synthesis and characterization of silica-based hyper-crosslinked sulfonate-modified reversed...
TRANSCRIPT
A
sgatrczse©
K
1
s[taenvpswcp
0
0d
Available online at www.sciencedirect.com
Journal of Chromatography A, 1182 (2008) 41–55
Synthesis and characterization of silica-based hyper-crosslinkedsulfonate-modified reversed stationary phases
Hao Luo 1, Lianjia Ma 2, Yu Zhang, Peter W. Carr ∗Department of Chemistry, University of Minnesota, Smith and Kolthoff Hall, 207 Pleasant Street SE, Minneapolis, MN 55455, USA
Received 24 April 2007; received in revised form 21 November 2007; accepted 30 November 2007
bstract
A novel type of silica-based sulfonate-modified reversed phase (−SO3-HC-C8) has been synthesized; it is based on a newly developed acidtable “hyper-crosslinked” C8 derivatized reversed phase, denoted HC-C8. The −SO3-HC-C8 phases containing controlled amounts of sulfonylroups were made by sulfonating the aromatic hyper-crosslinked network of the HC-C8 phase at different temperatures. The −SO3-HC-C8 phasesre only slightly less hydrophobic than the parent HC-C8 phase. The added sulfonyl groups provide a unique strong cation-exchange selectivityo the hydrophobic hyper-crosslinked substrate as indicated by the very large C coefficient as shown through Snyder’s hydrophobic subtractioneversed-phase characterization method. This cation-exchange activity clearly distinguishes the sulfonated phase from all other reversed phases asonfirmed by the very high values of Snyder’s column comparison function Fs. In addition, as was found in previous studies of silica-based and
irconia-based reversed phases, a strong correlation between the cation-exchange interaction and hydrophobic interaction was observed for theseulfonated phases in studies of the retention of cationic solutes. The overall chromatographic selectivity of these −SO3-HC-C8 phases is greatlynhanced by its high hydrophobicity through a “hydrophobically assisted” ion-exchange retention process. 2007 Elsevier B.V. All rights reserved.vity; B
pataap
abaTi
eywords: HPLC; Stationary phase; Reversed phase; Cation exchange; Selecti
. Introduction
Cation-exchange chromatography is widely applied to theeparation of different types of analytes including metal ions1–3], basic drugs [4,5], proteins and peptides [6–9]. As one ofhe most important retention modes, cation exchange providesn intrinsically different retention mechanism and thus differ-nt selectivity from the most popular mode of chromatography,amely reversed-phase liquid chromatography (RPLC). Theery different selectivities of cation exchangers and reversed-hase materials make them an attractive pair of chromatographicystems for use in two-dimensional liquid chromatography,
hich is becoming more widely applied to the separation ofomplex samples, such as the tryptic peptides encountered inroteomic studies [10–12]. More importantly for our current
∗ Corresponding author. Tel.: +1 612 624 0253; fax: +1 612 626 7541.E-mail address: [email protected] (P.W. Carr).
1 Present address: 1011 Morris Avenue, Union, NJ 07083, USA.2 Present address: Building 105, One Squibb Drive, New Brunswick, NJ8903, USA.
rcfsiftar
021-9673/$ – see front matter © 2007 Elsevier B.V. All rights reserved.oi:10.1016/j.chroma.2007.11.104
asic analytes
urposes, the introduction of a cation-exchange interaction intopredominantly reversed-phase material can greatly enhance
he retention of highly hydrophilic cations, including bio-activemines [13,14], drugs of abuse [14], water-soluble vitamins [13]nd anthocyanins [15], that are not well retained and thus areoorly separated by RPLC.
Furthermore, the importance of having both ion-exchangend hydrophobic contributions to selectivity was studied onoth octadecylsilane-bonded silica (ODS) phases [16–18] andpolybutadiene-coated zirconia (PBD-ZrO2) phase [16,18].
he ion-exchange sites on silica-based ODS phases are ion-zed silanol groups. For PBD-ZrO2 phase, the ion-exchangeesults from the adsorption of phosphate ions when phosphateontaining buffers are used. Regardless of the very different sur-ace chemistry of silica-based and zirconia-bases phases, theame two types of interactions, namely pure reversed-phasenteraction and “hydrophobically assisted ion exchange”, were
ound on both types of stationary phases [17,18]. More impor-antly the much higher contribution from the “hydrophobicallyssisted ion-exchange” process on PBD-ZrO2 relative to theeversed-phase interaction gave it radically different selectivity
4 atogr.
f[
rm(tassetip
wptdamsfopbtoaltno[biccrtpiuigototst[phi[t
rmhfiamacef
rCodphUborrsdcs
2
2
2 H. Luo et al. / J. Chrom
or basic analytes compared to various silica-based ODS phases16].
Despite the many advantages of cation-exchange chromatog-aphy, its further development to meet the requirements ofodern separations (e.g. high efficiency and mass spectrometry
MS) compatibility) is greatly challenged by difficulties in syn-hesizing high-efficiency cation exchangers. Cation exchangersre predominantly polymer based [19] because of their good pHtability and the relative ease of introducing cation-exchangeites. However, the crosslinked organic polymer backbone gen-rally gives poor efficiency caused by the slow mass transferhrough the chromatographic beads and mechanical instabil-ty due to the swelling and shrinking upon variation in eluent,ressure and temperature [20,21].
Silica has several inherent advantages compared to polymershich make it the most popular substrate for HPLC stationaryhases. The synthesis and/or size classification of silica par-icles is easily done to provide monodisperse particles of theesired pore size [22–24]. In addition, a large variety of silanesre available to modify silica’s surface through silanization andake different bonded phases [25,26]. The high mechanical
trength and the high efficiency ensured by the fast mass trans-er characteristics of bonded phases are also major advantagesf silica-based stationary phases over polymer-based stationaryhases [27–29]. Nevertheless, the biggest limitation of silica-ased phases is their poor chemical stability. The utility ofraditional silica-based stationary phases at extreme pHs is seri-usly limited by (1) the accelerated hydrolysis of siloxane bondsnd the subsequent removal of bonded stationary phase at pHess than about 2 [30] and (2) the greatly accelerated dissolu-ion of the silica substrate at pH > 8–9 [31]. Recently severalew reversed phases show a quite wide range in pH (2–11)ver which they are stable at least at moderate temperatures32,33]. Chemical instability at low pH is more serious for silica-ased ion exchangers. However, extreme pHs are sometimesndispensable in ion-exchange separations. For example, in theation-exchange separations of pyridines and anilines (espe-ially those with electron-withdrawing groups on the aromaticing), the separations must be done under quite acidic condi-ions to ensure that the amines with low pKa are completelyrotonated. In addition, their chemical instability also makest difficult to synthesize silica-based cation exchangers, whichsually takes place under very harsh conditions [34–36]. Fornstance, silica-based strong cation exchangers with sulfonylroups as the ion-exchange site can be prepared by three meth-ds. In the first method, the stationary phase is sulfonated afterhe surface is silanized to bond phenyl groups to it [34]. Seri-us losses of the bonded ligands occur during sulfonation due tohe use of the strongly acidic conditions, such as concentratedulfuric or chlorosulfonic acid to add the sulfonyl groups. Inhe second method, bare silica is reacted with sulfonated silanes34]. Ligand cleavage is avoided; but as the cation exchangerrepared by the first method, the phase is very hydrophilic and
as poor acid stability. In the third approach, sulfonyl groups arentroduced to the surface by using polymer coating techniques37,38], which can easily cause pore blockage and slow massransfer [39].A 1182 (2008) 41–55
A novel type of acid stable silica-based reversed phase wasecently developed in our laboratory [40–43]. Orthogonal poly-er forming reactions are used to prepare a very thin layer of
yper-crosslinked (HC) aromatic network that is entirely con-ned to the silica surface. Outstanding acid stability has beenchieved due to the formation of extensively networked poly-ers. The new HC phases are much more stable than the most
cid stable commercial silica-based phase, namely the “steri-ally protected” type C18 phases. At the same time, the excellentfficiency of bonded phases is preserved and no adverse effectsrom crosslinking have been observed.
In this paper, a novel type of silica-based, sulfonate-modified,eversed phase based on the previous HC reversed phase (HC-8) will be described. The hyper-crosslinked aromatic networkf HC-C8 is sulfonated with chlorosulfonic acid. The sulfonationegree can be easily controlled by adjusting the reaction tem-erature and a convenient low sulfonyl group surface coverageas been achieved by reaction at a low temperature (−61 ◦C).nlike the sulfonation of a conventional, that is monomeric,onded phenyl phase, no significant loss of bonded phase wasbserved upon use of the strong acid. The new sulfonatedeversed phase shows excellent efficiency and a mixed-modeetention mechanism with the presence of both cation-exchangeites and hydrophobic interaction sites. Interesting selectivityifference, especially towards cationic solutes, was observed inharacterization studies using Snyder’s method [44] and linearolvation energy relationships (LSERs) [45,46].
