Introduction to the theory of HPLC HPLC (High Performance Liquid Chromatography) depends on interaction of sample analytes with the stationary phase (packing) and the mobile phase to effect a separation. Following are explanations of the separation mechanisms commonly used in HPLC. In adsorption chromatography the stationary phase, properly speaking, is the liquid-solid interface Molecules are reversibly bound to this surface by dipole-dipole interactions. Since the strength of interaction with the surface is different for different compounds, residence time at the stationary phase varies for different substances thus achieving separation. Liquid-solid adsorption chromatography is most often used for polar, non-ionic organic compounds. Partition chromatography is the fundamental distribution mechanism in liquid-liquid chromatography, i. e. when both mobile phase and stationary phase are liquids. Separation by distribution is based on the relative solubility of the sample in the two phases. In normal phase partition chromatography the stationary phase is more polar than the mobile phase, in reversed phase (RP) chromatography the mobile phase is more polar than the stationary phase. Stationary phases may be either coated on to a support, or they may be chemically bonded to the surface. Normal phase partition chromatography is used for very polar organic compounds, while reversed phase chromatography is commonly used for nonpolar or weakly polar substances. Ionic compounds are often better separated by ion exchange chromatography (IEC). In this case, the stationary phase consists of acidic or basic functional groups bonded to the surface of a polymer matrix (resin or silica gel). Charged species in the mobile phase are attracted to appropriate functional groups on the ion exchanger and thereby separated. Ion pairing chromatography is an alternative to ion exchange chromatography. Mixtures of acids, bases and neutral substances are often difficult to separate by ion exchange techniques. In these cases ion pairing chromatography is applied. The stationary phases used are the same reversed phases as developed for reversed phase chromatography. An ionic organic compound, which forms an ionpair with a sample component of opposite charge, is added to the mobile phase. This ion-pair is, chemically speaking, a salt which behaves chromatographically like a non-ionic organic molecule that can be separated by reversed phase chromatography. Size exclusion chromatography (SEC) or gel permeation chromatography (GPC) uses as the stationary phase a porous matrix which is permeated by mobile phase molecules. Sample molecules small enough to enter the pore structure are retarded, while larger molecules are excluded and therefore rapidly carried through the column. Thus size exclusion chromatography means separation of molecules by size. The chromatogram below illustrates the most important parameters which characterise a separation. These parameters will be explained in the following paragraphs. t 0 tr1 w 1/2 w t R1 t R2 t R2 retention time 10% of peak height Explanation of the most important parameters to characterise a separation: peak widths: w 1/2 = peak width at half height w = band width of the peak (intersection point of the inflectional tangents with the zero line) peak symmetry is measured at 10% or peak height symmetry parameters: A = peak front at 10% of peak height to peak maximum B = peak maximum to peak end at 10% of peak height retention times t 0 = t R1 Injection dead time of a column = retention time of an unretarded substance,.. = retention times of components 1, 2.. t R2 t R1, t R2.. = net retention times of components 1, 2.. Retention: In an elution chromatographic separation substances differ from each other only in their residence time in or at the stationary phase. From this the following time definitions arise: The total retention time ( t R1 or t R2 ) is the time, which is needed by a sample component to migrate from column inlet (sample injection) to the column end (detector). The dead time t 0 is the time required by an inert compound to migrate from column inlet to column end without any retardation by the stationary phase. Consequently, the dead time is identical with the residence time of the sample compound in the mobile phase. A B 174
