Agilent Optimizing the Agilent 1100 Series high throughput LC/MS system
Contents
1. Optimizing the Agilent 1100 Series high throughput LC MS system Application Mark Stahl Abstract This Application Note describes the optimization of the Agilent 1100 Series high throughput LC MS system for different analytical tasks We describe in detail how to optimize chromatography diode array and MSD detection to archieve highest sample throughput in combination with substance identification how to acquire quantitative information while maintaining very high sample throughput and how to achieve fast analysis of very low sample amounts one Agilent Technologies Introduction Due to increasing automation and the enormous progress in combinatorial chemistry drug discovery and many other fields in organic and biological chemistry the amount of samples arising per day has been and still is growing enormously Especially central analytical labs of large enterprises have to analyze sample numbers of up to several thousands per day These samples result from totally different origins thus lead ing to totally different demands for the analysis In many cases substance identification is the goal in others an exact quantifica tion may be required For most issues sensitivity is not a limiting factor but for many biochemical analysis it is one of the major requirements In this Application Note analyses have been grouped into three different categories depending on the information required 1 High2. 2 081 0 13 2 245 0 13 Area 15750000 2 76 6415000 2 43 Mean Peak 3 Standard deviation Mean Peak 4 Standard deviation Peak 3 Peak 4 UV 254 nm RT min 2 591 0 07 2 591 0 13 Area 1132 0 78 0 07 0 26 MSD EIC RT min 2 623 0 07 3 49 0 05 Area 5555000 2 25 0 13 2 55 Table 4 Precisions of high throughput analysis on two parallel columns 10 leads to more measurement time per single step thus resulting in high sensitivity and accuracy A stepsize around 0 2 Da may be an acceptable compromise for quantitative analysis When using scan mode for quantification the creation of extracted ion chromatograms for the substances of interest is necessary An even better way resulting in much higher sensitivity and accuracy is the use of the SIM selected ion monitoring mode Here the quadrupole is not scanned but locked at the masses of interest thus leading to higher sensitivity and exacter quantification For an ion trap MS accumulation time and space charge density also have to be controlled Here a compromise between ion accumulation and amount of data points per chro matographic peak has to be drawn Additionally you have to ensure that not too many ions are loaded into the trap thus resulting in space charge effects leading to poorer resolution mass defects and less exact quantification To ensure a reliable and specific quantification the MRM mode can be used This allows sensitive determination of an intensive quantifie
3. with the gradient Thus k is not a good parameter to characterize gradient separations and it is replaced by k Plate height Figure 1 k 0 87tg ram with Vy column void volume A B gradient percent range S constant for a given solute solvent combination F flow rate and tg gradient time Due to this the formula for the selectivity a changes to ke ky These formulas show that in gradient elution changing flow rates has the same effect as changing the gradient For example at half flow rate the same resolution is achieved with a doubled gradient length Thus to achieve comparable gradient separations the gradient volumes gradient time multiplied with flow rate have to be held constant As high flow rates are used for high throughput analysis to achieve lowest run times this in this case leads to the advantage that shorter gradients can be used without loosing much chromato graphic resolution As the flow rate is closely related to the column diameter this has also to be taken into consideration for up and downscaling of methods e g from a 4 6 mm column to a l or a 0 3 mm capillary column Q imi 5 0 um SB C18 0 025 3 5 um SB C18 i T A 1 8 um SB C18 Velocity mm sec Column particle size comparison Van Deemter curve for acetonitrile water 1 Highest sample throughput in combination with substance identification To achieve fastest cycle times also called i
4. 16 nm Apex Slope on and Baseline to reduce data amount All or every on second allows summation of spectra in the ChemStation thus leading to higher sensitivity for the spectra of a peak 1 FWHM full width at half maximum of the absorbance band To reduce the high data amount the step size is set to 4 nm still leading to good identification results 3 To save the time necessary for balancing and due to the DAD drift being minimal balancing is not necessary prior to each run 16 MSD Application Identification in combination with minimal analysis time Peakwidth Narrowest peak width at inflection points Fast Scan FS Data Reconstruction DR Time Filter TF Activate FS DR and TF to sacrifice some sensitivity for fast scan cycles in combination with high spectral resolution and acceptable sensitivity Scan Range A narrow scan range is recom mended to ensure increased spectral resolution and sensitivity Scan Step Size The lower the step size the higher the resolution but less analysis time can be spent per single step thus resulting in a decreased sensitivity and vice versa SIM Fewer ions means more analysis time is spent for a single ion thus resulting in higher sensitivity Relative Dwell time for SIM The higher the dwell time percentage the more analysis time is spent on this ion result ing in higher sensitivity Identification and mo