. Experimental
.1. Chemicals
Reaction reagents: Chlorosulfonic acid (99%) was obtainedfrom Aldrich (Milwaukee, WI, USA).Reaction solvents: Dichloromethane was obtained fromMallinkrodt (HPLC grade, Phillipsburg, NJ, USA) and driedby MB-SPS (Solvent Purification System from MBRAUN,Stratham, NH, USA) before use. ACS grade tetrahydrofuran(THF), acetone and HPLC-grade isopropanol (IPA) were alsoobtained from Mallinkrodt.HPLC buffers and solvents: Trifluoroacetic acid (TFA,ReagentPlus, 99%), ACS grade lithium chloride and potas-sium chloride were obtained from Aldrich, 50% formic acidand triethylamine hydrochloride (TEA·HCl) are HPLC-gradereagents from Fluka (Allentown, PA, USA). HPLC acetoni-trile (MeCN) was obtained from Sigma–Aldrich (St. Louis,MO, USA). HPLC water was prepared by purifying laboratory-deionized water with a Barnstead Nanopure II deionizingsystem with an organic-free cartridge followed by a 0.45 �mMini Capsule filter from PALL (East Hills, NY, USA).HPLC solutes: Morphine, hydromorphone, cathinone,ephedrine, 3,4-methylenedioxyamphetamine and metham-
phetamine were purchased from Cerilliant (Round Rock,TX, USA) as 1 mg/mL solutions of drug in methanol. Allother chromatographic solutes were obtained from Aldrich orSigma.
atog
2
tBcc
2
ropfAppidpwoanddpwpw
2
wdtthbae
2
s
Aa
2
fICHpfiw
2
Ha5HpMccCocwInuc
ccctfsdgd
2
Hewlett-Packard 1090 chromatography system, equipped witha binary pump, an autosampler, a temperature controller and adiode array detector (Hewlett-Packard, Wilmington, DE, USA).
H. Luo et al. / J. Chrom
Silica substrate: Type-B Zorbax silica particles and SB C18 par-ticles were gifts from Agilent Technologies (Wilmington, DE,USA). The particle diameter, surface area, and pore diameterof the particles are 5.0 �m, 180 m2/g, and 80 A, respectively.The 5.0 cm × 0.46 cm Primesep 200 (5 �m, 100 A) column waspurchased from SIELC Technologies (Prospect Heights, IL,USA).
.2. Synthesis of HC-C8 phase
RPLC phase HC-C8 was prepared by a series ofhree SnCl4 catalyzed Friedel–Crafts alkylations on type-
Zorbax silica that has been silylated with dimethylhloromethylphenylethylchlorosilane. The detailed reactiononditions can be found in our previous publications [40].
.3. Sulfonation of HC-C8 phase
A 3-g amount of HC-C8 was put into a 250 mL one-neckound-bottom flask and dried at 80 ◦C in a vacuum ovenvernight and then cooled to room temperature under argonrior to use. Then, 117 mL of dry dichloromethane were trans-erred into the flask containing HC-C8 under argon protection.fter sonication for 10 min, the HC-C8 slurry sealed in the argonrotected flask was immersed into the appropriate low tem-erature bath (MeCN/dry ice for −41 ◦C and chloroform/dryce for −61 ◦C [47]). 19.5 mL of 0.75 M chlorosufonic acidichloromethane solution prepared in a glove bag under argonrotection was injected into the sealed stirring HC-C8 slurry,hich had been cooled to the bath temperature. After 30 minf sulfonation, the reaction was quenched by injecting a largemount of ethanol into the slurry. Then, the sulfonated particles,amed −SO3-HC-C8, were washed sequentially with 350 mL ofichloromethane, THF, H2O, IPA and acetone before being airried overnight at room temperature. The −SO3-HC-C8 phaserepared at −41 ◦C with a relatively higher degree of sulfonationas designated −SO3-HC-C8-H and the −SO3-HC-C8 phaserepared at −61 ◦C with relatively a lower sulfonation degreeas designated −SO3-HC-C8-L.
.4. Acid pre-treatment
After synthesis, the hyper-crosslinked phase −SO3-HC-C8as pre-conditioned by gradient acid washing at 150 ◦C asescribed in our previous publications [40]. The purpose ofhis aggressive gradient acid washing is to remove tin(IV) con-amination introduced during the Friedel–Crafts reactions, toydrolyze residual chloromethyl groups and the labile Si–O–Sionds to prevent their slow hydrolysis over time during use,nd to eliminate incompletely crosslinked organic surface moi-ties.
.5. Elemental analysis
A small amount of stationary phases after each step wasent for carbon, hydrogen, and chlorine analysis conducted by
DC50
r. A 1182 (2008) 41–55 43
tlantic Microlabs (Norcross, GA, USA). Sulfur analysis waslso done after the sulfonation step.
.6. Column packing
HC particles were packed into a 5.0 cm × 0.46 cm columnor further characterization. The particles were slurried inPA (1.0 g/8.0 mL) and sonicated for 20 min prior to packing.olumns were packed by the downward slurry technique usinge gas to drive the pump and push IPA through the column. Theacking pressure was increased from 3000 to 6000 psi within therst 5 s and then kept at 6000 psi until about 90 mL of solventere collected before the pressure was released.
.7. Characterization of the cation-exchange capacity
To determine the cation-exchange capacities of the −SO3-C-C8 phases under the chromatographic conditions, a large
mount of 50/50 MeCN/H2O with 0.1% formic acid andmM potassium chloride was first passed through the −SO3-C-C8 column to saturate the cation-exchange sites withotassium ions. The potassium ions were then eluted with 50/50eCN/H2O containing 0.1% formic acid and 5 mM lithium
hloride. The first 25 or 10 mL of eluent flushed out of theolumn were collected for the −SO3-HC-C8-H and −SO3-HC-8-L phases, respectively. The amount of potassium ions in eachf the collected eluents was determined by quantitative cationhromatography, using a Dionex ICS-2000 system equippedith a CS16 analytical column and a CMMS III suppressor.
n addition, to eliminate the amount of free potassium ionsot associated with cation-exchange sites in the void volume,racil was injected to measure the void volumes of the twoolumns.
In order to keep the conditions for the determination of theation-exchange capacity consistent with the chromatographiconditions in the study of the retention of basic compounds, theoncentrations of acetonitrile (50%) and formic acid (0.1%) inhe eluents were kept the same. Additionally, the use of 0.1%ormic acid significantly reduced the contribution from residualilanols to the cation-exchange capacity, which allowed us toetermine the cation-exchange capacity due to only the sulfonylroups. The cation exchange properties will be described in moreetail in future studies.
.8. Chromatographic conditions
All chromatographic experiments were carried out on a
ata were collected and processed using Hewlett-Packardhemstation software. The solutes were prepared in ca. 1 mM0/50 MeCN/H2O solutions and the injection volume was.5 �L.
44 H. Luo et al. / J. Chromatogr. A 1182 (2008) 41–55
for t
3
3
CatT
ssFditt0act
fncw
sbftucmiwttSmcgtamtebd
Fig. 1. Synthesis scheme
. Results and discussion
.1. Elemental analysis
The synthesis scheme from the previously described [40] HC-8 phase to the final −SO3-HC-C8 phase and the phase structuret each step are shown in Fig. 1. The product at each stage inhe preparation was characterized by elemental analysis (seeable 1).
As shown in Fig. 1, the HC-C8 phase has an exten-ive aromatic network completely confined to the particleurface; the network was generated by a series of tworiedel–Crafts reactions starting with silica that is silylated withimethyl chloromethylphenylethylchlorosilane [40]. Approx-mately 0.9 ± 0.1 �mol/m2 of octyl groups are attached tohe aromatic network through a third Friedel–Crafts alkyla-ion [40]. In addition to the octyl groups, there are roughly.9–1.1 �mol/m2 of residual chloromethyl groups left in theromatic network. These chloromethyl groups are ultimatelyonverted to hydroxymethyl groups in the final phase pre-reatment step, that is, gradient hot acid washing [40,48].
