Effect of pressure drop on solute retention and column efficiency in supercritical fluid chromatography

 Hinduismus

 5 views
of 11
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Description
Effect of pressure drop on solute retention and column efficiency in supercritical fluid chromatography
Share
Tags
Transcript
  J.Sep.Sci.2008, 31 ,1279–1289 A.Rajendran etal.  1279 ArvindRajendran 1 TimothyS.Gilkison 2 MarcoMazzotti 2 1 School of Chemical andBiomedical Engineering,Nanyang TechnologicalUniversity, Singapore 2 Institute of Process Engineering,Department of Mechanical andProcess Engineering, ETHZurich, Zurich, Switzerland OriginalPaper EffectofpressuredroponsoluteretentionandcolumnefficiencyinsupercriticalfluidchromatographyPart2:Modifiedcarbondioxideasmobilephase* The effect of pressure drop on the performance of supercritical fluid chromatogra-phy systems using a modified mobile phase (carbon dioxide + ethanol) was studied.Experiments were performed on a Lichrospher-RP-18 column with phenanthrene asa solute. A wide range of back pressures (130 to 210 bar) and modifier concentra-tions (2 to 7% w/w) have been explored. Experiments yielding both small and largepressure drops were performed. From these experiments, parameters to describepressure drop, retention, and column efficiency were extracted, and were used tosimulate the dynamics of the chromatographic column. A good match between theexperimentally measured and calculated values of pressure drop, retention times,and column efficiency was observed. At low back pressure and modifier composi-tion, significant loss of column efficiency was observed. Keywords:  Efficiency / Modeling / Pressure drop / Simulation / Supercritical fluid chromatogra-phy /Received:October10,2007;revised:December5,2007;accepted:December7,2007DOI10.1002/jssc.200700488 1 Introduction The use of environmentally benign “green” solvents isbeing intensely pursued in the chemical process indus-try. In particular, supercritical carbon dioxide is success-fully substituting organic solvents for several applica-tions. Supercritical CO 2 due to its non-toxic nature, mildcritical properties, and low cost has found applicationsin extractions, heterogeneous catalysis, polymer process-ing, and chromatography. In particular, supercriticalfluid chromatography (SFC) has matured from being ananalytical tool to an effective process for preparative sep-arations [2–5]. Further, the use of supercritical CO 2  as asolvent for multicolumn simulated moving bed chroma-tography(SMB)hasalsobeendemonstrated[6–10].The applicability of supercritical fluids for chromatog-raphy stems from their unique properties. Under typicaloperating conditions, these fluids show liquid-like den-sities and gas-like viscosities. Hence, compared to highperformance liquid chromatography (HPLC), it is possi-ble to operate preparative SFC units at lower pressuredrops without compromising on column efficiency. Thesolvent power of pure supercritical fluids is a function of the fluid phase pressure. This property can be skillfully usedtoconcentratetheproductbyreducingthepressureand passing it through a cyclone. This results in reducedpowerconsumptionfortheconcentrationoftheproduct[11]. The retention time of solutes in SFC is greatly influ-enced by the density of the mobile phase. In general, sol-utes show weaker retention at higher densities. Thisbehavior stems from the fact that the solvent showsincreased solubilities at high pressures and competes foradsorption sites with the solute [12–15]. This is a uniqueproperty that allows the implementation of gradients inpreparative applications, which can potentially lead toincreasedproductivities[8,16,17].Though CO 2  possesses the above mentioned advan-tages,owingtoitsnon-polarnatureitshowspoorsolubil-itytowardspolarsolutes.Inordertoovercomethisshort-coming, polar