5. and DAD 3 Automation interface of the well plate sampler 4 Well plate handler 5 External valves 6 Mass selective detector based on quadropole or ion trap technology Standard assembly Standard without standard assembly mixer without standard mixer 170 um ID capillary 400 uL binary pump without standard mixer Standard assembly 170 um ID capillary 170 um ID capillary 500 um 170 um ID ID capillary capillaries 200 uL 15 x 4 6 mm columns 170 um ID capillary Active or Passive splitter at flow rates gt 1mL min 250 um ID capillary Figure 3 Schematic diagram of the Agilent 1100 Series high throughput LC MS system columns below the length of 30 mm are recommended The internal diameter of these columns should be relatively high gt 3 9 mm in order not to generate too much backpressure thus allowing to perform separations with high flow rates of 4 or 5 mL min These combinations ensure that the delay time of the gradient time gradient needs to reach the detector is kept to a minimum Some possible column flow combinations are shown in table 1 To maintain res olution smaller particle size mate rials should be used but again the resulting backpressure is the limi tation A good solution is to use the Agilent 1100 sub two micron particles particle size of 1 8 pm however 2 5 um or 3 5 m particles may also be used In general all column types can be us
6. expected absorbance range RB Use a wide reference bandwidth of 80 or 100 nm Slit Step For optimum 1nm spectral conformation choose a slit such that FWHM Slit gt 10 Store Auto balance Allinpeak off to reduce data amount Identification and most accurate quantification Identification of lowest sample amounts Narrowest peak width at inflection points Typical peak width at inflection points SW Choose an appropriate wavelength for each compound BW a No coelution For each compound choose BW peak width at inflection points b Coelution For quantification choose a bandwidth with no absorbance of other compounds SW Choose an appropriate wavelength for each compound BW For each compound choose BW peak width at inflection points RW Choose a reference wavelength outside but also as close as possible to the expected absorbance range RB Use a wide reference bandwidth of 80 or 100 nm RW Choose a reference wavelength outside but also as close as possible to the expected absorbance range RB Use a wide reference bandwidth of 80 or 100 nm For optimum 1nm spectral conformation choose a slit such that FWHM Slit gt 10 In case FWHM Slit gt max slit width then set slit to 8 nm For optimum 4nm spectral conformation choose a slit such that FWHM Slit gt 10 In case FWHM Slit gt max slit width then set slit to
7. the chromatographic separation required for accurate quantification A slightly longer column e g 50 mm may be used in combination with a lower flow rate 1 or 2 mL min and a shallower gradient This results in better separations and thus enables the possibility for quantification Additionally a lower noise level is generated thus resulting in more accurate integrations and a higher sensitivity This can be supported by using the standard mixer within the binary pumps again sacrificing some time The well plate autosampler can be used as stated above with some slight changes The time until the injection valve is switched into bypass can be elongated somewhat thus cleaning of the needle interior is done more thoroughly Also the needle wash should be extended to 20 seconds to ensure very low carry over For this purpose the valve switching procedure for cleaning the injector valve grooves may also be applied The use of flow rates of 1 mL min only has the advantage that no split is necessary prior to the MSD which results in good peak shapes for quantification For higher flow rates the use of an active splitter is recommended for quantitative analysis since much better peak shapes are obtained than with a passive splitter Additionally a fraction collector may be added to collect the purified substances In general quantification with a UV or a DAD detector leads to much better precisions than those with an MSD but th