In sulfonating the HC-C8 phase, different amounts of sul-
onyl groups could be introduced in the hydrophobic aromaticetwork by controlling the temperature at which sulfonation wasarried out. At −41 ◦C, 1.1 ± 0.6 �mol/m2 of sulfonyl groupsere added to the surface based on sulfur content analysis. ThetbbC
he −SO3-HC-C8 phase.
urface coverage of sulfonyl group can be significantly loweredy decreasing the sulfonation temperature. At −61 ◦C, the sul-onation degree becomes so low that the sulfur content is lowerhan the detection limit by elemental analysis. Given the largencertainty and the high detection limit of the sulfur analysis, theation-exchange capacities of −SO3-HC-C8 phases were deter-ined by an ion-exchange approach and the results are discussed
n Section 3.2. After the synthesis was complete, the HC phasesere extensively pre-treated by “gradient” flushing (see Sec-
ion 2 for details) at pH 0.5 and 150 ◦C to (1) remove tin(IV)hat was left by the catalyst of the Friedel–Crafts reactions,nCl4, (2) eliminate incompletely crosslinked organic surfaceoieties, (3) hydrolyze the siloxane bonds under the extensively
rosslinked networks, and (4) convert residual chloromethylroups to hydroxymethyl groups. We believe that these reac-ions happen slowly under normal chromatographic conditionsnd will cause gradual loss of some retention. These reactionsust be driven to completion before the column is used. After
he aggressive acid washing during the high temperature gradi-nt pre-treatments, almost all the labile bonds and groups wille broken or reacted. Obviously this will greatly decrease theegree to which such processes can happen under normal separa-
ion conditions. Retention as measured by the k′ value decreasedy 10–30% after gradient acid washing. In contradistinction,ased on the change in the carbon content of the −SO3-HC-8-L phase (see Table 1) there was only a minimal loss (3%)
H. Luo et al. / J. Chromatogr. A 1182 (2008) 41–55 45
Table 1Summary of elemental analysis at each step in the synthesis of the various −SO3-HC-C8 phases
Step Phase designation Elemental analysis (wt/wt) Surface coverage (�mol/m2)
%C %Cl %S Octyl group Chloromethyl group Sulfonyl group
Starting phase HC-C8a 11.48 0.63 N/A 0.9b 1.1c N/A
Sulfonation −SO3-HC-C8-Hd 10.99 0.53 0.53±0.3 0.9b 0.9c 1.1±0.6e
−SO3-HC-C8-Lf 11.42g 0.63 <0.3h 0.9b 1.1c NDi
Gradient acid washing −SO3-HC-C8-Lf aftergradient acid washing
11.13g <0.25j <0.3h 0.9b NDi NDi
a HC-C8 phase prepared on Zorbax silica using SnCl4 as the Friedel–Crafts catalyst in the phase synthesis [40].b Surface coverage of octyl groups based on carbon analysis.c Surface coverage of chloromethyl groups based on chlorine analysis.d Phase with higher sulfonation degree obtained from sulfonating HC-C8 at −41 ◦C.e Surface coverage of sulfonyl groups based on sulfur analysis.f Phase with lower sulfonation degree obtained from sulfonating HC-C8 at −61 ◦C.g The weight %C decreases by only 3% after gradient acid washing. However, the retentions show significant decrease of 10–30% based on the k′ change of
different solutes under various chromatographic conditions.
isiftisbeNtat
3
Hu(
Λ
ot
erac
esiapHmst1aa
3
i
TT
S
−−−
h Detection limit is 0.30% (wt/wt).i Non-detectable.j Detection limit is 0.25% (wt/wt).
n the amount of bonded phase as a result of this very aggres-ive acidic washing procedure. This indicates that the decreasen retention is not due to a decrease in the phase ratio as it isor conventional, i.e. non-hyper-crosslinked phases, but is dueo the modification of the stationary phase chemistry. Specif-cally, we believe it is primarily due to the hydrolysis of theiloxane bonds and chloromethyl groups. Similar trends haveeen observed with the HC-C8 phase [49]. This confirms thexistence of hyper-crosslinked networks on silica surface [41].early all chlorine was removed after the high temperature acid
reatment, which indicates that the chloromethyl groups werelmost completely converted to hydroxymethyl groups duringhe treatment.
.2. Characterization of cation-exchange capacity
The total amount of potassium ions eluted from the −SO3-C-C8 columns and the column void volumes measured byracil are listed in Table 2. The cation-exchange capacity Λ
�mol/m2) can be calculated through Eq. (1).
= NK+ − CK+V0 (1)
wAHere NK+ and V0 are the amount of potassium ions eluted outf the column and the column void volume, respectively. CK+ ishe concentration of potassium ion in the void volume, which is
ruft
able 2he cation-exchange capacity of the various −SO3-HC-C8 phases
tationary phases Gradient washing Void volume (mL) Total amount
SO3-HC-C8-Hb No 0.491 41.89SO3-HC-C8-Lb No 0.464 9.86SO3-HC-C8-Lb Yes 0.476 11.99
a The cation-exchange capacity was calculated based on Eq. (1) assuming that eachb Phase designation is the same as in Table 1.
qual to the concentration in the eluent (5 mM). The term A rep-esents the specific surface area of the silica particles (180 m2/g)nd w is the weight of the particles in the 5.0 cm × 0.46 cmolumn, which is assumed to be 0.5 g.
As shown in Table 2, only very small numbers of ion-xchange sites were introduced into the HC-C8 phases uponulfonation at the low temperatures. The cation-exchange capac-ty is no more than 0.5 �mol/m2 on the −SO3-HC-C8-H phasend is only approximately 0.1 �mol/m2 on the −SO3-HC-C8-Lhase. The slight increase in cation-exchange capacity on −SO3-C-C8-L phase after acid treatment (from 0.08 to 0.11 �mol/m2)ay be due to the additional silanol groups generated from
iloxane bonds. It is of great importance to be able to controlhe ion-exchange capacity to a very low level (on the order of�mol/m2 or less). Thus, the −SO3-HC-C8 phases are almosts hydrophobic as their parent phase (HC-C8) and can be usednd characterized as ordinary reversed phases.
.3. Hydrophobicity of the −SO3-HC-C8 phases
The hydrophobicity of a RPLC stationary phase is anmportant property which controls its overall retention in
eversed-phase mode. A phase’s hydrophobicity can be eval-ated based on the free energy of transfer of a methylene grouprom the stationary phase to the mobile phase [50]. A quanti-ative comparison of this free energy leads to a ranking of theof K+ eluted from column (�mol) Cation-exchange capacity (�mol/m2)a
0.440.080.11
column was packed with 0.5 g of particles.
46 H. Luo et al. / J. Chromatogr. A 1182 (2008) 41–55
Table 3The slopes and intercepts of log k′ vs. nCH2 , and �G0
CH2obtained on different stationary phases
Stationary phases Gradient washing Slopea Interceptb R2c S.E.d �G0CH2
e (cal mol−1)
SB C18 N/A 0.226±0.0006 −0.073±0.003 0.999985 0.0019 −324±0.9
HC-C8f No 0.195±0.0006 0.100±0.003 0.999980 0.0014 −280±0.9
−SO3-HC-C8-Hf No 0.172±0.0004 −0.163±0.002 0.999991 0.0011 −246±0.5−SO3-HC-C8-Lf No 0.189±0.0001 0.0638±0.0006 0.999999 0.00034 −271±0.2−SO3-HC-C8-Lf Yes 0.183±0.0002 −0.0327±0.001 0.999997 0.00070 −262±0.3
Primesep 200g N/A 0.145±0.0007 −0.194±0.003 0.999952 0.0016 −208±1.0
a The slope of the linear regression of log k′ vs. nCH2 based on the data in Fig. 2.b The intercept of the linear regression of log k′ vs. nCH2 based on the data in Fig. 2.c The squared correlation coefficient of the linear regression of log k′ vs. nCH2 based on the data in Fig. 2.d The standard error of the linear regression of log k′ vs. nCH2 based on the data in Fig. 2.e The free energy of transfer per methylene group calculated from Eq. (2) using the corresponding slope given in Table 3.
IL).
hc
�
Boe
l
n
sabTlcT
Fd5soH(
hpuafats�
orase
f Phase designation is the same as in Table 1.g A commercial mixed-mode phase (SIELC Technologies, Prospect Heights,
ydrophobicities of different stationary phases. The free energyan be estimated via the following equation:
G0CH2
= −2.3RTB (2)
is obtained by linear regression of log k′ against the numberf CH2 groups in a homolog series by the use of the Martinquation:
og k′ = A + BnCH2 (3)
CH2 is the number of methylene groups in a solute of a homologeries. The log k′ values were plotted against the nCH2 forlkylphenones on SB C18, Primesep 200 (a commercial silica-ased mixed-mode phase), and different HC phases (see Fig. 2).
he slopes, intercepts and free energies of transfer per methy-ene group for the various phases together with the correlationoefficients and standard errors of the regressions are listed inable 3. Due to the presence of C18 chains, SB C18 shows a
ig. 2. Plot of log k′ vs. number of methylene groups. Chromatographic con-itions: 50/50 MeCN/H2O with 0.1% formic acid and 10 mM TEA·HCl,.0 cm × 0.46 cm columns, T = 40 ◦C, F = 1.0 mL/min. Alkylphenone homologolutes: butyrophenone, valerophenone, hexanophenone, heptanophenone andctanophenone. (*) SB C18; (�) HC-C8 before gradient washing; (�) −SO3-C-C8-L before gradient washing; (�) −SO3-HC-C8-L after gradient washing;�) −SO3-HC-C8-H before gradient washing; (+) Primesep 200.
