of ACN to 35% for phenyl-amine and 30% on phenyl-amide stationary phase. The obtained chromatograms are shown in Fig. 3E and F. Phenyl-bonded stationary phases were also compared according to size selectivity. In Fig. 5, the retention (log k) of solutes with different number of aromatic rings was compared. For all stationary phases, the linear dependences were found with the increasing number of phenyl ring in the solute molecule. Phenyl, phenyl-amine, and phenyl-amide exhibit parallel trends. Phenoxy stationary phases show a stronger increase in the retention with an increasing number of phenyl rings compared to other tested materials.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
It may be concluded that all phenyl-bonded stationary phases provide good size selectivity and are able to separate PAHs. However, it has to be remembered that phenyl-amine and phenyl-amide require lower concentration of organic solvent for obtaining the same retention and selectivity compared to phenyl and phenoxy stationary phases.
4 Concluding remarks Different phenyl-bonded stationary phases for LC were synthesized. Obtained materials were investigated using instrumental analysis. The structures of synthesized materials were confirmed using the IR and NMR spectroscopies. New stationary phases were tested according to their application in RP chromatography. The presence of polar-embedded functional groups in the structure of the bonded ligands offers www.jss-journal.com
Liquid Chromatography
J. Sep. Sci. 2014, 37, 3435–3442
3441
Table 4. Chromatographic parameters of the PAHs separation on phenyl-bonded stationary phases using 40% ACN in water as a mobile phase
Compound
Phenoxy Benzene Naphthalene Phenantrene Antracene Pyren Chryzene Benzo[a]pyrene Phenyl Benzene Naphthalene Phenantrene Antracene Pyren Chryzene Benzo[a]pyrene Phenyl-amide Benzene Naphthalene Phenantrene Antracene Pyren Chryzene Benzo[a]pyrene Phenyl-amine Benzene Naphthalene Phenantrene Antracene Pyren Chryzene Benzo[a]pyrene
Retention factor (k)
Number of theoretical plates (N)
Resolution (RS )
Selectivity (␣)
1.002 2.528 5.612 6.188 7.727 12.305 17.357
6529 8286 8850 8530 8269 8042 9262
– 12.008 14.142 1.947 4.427 9.368 7.448
– 2.523 2.220 1.103 1.249 1.592 1.411
1.01 2.475 5.243 5.698 7.114 10.802 14.763
7184 8558 8896 8786 8964 7850 8437
– 11.953 13.339 1.651 4.506 8.422 6.498
– 2.450 2.118 1.087 1.249 1.518 1.367
0.437 0.949 1.827 1.939 2.411 3.469 4.667
4366 5371 4847 5342 6258 6531 6018
– 5.291 6.534 0.689 2.837 5.373 4.669
– 2.170 1.925 1.061 1.244 1.439 1.345
0.66 1.468 2.955 3.181 3.944 5.865 8.024
5255 5788 5465 5734 5888 5605 5957
– 7.299 8.647 1.042 3.189 6.151 5.173
– 2.224 2.012 1.077 1.240 1.487 1.368
Figure 4. Selectivity of the isomer pair anthracene/phenanthrene.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 5. Changes of the retention (log k) with increasing number of phenyl rings in the molecule; mobile phase: 40% ACN in water.
www.jss-journal.com
3442
S. Bocian and B. Buszewski
specific properties of the adsorbents, such as relatively low hydrophobicity. Lower hydrophobicity of phenyl-amine and phenyl-amide stationary phases requires lower concentration of organic solvent to perform sufficient separation of hydrophobic compounds. In less concentrated mobile phases, phenyl-amine and phenyl-amide stationary phases exhibit selectivity similar to phenyl and phenoxy stationary phases at higher concentration of organic solvent. This work was supported by the Ministry of Science and Higher Education, Grant no. NCN 2013/09/D/ST4/03807 for the period 2014–2017. The authors thank Akzo Nobel (Bohus, Sweden) for the kind donation of silica gel Kromasil 300 used in the study. The authors have declared no conflict of interest.
5 References [1] Buszewski, B., Suprynowicz, Z., Lodkowski, R., Nasuto, ´ R., Szymanska, K. Chem. Anal. (Warsow) 1983, 28, 731–737. [2] Kimata, K., Hosoya, K., Kuroki, H., Tanaka, N., Barr, J. R., McClure, P. C., Patterson, D. G. J., Jakobsson, E., Bergman, A. J. Chromatogr. A 1997, 786, 237–248.