8. the gradient onto the column the sample flush out factor should be reduced to 2 because otherwise a part of the gradient may be trapped in the sample loop This reduces injec tion to injection time to 2 2 min and only little sensitivity is sacri ficed due to the relatively high flow rate and the MS being a con centration sensitive detector rather than an amount sensitive detector For the DAD the settings in figure 13 are recommended to ensure highest detection sensitivity After having chosen an appropriate detection wavelength e g 215 and 254 nm and a bandwidth of 30 nm to ensure most sensitive detection an appropriate reference wavelength e g 360 nm in combination with a wide bandwidth e g 80 nm should be defined If you want to save spectra you should save all spectra to acquire all spectra during a chromatographic run thus allowing manual averaging of the spectra recorded which leads to a better sensitivity for spectra To obtain not too large file sizes in combination with still relatively good spectral resolution e g for library searches the step size should be set to 4 nm and the range should be as narrow as possible To obtain good chromatographic peak shapes the Agilent ChemStation tries to acquire at least 20 spectra Figure 14 MSD settings for the detection of lowest sample amounts during a chromatographic peak Therefore the expected chromatographic peak width has to be entered in the DAD s
9. creen To obtain high sensitivity in combination with acceptable chromatographic peak shapes the expected peak width at half height should be set to a typical value for the majority of the peaks and not to that of the narrowest ones Typical values are 0 1 min The slit width should be set to 16 nm to obtain the highest sensitivity possible To also ensure sensitive measurements here an autobalance should be performed prior to each run For the ESI source of the MSD the micro sprayer is used to ensure optimized ionization condi tions and thus highest sensitivity The settings in figure 14 are recommended to ensure most sensitive detection In the MSD screen a peak width similar to that used for the DAD is recommended To ensure highest sensitivity fast scan and data reconstruction should be disabled To further support this the scan range should be narrowed as much as possible and the step size should be increased thus allowing to obtain higher sensitivity while sacrificing spectral resolution In many cases a step size of 0 3 Da may be an acceptable compromise between sensitivity and mass spec tral accuracy To obtain highest sensitivity the use of SIM mode is recommended Here the quadru pole is locked at specified masses thus resulting in highest sensitivity and best chromatographic peak shapes For an ion trap accumula tion time and space charge density also have to be controlled Here relatively high values should b
10. e A Intensity gt Sample Sulfamethizole 1 Sulfamethazine 2 Sulfachloropyridazine 3 x106 3 Sulfadimethoxine 4 10 ng each 51 0 Gradient 0 5 min from 5 95 ACN x10 0 0 Flow 20 pL min a Column Zorbax Extend C18 capillary column x10 0 3 x 50 mm 3 5 um particle size 0 7 B 200 250 300 350 400 450 Sample Sulfamethizole 1 Sulfamethazine 2 Mass charge m z Sulfachloropyridazine 3 Sulfadimethoxine 4 Tic 500 ng each Gradient 0 5 min from 5 95 ACN x107 142 Capillary LC MS Flow 4 mL min 3 0 ap tary Column Zorbax STM SB C18 column 4 6 x 15 mm particle size 1 8 ym 2 5 10 ng each used to store as many ions as pos sible and thus generate highest 0 5 sensitivity This to some extent sacrifices chromatographic and Vr Time min 08 1 0 mass spectrometric resolution and accuracy but optimizes sensitivity B For the detection of known com TIC MSD pounds the MRM mode can be used This allows choosing an 500 ng each intensive quantifier ion in combi l nation with qualifier ions thus resulting in high sensitivity in combination with secure identifi cation and quantification A separation of the same sub stances shown in figure 7 and 12 analyzed for sensitivity purposes 0 0 1 0 2 0 3 0 4 05 0 6 is shown in figure 15 Although Time min o d no limit of detection has been determined signal intensities Figure 15 Oo A Capillary LC lon trap high throug
11. e Wait 0 45 min Valve bypass Draw sample 20 sec needle wash Needle into seat Valve mainpass Eject sample Wait 0 45 min Continue to 8 injection cycles 1 Pump 0 min 5 ACN 1 min 95 ACN 1 01 min 5 ACN 1 3 min 5 ACN 2 3 min 95 ACN 2 31 min 95 ACN 2 6 min 5 ACN Continue to 8 cycles up to 10 6 min Flow 4 mL min 1 min gradient but can also be performed with shorter gradients 2 Pump isocratic reequilibration with 4 ml min at 5 ACN Valve 1 3 min Next positi 2 6 min Next positi 3 9 min Next positi 5 2 min Next positi 6 5 min Next positi 7 8 min Next positi 9 1 min Next positi 10 4 min Next posi tion Detectors 0 min Start 10 6 min Stop Table 3 Exemplary timetables for the acquisition of 8 injections into 1 data file Sample Caffeine 500 ng Gradient 1 min from 5 95 ACN Flow 4 mL min Column Zorbax STM SB C18 column 4 6 x 15 mm particle size 1 8 um within a complex action timing network of the overall instrument This makes it quite complicated to change and adopt the method to a specific problem Table 3 shows timetables for a method with 8 injections into one data file and figure 9 shows the corresponding chromatogram 2 Best quantification while maintaining a very high sample throughput The instrument described above can also be used for quantification Only minor adjustments have to be done to sacrifice some time to get