Introduction to the theory of HPLC The net retention time ( t R1 or t R2 ) is the difference between total retention time and dead time. That is the time the sample component remains in the stationary phase. The capacity factor k is a measure of the position of a sample peak in the chromatogram. It is specific for a given substance. k depends on the stationary phase, the mobile phase, the temperature, quality of the packing etc. The relative retention α, also known as separation factor, is the ratio between two capacity factors, where the figure in the denominator is the reference compound. The relative retention describes the ability of a system of stationary and mobile phase to discriminate between two compounds. It is independent of column construction (length, quality of packing) and flow velocity. It depends on the temperature and the properties of the mobile and stationary phases. Impurities in the mobile phase (e.g. water content) strongly influence the relative retention. Instead of the mobile phase volumetric flow rate F (ml/min) it is advantageous to use the linear velocity u (cm/sec). The linear velocity is independent of the cross section of the column and proportional to the pressure drop in the column. The linear velocity can be calculated by means of the dead time, where L is the column length in cm and t 0 the dead time in sec. The permeability K of a column describes its transmittance for a mobile phase and characterises the hydraulic resistance. The permeability of a column depends on mobile phase, temperature, column length and pressure. A change in permeability indicates a change in the packing (e.g. swelling of ion exchangers, silica gel etc.). The number of theoretical plates n characterises the quality of a column packing and mass transfer phenomena. The larger n, the more complicated sample mixtures can be separated with the column. The height equivalent of a theoretical plate h, HETP, is the length, in which the chromatographic equilibrium between mobile and stationary phase is established. Since a large number of theoretical plates is desired, h should be as small as possible. There are, of course, no real plates in a chromatographic column, since the packing is homogeneous. The value of h is a criterion for the quality of a column. h values depend on the particle size, the flow velocity, the mobile phase (viscosity) and especially on the quality of a packing. For practical reasons, the peak symmetry is measured at 10% of peak height, where A is the distance from peak front to peak maximum and B is the distance from peak maximum to peak end. Ideally symmetry should be 1, i. e. A = B. With values below 1 one speaks of fronting, with values above 1 there is peak tailing. t R1 = t R1 t 0 t R2 = t R2 t 0 t R1 t 0 t R2 t 0 k 1 = ----------------- k t 2 = ----------------- 0 t 0 α u k = ------ 2 k 1 = L --- t 0 t R1 n 16 ------ 2 t R1 2 = or n = 5.54 ------------ w w 1 2 h = L -- n B symmetry = --- A 175
ideal peak shape symmetry = 1 The following table lists frequent problems in HPLC and possible solutions for: uncommon peak shapes lack of sensitivity poor sample recovery pressure problems baseline problems leaks changing retention times As mentioned above, the ideal peak symmetry is 1.0. Though broad peaks may have a symmetry value of 1, their peak shape indicates a problem in the chromatographic system. If the symmetry deviates largely from 1, it may provide valuable information about problems in a separation system. A uncommon peak shapes 1. broad peaks 1.1. early eluting analyte due to column dilute sample 1:10 and repeat the separation overload 1.2. injection volume too large inject smaller volumes or reduce solvent strength for injection to focus the sample components 1.3. viscosity of mobile phase is too high increase column temperature or use a solvent of lower viscosity 1.4. retention times too long use gradient elution or a stronger mobile phase for isocratic elution 1.5. poor column efficiency use mobile phases of lower viscosity, elevated column temperature, lower flow rate or a packing with smaller particle size 1.6. peak broadening in the injection valve decrease size of the sample loop, or introduce an air bubble in front and back of the sample in the loop 1.7. extra column volume of the LC system too large use zero dead volume fittings and connectors; use smallest possible tubing diameter (<0,25 mm) and matched size of fittings 1.8. volume of detector cell too large use smallest possible cell volume for the sensitivity required; use a detector without heat exchanger in the system 1.9. detector time constant too slow adjust the time constant to the peak width 1.10. sampling rate of the data system is too low increase the sampling rate 1.11. only some peaks broad: late elution of analytes from a previous run flush the column with a strong eluent after each run, or end gradient at a higher concentration broad peak symmetry ~ 1 2. peak fronting 2.1. column overload decrease sample amount; increase column diameter; use a stationary phase with higher capacity 2.2. formation of channels in the column buy a new column or have the column repacked (we will be glad to inform you about our refill service) symmetry < 1 fronting 176