J. Sep. Sci. 2014, 37, 3435–3442
[8] Miyabe, K., Sakai, Y., Sarashina, M., Yokokaw, C. Analyst 2003, 128, 1425–1433. [9] Stevenson, P. G., Mayfield, K. J., Soliven, A., Dennis, G. R., Gritti, F., Guiochon, G., Shalliker, R. A. J. Chromatogr. A 2010, 1217, 5358–5364. [10] Stevenson, P. G., Soliven, A., Dennis, G. R., Gritti, F., Guiochon, G., Shalliker, R. A. J. Chromatogr. A 2010, 1217, 5377–5383. [11] Hanai, T., Ahmed, F., Rustamov, I., Babusis, D. J. Liq. Chromatogr. Relat. Technol. 1999, 22, 501–511. [12] Jinno, K., Nagoshi, T., Tanaka, N., Okamoto, M., Fetzer, J. C., Biggs, W. R. J. Chromatogr. A 1987, 286, 123–135. [13] Ellwanger, A., Brindle, R., Albert, K. J. High Resolut. Chromatogr. 1997, 20, 39–45. ¨ [14] Haas, A., Kohler, J., Hemetsberger, H. Chromatographia 1981, 14, 341–344. [15] Pesek, J. J., Matyska, M. T., Kim, A. M. J. Sep. Sci. 2013, 36, 2760–2766. [16] Pesek, J. J., Boysen, R. I., Hearn, M. T. W., Matyska, M. T. Anal. Method. 2014, 6, 4496–4503. [17] Layne, J. J. Chromatogr. A 2002, 957, 149–164. [18] Bocian, S., Nowaczyk, A., Buszewski, B. Anal. Bioanal. Chem. 2012, 404, 731–740. [19] Galushko, S. V. Chromatographia 1993, 36, 39–42.
[3] Horak, J., Maier, N. M., Lindner, W. J. Chromatogr. A 2004, 1045, 43–58.
[20] Molikova, M., Jandera, P. J. Sep. Sci. 2010, 33, 453–463.
[4] Gadzała-Kopciuch, R. M., Kluska, M., Wełniak, M., ´ Kroszczynski, W., Buszewski, B. Pol. J. Environ. Stud. 1999, 8, 383–387.
[22] Jandera, P., Novotna, K., Beldean-Galea, M. S., Jisa, K. J. Sep. Sci. 2006, 29, 856–871.
[5] Chan, F., Yeung, L. S., LoBrutto, R., Kazakevich, Y. V. J. Chromatogr. A 2005, 1069, 217–224.
[21] Jandera, P. J. Chromatogr. 1984, 352, 13–36.
[23] Sentell, K. B., Dorsey, J. G. J. Chromatogr. A 1989, 461, 193–207.
[6] Hanai, T., Hubert, J. J. Chromatogr. A 1984, 291, 81–89.
[24] Buszewski, B., Jezierska, M., Wełniak, M., Kaliszan, R. J. Chromatogr. A 1999, 845, 433–445.
´ [7] Heron, S., Tchapla, A. J. Chromatogr. A 1996, 725, 205–218.
[25] Berendsen, G. E., de Galan, L. J. Chromatogr. 1980, 196, 21–37.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.jss-journal.com
J. Sep. Sci. 2014, 37, 3435–3442
Szymon Bocian Bogusław Buszewski Faculty of Chemistry, Department of Environmental Chemistry & Bioanalytics, Nicolaus Copernicus University, Torun, Poland Received July 17, 2014 Revised August 18, 2014 Accepted September 9, 2014
Research Article
Phenyl-bonded stationary phases—The influence of polar functional groups on retention and selectivity in reversed-phase liquid chromatography The chromatographic properties of four phenyl-bonded phases with different structures were studied. The columns used were packed with a stationary phase containing a phenyl ring attached to the silica surface using different types of linkage molecules. As a basic characteristic of the bonded phases, the hydrophobicity and silanol activity (polarity) were investigated. The presence of the polar amino and amide groups in the structure of the bonded ligand strongly influences the polarity of the bonded phase. Columns were compared according to methylene selectivity using a series of benzene homologues and according to their shape and size selectivity using polycyclic aromatic hydrocarbons. The measurements were done using methanol/water and acetonitrile/water mobile phases. The presented results show that the presence of polar functional groups in the ligand structure strongly influences the chromatographic properties of the bonded phase. Keywords: Liquid chromatography / Phenyl-bonded phases / Polar functional groups DOI 10.1002/jssc.201400764
1 Introduction Phenyl-bonded stationary phases are unique separation materials that are gaining popularity as packing materials in HPLC analyses [1–3]. Phenyl-bonded stationary phases have been available for many years [4, 5]. The idea of this type of material is to promote the – interaction of the aromatic rings with a solute as an additional force, which modifies the RP retention mechanism. The presence of phenyl-bonded ligands also changes the solvation process and offers different selectivity of the chromatographic system in comparison with octadecyl (C18 ) stationary phases [6–8]. This special selectivity will be observed when analyzed compounds possess aromatic functionalities in their structure, which enable – interactions. As a result of these interactions, solutes with -electrons will display a different retention behavior on phenyl columns than on alkyl RP phases, while solutes without -electrons or with sterically hindered -electron systems will be retained by the classical C8 /C18 RP-like mode [3]. The methylene and aromatic selectivity of phenyl-bonded stationary phases depends on the