modifiers such as alcohols are added tothe mobile phase. The modifier has a twofold influenceon the retention of the solute. First, the addition of amodifier increases the solvent power of the mobilephase. Secondly, the modifier competes for adsorptionsites on the stationary phase, hence lowering the Henry constant of the solute. These effects have been observed Correspondence:  Professor Arvind Rajendran, School of Chemi-cal and Biomedical Engineering, Nanyang Technological Univer-sity, 62 Nanyang Drive, Singapore 637459 E-mail:  arvind@ntu.edu.sg Fax:  +65-6794-7553 i  2008WILEY-VCHVerlagGmbH&Co.KGaA,Weinheim www.jss-journal.com * Part 1 of this study: see [1].  1280 A.Rajendran etal.  J.Sep.Sci.2008, 31 ,1279–1289 in the literature and semi-empirical relations have beenusedtodescribethem[18–20].The effect of pressure drop on retention and columnefficiency has been the focus of various studies [1, 21–26]. The challenges in describing the effect of pressuredrop on retention and column efficiency arise from thefact that pressure drop results in the axial variation of the fluid properties,  e.g. , density, viscosity, diffusivity,and fluid velocity. The variation of these propertiesdirectly influences the retention of the solute and thecolumnefficiency.Inaddition,theexpansionofthefluidalong the column can lead to the development of axialand radial temperature gradients that result in reducedcolumn efficiencies [25–28]. Most of the studies thatdealt with pressure drop effects used pure CO 2  as amobile phase. However, most practical applications usemodified mobile phases and hence it is important tostudytheseeffectsundermorepracticalscenarios.This study is a continuation of a previous one thatinvestigated the effect of pressure drop on the retentionand column efficiency of the system consisting of phe-nanthrene (solute) and Lichrospher RP-18 column withpure CO 2 as a mobile phase [1]. The present study retainsthe solute and the column of the previous one whileinvestigating the use of a modified mobile phase consist-ing of CO 2  and ethanol. Experiments have been carriedout at four back pressure settings, namely, 130, 150, 180,and210bar.Ateachbackpressuresettingthreemodifierconcentrations,  viz . 2, 5, and 7% (w/w of ethanol), havebeen chosen and a wide range of flow rates have beenconsidered.Allexperimentswerecarriedoutatatemper-ature of 65 8 C. The pressure drop, retention time and theheight equivalent to a theoretical plate (HETP),  i.e. , key parameters that relate to the hydrodynamics, retention,and column efficiency, respectively, have been meas-ured. By using parameters obtained from experiments atlow pressure drops, the performance of the chromato-graphic column at larger pressure drops is studied. Theexperimental and calculated values are compared andtheresultsarediscussed. 2 Materialsandmethods 2.1 Materials Phenanthrene (purity   A 97%), toluene (purity 99.7%), andanalytical grade ethanol were obtained from FlukaChemie, Buchs, Switzerland. Carbon dioxide (99.995%pure) was obtained from Pangas, Luzern, Switzerland. A Lichrospher RP-18 column (4 mm diameter, 125 mmlength, 5  l m particle size) obtained from Merck, Darm-stadt,Germanywasusedfortheexperiments. 2.2 Experimentalset-upandprocedure The experimental set-up used for this study has beendescribed before [10, 29]. However, for the sake of com-pleteness the unit shown in Fig. 1 is described briefly.Two syringe pumps, ISCO 260D and ISCO  l L-500 (ISCO,Nebraska, USA), were used to deliver the CO 2  and themodifier streams, respectively. The two streams weremixed at a mixing tee followed by a short column filled withglassbeadstoensurecompletemixing.Amotorizedinjectionvalve(ValcoC14W,ValcoInstruments,Houston,TX, USA) with a 60-nL internal sample loop was used toinject a pulse