wCa2oo2mHc
orTFg0g−rtdtpo
igher slope and thus a higher �G0CH2
than any of the HChases. In addition, the HC phase’s hydrophobicity decreasespon the introduction of polar sulfonyl groups but it is onlyvery small change. With an additional 0.44 �mol/m2 of sul-
onyl groups, the �G0CH2
of −SO3-HC-C8-H only decreases bypproximately 12% compared to the HC-C8 phase; however,he −SO3-HC-C8-L phase with the much lower sulfonyl groupurface coverage (∼0.08 �mol/m2), shows an almost identicalG0
CH2to that of HC-C8. After gradient washing, the �G0
CH2
f −SO3-HC-C8-L decreases a bit further as the result of theemoval of isolated organic moieties and the hydrolysis of silox-ne bonds and residual chloromethyl groups to more polarilanol and alcohol groups. After both sulfonation and gradi-nt washing, −SO3-HC-C8-L still shows a rather high �G0
CH2,
hich is around 80% of that of SB C18 and 94% of that of HC-8. The retentions of the alkylphenone homolog solutes werelso measured on a commercial mixed-mode phase, Primesep00, which possesses both alkyl chains and negative chargesn its surface. It is clearly shown in Fig. 2 and Table 3 that allf the −SO3-HC-C8 phases have higher �G0
CH2than Primesep
00 and therefore are more hydrophobic. As will be discussed inore detail subsequently, the high hydrophobicity of the −SO3-C-C8 phases is very important for the separation of hydrophilic
ations.We next examine the intercept A of Eq. (3). The intercept
f the plot depends upon the phase ratio and the extrapolatedetention of the alkylphenones series at nCH2 equals to zero.his is equivalent to the k′ of benzaldehyde. As shown in bothig. 2 and Table 3, the intercept decreases upon sulfonation andradient washing. The most significant intercept decrease (from.100 to −0.163) was observed when 0.44 �mol/m2 of sulfonylroups were introduced into the parent HC-C8 phase to make theSO3-HC-C8-H phase. Since it is very unlikely that the phase
atio changes significantly after sulfonation or gradient washing,he most probably reason for the decrease of the intercept is the
ecrease in the retention of hypothetical species correspondingo the intercept, i.e. benzaldehyde, due to the introduction ofolar groups into the aromatic HC networks by either sulfonationr gradient washing (silanol and benzylic hydroxyl groups).
H. Luo et al. / J. Chromatogr. A 1182 (2008) 41–55 47
F
3−
ttp
Fig. 4. Plot of k′ vs. 1/[TEAH+] on −SO3-HC-C8-H. Chromatographic con-ditions: 50/50 MeCN/H2O with 0.1% formic acid and 20, 40, 60 or 80 mMTEA·HCl, 5.0 cm × 0.46 cm column, T = 40 ◦C, F = 1.0 mL/min. Cationicsolutes: (�) morphine; (*) hydromorphone; (�) cathinone; (♦) ephedrine; (+)3
hc
k
TT
S
−
−
−
s
l
t
ig. 3. The structures of the basic solutes used in the ion-exchange study.
.4. Hydrophobically assisted cation exchange onSO3-HC-C8 phases
According to the “three-site” model for a mixed-mode reten-ion mechanism of basic compounds on RPLC [17,18,51–53],he total retention of a basic compound is the sum of the reversed-hase retention k′
RP, pure ion-exchange retention k′Pure,IEX, and
etiE
able 4he total retention and the percent cation-exchange contribution on −SO3-HC-C8 ph
tationary phases Gradient washing [TEAH+] (mM) Retentio
SO3-HC-C8-H No 120 k′total
b
%k′IEX
c
12 k′total
b
%k′IEX
c
SO3-HC-C8-L No 20 k′total
b
%k′IEX
c
2 k′total
b
%k′IEX
c
SO3-HC-C8-L Yes 20 k′total
b
%k′IEX
c
2 k′total
b
%k′IEX
c
a Solutes—A: morphine; B: hydromorphone; C: cathinone; D: ephedrine; E: 3,4-metructures).b The total retention at different [TEAH+] calculated from Eq. (5) using the BIEX ac The percentage of cation-exchange contribution to the total retention at differen
inear regression of k′ vs. 1/[TEAH+].d A percentage slightly higher than 100% is obtained because k′
RP is a small negathe very small retention of the dead time marker uracil.
,4-methylenedioxyamphetamine; (�) methamphetamine.
ydrophobically assisted ion-exchange retention k′RP,IEX as indi-
ated in Eq. (4).
′total = k′
RP + k′Pure,IEX + k′
RP,IEX (4)
Based on the stoichiometric displacement model [1], forqually charged analyte and displacer, the ion-exchange reten-
ion is inversely proportional to the concentration of the displacern the mobile phase. For a singly charged analyte and displacer,q. (4) can be extended to Eq. (5), where [C+]m represents theases
n Solutesa
A B C D E F
1.26 1.67 2.07 2.38 3.43 3.98100d 100d 85 86 81 84
12.82 17.35 17.87 20.75 28.36 33.94100 100 98 98 98 98
1.29 1.91 2.04 2.38 3.17 3.9291 87 83 83 80 81
11.86 16.85 17.29 20.23 26.05 32.5199 99 98 98 98 98
1.03 1.50 1.57 1.83 2.38 2.87100d 95 94 92 89 90
10.51 14.33 14.81 17.03 21.51 26.10100 100 99 99 99 99
thylenedioxyamphetamine; F: methamphetamine (see Fig. 3 for the compound
nd k′RP obtained from the linear regression of k′ vs. 1/[TEAH+].
t [TEAH+] calculated from Eq. (5) using the BIEX and k′RP obtained from the
ive value, which could be caused by either the error of the linear regression or
48 H. Luo et al. / J. Chromatogr. A 1182 (2008) 41–55
Fig. 5. Plot of BIEX vs. k′RP on the −SO3-HC-C8 phases. (A) −SO3-HC-C8-H; (B) −SO3-HC-C8-L; (C) −SO3-HC-C8-L at different acetonitrile composi-
tions. (�) −SO3-HC-C8-H phase before gradient washing (50% MeCN): BIEX = 179±28 + 256±65 × k′RP, R2 = 0.79, S.E. = 46 (n = 6); (�) −SO3-HC-C8-L phase
before gradient washing (50% MeCN): BIEX = 16±2 + 61±5 × k′RP, R2 = 0.97, S.E. = 2.7 (n = 6); (♦) −SO3-HC-C8-L phase after gradient washing (50% MeCN):
BIEX = 22±2 + 93±9 × k′RP, R2 = 0.97, S.E. = 2.2 (n = 6); (�) −SO3-HC-C8-L phase before gradient washing (80% MeCN): BIEX = 31±2 + 15±12 × k′
RP, R2 = 0.26,S.E. = 3.8 (n = 6). BIEX and k′ are obtained from the linear regression of k′ vs. 1/[TEAH+]. The solid lines in each plot are the least squares fitting of thec
c
k
asBickap(−avcct
tfk
1rcoitrh
ik
p
RPorresponding data.
oncentration of the displacer in the mobile phase.
′total = k′
RP + BPure,IEX
[C+]m+ BRP,IEX
[C+]m= k′
RP + BIEX
[C+]m(5)
Here k′total is linearly related to 1/[C+]m. The intercept k′
RP, isn “ion-exchange-free” contribution to retention since it corre-ponds to the k′ at infinite displacer concentration, and the slopeIEX, is proportional to the overall strength of the ion-exchange
nteraction including both pure ion exchange and hydrophobi-ally assisted ion exchange. The linear relationship between the′ of basic compounds (see Fig. 3 for the compound structures)nd 1/[C+]m based on Eq. (5) as observed on −SO3-HC-C8hases is shown in Fig. 4, where the protonated triethylamineTEAH+) was used as the displacer and the data obtained onSO3-HC-C8-H are shown as examples. The choice of TEAH+
s the displacement cation was originally based on its high
olatility and strong cation-exchange ability to be used at lowoncentrations, which makes the mobile phase more LC–MSompatible. In addition, TEAH+ provided good peak shape forhe basic compounds studied.[pft
The reversed-phase and ion-exchange contributions to theotal retention at different displacer concentrations can beurther separated using Eq. (5) by applying the BIEX and′RP values obtained from the linear regression of k′ versus/[TEAH+]. The percent contributions of cation exchange toetention on the different phases as a function of the TEAH+
oncentration are listed in Table 4. Over a very wide rangef retentions on all three sulfonated phases, cation-exchangenteractions predominate (>80%) and play a more impor-ant role at lower displacer concentrations and for the moreetained cations under the chromatographic conditions studiedere.
More importantly the hydrophobic assistance to ion exchanges clearly evident in the correlation between the two parameters′RP and BIEX (see Fig. 5); this is very similar to what we observedreviously on silica-based Alltima ODS and PBD-ZrO2 phases
18]. Ionized silanol groups and dynamically adsorbed phos-hate groups on zirconia serve as the cation-exchange sitesor Alltima ODS and PBD-ZrO2, respectively. Based on thehree-site model developed in our laboratory [18], the slope
atog
abigatpofaensottwhtchh
ledbthabTiaii
iaBrhtainvciaasirpsp
3Rm
mtsgv(
l
k
eabsipbiκ
bospraFFSci
(CCoctscCrcoeo
i
H. Luo et al. / J. Chrom
nd intercept of BIEX versus k′RP are proportional to the num-
ers of the hydrophobically assisted ion-exchange site and pureon-exchange site, respectively, under the examined chromato-raphic conditions. For the Alltima ODS phase, the BIEX valuest the intercepts of the plots were close to zero which implieshat almost no “pure” ion-exchange site exists on the reversedhase. However, finite intercepts, i.e. BIEX values, were observedn all three −SO3-HC-C8 phases (see Fig. 5). The highly sul-onated phase, −SO3-HC-C8-H gives an intercept as high as 180nd the low sulfonated phase, −SO3-HC-C8-L shows an almostqual intercept of 20 before and after gradient washing. This sig-ificant difference from the previously studied reversed phasesuggests that −SO3-HC-C8 phases have much larger numbersf pure ion-exchange sites and consequently much higher reten-ions are expected for completely non-hydrophobic cations onhese sulfonated phases. This is deemed very important as itill provide a reasonable amount of retention for even the mostydrophilic organic cations. In addition, both the intercept andhe slope were found to be much higher on −SO3-HC-C8-Hompared to −SO3-HC-C8-L, as should be the case given itsigher level of sulfonation and therefore larger numbers of bothydrophobically assisted and pure ion-exchange sites.