Correspondence: Professor Bogusław Buszewski, Faculty of Chemistry, Department of Environmental Chemistry & Bioanalytics, Nicolaus Copernicus University, Gagarin 7 St., 87-100 Torun, Poland E-mail: [email protected] Fax: +48-56-611-4837
Abbreviations: CP MAS, cross-polarization magic-angle spinning; PAH, polycyclic aromatic hydrocarbon C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
spacer length, which binds the phenyl ring with the silica surface [9]. The surface coverage of bonded phenyl ligands is also an important parameter that influences the selectivity of the stationary phase [10]. Because of the manifold possibilities to create – active stationary phases, numerous types of – active RP phases were already designed and investigated. Early and moderately – active phases contained mono-aromatic [11] or polyaromatic [12] moieties and stationary phases employing heteroaromatic ring systems [13]. Other types of phenylbonded stationary phases are fluorinated phenyl phases. This material is especially recommended for solutes that possess some halogens in their structure, as many pesticides, for example [14]. Recently, a novel type of aromatic stationary phase was introduced in the market that contains biphenyl ligands and offers selectivity significantly different from alkyl stationary phases. Another type of bonded stationary phase that contains phenyl ligand is a phenyl-hydride material [15, 16]. This type of stationary phase may be applied in both RP and aqueous normal-phase chromatography. Nowadays, polar-endcapped and polar-embedded RP phases with incorporated different polar functional groups are in the center of interest [17]. Due to their ability to undergo hydrogen bonding and other polar interactions with water molecules, such materials are significantly more stable in water-rich mobile phases than the conventional hydrophobic RP adsorbents. Such materials may also be stable in purely aqueous mobile phases [17, 18]. Colour Online: See the article online to view Fig. 2 in colour. www.jss-journal.com
3436
J. Sep. Sci. 2014, 37, 3435–3442
S. Bocian and B. Buszewski
The goal of our work was to synthesize a series of phenyl-bonded stationary phases with different polar functional groups in their structures and compare their chromatographic properties. Typical propyl-phenyl adsorbents were compared with polar-endcapped and polar-embedded phenyl-bonded stationary phases. The influence of the polar group on the retention and selectivity was determined. Methylene selectivity using a series of alkyl benzene derivatives and shape/size selectivity obtained from polycyclic aromatic hydrocarbon (PAH) separations in MeOH/water and ACN/water was compared.
2 Materials and methods 2.1 Instruments The liquid chromatograph was a Shimadzu Prominence system (Tokyo, Japan) equipped with ternary gradient pump (LC20AD), diode array detector (SPD-M20A), an autosampler (SIL-20A), and a column thermostat (CTO-10AS VP). Data were collected using LabSolutions software (Tokyo, Japan). The degree of coverage of the surface by alkylsilyl ligands (␣RP ) was calculated on the basis of the carbon percentage determined on a Model 240 CHN analyzer (Perkin Elmer, Norwalk, USA). Solid-state NMR measurements were performed on a Bruker Avance III 700 MHz (Karlsruhe, Germany). The 13 C cross-polarization magic-angle spinning (CP MAS) NMR spectra were obtained with rotation frequency 8 kHz, pulse time 2 ms, acquisition time 0.01643 s, and relaxation time 6 s. All spectra were externally referenced with liquid tetramethylsilane and the chemical shifts (␦) were given in parts per million (ppm). Adsorbents were packed using laboratory-made apparatus equipped with Haskel packing pump (Burbank, CA, USA) into 125 × 4.6 mm id stainless steel columns using the slurry method. About 1.5 g of the modified silica was prepared as a slurry with 15 mL of chloroform and placed into the packing apparatus. Methanol was used as a packing pressurizing solvent during the filling process. Columns were packed under a constant pressure of 40 MPa.
Aldrich Chemie (Steinheim, Germany). The concentration of test compounds was in the range of 10–40 g/mL. The injection volumes were in the range 2–5 L.