of the sample that consisted of the solutephenanthrene dilutedintoluene (2% w/w).AUVdetector(JASCO UV-1570, Omnilab, Mettmenstetten, Switzerland) was used to measure the elution profile while a back pressure regulator (JASCO BP-1580-81, Onmnilab, Mett-menstetten, Switzerland) located downstream of the UV detector was used to regulate the pressure of the system.The chromatographic column and the injection valve were housed in a temperature controlled water bath.Pressure transducers (Trafag-8891, Trafag, Maennendorf,Switzerland), with an accuracy of   l 0.4 bar, were placedboth upstream of the injection valve and downstream of the UV detector. Note that the capillaries connecting thecolumn to the injection valve and to the UV detector hadaninner diameter of0.5 mm.Thiswasdifferentfrom theprevious work where capillaries with an inner diameterof0.12 mmwereused[1].The experiments were initiated by setting the back pressure regulator at a desired level and by starting theCO 2  pump and the modifier pump. The system was thenallowed to reach a hydrodynamic steady state after which the solute pulse was injected. Under each experi-mental condition, the injections were repeated at leastthreetimesinordertoensurereproducibility. 3 Resultsanddiscussion 3.1 Pressuredrop The pressure profile along the column directly affectstransport properties and retention characteristics.Hence, the pressure drop in the column is a key param-eter that has to be described accurately. The pressuretransducers were positioned upstream of the injection valveanddownstreamoftheUVdetectorinordertomin-imize extra-column effects. Hence, under the operatingconditions, the measured pressure drop corresponded tothe contributions from the chromatographic columnand the capillaries connecting the column to the injec-tion valve and to the UV detector. The method for esti-mating  D  P  col | expt , the pressure drop of the column alone,has been discussed earlier [1]. This involves estimating D  P  us | expt  and  D  P  ds | expt , the pressure drop contributionsfrom the upstream and downstream respectively andsubtracting them from the measured pressure drop, D  P  expt . The pressure drops thus calculated are listed inTables 1,2,and3.Itcanbeobservedthatforafixedmodi- i  2008WILEY-VCHVerlagGmbH&Co.KGaA,Weinheim www.jss-journal.com  J.Sep.Sci.2008, 31 ,1279–1289 OtherTechniques 1281 fier concentration, the pressure drop at a particular flow rate increases with increasing back pressure level. Fur-ther, for a fixed back pressure setting, the pressure dropat a particular flow rate increased with increasing modi-fier concentration. This can be rationalized, as theincrease of either the modifier concentration at a fixedpressure, or the increase of the pressure at a fixed modi-fierconcentration,leadstoanincreaseindensityandvis-cosity of the mobile phase, which will result in anincreasedpressuredrop.From the calculated  D  P  col | expt , it is possible to estimatethe value of   b  in the Darcy's equation [1]. A value of  b  = 3.83 6 10 10 cm –4  was obtained by fitting the Darcy'sequation to  D  P  col | expt . This is larger than  b  =2.54 6 10 10 cm –4 , the value obtained in the previousstudy [1].Theprolongedoperation ofthe column atlargepressure drops could have led to the breakage of station-ary phase particles and can possibly explain a higher value of   b . All calculations reported in the current work arebasedon b  = 3.83 6 10 10 cm –4 . i  2008WILEY-VCHVerlagGmbH&Co.KGaA,Weinheim www.jss-journal.com Table 1.  