As shown in Fig. 5B, gradient washing of −SO3-HC-C8-Leads to a decrease in both BIEX and k′
RP. The decrease can bexplained by the hydrolysis reactions involved in the washing asiscussed in Sections 3.1 and 3.3, which can reduce hydropho-icity and thus the closely related BIEX and k′
RP. At the sameime, BIEX became more dependent on k′
RP as indicated by theigher slope. This increased contribution of hydrophobicallyssisted ion exchange might result from an increase in the num-er of the silanol groups generated during gradient washing (seeable 2). Although the reason for this change in the dependence
s not clear, the separation selectivity of basic compounds inretention mechanism dominated by ion exchange should be
mproved by the hydrophobic contribution to selectivity evidentn the relationship between BIEX and k′
RP.The importance of hydrophobic assistance is clearly shown
n Fig. 5C, where BIEX and k′RP measured at 50% and 80%
cetonitrile are compared. For any compound, both k′RP and
IEX decreased as the acetonitrile concentration was increasedesulting from the reduced hydrophobic interaction and reducedydrophobic assistance, respectively. The plot also shows thathe range in k′
RP and BIEX among the solutes is much smallert 80% as compared to 50% acetonitrile. Clearly, selectiv-ty will be lower at 80% acetonitrile. In addition, we alsooticed a small increase in the intercept of the plot of BIEXersus k′
RP with the increase of acetonitrile composition. Thisould result from (1) an increase in the strength of the pureon-exchange interaction or (2) some of the hydrophobicallyssisted ion-exchange sites became pure ion-exchange sitest the higher acetonitrile concentration. For equally chargedolutes, the increase in pure ion-exchange retention cannotmprove the selectivity, but only contributes to the overall
etention. As a conclusion, hydrophobic interaction superim-osed on ion exchange is the key factor to provide highelectivity for different basic compounds on the −SO3-HC-C8hases.mfci
r. A 1182 (2008) 41–55 49
.5. The comparison of −SO3-HC-C8 phases and otherPLC phases by Snyder’s reversed-phase characterizationethod
Snyder et al. developed the “hydrophobic subtractionethod” to assess and compare the selectivity of RPLC sta-
ionary phases; it uses a set of 16 judiciously selected probeolutes [44,54–60]. In this method, the relative retention of aiven solute k′/k′
EB can be dissected into contributions fromarious types of intermolecular interactions as expressed in Eq.6).
og
(k′
k′EB
)≡ log α = η′H
(i)− σ′S∗
(ii)+ β′A
(iii)+ α′B
(iv)+ κ′C
(v)(6)
′EB is the retention factor of the non-polar reference solutethylbenzene. (i)–(v) designates the major types of interactionttributed to the retention difference and they are hydropho-ic interactions (i), steric resistance to insertion in the modifiedurface (ii), hydrogen bonding (iii) and (iv), and coulombicnteraction (v). H, S*, A, B, and C are the column selectivityarameters which denote hydrophobicity, steric resistance toulky molecules, hydrogen-bond acidity, hydrogen-bond basic-ty and coulombic interaction, respectively. η′, σ′, β′, α′ and′ are the corresponding solute properties. This method haseen proved very useful in the characterization and classificationf more than 300 reversed phases of different types. The newulfonate-modified reversed-phase −SO3-HC-C8 phases and thearent HC-C8 phase were characterized by this method and theesulting column parameters together with the average data offew relevant commercial phases are listed and compared inig. 6. As indicated by the regression results (see the caption ofig. 6), the two −SO3-HC-C8 phases showed excellent fits usingynder’s reversed-phase characterization method. The squaredorrelation coefficients are very close to one and the correspond-ng standard errors are quite small.
We see from the results in Fig. 6, the hydrophobicityrepresented by H) follows the order: phenyl phase < −SO3-HC-8-H < EPG ∼ −SO3-HC-C8-L < HC-C8 < type-A C8 < type-B8. The phenyl phases show the lowest H due to the absencef hydrophobic alkyl chains. The lower H of our HC phasesompared to the average H of commercial C8 phases (on eitherype-A or type-B silica) results from the relatively lower C8urface density: approximately 0.9 �mol/m2 versus the typi-al surface density of 2–3 �mol/m2 on commercial C8 phases.omparing the H of the three HC phases, the hydrophobicity is
educed upon the introduction of polar sulfonyl groups, which isonsistent with the results of Section 3.3. More decrease in H isbserved with the higher degree of sulfonation. −SO3-HC-C8-Lxhibits a hydrophobicity close to the average hydrophobicityf the phases with embedded polar groups (EPG).
By comparing the steric resistance, i.e. the S* coefficientsn Fig. 6, we see that all three of the HC phases show the
ost negative values. This means that the marker compoundsor S* are more strongly retained on the HC phases than ononventional RPLC phases. According to Eq. (6), this resultndicates that the HC phases have much lower steric resistance
50 H. Luo et al. / J. Chromatogr.
Fig. 6. The column selectivity parameters of different phases measured by Sny-der’s method. C(2.8) is column selectivity parameter of coulombic interaction atpH 2.8. The C(2.8) of −SO3-HC-C8-H should be very high. The exact value can-not be calculated because the two drugs (nortriptyline and amitriptyline), whichcontribute most to the C(2.8) measurement, are strongly retained and cannot beeluted in a reasonable time scale. The squared correlation coefficients and thestandard errors of the regression of Eq. (6) indicate good fits for all three HCphases: HC-C8 (after gradient washing), R2 = 0.999, S.E. = 0.022; −SO3-HC-C8-L (after gradient washing), R2 = 0.995, S.E. = 0.060; −SO3-HC-C8-H (beforegradient washing), R2 = 0.990, S.E. = 0.059. The data for type-B C8 phases (C8
phases synthesized on type-B silica), type-A C8 phases (C8 phases synthesizedop5
tvb0ncsssta
ogycaaIatba
−m
thpb[−sp
pcrdtcbEo
p3hFiNptaupothe ionization of highly acidic silanols on type-A silica substrate,respectively. Even for the above phases showing the most sim-ilar selectivity to −SO3-HC-C8-L, the Fs values are still higherthan 40 [57,59], which means significant difference in selec-
Fig. 7. The log(k′/k′ethylbenzene) of the 16 solutes in Snyder’s phase character-
ization method [58] on different phases. Chromatographic conditions are the
n type-A silica), EPG phases (phases with embedded polar groups) and phenylhases were obtained from reference [44,54–60]. Chromatographic conditions:0/50 MeCN/60 mM phosphate buffer (pH 2.8), T = 35 ◦C, F = 1.0 mL/min.
han all the other phases investigated by Snyder [44,54–60]. Theery low steric resistance on our HC phase is probably causedy the rather low surface density of C8 chains (approximately.9 �mol/m2 versus a typical 2–3 �mol/m2). However, we alsooticed that the two solutes which are major determinants of S*,is- and trans-chalcone, have high aromaticity. From the phasetructures shown in Fig. 1, we can see that our HC phases pos-ess a large aromatic portion. Therefore, it is very possible thattrong �–� interactions (not included in Eq. (6)) might also con-ribute significantly to the high retention of the two chalconesnd consequently lead to the very negative S*.
After the post-synthesis gradient washing there are two typesf active hydrogen-bond donors present, namely the silanolroups released upon siloxane bond breaking and hydrox-methyl groups generated from the hydrolysis of residualhloromethyl groups. Therefore, the gradient washed HC-C8nd −SO3-HC-C8-L phases show the highest hydrogen-bondcidities of all phases tested (see the A coefficients in Fig. 6).n addition, the −SO3-HC-C8-H phase before gradient washinglso exhibits high hydrogen-bond acidity which is comparableo those of the above two gradient washed HC phases. This maye caused by the presence of hydroxymethyl groups that resulted
fter prolonged use from residual chloromethyl groups.The hydrogen-bond basicity (B coefficient) of HC-C8 andSO3-HC-C8-L are similar to the average B values of com-ercial C8 and phenyl phases. However, the HC phase with
sair
A 1182 (2008) 41–55
he higher sulfonation degree, −SO3-HC-C8-H, shows a muchigher B and it is even comparable to the average B of the EPGhases. According to Snyder, the B coefficient is postulated toe closely related to adsorbed water on the stationary phase44,54–60]. Thus, the relatively high hydrogen-bond basicity ofSO3-HC-C8-H may be caused by the sulfonyl groups them-
elves and the increased adsorption of water into the bondedhase resulting from the higher degree of sulfonation.