3. peak tailing 3.1. basic analytes: interactions with silanol groups 3.2. sample components which can form chelates: metal traces in the packing 3.3. silica-based column: silanol interactions 3.4. silica-based column: degradation at high ph values 3.5. silica-based column: degradation at high temperatures use silica-based base deactivated RP phases (e.g. NUCLEOSIL PROTECT I, NUCLEOSIL HD or NUCLEOSIL AB); use a competing base such as triethylamine; use a stronger mobile phase; switch to polymer-based columns (e.g. NUCLEOGEL RP) only use high-purity silica packings (e.g. NUCLEOSIL, NUCLEOPREP ) with their very low content of metal ions; add EDTA or another chelating compound to the mobile phase; switch to polymer columns (e.g. NUCLEOGEL ) decrease the ph value of the mobile phase to suppress ionisation of the silanol groups; increase the buffer concentration; derivatise the sample to avoid polar interactions use RP columns with good surface shielding, polymer columns or sterically protected phases; also see A. 3.1. use temperatures below 50 C 3.6. dead volume at the column head rotate injection valve quickly; use an injection valve with pressure bypass; avoid pressure pulses / replace the deteriorated column, or, if possible, open the upper end fitting and fill the void with the column packing or some silanised glass fibre wadding; have the column repacked. With our VarioPrep columns for preparative HPLC you can compensate dead volumes with the adjustable end fitting. 3.7. unswept dead volume minimise the number of connections; ensure that the rotor seal is tight; check whether all fittings are tight 3.8. beginning of peak doubling see under double peaks symmetry > 1 tailing 4. double peaks 4.1. simultaneous elution of an interfering substance 4.2. simultaneous late elution of a substance from a previous run use sample clean-up or fractionation prior to injection (e. g. SPE with CHROMABOND or CHROMAFIX ); improve selectivity by choice of another mobile or stationary phase flush the column with a strong eluent after each run, or end gradient at a higher concentration 4.3. column overload see A. 2.1. 4.4. injection solvent too strong use a weaker solvent for the sample or a stronger mobile phase 4.5. sample volume too large if the sample is dissolved in the mobile phase, the injection volume should be smaller than one-sixth of the column volume 4.6. dead volume or formation of channels in the column replace the column or, if possible, open the upper end fitting and fill the void with the same packing; have the column repacked 4.7. plugged frit install an in-line filter with 0.5 µm pore size between pump and injector to remove solids from the mobile phase, or between injector and column, to filter particulate matter from the sample / if possible, clean or replace the plugged frit 4.8. unswept volume in the injector replace the rotor of the injection valve double peak 177
5. negative peaks 5.1. RI detector: refractive index of the analyte lower than that of the mobile phase 5.2. UV detector: absorption of the analyte lower than absorption of the mobile phase reverse detector polarity to obtain positive peaks use a mobile phase with lower UV absorption; if recycling solvent, use fresh HPLC grade eluent when the recycled mobile phase starts to affect detection negative peak 6. ghost peaks 6.1. contamination only use HPLC grade solvents / flush the column to remove impurities 6.2. late elution of an analyte from a previous see A. 4.2. run 6.3. unknown interfering substances in the sample 6.4. in ion pairing chromatography: disturbed equilibrium 6.5. in peptide mapping: oxidation of trifluoroacetic acid 6.6. in RP chromatography: contaminated water 7. spikes use sample clean-up or fractionation prior to injection (e. g. SPE with CHROMABOND or CHROMAFIX ) prepare the sample in the mobile phase; reduce the injection volume prepare fresh trifluoroacetic acid solution daily; add an antioxidant check the suitability of the water by passing different amounts through the column and measure the peak height of the impurity as a function of enrichment time; purify the water by running it through an old RP column or use HPLC grade water 7.1. air bubbles in the mobile phase degas the mobile phase; install a back pressure restrictor at the detector outlet; ensure that all fittings are tight 7.2. column was stored without endcaps always store columns tightly capped; flush reversed phase columns with degassed methanol Spikes B lack of sensitivity 1. detector attenuation set too high reduce detector attenuation 2. not enough sample injected increase amount of sample for injection 3. sample loop of injector underfilled overfill loop with sample 4. sample loss during sample preparation use an internal standard for sample preparation and optimise your method 5. sample loss on column see paragraph C: poor sample recovery 6. autosampler line blocked check the flow and clear any blockages 7. peaks outside the linear range of the detector 8. only during first few injections: sample absorption in sample loop of injector or column dilute or enrich the sample until the concentration is in the linear range of the detector condition sample loop and column with concentrated sample 178