2.4 Methods Silanol activity (SAG ) and the hydrophobicity (HG ) of the stationary phases were determined according to the method described by Galushko [19] and commonly used for stationary phase characterization [20]. To determine these factors, the analysis of aniline, phenol, benzene, and toluene retention was done using a mobile phase containing 60% methanol in water. The silanol activity (SAG ) and hydrophobicity (HG ) can be calculated as follows using solute retention factor k: SAG = 1 + 3 kaniline /kphenol − 1
(1)
HG = (ktoluene + kbenzene ) /2
(2)
Methylene selectivity tests were performed in MeOH/water and ACN/water conditions using benzene, toluene, ethylbenzene, propylbenzene, and butylbenzene [20–22]. log k = log  + n log ␣
(3)
where ␣ is the measure of methylene (liphophilic) selectivity, n is the number of repeat methylene units, and  describes the aromatic (phenyl) contribution to the retention [23]. In the shape and size selectivity tests, six PAHs were used: naphthalene, phenanthrene, anthracene, pyrene, chrysene, and benzo[a]pyrene. In this group, phenanthrene and anthracene as well as pyrene and chrysene, are pairs of isomers with the same number of aromatic rings. The retention of PAHs was performed in ACN/water mixtures as a mobile phase. In all methods, the flow rate was 1 mL/min. Measurement was carried at 298 K. Column dead volumes were measured using thiourea as a marker.
2.5 Synthesis procedure 2.2 Materials As a support for the synthesis, the silica gel Kromasil 300 (Akzo Nobel, Bohus, Sweden) was used with particle size 5 m and pore diameter 300 Å. Two different mobile phase systems were used in the measurements: methanol/water and acetonitrile/water. Organic solvents (methanol and acetonitrile) were of high purity “for HPLC” isocratic grade from J. T. Baker (Deventer, The Netherlands). Water was purified using a Milli-Q system (Millipore, El Paso, TX, USA) in our laboratory. The standard test compounds—aniline, phenol, homologues of benzene—and PAHs were obtained from Sigma– C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Before the chemical modification reaction of bare silica gel, a sample of adsorbent was placed in a specially designed glass reactor protecting against the contact of the reagents with the external environment. Silica gel was treated at 180⬚C under vacuum (10–2 Pa) for 10 h to remove physically adsorbed water. Then, the temperature was decreased to 120⬚C and proper silane was added. The silanes, (3-phenylpropyl) dimethylchlorosilane, 3-phenoxypropyldimethylchlorosilane, and Nphenylaminopropyltrimethoxysilane, were used to obtain phenyl, phenoxy, and phenyl-amine stationary phases, respectively. Silica support surface modification with phenyl www.jss-journal.com
Liquid Chromatography
J. Sep. Sci. 2014, 37, 3435–3442
3437
density (␣RP ) on the silica surface. Calculation was performed using the Berendsen equation [25]: ␣RP I =
106 PC 1 (mol/m2 ) · ( ) 1200nC − PC M1 − nx SBET
(4)
where ␣RP I is the coverage density (mol/m2 ); PC , the percentage of carbon (%); nC , the number of carbon atoms in the ligand; M1 , molar mass of the ligand; nx , the number of functional groups in reactive group of the silane; and SBET is the specific surface area (m2 /g). Table 1 shows that three phases, phenyl, phenoxy, and phenyl-amide, have almost identical carbon load and as a result almost the same coverage density of phenyl ligands on the silica gel surface. Phenyl-amine stationary phase has a slightly lower carbon load and lower surface coverage. However, the difference is relatively low, about 5%. Thus, in this work we assume that the surface coverage is the same for all phases and the main parameter that differentiates the stationary phases are different polar groups (atoms) in the structure of bonded ligands. During the synthesis of phenyl-amide stationary phase, some of the amine ligands may become unreacted. According to elemental analysis, the coverage density of amine ligands was 3.59 mol/m2 and coverage of phenyl rings was 3.55 mol/m2 . It may be concluded that the differences in the coverage density are in the range of measurement error and there are almost any residual amines in the final stationary phase. Figure 1. phases.
13
C CP MAS NMR spectra of synthesized stationary
silanes was carried out in nonsolvent conditions as described in detail in Refs. [1, 24]. For the synthesis of phenyl-amide stationary phase, silica was first modified using ␥-aminopropyltrimethoxysilane. After 12 h, the reaction products were washed out with toluene, methanol, and hexane and dried. Next, the aminopropyl silica was placed in a glass reactor and heated up to 100⬚C. Further, the aminopropyl silica was modified using benzoyl chloride in toluene solution with an addition of triethylamine at 50⬚C for 12 h. The reaction products were washed out with toluene, methanol, and hexane and dried. Structures of synthesized phenyl-bonded stationary phases are illustrated in Fig. 1.