12. e MSD is more sensi tive Therefore wherever possible quantification should be based on UV rather than on MSD For the DAD the following settings are rec ommended to ensure most accu rate detection figure 10 After having chosen an appropriate detection wavelength e g 215 and 254 nm and a bandwidth of 10 nm to ensure specific detection an appropriate reference wavelength e g 360 nm in combination with a wide bandwidth e g 80 nm should be defined To save spectra you should save all spectra in peak or apex slope and baseline spectra to get spectral information in combination without creating large file sizes To obtain good spectral resolution e g for library searches the step size should be set to 1 nm To obtain good DAD Signals Instrument 1 x r Signals Timo epee panao EN Stoptime no Limit 4 min ar fers fio peo fpo Sh om Bh fesa fio fso fo nm Postime ae y a c 7 feso fio 360 jo 3 D C feso fio feo feo 8 rn Required Lamps K UV W Vis Spectrum Peakwidth Responsetime Store Range 190 to 300 nm Autobalance Slit Step ho nm F Prerun X Threshold 1 000 mAU Mo Timetable Total Lines 0 ranean cy a es el Cancel Help rE pi Figure 10 DAD settings for exact quantification Set Up MSD Signals MSD Control MSD Signal Settings F Use MSD Si 3 Frag Ramp ae Gee eee Mode Poiariy et
13. e ume 00T On Mass Range Frag Thres Step a Hiii Se Sey pee CoA ja a Tune File 1 oo F 200 00 1000 00 70 1 00 100 0 20 lon Mode APES Peakwidth 905 min Cycle Time 0 48 sec cycle Son Inser and cut CG Paste F Fast Sean me Dats Reconstruction Time Filter Signal e z Frag Ramp Scan Data Storage Mode Polarity cycle time Full z On Mass Range Frag Thes Step Time min Gain 5 A Signals Off Low High mentor hold size A 1 nao F 20000 100000 70 100 100 020 rz ra r4 Acquisition Parameters Sort Insert Append Cut Copy Paste C Display EIC Parameters cancel me Figure 11 MSD settings for exact quantification Sample Sulfamethizole 1 Sulfamethazine 2 Sulfachloropyridazine 3 Sulfadimethoxine 4 500 ng each Gradient 10 min from 5 95 ACN Flow 1 ml min Column Zorbax STM SB C18 column 4 6 x 30 mm particle size 1 8 pm chromatographic peak shapes the Agilent ChemStation tries to acquire at least 20 spectra during a chromatographic peak Therefore the expected chromatographic peak width has to be entered in the DAD screen To obtain as many single measurements as possible during a chromatographic peak the expected peak width at half height should be set to a value slightly lower than that of the narrowest peak A typical value may be 0 05 min For the slit width it is important to ensure spectral resolution The lower the slit width the better t
14. ed but the most common the most generally applicable and the ones with the highest peak capacity are reversed phase columns To achieve short cycle times short gradients have to be used For this purpose a 0 5 or a 1 minute gradient can be regarded as standard gradients To keep cycle times low only a very short if any isocratic step should be used before the gradient To ensure the use of a generic method the gradient should include both extreme regions e g 5 to 95 ACN Using ESI as ionization method the addition of ion pairing reagents should be kept as low as possible and only volatile buffers for example formic acid ammonium hydroxide or ammonium carbonate should be used Two typical gradients are shown in table 2 Column Dimension Flow Rate Cycle Time for a 0 5 min gradient 50 x 4 6 mm 1 mL min 2 5 min 50 x 4 6 mm 4 mL min 1 4 min 30 x 4 6 mm 1 mL min 2 2 min 15 x 4 6 mm 1 mL min 2 0 min 15x 2 1 mm 1 mL min 2 0 min 15 x 4 6 mm 5 mL min 1 1 min Table 1 Injection to injection times that can be achieved with different column dimensions and flow rates Short high throughput gradient Standard gradient 0 min 5 ACN Omin 5 ACN 0 5 min 95 ACN 3min 5 ACN 0 51 min 5 ACN 17 min 95 ACN 0 7 min 5 ACN 19min 5 ACN 25min 5 ACN Flow 5 mL min Flow 1 mL min Table 2 Comparison of a standard gradient and a high throughput gradient Setup Injector Instrument 1 Figure 4 Injecto