C poor sample recovery 1. adsorption on stationary phase increase mobile phase strength to minimise adsorption; for basic compounds add a competing base or use a base deactivated packing like NUCLEOSIL HD, NUCLEOSIL PROTECT I or NUCLEOSIL AB 2. chemisorption on stationary phase ensure no reactive groups are present; use a column or separation mechanism better suiting the problem; use polymer-based columns like e. g. NUCLEOGEL 3. acidic substances: <90% yield, irreversible adsorption on active groups use endcapped or polymeric stationary phases; acidify the mobile phase 4. basic substances: <90% yield, irreversible adsorption on active groups 5. hydrophobic interactions between stationary phase and biomolecules use endcapped, base deactivated, sterically protected phases with dense surface coverage or polymer-based reversed phase materials (NUCLEOSIL HD, NUCLEOGEL RP); add a competing base to the mobile phase use short-chain reversed phase packings with 300 Å pore size; as an alternative you may use hydrophilic stationary phases or ion exchangers 6. adsorption of proteins use another HPLC mode to reduce nonspecific interactions; use a mobile phase containing reagents which enhance solubility of the proteins, strong acids or bases (only with polymer-based columns) or detergents like SDS 7. adsorption on tubing and other hardware components use inert tubing and fittings made from e.g. PEEK or titanium D pressure problems 1. high back pressure 1.1. viscosity of mobile phase too high use a solvent of lower viscosity or increase the temperature 1.2. particle size of packing too small use a packing with larger particle size (e. g. 7 µm instead of 5 µm) 1.3. for polymer-based columns: swelling of the adsorbent caused by eluent changes use only solvents compatible with the column; check proper eluent composition; consult instructions for use for solvent compatibility; use a column with a higher degree of cross-linking 1.4. salt precipitation especially in reversed phase chromatography with high proportions of organic solvents in the mobile phase; ensure that the solvent composition is compatible with the buffer concentration; reduce the ionic strength and the ratio organic : aqueous in the mobile phase; premix the mobile phase 1.5. contamination at the column inlet improve sample clean-up; use guard columns; backflush column with a strong solvent in order to dissolve the impurity 1.6. microbial growth in the column use a mobile phase with at least 10% organic solvent; prepare fresh buffer daily; add 0.02% sodium azide to aqueous mobile phases; for storage equilibrate the column with at least 25% organic solvent and without buffer 1.7. plugged frit in in-line filter or guard replace frit or guard column column 1.8. plugged frit at column inlet replace the end fitting or the frit 1.9. when the injector is disconnected from the column: plugged injector clean the injector or replace the rotor 2. pressure fluctuations 2.1. air bubbles in the pump degas the solvent; flush the solvent with helium 2.2. leak in liquid lines between pump and column tighten all fittings; replace defective fittings; tighten rotor in the injection valve 179
3. increasing pressure 3.1. accumulation of solids at the column head 3.2. in aqueous / organic solvent systems: precipitation of buffer components filter sample and mobile phase; use an 0.5 µm in-line filter; disconnect the contaminated column and clean it by back-flushing; replace plugged inlet frits; replace the guard column ensure that the solvent composition is compatible with the buffer concentration; reduce the ionic strength and the ratio of organic : aqueous in the mobile phase 3.3. plugged liquid lines systematically disconnect system components from the detector end to the blockage; clean or replace the plugged component 4. decreasing pressure 4.1. insufficient flow from the pump loosen the cap on the mobile phase reservoir 4.2. leak in liquid lines between pump and column tighten all fittings; replace defective fittings; tighten the rotor in the injection valve 4.3. leaking pump check valve or seals clean the check valve; replace defective check valves or seals 4.4. air bubbles in the pump degas all solvents; check for blockage between solvent reservoir and pump; if necessary replace the frit in the inlet line E baseline problems 1. baseline drifting to lower absorption 1.1. with gradient elution: UV absorption of mobile phase A use non-uv-absorbing HPLC grade solvents for your mobile phases; if a UV-absorbing solvent is inevitable, use a UV-absorbing additive in mobile phase B 2. baseline drifting to higher absorption 2.1. accumulation and elution of impurities use sample clean-up or fractionation prior to injection; use only HPLC grade solvents; clean the contaminated column with a strong solvent 2.2. with gradient elution: UV absorption of mobile phase B use a higher wavelength of the UV detector; use non- UV-absorbing HPLC grade solvents for your mobile phases; if a UV-absorbing solvent is inevitable, use a UV-absorbing additive in mobile phase A 3. undulating baseline 3.1. temperature changes in the room monitor or avoid changes in room temperature; isolate the column or use a column oven; cover the RI detector to protect it from air currents 180