In Fig. 6, we see very high C coefficients for sulfonatedhases even at a pH of 2.8. On −SO3-HC-C8-H, the two basicompounds (nortriptyline and amitriptyline) were so stronglyetained that their retention could not be measured under Sny-er’s experimental conditions. Due to the lack of the data of thewo major C determinants, the C coefficient of −SO3-HC-C8-Hould not be calculated. However, it should be a very high valueecause of the presence of about 0.44 �mol/m2 sulfonyl groups.ven for −SO3-HC-C8-L with a much lower surface coveragef sulfonyl groups, the C(2.8) is still as high as 2.6.
We further compared the selectivity of the −SO3-HC-C8-Lhase with all the reversed-phases studied by Snyder (more than00 phases) using the column comparison function Fs [57]. Veryigh Fs values were found in all the comparisons. The averages is 210. −SO3-HC-C8-L shows the most different selectiv-
ty from phases that have very negative C(2.8), such as ECucleosil (C(2.8) = −3.2; Fs = 486). Only three zirconia-basedhases (ZirChrom-PBD, ZirChrom-PS and ZirChrom-EZ) andhree type-A silica-based C18 phases (Apex II C18, Resolve C18nd Supelcosil LC-18) with C(2.8) higher than 1.5 show Fs val-es lower than 95. The high C coefficients of zirconia-basedhases and type-A silica-based phases are due to the adsorptionf phosphate under the testing chromatographic conditions and
ame as in Fig. 6. (�) SB C18; (�) HC-C8; (�) −SO3-HC-C8-L after gradientcid washing; (+ and dotted line): −SO3-HC-C8-H before gradient acid wash-ng (nortriptyline and amitriptyline are strongly retained and not eluted in aeasonable time).
atog
tp
pnHppvttooltoohm
3R
pa(btee
l
lsrrt
amTptibrb“itracraoto[
auHTadewrds
tp
TL
S
H−ZPP
H. Luo et al. / J. Chrom
ivity exist between the −SO3-HC-C8-L phase and these otherhases.
The very high cation-exchange activity of the sulfonatedhase is also clearly shown in Fig. 7, where the logarithm of theormalized retentions (k′/k′
EB) of SB C18 phase and the threeC phases are plotted. The first 14 solutes on the left of thelot are either non-electrolytes or acids that are almost com-letely protonated at pH 2.8. These uncharged solutes showery similar retention trends on the four columns tested. Thehree HC phases are almost overlapped due to the similari-ies in the reversed-phase properties that govern the retentionf the 14 uncharged solutes. The most significant differenceccurs in the retention of the two basic compounds: nortirpty-ine and amitriptyline. They are substantially more retained onhe two sulfonated phases. This strong cation-exchange activityf −SO3-HC-C8 phases provides very different selectivity fromther reversed phases especially towards complex mixtures ofigh diversity with charged species, which will be discussed inore detail in our next paper.
.6. The comparison of −SO3-HC-C8 phases and otherPLC phases in LSER study
The non-electrolyte selectivity of the weakly sulfonatedhase −SO3-HC-C8-L after gradient washing was further char-cterized by the linear solvation energy relationship methodLSER) [45,46,61–65]. In this approach, the relationshipetween the free energy of retention and the various fundamen-al types of energy change involved in the solute distributionquilibrium between the stationary phase and mobile phase isxpressed by the following equation.
og k′ = log k′0 + νV2 + sπ∗
2 + a∑
αH2 + b
∑βH
2 + rR2 (7)
In Eq. (7), V2, π∗2,
∑αH
2 ,∑
βH2 , and R2 are five molecu-
ar descriptors. V2 is the computed molecular volume [66]; π∗
2tands for solute dipolarity/polarizability;∑
αH2 and
∑βH
2 rep-esent the solute’s overall hydrogen-bond acidity and basicity,espectively; and R2 is its excess molar refraction. Twenty-wo chromatographic solutes [61] were judiciously chosen from
THwm
able 5SER coefficients on different stationary phases in 50/50 MeCN/H2O
tationary phase log k′0
a LSER coefficients
v s
C-C8d,e,f −0.33±0.06 1.27±0.07 −0.16±0.05
SO3-HC-C8-Ld,e,g −0.47±0.05 1.22±0.06 −0.14±0.04
orbax C8f,h −0.25±0.05 1.41±0.06 −0.27±0.05
RPf,h 0.03±0.07 1.35±0.08 −0.39±0.06
henylf,h −0.22±0.06 0.83±0.07 −0.15±0.05
a The intercept of the regression of Eq. (7).b The squared correlation coefficient of the regression of Eq. (7).c The standard error of the regression of Eq. (7).d Phase designation is the same as in Table 1.e HC phase after gradient washing.f Chromatographic conditions: 50/50 MeCN/H2O, T = 30 ◦C, other chromatographg Chromatographic conditions: 50/50 MeCN/H2O with 0.02% TFA, T = 30 ◦C, F =h Data adapted from Refs. [61,65].
r. A 1182 (2008) 41–55 51
much larger set of solutes [62] to characterize the chro-atographic system by five coefficients, ν, s, a, b, and r.he five coefficients of the chromatographic system are com-limentary to the above five solute descriptors: ν representshe systems cohesiveness/dispersiveness; s is determined byts dipolarity/polarizability; a and b represent the hydrogen-ond acceptor basicities and hydrogen-bond donor acidities,espectively; r is related to the �- and n-electron interactionsetween the system and the solute [67]. The rR2 term is acorrection” term that accounts for the inadequacy of lump-ng dipolarity and polarizability into a single term sπ∗
2 andhe regression intercept log k′
0 is associated with the phaseatio. The chemical meaning of these terms is very clearlynd definitively elucidated in a recent review [64]. The fivehromatographic system coefficients are determined by the cor-esponding property differences between the stationary phasend the mobile phase. For the same mobile phase, the propertiesf the stationary phases can be indicated by the five coefficientshat are calculated from the multi-variable linear regressionf Eq. (7) using the solute descriptors of the 22 compounds61].
The coefficients of the different stationary phases of interestre listed in Table 5. Unlike the other RPLC phases characterizedsing unbuffered 50/50 MeCN/water in the LSER study, −SO3-C-C8-L was tested in 50/50 MeCN/water with 0.02% (v/v)FA to ensure that the charged surface is adequately bufferednd all the sulfonyl groups are balanced by protons. The stan-ard deviations for each regression coefficient, the standardrror (S.E.) and the squared correlation coefficient R2 of thehole regression are comparable to the fitting results of the
eversed-phase HC-C8 and other commercial reversed phases,emonstrating the applicability of LSER theory on this specialulfonate-modified reversed phase.
The cohesiveness/dispersiveness coefficient ν combined withhe molecular volume of the solute V2 (as shown in Eq. (7))lays a dominant role in non-electrolyte retention. As shown in
able 5, the ν coefficient follows the order of Phenyl < −SO3-C-C8-L < HC-C8 < PRP < Zorbax C8, which agrees very wellith the order of hydrophobicity coefficient H in Snyder’sethod (see Section 3.5).R2b S.E.c
a b r
−0.58±0.05 −1.48±0.07 0.20±0.06 0.991 0.04−0.55±0.04 −1.42±0.06 0.20±0.06 0.991 0.04−0.42±0.05 −1.61±0.07 0.02±0.06 0.993 0.04−0.94±0.06 −1.89±0.08 0.48±0.07 0.994 0.05−0.34±0.05 −0.99±0.07 0.09±0.07 0.980 0.04
ic conditions see Refs. [61,65].1.0 mL/min.
52 H. Luo et al. / J. Chromatogr. A 1182 (2008) 41–55
Fig. 8. Selectivity comparison of different phases via log k′ vs. log k′ plots based on the retention of the 22 LSER solutes. Chromatographic conditions are the samea 65]. (− HC-Cv
shcahCSdg
prdcocra
falaw
omapitb
bsdir0icecPt[
s in Table 5. Data of Zorbax C8, Phenyl and PRP phases are obtained from [SO3-HC-C8-L vs. PRP, R2 = 0.974, S.E. = 0.06, slope = 0.67±0.02; (C) −SO3-s. Zorbax C8, R2 = 0.957, S.E. = 0.08, slope = 0.88±0.04.
For all the five stationary phases studied, the second mostignificant contributor to the non-electrolyte retention is theydrogen-bond acidity represented by the b coefficient. The boefficient of −SO3-HC-C8-L is very close to that of HC-C8nd much more positive than those of Zorbax C8 and PRP. Thisigher hydrogen-bond acidity on HC phases compared to Zorbax8 and PRP is consistent with the results of Snyder’s method inection 3.5 and can be attributed to the relatively higher surfaceensity of silanol groups and the presence of hydroxymethylroups after gradient washing.