Measured and calculated values of pressure drops at 2.0% (w/w) of modifier. Backpressure(bar)  P  out (bar) FlowrateatCO 2 pump(mL/min) D  P  us|expt (bar)  D  P  col|expt (bar)  D  P  ds|expt (bar)  D  P  expt (bar)  D  P  pred (bar)130 137.5 0.3 0.0 1.5 0.0 1.5 1.6137.6 0.5 0.1 2.5 0.1 2.7 2.8137.4 0.7 0.1 3.5 0.2 3.8 4.0137.4 1.0 0.2 4.8 0.3 5.3 5.9137.6 2.0 0.7 9.1 1.2 10.9 12.7137.7 3.0 1.4 13.4 2.4 17.2 20.2150 157.6 0.3 0.0 1.6 0.0 1.7 1.7157.5 0.5 0.1 2.5 0.1 2.6 2.9157.6 0.7 0.1 3.5 0.2 3.8 4.1157.5 1.0 0.2 4.8 0.3 5.4 6.0157.4 2.0 0.7 9.7 1.1 11.4 13.0157.7 3.0 1.3 14.1 2.2 17.6 20.6180 187.6 0.3 0.0 1.7 0.0 1.7 1.7187.2 0.5 0.1 2.8 0.1 3.0 3.0187.1 0.7 0.1 3.5 0.2 3.8 4.3186.9 1.0 0.2 5.7 0.3 6.2 6.4187.1 2.0 0.6 11.4 1.0 13.0 13.6187.1 3.0 1.3 17.7 2.1 21.0 21.5187.5 4.0 2.1 24.0 3.5 29.6 30.1187.5 5.0 3.1 32.6 5.2 40.9 39.3210 217.0 0.3 0.0 1.8 0.0 1.8 1.8216.8 0.5 0.1 3.1 0.1 3.3 3.2217.4 0.7 0.1 4.0 0.2 4.3 4.6217.3 1.0 0.2 5.9 0.3 6.4 6.7217.2 2.0 0.6 12.4 1.0 14.1 14.2217.4 3.0 1.3 19.2 2.1 22.5 22.5217.9 3.5 1.6 23.8 2.7 28.2 26.9218.6 4.0 2.1 26.7 3.4 32.2 31.4 Figure 1.  Schematic of the SFC plant used in the study. Thick lines between the injection valveand column inlet and between column outlet and UV detector are capillaries with an internal diame-ter of 0.5 mm. The equipment indicated within dotted lines were thermostated at 65 8 C.  1282 A.Rajendran etal.  J.Sep.Sci.2008, 31 ,1279–1289 For the case when the mobile phase consisted of a mix-ture of CO 2  and ethanol, a Peng–Robinson equation of state with a Melhelm mixing rule was used to calculatethefluidphasedensity[30].Themixtureviscositywascal-culatedusingtheequivalentCO 2  viscosity correspondingto the mixture density. Using the parameters listedabove, the pressure drop was calculated by the methoddescribedin Rajendran  et al.  [1].For the calculationof thepressuredrop,bothdensityandviscositywerecalculatedalong the column length. The predicted pressure drops, D  P  pred , along with the experimentally measured ones, arelisted in Tables 1, 2, and 3. It can be seen that the experi-mentally measured pressure drop is well predicted by the model. The average deviation between the predictedand calculated pressure drops is 5%. The solution of thehydrodynamics equations also yields the profiles of den-sity, viscosity, and velocity along the column that shallbe used for the determination of the retention time andtheHETP. 3.2 Retentiontime In an SFC system operated at a constant inlet mass flow rate, the volumetric flow rate varies along the column withthepressureasthemassflowrateremainsconstant.Hence, in this study, key parameters such as the resi-dence time and the HETP shall be analyzed as functionsofmassflowrate.The experimentally measured retention time,  t   Ri  expt corresponds to the first moment of the chromatogramandistheeffectoftwocontributions: t   Ri  expt ¼ t  R  i  þ t  R  i  extra  ð 1 Þ  where  t   Ri  expt  and  t   Ri  extra  represent the retention times inthe column and in the extra-column volumes, respec-tively. It is important to correct for the time spent by thesolute in the extra-column volumes that include the con-necting capillaries and the detector. In order to estimatethe extra-column contribution, the column was replacedby a zero-dead volume connector and pulse injections were made. From the retention time of these pulses, theextra-column dead volume was estimated. Using thedead volume and the volumetric flow rates, the timesspent in the extra-column volumes were calculated andthen subtracted from the experimentally measuredretention times. The experimentally measured retentiontimes corrected for dead times in the capillaries areshown in Fig. 2. It can be seen that for a particular mass i  2008WILEY-VCHVerlagGmbH&Co.KGaA,Weinheim www.jss-journal.com Table 2.  