4. baseline noise 4.1. continuous: detector lamp problem or replace the UV lamp or clean the detector cell dirty detector cell 4.2. periodic: pump pulses repair or replace the pulse damper; purge any air from the pump; clean or replace the check valves 4.3. random: accumulation of impurities use sample clean-up or fractionation prior to injection; use only HPLC grade solvents / backflush contaminated column with a strong solvent 4.4. spikes: air bubble in the detector see A. 7.1. 4.5. spikes: column temperature higher than the boiling point of the solvent use lower working temperature 4.6. occasional sharp spikes: external electrical interferences use a voltage stabiliser for your LC system or use an independent electrical circuit for your chromatography equipment 5. baseline noise during gradient elution or isocratic proportioning 5.1. insufficient solvent mixing mix by hand, or if possible use solvents of lower viscosity; monitor proportioning precision by spiking one solvent with a UV-absorbing substance and measure the resulting detector output 5.2. malfunctioning proportioning valves clean or replace the proportioning valve; use partially premixed solvents 6. disturbance at dead time 6.1. air bubbles in the mobile phase degas the mobile phase or use premixed eluents 6.2. difference in refractive index between injection solvent and mobile phase normal with many samples; if possible, use the mobile phase as solvent for the sample F leaks 1. column loses stationary phase replace column! 2. serious leaks at column or fittings tighten loose fittings or use new fittings 3. serious leak at the detector replace defective detector seals or gaskets 4. serious leak at the injector replace worn or scratched valve rotors 5. serious leak at the pump replace defective pump seals; check the piston for scratches and replace piston, if necessary G changing retention times 1. decreasing retention times 1.1. column overloaded with sample reduce the amount of sample or use a column with larger diameter 1.2. increasing flow rate check and if necessary adjust the pump flow rate 1.3. active groups at the stationary phase use a mobile phase containing an organic solvent (modifier) or a competing base, increase the buffer strength; use a packing with higher surface coverage 1.4. loss of bonded stationary phase replace column; for silica adsorbents use mobile phases between ph 2 and ph 8 181
2. increasing retention times 2.1. changing mobile phase composition cover the solvent reservoirs; ensure that the gradient system supplies the proper composition; if possible, mix the mobile phase by hand 2.2. decreasing flow rate check and if necessary adjust the pump flow rate; check for pump cavitation; check for leaking pump seals and other leaks in the system 2.3. loss of bonded stationary phase for silica adsorbents use mobile phases between ph 2 and ph 8 3. fluctuating retention times 3.1. only during first few injections: active condition the column with concentrated sample groups 3.2. insufficient buffer capacity use buffer concentrations above 20 mm 3.3. insufficient mixing of the mobile phase ensure that the gradient system supplies a mobile phase with constant composition; compare with manually mixed eluents; use partially premixed mobile phases 3.4. selective evaporation of one component from the mobile phase cover the mobile phase reservoirs; avoid vigorous flushing with helium; prepare fresh mobile phase 3.5. accumulation of impurities flush the column occasionally with a strong solvent, replace the guard column more frequently 3.6. fluctuating column temperature ensure that the room temperature is constant; if necessary, thermostat or isolate the column 3.7. Leaks see paragraph F 4. changing retention times resulting from insufficient equilibration 4.1. isocratic separation pass 10 to 15 column volumes of mobile phase through the column for equilibration 4.2. gradient elution increase equilibration time with mobile phase A in order to obtain constant retention times for early peaks, also: pass at least 10 column volumes of eluent A through the column for gradient regeneration 4.3. reversed phase ion-pairing chromatography increase the equilibration time; in ion-pairing chromatography sometimes 50 column volumes may be required for equilibration; long-chain ion-pairing reagents require more time; if possible, use ion-pairing reagents with shorter alkyl chains 182