The increased importance of �–� interactions on our HChases compared to conventional aliphatic phases (Zorbax C8,= 0.02) and even aromatic phases (Phenyl, r = 0.09) is clearlyemonstrated by the higher r coefficients (r = 0.20). The r coeffi-ients of −SO3-HC-C8-L and HC-C8 are almost identical to eachther considering the errors, and are only lower than the r coeffi-ient of the highly aromatic PRP phase (r = 0.48). The rather highcoefficients of our HC phases result from the hyper-crosslinkedromatic network.
To further qualitatively determine the overall selectivity dif-erence, �–� plots [68] based on the retention of the 22 solutes
re generated and compared in Fig. 8. �–� plot is a plot ofog k′ from one chromatographic system versus the log k′ fromnother chromatographic system. According to Horvath and co-orkers [68], a good linear correlation between the two setsdao−
A) −SO3-HC-C8-L vs. HC-C8, R2 = 0.998, S.E. = 0.02, slope = 0.95±0.01; (B)
8-L vs. Phenyl R2 = 0.974, S.E. = 0.06, slope = 1.40±0.05; (D) −SO3-HC-C8-L
f log k′ means that the retention mechanisms of the two chro-atographic systems are similar. A slope close to unit indicates
lmost identical energetics of retention. On the other hand, aoor linear correlation of the �–� plot implies different selectiv-ty. Under the same elution conditions, the difference betweenhe two chromatographic systems reflects directly the differenceetween the two stationary phases.
In Fig. 8, the selectivity difference towards the 22 solutesetween −SO3-HC-C8-L and other four phases are clearlyhown in the �–� plots. The correlation coefficient R2, stan-ard error, and the slopes of the linear regression are also listedn the caption of Fig. 8. −SO3-HC-C8-L shows a very good cor-elation (R2 = 0.998) with its precursor HC-C8 and a slope of.95 is obtained, which means that the two phases are very sim-lar in the separation of these non-ionic solutes. For the threeommercial phases, −SO3-HC-C8-L exhibits the most differ-nt selectivity from the aliphatic phase Zorbax C8 and morelosely resembles the two aromatic phases: Phenyl phase andRP. The trend of selectivity similarity of −SO3-HC-C8-L to
he three commercial phases is the same as the HC-C8 phases49]. The above results demonstrate that the weak sulfonation
oes not significantly change the surface non-ionic propertiesnd the unique reversed-phase selectivity from the combinationf aromatic networks and aliphatic C8 chains is preserved on theSO3-HC-C8-L.
H. Luo et al. / J. Chromatog
Fig. 9. Selectivity comparison of various phases based on the retentions of the22 non-electrolyte LSER solutes via a plot of normalized k′ ratio. (�) ZorbaxCcp
intfiFomsaadHtfpt
3
−CFeiraichd1s
sp
ootpecatatrlower efficiency on the −SO3-HC-C8-H phase is likely causedby its poorer packing quality. In addition, the efficiency of basiccompounds relative to the alkylphenones clearly increases withretention. We believe this suggests that “kinetic resistance for
Fig. 10. The efficiency of bases on −SO3-HC-C8 phases. (A) Plate count vs.[TEAH+]; (B) plate count normalized to the average plate count of alkylphenonesvs. k′. Solute symbols are the same as in Fig. 4. Bold solid lines: −SO3-HC-C8-H
8; (�) Phenyl; (�) PRP; (�) HC-C8; (♦) −SO3-HC-C8-L. Chromatographiconditions are the same as in Table 5. Data on Zorbax C8, Phenyl and PRPhases are obtained from [65].
The difference in selectivity towards each solute on the var-ous phases is clearly shown in Fig. 9. The retentions were firstormalized to the retention of benzene on each phase. Afterhat, the retention ratios of each solute on different phases wereurther normalized to the ratios on Zorbax C8. Then, the normal-zed k′ ratios for each solute on different phases were plotted inig. 9. The solutes were sorted from left to right in the ascendingrder of retention on Zorbax C8. Therefore, the solutes becomeore hydrophobic from left to right. For a phase that has very
imilar properties to Zorbax C8, we should expect a flat linet 1.00. However, we see different patterns and many fluctu-tions in Fig. 9, which clearly demonstrates the existence ofifferences in selectivity among these stationary phases. TheC phases are closer to Phenyl phase in the selectivity towards
he more hydrophilic solutes and become more similar to PRPor the more hydrophobic compounds. Additionally, the two HChases, with or without sulfonyl groups, behave very similarlyo each other for all non-electrolytes.
.7. The efficiency of bases on −SO3-HC-C8 phases
Column performance as represented by the plate count onSO3-HC-C8-H prior to acid gradient washing, and −SO3-HC-8-L both before and after acid gradient washing are shown inig. 10. It should be noted that the plate counts reported arexceptionally good compared to the vast majority of classicalon exchangers based on organic polymers. We believe that thisesults from the open pore structure of the silica frameworknd the fact that the synthetic procedure (“orthogonal polymer-zation” [41]) used to make the polymer leads to a polymeroating which does not plug any pores and thus a phase that
as fast mass transfer. Keeping in mind that these are 5 �miameter particles and the plate counts are in some cases over20,000 m−1. Some of these materials with certain solutes arehowing better efficiency than the HC-C8 phase for non-polarbH−gM
r. A 1182 (2008) 41–55 53
olutes. Thus these seem to be extraordinarily efficient stationaryhases.
In Fig. 10A, the plate counts are plotted against the logarithmf the concentration of displacer (TEAH+). We see a clear trendf a decrease in plate count with increasing TEAH+ concentra-ion and that the −SO3-HC-C8-H phase has an overall poorererformance than either of the −SO3-HC-C8-L phases. How-ver, the retention of basic compounds changes with TEAH+
oncentration and the efficiencies of the neutral alkylphenonesre also much lower on the −SO3-HC-C8-H phase. Therefore,he plate counts of the basic compounds were normalized to theverage plate count of the alkylphenones and plotted againsthe k′. As shown in Fig. 10B, the normalized plate counts areoughly the same for all three −SO3-HC-C8 phases. Thus, the
efore gradient washing, (N/m)alkylphenones = 81,000; thin solid lines: −SO3-C-C8-L before gradient washing, (N/m)alkylphenones = 102,000; dotted lines:SO3-HC-C8-L after gradient washing, (N/m)alkylphenones = 100,000. Chromato-raphic conditions: 5.0 cm × 0.46 cm columns, T = 40 ◦C, F = 1.0 mL/min, 50/50eCN/H2O with 0.1% formic acid and TEA·HCl.
5 atogr.
eiaabgaii
q1c
4
paagawouthsnmsshsam
udwRbbtahaTsws−fEa1
A
sD
R
[
[
[
[[[
[[
[[
[[[
[[
[[[[[[[[
[[
[[[
4 H. Luo et al. / J. Chrom
lute binding” [69] as described by Horvath contributes signif-cantly to band broadening of the bases. According to Horvathnd Lin, the slow association and dissociation processes betweensolute and the stationary phase contribute to the plate heighty Hkin. This term decreases with increased retention when k′ isreater than 1. When the association or dissociation processesre slow and Hkin becomes important to the total band broaden-ng, we will see the efficiency increase with retention as shownn Fig. 10B.
More importantly, as shown in Fig. 10A, with a good packinguality (the average plate count of neutral compounds is around00,000 m−1), −SO3-HC-C8-L overall provides excellent effi-iency for basic compounds.
. Conclusions
A novel type of silica-based sulfonate-modified reversed-hase −SO3-HC-C8 has been synthesized by sulfonating theromatic hyper-crosslinked networks of a previously developedcid stable material [40]. Different surface coverages of sulfonylroups can be achieved by controlling the sulfonation temper-ture: approximately 0.44 and 0.1 �mol/m2 of sulfonyl groupere obtained at −41 and −61 ◦C, respectively, after 30 minf reaction. Based on the free energy of transfer per methylenenit �G0
CH2, the H in Snyder’s characterization method, and
he ν coefficient in the LSER study, it is clear that most of theydrophobicity of the HC-C8 phase has been preserved afterulfonation. The HC-C8 phase’s differential selectivity towardson-electrolyte solutes from conventional RPLC phases is alsoaintained. The special properties that provide the different
electivity from conventional RPLC phases include: the fairlytrong �-interaction from the aromatic network indicated by theigher r coefficient in LSER fits compared to the conventionalilica-based Phenyl phase; and the rather high hydrogen-bondcidity demonstrated by the quite large A coefficient in Snyder’sethod and the less negative b coefficient in the LSER analysis.Much more importantly, these −SO3-HC-C8 phases offer
nique high cation-exchange activity as shown by the extraor-inarily large C coefficient as determined by Snyder’s method,hich clearly distinguishes the sulfonated phase from all otherPLC phases. Additionally, the cation-exchange interactionetween cationic solutes and −SO3-HC-C8 phase was found toe closely correlated with the hydrophobic interaction betweenhe solutes and the stationary phase. The hydrophobicallyssisted cation exchange will greatly benefit the separation ofighly hydrophilic basic compounds, which are very hard to sep-rate by the most frequently used conventional RPLC phases.he successful application of our −SO3-HC-C8-L phase to theeparation of catecholamines, serotonin and drug metabolitesill be discussed in detail in our next paper. Very different
electivities from conventional ODS columns are found on ourSO3-HC-C8-L. High orthogonality is one of the prerequisites
or successful two-dimensional chromatographic separations.xcellent efficiency for both neutral and basic compounds islso demonstrated by the very high plate counts that are close to00,000 m−1.
[[[
A 1182 (2008) 41–55
cknowledgements
We thank the National Institutes of Health for the financialupport. We also thank Agilent Technologies Inc. (Wilmington,E, USA) for the donation of Zorbax silica and SB C18 phases.
eferences
[1] J.S. Fritz, D.T. Gjerde, Ion Chromatography, third ed., Wiley, New York,2000.
[2] J.S. Fritz, J. Chromatogr. A 1039 (2004) 3.[3] J.S. Fritz, LC Mag. 2 (1984) 446.[4] B.A. Bidlingmeyer, J.K. Del Rios, J. Korpi, Anal. Chem. 54 (1982)
442.[5] K. Croes, P.T. McCarthy, R.J. Flanagan, J. Chromatogr. A 693 (1995)
289.[6] C.T. Mant, J.R. Litowski, R.S. Hodges, J. Chromatogr. A 816 (1998) 65.[7] D.L. Crimmins, Anal. Chim. Acta 352 (1997) 21.[8] A.J. Alpert, J. Chromatogr. 499 (1990) 177.[9] T. Yoshida, J. Biochem. Biophys. Methods 60 (2004) 265.10] J. Blonder, M.C. Rodriguez-Galan, K.C. Chan, D.A. Lucas, L.-R. Yu, T.P.
Conrads, H.J. Issaq, H.A. Young, T.D. Veenstra, J. Proteome Res. 3 (2004)862.
11] K. Fujii, T. Nakano, T. Kawamura, F. Usui, Y. Bando, R. Wang, T.Nishimura, J. Proteome Res. 3 (2003) 712.
12] H. Saito, Y. Oda, T. Sato, J. Kuromitsu, Y. Ishihama, J. Proteome Res. 5(2006) 1803.
13] Primesep Columns, Methods, Applications, SIELC, 2004.14] H. Luo, L. Ma, C. Paek, P.W. Carr, submitted.15] J.L. McCallum, R. Yang, J.C. Young, J.N. Strommer, J. Chromatogr. A
1148 (2007) 38.16] J. Dai, X. Yang, P.W. Carr, J. Chromatogr. A 1005 (2003) 63.17] U.D. Neue, K. Tran, A. Mendez, P.W. Carr, J. Chromatogr. A 1063 (2005)
35.18] X. Yang, J. Dai, P.W. Carr, J. Chromatogr. A 996 (2003) 13.19] B. Paull, P.N. Nesterenko, Analyst (Cambridge, United Kingdom) 130
(2005) 134.20] A. Klingenberg, A. Seubert, J. Chromatogr. A 946 (2002) 91.21] A. Klingenberg, A. Seubert, J. Chromatogr. 640 (1993) 167.22] R.K. Iler, The Chemistry of Silica: Solubility, Polymerization, Colloid and
Surface Properties and Biochemistry, Wiley, New York, 1979.23] K. Unger, J. Schick-Kalb, K.F. Krebs, J. Chromatogr. 83 (1973) 5.24] K.K. Unger, Porous Silica, Its Properties and Use as Support in Column
Liquid Chromatography (Journal of Chromatography Library, vol. 16),Elsevier, Amsterdam, New York, 1979.
25] T. Eimer, K.K. Unger, T. Tsuda, Fresenius’ J. Anal. Chem. 352 (1995) 649.26] H. Engelhardt, G. Ahr, Chromatographia 14 (1981) 227.27] A. Berthod, J. Chromatogr. 549 (1991) 1.28] D.J. Anderson, Anal. Chem. 67 (1995) 475R.29] L.C. Sander, S.A. Wise, Crit. Rev. Anal. Chem. 18 (1987) 299.30] L. Ma, W. Carr Peter, Anal. Chem. 79 (2007) 4681.31] H.A. Claessens, M.A. van Straten, J. Chromatogr. A 1060 (2004) 23.32] K.D. Wyndham, J.E. O’Gara, T.H. Walter, K.H. Glose, N.L. Lawrence,
B.A. Alden, G.S. Izzo, C.J. Hudalla, P.C. Iraneta, Anal. Chem. 75 (2003)6781.
33] US Patent 6,686,035 B2.34] J.B. Crowther, P. Griffiths, S.D. Fazio, J. Magram, R.A. Hartwick, J. Chro-
matogr. Sci. 22 (1984) 221.35] T.K. Chambers, J.S. Fritz, J. Chromatogr. A 797 (1998) 139.36] P.A. Asmus, C.-E. Low, M. Novotny, J. Chromatogr. 123 (1976) 109.37] A. Ohkubo, T. Kanda, Y. Ohtsu, M. Yamaguchi, J. Chromatogr. A 779
(1997) 113.38] G. Huhn, H. Mueller, J. Chromatogr. 640 (1993) 57.39] A.V. Neimark, M. Hanson, K.K. Unger, J. Phys. Chem. 97 (1993) 6011.40] L. Ma, H. Luo, J. Dai, W. Carr Peter, J. Chromatogr. A 1114 (2006)
21.
atog
[
[
[
[
[[[
[[[
[[
[[
[
[
[
[
[
[
[[[
[
H. Luo et al. / J. Chrom
41] B.C. Trammell, L. Ma, H. Luo, M.A. Hillmyer, P.W. Carr, J. Am. Chem.Soc. 125 (2003) 10504.
42] B.C. Trammell, L. Ma, H. Luo, M.A. Hillmyer, P.W. Carr, J. Chromatogr.A 1060 (2004) 61.
43] B.C. Trammell, L. Ma, H. Luo, D. Jin, M.A. Hillmyer, P.W. Carr, Anal.Chem. 74 (2002) 4634.
44] N.S. Wilson, M.D. Nelson, J.W. Dolan, L.R. Snyder, R.G. Wolcott, P.W.Carr, J. Chromatogr. A 961 (2002) 171.
45] M.J. Kamlet, R.W. Taft, J. Am. Chem. Soc. 98 (1976) 377.46] M.J. Kamlet, R.W. Taft, Acta Chem. Scand. B B39 (1985) 611.47] A.J. Gordon, R.A. Ford, A Chemist’s Companion—A Handbook of Prac-
tical Data, Techniques, and References, Wiley, New York, 1972.48] K. Tanabe, T. Sano, J. Res. Inst. Catal. 10 (1962) 173.49] L. Ma, Ph.D. Thesis, University of Minnesota, 2005.50] L.R. Snyder, J.J. Kirkland, J.L. Glajch, Practical HPLC Method Develop-
ment, second ed., Wiley, New York, 1997.51] A. Sokolowski, K.G. Wahlund, J. Chromatogr. 189 (1980) 299.
52] K.E. Bij, C. Horvath, W.R. Melander, A. Nahum, J. Chromatogr. 203 (1981)65.53] A. Nahum, C. Horvath, J. Chromatogr. 203 (1981) 53.54] N.S. Wilson, M.D. Nelson, J.W. Dolan, L.R. Snyder, P.W. Carr, J. Chro-
matogr. A 961 (2002) 195.
[[[[[
r. A 1182 (2008) 41–55 55
55] N.S. Wilson, J. Gilroy, J.W. Dolan, L.R. Snyder, J. Chromatogr. A 1026(2004) 91.
56] N.S. Wilson, J.W. Dolan, L.R. Snyder, P.W. Carr, L.C. Sander, J. Chro-matogr. A 961 (2002) 217.
57] J.J. Gilroy, J.W. Dolan, L.R. Snyder, J. Chromatogr. A 1000 (2003)757.
58] J.J. Gilroy, J.W. Dolan, P.W. Carr, L.R. Snyder, J. Chromatogr. A 1026(2004) 77.
59] D.H. Marchand, K. Croes, J.W. Dolan, L.R. Snyder, J. Chromatogr. A 1062(2005) 57.
60] D.H. Marchand, K. Croes, J.W. Dolan, L.R. Snyder, R.A. Henry, K.M.R.Kallury, S. Waite, P.W. Carr, J. Chromatogr. A 1062 (2005) 65.
61] J. Zhao, P.W. Carr, Anal. Chem. 70 (1998) 3619.62] L.C. Tan, P.W. Carr, M.H. Abraham, J. Chromatogr. A 752 (1996) 1.63] P.C. Sadek, P.W. Carr, R.M. Doherty, M.J. Kamlet, R.W. Taft, M.H. Abra-
ham, Anal. Chem. 57 (1985) 2971.64] M. Vitha, P.W. Carr, J. Chromatogr. A 1126 (2006) 143.
65] J. Zhao, Ph.D. Thesis, University of Minnesota, 1999.66] J.C. McGowan, J. Chem. Technol. Biotechnol. 34A (1984) 38.67] M.H. Abraham, Chem. Soc. Rev. 22 (1993) 73.68] W. Melander, J. Stoveken, C. Horvath, J. Chromatogr. 199 (1980) 35.69] C. Horvath, H.J. Lin, J. Chromatogr. 149 (1978) 43.