3.2 NMR spectroscopy The structures of the synthesized materials were confirmed using 13 C and CP MAS NMR spectroscopy. Figure 1 displays the obtained 13 C CP MAS NMR spectra. The carbon atoms of the bonded ligands for each observed signal are shown in the figure. The methyl group bonded to the silicon atom gives the signal at ␦ = 0 ppm and the carbon atom connected to silicon gives a signal indicated by peak at ␦ = 11 ppm. Carbon atoms from phenyl ring provide signals in the range of 111–128 ppm, excluding carbon atom bonded with the main chain whose position in the spectra varies depending on the polar functionalities in the chain. Carbonyl groups may be identified as a signal at ␦ = 170 ppm.
3.3 Polarity/hydrophobicity investigation
3 Results and discussion 3.1 Elemental analysis Synthesized adsorbents were a subject of elemental analysis. Table 1 presents the results of chemical modification of the silica gel surface, i.e. content of carbon, nitrogen, and hydrogen after each bonding reaction determined by elemental analysis. It allows the calculation of bonded ligands coverage C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
The retention of four standard compounds (aniline, phenol, benzene, and toluene) on the tested stationary phases was compared using mobile phase containing 60% methanol in water. Hydrophilic and silanol activity and stationary phase hydrophobicity were calculated according to Galushko test [19]. The results are presented in Table 2. These two parameters obtained from Galushko test differentiate phenyl-bonded stationary phases. The lowest polarity (including silanol activity) is observed on the phenyl-amide stationary phase and the www.jss-journal.com
3438
J. Sep. Sci. 2014, 37, 3435–3442
S. Bocian and B. Buszewski
Table 1. Physicochemical properties of stationary phases
Property
Phenyl
Phenoxy
Phenyl-amide
Phenyl-amine
Carbon percentage Hydrogen percentage Coverage density (mol/m2 )
4.856 0.921 3.57
4.738 0.936 3.50
4.501 0.837 3.55
3.731 0.955 3.36
Table 2. Silanol activity and hydrophobicity of synthesized phenyl-bonded stationary phases
Property
Phenyl
Phenoxy
Phenyl-amide
Phenyl-amine
Silanol activity Hydrophobicity
0.85 0.69
0.90 0.59
0.55 0.19
1.38 0.24
highest silanol activity exhibits phenyl-amine adsorbent. On the other hand, phenyl and phenoxy materials exhibit much higher hydrophobicity than the materials containing amine and amide groups. As a result, it may be expected that phenyl and phenoxy stationary phases should exhibit higher retention in RP conditions. However, the presence of functional groups capable of polar interaction should result in different selectivity.
3.4 RP retention and selectivity All synthesized chemically bonded stationary phases were tested in RP-LC conditions. As a test sample, the mixture of alkyl benzene derivatives was used. The retention factors of alkyl benzene derivatives in 50% of organic solvent in the mobile phase, both MeOH and ACN, are shown in Fig. 2. Phenyl and phenoxy stationary phases exhibit significantly higher retention of hydrophobic compounds as compared with phenylamine and phenyl-amide materials. This observation is in agreement with hydrophobicity of synthesized materials (Table 2). More hydrophobic materials provide higher retention. The highest retention in MeOH environment provides a phenyl stationary phase that does not possess any polar group in the structure. The presence of polar groups reduces the retention of hydrophobic compounds. In ACN environment, the highest retention is observed on phenoxy column, which is less hydrophobic than the phenyl column. Additionally, the differences of retention between columns are not so significant as in the case of MeOH. These may be caused by differences in the solvation process of phenyl stationary phases with polar functional groups. Probably, in MeOH environment, polar phenyl phases exhibit lower retention of hydrophobic compounds as compared with mobile phases containing ACN, as a result of MeOH interaction with polar functionalities and relatively stronger elution strength of MeOH. It may be explained that methanol shows higher specific hydrogen bonding interactions compared to acetonitrile. C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 2. Retention factor of alkyl benzene derivatives in 50% MeOH in water (A) and 50% ACN in water (B).