Measured and calculated values of pressure drops at 5.0% (w/w) of modifier. Backpressure(bar)  P  out (bar) FlowrateatCO 2 pump(mL/min) D  P  us|expt (bar)  D  P  col|expt (bar)  D  P  ds|expt (bar)  D  P  expt (bar)  D  P  pred (bar)130 137.4 0.3 0.0 1.6 0.0 1.6 1.7137.3 0.5 0.1 2.5 0.1 2.6 2.9137.4 0.7 0.1 3.5 0.2 3.8 4.2137.4 1.0 0.2 4.8 0.3 5.4 6.1150 157.6 0.3 0.0 1.6 0.0 1.7 1.7157.5 0.5 0.1 2.8 0.1 3.0 3.0157.1 0.7 0.1 4.0 0.2 4.3 4.3157.3 1.0 0.2 5.5 0.3 6.1 6.3157.1 2.0 0.7 11.9 1.1 13.7 13.6157.1 3.0 1.3 18.9 2.2 22.5 21.6157.3 4.0 2.2 26.5 3.7 32.5 30.3158.1 5.0 3.2 35.1 5.6 43.9 39.7180 187.2 0.3 0.0 1.8 0.0 1.8 1.8187.2 0.5 0.1 3.0 0.1 3.2 3.2187.2 0.7 0.1 4.2 0.2 4.4 4.6187.2 1.0 0.2 5.9 0.3 6.4 6.7187.2 2.0 0.6 12.2 1.1 13.9 14.3187.3 3.0 1.3 18.5 2.2 22.0 22.6187.6 4.0 2.2 25.3 3.6 31.1 31.5187.7 5.0 3.2 33.4 5.4 42.0 41.2210 216.7 0.3 0.0 2.1 0.0 2.1 1.9216.9 0.5 0.1 3.2 0.1 3.4 3.3216.9 0.7 0.1 4.4 0.2 4.7 4.8217.0 1.0 0.2 6.3 0.3 6.8 7.0217.2 2.0 0.6 12.8 1.0 14.5 14.9217.1 3.0 1.3 19.8 2.1 23.2 23.6217.0 3.5 1.7 23.8 2.8 28.3 28.1217.5 4.0 1.7 22.4 3.6 27.7 32.9  J.Sep.Sci.2008, 31 ,1279–1289 OtherTechniques 1283 flow rate, the retention time decreases with increasingback pressure setting. Further, at a particular back pres-surelevel,theadditionofthemodifierleadstodecreasedretentiontimes.The adsorption isotherm for a system in which the sol-uteishighlydilutedcanberepresentedas n  i  ¼  H  i c i  ð 2 Þ  where  H  i  is the Henry constant. In SFC, as discussed ear-lier, the Henry constant is dependent on the density of the mobile phase. Hence, under the conditions wherethe density of the mobile phase is not uniform along thecolumn length, the retention time of the solute is givenby: t  R  i  ¼ Z   L 0 1 x ð z Þ  dz ¼ Z   L 0 1 v ð z Þ  1 þ  1  e b e b  H  i ð z Þ   dz  ð 3 Þ  where x ( z )isthewavevelocityofthesolute[1].Itisworthnoting that the above equation is derived under theassumption that the adsorption of the solute does notchange the mobile phase flow rate. In fact, this assump-tion can be validated when dilute mixtures are injectedinto the column [1]. Under operating conditions wherethe density variation along the column is negligible, thefollowingapproximationscanbemade: q ð z Þ  qq  ð 4a Þ v ð z Þ  vv  ð 4b Þ  H  i ð z Þ  H  i ð  qq Þ ð 4c Þ  where   qq  is the arithmetic mean of the inlet and exit den-sities. Using the above assumptions, Eq. (3) can bereducedto t  R  i  ¼  Lv ð  qq Þ  1 þ  1  e b e b  H  i ð  qq Þ   dz  ð 5 Þ Hence, from chromatographic runs performed at con-ditions where the density gradient across the column isnegligible, Eq. (5) can be used to calculate  H  i . This can beachieved by operating the unit at low flow rates. In thepresent study, density drops less than 3% across the col- i  2008WILEY-VCHVerlagGmbH&Co.KGaA,Weinheim www.jss-journal.com Table 3.  Measured and calculated values of pressure drops at 7.0% (w/w) of modifier. Backpressure(bar)  P  out (bar) FlowrateatCO 2 pump(mL/min) D  P  us|expt (bar)  D  P  col|expt (bar)  D  P  ds|expt (bar)  D  P  expt (bar)  D  P  pred (bar)130 137.5 0.3 0.0 2.0 0.0 2.1 1.7137.6 0.5 0.1 3.0 0.1 3.2 3.0137.7 0.7 0.1 4.1 0.2 4.4 4.3137.7 1.0 0.2 5.8 0.3 6.3 6.3137.7 2.0 0.7 11.7 1.1 13.6 13.6137.4 3.0 1.4 17.9 2.3 21.6 21.6137.9 4.0 2.3 24.8 3.9 30.9 30.4138.5 5.0 3.3 32.2 5.8 41.4 39.9150 157.3 0.3 0.0 1.9 0.0 1.9 1.8157.3 0.5 0.1 3.1 0.1 3.2 3.1157.1 0.7 0.1 4.4 0.2 4.7 4.5157.0 1.0 0.2 6.2 0.3 6.8 6.6157.5 2.0 0.7 12.7 1.1 14.4 14.1157.5 3.0 1.4 19.8 2.3 23.4 22.3158.0 4.0 2.2 27.1 3.8 33.1 31.3158.5 5.0 3.3 35.1 5.6 44.0 41.0180 187.6 0.3 0.0 2.1 0.0 2.2 1.9187.6 0.5 0.1 3.4 0.1 3.6 3.3187.2 0.7 0.1 4.7 0.2 5.0 4.7187.2 1.0 0.2 6.8 0.3 7.3 6.9187.3 2.0 0.7 13.8 1.1 15.5 14.8187.7 3.0 1.3 21.3 2.2 24.9 23.3187.7 4.0 2.2 29.5 3.7 35.4 32.6188.5 5.0 3.3 37.3 5.5 46.1 42.6210 217.5 0.3 0.0 2.2 0.0 2.2 2.0217.2 0.5 0.1 3.6 0.1 3.7 3.5217.3 0.7 0.1 4.9 0.2 5.1 4.9217.0 1.0 0.2 7.3 0.3 7.8 7.2217.1 2.0 0.7 14.8 1.1 16.5 15.4217.2 3.0 1.3 22.7 2.2 26.2 24.4217.8 3.5 1.8 26.3 2.9 30.9 29.1217.5 4.0 2.2 31.3 3.7 37.2 34.0
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks