49Lecture C1 – Part 3:Introduction and Theory of ChromatographyCU- Boulder CHEM 5181Mass Spectrometry & ChromatographyProf. Jose L. JimenezFall 2006Reading: Braithwaite & Smith Chapters 1 & 250Retention (or Capacity) Factor• k is like K except:– k is the ratio of total amounts, rather than of concentrations– Describes the ability of the stationary phase to retain components (same as K)– But it is a measure of actual retention propertiesMMssCVCVk =MCCKs=MMRMRtttttk−=='MsVVKk =⇒51Separation Factor•α : ability of a given stationary phase to separate two components– Substituting:– For given A & B, α= f(stationary phase, mobile phase, T)–α> 1, but for GC and HPLC, α~ 1 (because very narrow peaks)ABkk=αMRttk'=''RARBtt=α52Resolution & Selectivity• It can be shown (see book):• Very important for column design– Large changes in N are need to improve RS– Separation can be increased by increasing L– But better to improve k and αfirst!• Change the choice of stationary & mobile phase for separation of analytes of interest (A & B)• Agilent chromatography library: http://www.chem.agilent.com/scripts/chromatograms.asp⎟⎠⎞⎜⎝⎛−⎟⎠⎞⎜⎝⎛+=αα114 kkNRSABkk=α53Band Broadening Processes• What physical and/or chemical processes cause broadening of the peaks in chromatography?54Band Broadening Processes• Non-column broadening– Dispersion of analyte in• Dead volume of injector• Connection between injector & column• Connection between column & detector– Emphasis on minimizing dead volume (injectors, fittings…)• Column broadening: Van Demteer modelStationary PhaseStationary PhaseMobile PhaseDetectorLaser55Schematic of Column Chromatography III• If analyte has some affinity to the stationary phase, it will be retarded– Equilibrium– Kinetics• Molecular mass transfer: diffusion– Emerge at the detector after “retention time” tRSPMCCKs=56GC vs. HPLC ColumnsGC HPLC• Most GC columns do NOT have particles• Most HPLC columns do have particles• Why?A. Particles are needed to prevent liquid flow for being too fastB. Particles are needed in HPLC to shorten diffusion distance in MPC. Particles are not needed in GC because diffusion is very fastD. B & CE. I don’t know57Effect of Mobile Phase velocity on H• Experiment– Repeat the same separation, same column and mobile phase– Determine H vs. flow rate– Observe an optimum, H increases to both sides– Goethe: “there is nothing more practical than a good theory”Skoog & Leary, 4thEd58Van Deemter Model: “A Term” Broadening• ‘Eddy diffusion’ &unequal pathways• Molecules may travel unequal distances• Particles (if present) cause eddies & turbulence• A depends on size of stationary particles (want small) and their packing (want uniform) (or coating in TLC plate)– GC: 150 µm, HPLC: 5-10 µmPdAλ=ClickerA. A term ↑ as u ↑B. A term ↓ as u ↑C. A term D. Don’t know)(uf≠59Van Deemter Model: “B” Term• Longitudinal Diffusion• Basically molecular diff., as ifmobile phase was not moving Clicker 1A. B term ↑ as u ↑B. B term ↓ as u ↑C. B term D. Don’t know)(uf≠Clicker 2. B term isA. more imp in GCB. More imp in HPLCC. Similar importanceD. I don’t know60Van Deemter Model: “B” Term• Model for B:• γ is hindrance factor due to packing (0.7 in packed – 1 in open) and DMis molecular diffusion coeff.• B terms dominates at low uMDBγ2=AB61Van Deemter Model: “C” Terms I• Accounts for finite time for mass transfer (equil. btwanalyte in stationary and mobile phase not instantaneous)• Most important effect in GC & HPLC• CSaccounts for stationary phase mass transfer– df: stationary phase film thickness– DS: diffusion coeff. of analyte in SP– Thinner films reduce mass transfer time& broadening• But also reduce capacity of the columnsfsDdC2=62Van Deemter Model: “C” Terms II• CMaccounts for mass transfer on themobile phase interface with the SP• In packed columns:– dpis particle diameter• In open columns– dcis column diameterMPMDdC2=MCMDdC2=Clicker 1A. C term ↑ as u ↑B. C term ↓ as u ↑C. C term D. Don’t know)(uf≠Clicker 2. CMterm isA. More important in GCB. More important in HPLCC. Similar importanceD. I don’t know63Van Deemter Model of Band Broadening• Tries to explain previous experiment• H: plate height• u: average linear velocity• H: as small as possible (calculate Hmin)• Some terms decrease, other increase with u– There should be optimum u• There are alternative models (see reading)uCuCuBAHMS+++= MtLu =64Optimum Mobile Phase Velocity•We want N highest, H lowest65Optimum Mobile Phase Velocity: GC & HPLCQ: differences in A, B, C between Supercritical Fluid Chrom. &