Alkyl benzene derivatives were also used for methylene selectivity test. In binary mobile phases containing various volume fractions, , of organic modifier in water, linear dependencies of log k on were found according to Eq. (3) [20–22], where a = a0 + a1 n
(5)
m = m0 + m1 n
(6)
The increments in the parameters a and m per methylene group, a0 , a1 , m0 , and m1 were determined using multilinear regression; n is the number of repeat methylene units in alkyl benzene derivatives homologues series. Measurements were performed using MeOH and ACN as an organic modifier. The results are listed in Table 3A and B. The parameter a1 characterizes the methylene selectivity and a0 the phenyl selectivity. Higher methylene selectivity was obtained in MeOH/water mobile phases. The phenyl stationary phase exhibits the highest methylene selectivity, which may be attributed to the highest hydrophobicity of this material. The lowest values were obtained for phenyl-amine www.jss-journal.com
Liquid Chromatography
J. Sep. Sci. 2014, 37, 3435–3442
3439
Table 3A. The parameters a0 , a1 , m0 , m1 , and SD and determination coefficient of Eqs. (5) and (6) in the homologous series of nalkylbenzenes (C0–C4) in MeOH/water mobile phase
MeOH
Phenyl
Phenoxy
Phenyl-amide
Phenyl-amine
a0 a1 R2 m0 m1 R2
1.328 ± 0.036 0.439 ± 0.015 0.9967 2.684 ± 0.045 0.405 ± 0.018 0.9938
1.295 ± 0.017 0.399 ± 0.007 0.9991 2.733 ± 0.028 0.359 ± 0.011 0.9970
0.338 ± 0.024 0.342 ± 0.010 0.9975 1.813 ± 0.033 0.389 ± 0.014 0.9964
0.260 ± 0.038 0.350 ± 0.016 0.9940 1.486 ± 0.086 0.358 ± 0.035 0.9722
Table 3B. The parameters a0 , a1 , m0 , m1 , and SD and determination coefficient of Eqs. (5) and (6) in the homologous series of nalkylbenzenes (C0–C4) in ACN/water mobile phase
ACN
Phenyl
Phenoxy
Phenyl-amide
Phenyl-amine
a0 R2 a1 m0 m1 R2
0.656 ± 0.015 0.9990 0.324 ± 0.006 2.308 ± 0.028 0.372 ± 0.0112 0.9971
1.082 ± 0.008 0.9995 0.280 ± 0.004 2.650 ± 0.015 0.267 ± 0.006 0.9985
0.369 ± 0.015 0.9990 0.331 ± 0.006 1.689 ± 0.035 0.470 ± 0.014 0.9972
0.719 ± 0.014 0.9972 0.331 ± 0.006 2.279 ± 0.034 0.449 ± 0.014 0.9992
and phenyl-amide stationary phases. In acetonitrile environment, the methylene selectivity of phenyl, phenyl-amide, and phenyl-amine stationary phases was almost identical and for phenoxy column it was a little bit lower. Phenyl and phenoxy columns exhibit significantly higher phenyl selectivity in MeOH solutions compared with phenylamide and phenyl-amine stationary phases. It may be connected with the structure of bonded ligands and significantly lower hydrophobicity. The presence of nitrogen atoms changes the conformation of the main chain. In the ACN environment, the phenyl selectivity is reduced. It is observed especially for phenyl stationary phase that has lower selectivity than phenyl-amine stationary phase. Such a change may be caused by solvation process, which is different in MeOH and ACN solution. Parameter m1 represents the influence of solvent of the methylene selectivity and m0 describes the influence of organic solvent on the phenyl selectivity. Although in MeOH solution, the parameter m1 is comparable for all tested stationary phases, in ACN environment the significant differences are observed. Methylene selectivity of phenyl-amide and phenyl-amine is more influenced by organic solvent concentration than phenyl and phenoxy columns. On the other hand, in MeOH solution, the concentration of organic solvent influences more strongly phenyl and phenoxy stationary phases, whereas the m0 value obtained for phenyl-amine stationary phase is relatively low. Synthesized stationary phases were also compared according to their size and shape selectivity. For such tests, the mixture of PAHs was applied. The separations of PAH isomers may be a good parameter for the comparison of stationary phase selectivity. The chromatograms of PAHs sepa C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
ration in mobile phase containing 40% of ACN in water are shown in Fig. 3. One can observe, that phenyl and phenoxy stationary phases offer about two times higher retention as a result of higher hydrophobicity. Significant hydrophobicity of these two phases causes the better selectivity that results in the separation of anthracene and phenanthrene, which was not possible on less hydrophobic phenyl-amine and phenylamide stationary phases. Detailed parameters of PAHs separation, such as retention factor, resolution, and selectivity are listed in Table 4. As it was mentioned above, higher hydrophobicity causes higher retention that results in better resolution of the PAHs separations. The lowest resolution provides phenyl-amide stationary phase that exhibits the lowest hydrophobicity. For a better investigation of stationary phase selectivity, a critical pair of isomers—anthracene/phenanthrene— was evaluated in the test mixture. Figure 4 shows that all compared stationary phases offer a good selectivity of this pair. However, it has to be emphasized that this selectivity is obtained in different concentrations of ACN in the mobile phase. For the mobile phase that contains 50% of ACN in water, only phenyl and phenoxy phases provide selectivity higher than 2. To obtain the separation of anthracene and phenanthrene to the baseline, it is necessary to decrease the ACN concentration to 40%. Similar selectivity may be obtained on phenyl-amine stationary phase, when concentration of ACN mobile phase is decreased to 45%. On the phenyl-amide stationary phase, the separation of anthracene and phenanthrene was possible with mobile phases containing 40% and less ACN. The decreasing concentration of organic solvent for the separation of anthracene and phenanthrene results from the lower
www.jss-journal.com
3440
S. Bocian and B. Buszewski
J. Sep. Sci. 2014, 37, 3435–3442
Figure 3. Separation of PAHs on different phenyl-bonded stationary phases (A–D); mobile phase 40% ACN/60% H2 O; and on phenyl-amide in 30% ACN (E) and on phenyl-amine at 35% ACN in the mobile phase (F); compounds in order of appearance: benzene, naphthalene, phenanthrene, anthracene, pyrene, chrysene, benzo[a]pyrene.
hydrophobicity of phenyl-amine and phenyl-amide stationary phases. However, all four phases exhibit good isomer selectivity in different mobile phases. To obtain a satisfactory separation, it is necessary to decrease the concentration of ACN to 35% for phenyl-amine and 30% on phenyl-amide stationary phase. The obtained chromatograms are shown in Fig. 3E and F. Phenyl-bonded stationary phases were also compared according to size selectivity. In Fig. 5, the retention (log k) of solutes with different number of aromatic rings was compared. For all stationary phases, the linear dependences were found with the increasing number of phenyl ring in the solute molecule. Phenyl, phenyl-amine, and phenyl-amide exhibit parallel trends. Phenoxy stationary phases show a stronger increase in the retention with an increasing number of phenyl rings compared to other tested materials.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
It may be concluded that all phenyl-bonded stationary phases provide good size selectivity and are able to separate PAHs. However, it has to be remembered that phenyl-amine and phenyl-amide require lower concentration of organic solvent for obtaining the same retention and selectivity compared to phenyl and phenoxy stationary phases.
4 Concluding remarks Different phenyl-bonded stationary phases for LC were synthesized. Obtained materials were investigated using instrumental analysis. The structures of synthesized materials were confirmed using the IR and NMR spectroscopies. New stationary phases were tested according to their application in RP chromatography. The presence of polar-embedded functional groups in the structure of the bonded ligands offers www.jss-journal.com
Liquid Chromatography
J. Sep. Sci. 2014, 37, 3435–3442
3441
Table 4. Chromatographic parameters of the PAHs separation on phenyl-bonded stationary phases using 40% ACN in water as a mobile phase
Compound
Phenoxy Benzene Naphthalene Phenantrene Antracene Pyren Chryzene Benzo[a]pyrene Phenyl Benzene Naphthalene Phenantrene Antracene Pyren Chryzene Benzo[a]pyrene Phenyl-amide Benzene Naphthalene Phenantrene Antracene Pyren Chryzene Benzo[a]pyrene Phenyl-amine Benzene Naphthalene Phenantrene Antracene Pyren Chryzene Benzo[a]pyrene
Retention factor (k)
Number of theoretical plates (N)
Resolution (RS )
Selectivity (␣)
1.002 2.528 5.612 6.188 7.727 12.305 17.357
6529 8286 8850 8530 8269 8042 9262
– 12.008 14.142 1.947 4.427 9.368 7.448
– 2.523 2.220 1.103 1.249 1.592 1.411
1.01 2.475 5.243 5.698 7.114 10.802 14.763
7184 8558 8896 8786 8964 7850 8437
– 11.953 13.339 1.651 4.506 8.422 6.498
– 2.450 2.118 1.087 1.249 1.518 1.367
0.437 0.949 1.827 1.939 2.411 3.469 4.667
4366 5371 4847 5342 6258 6531 6018
– 5.291 6.534 0.689 2.837 5.373 4.669
– 2.170 1.925 1.061 1.244 1.439 1.345
0.66 1.468 2.955 3.181 3.944 5.865 8.024
5255 5788 5465 5734 5888 5605 5957
– 7.299 8.647 1.042 3.189 6.151 5.173
– 2.224 2.012 1.077 1.240 1.487 1.368
Figure 4. Selectivity of the isomer pair anthracene/phenanthrene.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 5. Changes of the retention (log k) with increasing number of phenyl rings in the molecule; mobile phase: 40% ACN in water.
www.jss-journal.com
3442
S. Bocian and B. Buszewski
specific properties of the adsorbents, such as relatively low hydrophobicity. Lower hydrophobicity of phenyl-amine and phenyl-amide stationary phases requires lower concentration of organic solvent to perform sufficient separation of hydrophobic compounds. In less concentrated mobile phases, phenyl-amine and phenyl-amide stationary phases exhibit selectivity similar to phenyl and phenoxy stationary phases at higher concentration of organic solvent. This work was supported by the Ministry of Science and Higher Education, Grant no. NCN 2013/09/D/ST4/03807 for the period 2014–2017. The authors thank Akzo Nobel (Bohus, Sweden) for the kind donation of silica gel Kromasil 300 used in the study. The authors have declared no conflict of interest.
5 References [1] Buszewski, B., Suprynowicz, Z., Lodkowski, R., Nasuto, ´ R., Szymanska, K. Chem. Anal. (Warsow) 1983, 28, 731–737. [2] Kimata, K., Hosoya, K., Kuroki, H., Tanaka, N., Barr, J. R., McClure, P. C., Patterson, D. G. J., Jakobsson, E., Bergman, A. J. Chromatogr. A 1997, 786, 237–248.
J. Sep. Sci. 2014, 37, 3435–3442
[8] Miyabe, K., Sakai, Y., Sarashina, M., Yokokaw, C. Analyst 2003, 128, 1425–1433. [9] Stevenson, P. G., Mayfield, K. J., Soliven, A., Dennis, G. R., Gritti, F., Guiochon, G., Shalliker, R. A. J. Chromatogr. A 2010, 1217, 5358–5364. [10] Stevenson, P. G., Soliven, A., Dennis, G. R., Gritti, F., Guiochon, G., Shalliker, R. A. J. Chromatogr. A 2010, 1217, 5377–5383. [11] Hanai, T., Ahmed, F., Rustamov, I., Babusis, D. J. Liq. Chromatogr. Relat. Technol. 1999, 22, 501–511. [12] Jinno, K., Nagoshi, T., Tanaka, N., Okamoto, M., Fetzer, J. C., Biggs, W. R. J. Chromatogr. A 1987, 286, 123–135. [13] Ellwanger, A., Brindle, R., Albert, K. J. High Resolut. Chromatogr. 1997, 20, 39–45. ¨ [14] Haas, A., Kohler, J., Hemetsberger, H. Chromatographia 1981, 14, 341–344. [15] Pesek, J. J., Matyska, M. T., Kim, A. M. J. Sep. Sci. 2013, 36, 2760–2766. [16] Pesek, J. J., Boysen, R. I., Hearn, M. T. W., Matyska, M. T. Anal. Method. 2014, 6, 4496–4503. [17] Layne, J. J. Chromatogr. A 2002, 957, 149–164. [18] Bocian, S., Nowaczyk, A., Buszewski, B. Anal. Bioanal. Chem. 2012, 404, 731–740. [19] Galushko, S. V. Chromatographia 1993, 36, 39–42.
[3] Horak, J., Maier, N. M., Lindner, W. J. Chromatogr. A 2004, 1045, 43–58.
[20] Molikova, M., Jandera, P. J. Sep. Sci. 2010, 33, 453–463.
[4] Gadzała-Kopciuch, R. M., Kluska, M., Wełniak, M., ´ Kroszczynski, W., Buszewski, B. Pol. J. Environ. Stud. 1999, 8, 383–387.
[22] Jandera, P., Novotna, K., Beldean-Galea, M. S., Jisa, K. J. Sep. Sci. 2006, 29, 856–871.
[5] Chan, F., Yeung, L. S., LoBrutto, R., Kazakevich, Y. V. J. Chromatogr. A 2005, 1069, 217–224.
[21] Jandera, P. J. Chromatogr. 1984, 352, 13–36.
[23] Sentell, K. B., Dorsey, J. G. J. Chromatogr. A 1989, 461, 193–207.
[6] Hanai, T., Hubert, J. J. Chromatogr. A 1984, 291, 81–89.
[24] Buszewski, B., Jezierska, M., Wełniak, M., Kaliszan, R. J. Chromatogr. A 1999, 845, 433–445.
´ [7] Heron, S., Tchapla, A. J. Chromatogr. A 1996, 725, 205–218.
[25] Berendsen, G. E., de Galan, L. J. Chromatogr. 1980, 196, 21–37.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.jss-journal.com