15. est sample throughput in combination with substance identification 2 Best quantification while maintaining a high sample throughput 3 Highest sensitivity in combination with fast analyses Theoretical background To speed up analysis several factors have to be taken in consideration To create a minimal dead volume the column length should be minimized but this results in poorer chromatographic separations To retain as much performance as possible particle diameters are lowered but here the created backpressure is limiting Smaller particle diameters also offer the advantage that with increased flow rates theoretical plate heights do not increase as much as with larger particles figure 1 Therefore the use of small particle diameters below 2 pm is highly recommended for high throughput analysis To convert a common separation into a fast separation chromato graphic parameters have to be adjusted too Therefore isocratic and gradient separations have to be distinguished In isocratic separations the chromatographic separation is more or less independent of the flow With increasing flow rate dead time to and retention time tr are lowered proportionally which results in a constant retention factor k and thus also selectivity a Therefore higher flow rates result in lower run times with roughly the same resolution In gradient separations this is different Here the retention factor k changes continuously
16. g the SIM mode provides much higher sensi tivity than the scan mode The higher the dwell time per centage the more analysis time is spent on this ion resulting in higher sensi tivity does Data reconstruction can be used for singly charged ions only Table 6 Detector setup for known separations Comparison of the DAD and MS settings for ultra high throughput analysis for high throughput analysos in combination with quantification and for highest detection sensitivity during high throughput analysis 17 Mark Stahl is Application Chemist at Agilent Technologies Waldbronn Germany www agilent com chem The information in this publication is subject to change without notice Copyright 2003 Agilent Technologies All Rights Reserved Reproduction adaptation or translation without prior written permission is prohibited except as allowed under the copyright laws Published September 1 2003 Publication number 5988 9863EN Agilent Technologies
17. harge den sity target should also be set to relatively low values to obtain a good chromatographic resolution To further support this the averag ing values should be set to low values too The adjustments of scan range and step size are simi lar to those described above for the quadrupole instrument With such a system cycle times of 1 1 min can be achieved in Sample Sulfamethizole 1 Sulfamethazine 2 Sulfachloropyridazine 3 Sulfadimethoxine 4 500 ng each Gradient 0 5 min from 5 95 ACN Flow 4 mL min Column Zorbax STM SB C18 column 4 6 x 15 mm particle size 1 8 ym combination with a generic gradient from 5 to 95 acetonitrile in 0 5 min As an example a separation of four sulfonamides is shown in figure 7 and compared to a normal analy sis in figure 8 To further reduce cycle times you may acquire dif ferent injections in one data file This skips the initialization times of the instrument To do this you will have to use an injector program for the injections and complex timetables for pumps and valves Data analysis may be complement ed by a data analysis macro cutting this multiple injection file into a single file per injection The draw back of this approach is that method generation is no longer simple and that the different times can no longer be regarded as single timing events but are integrated HT Analysis Sample Sulfamethizole 1 Sulfamethazine 2 Sulfachloropyrida
18. he spectral resolution however your analyses are less sensitive For quantitative purposes a width of 8 nm may be used as starting point To obtain the best integrations possible an autobalance should be performed prior to each Absorbance mAU 350 300 250 200 150 100 50 DAD 254 nm gt T T 0 1 2 Figure 12 Analysis of 4 sulfonamides optimized for quantitative purposes run For the MSD the settings in figure 11 are recommended to ensure accurate detection In the MSD screen a peak width of 0 05 min should also be defined Also here Jast scan data reconstruction and time filter should be enabled These functions are used only when they are needed and the Agilent ChemStation automatically decides whether to use or not to use them Due to the data reconstruction Time min algorithm no spectral resolution is lost To obtain fast quadrupolar scan cycles in combination with high sensitivity and accurate quantification the scan range should be narrowed as much as possible For quantitative measurements a compromise of spectral resolution and MS cycle time has to be drawn On the one hand a lower step size leads to a better spectral resolution on the other hand a higher stepsize Standard deviation Mean Peak 1 Mean Peak 2 Standard deviation Peak 1 Peak 2 UV 254 nm RT min 2 051 0 13 2 212 0 10 Area 1392 0 09 1562 0 14 MSD EIC RT min
19. hput separation of 4 sulfonamides Y axes and injected amounts B LC MSD high throughput analysis of 4 sulfonamides Comparison of the y axes and the injected clearly demonstrate the high sample amounts of A and B clearly demonstrates the high sensitivity of the capillary LC MS sensitivity of the capillary LC MS system system 13 Conclusion Three different high throughput scenarios can be distinguished Analysis with shortest cycle times for the highest sample throughput possible quantitative analysis still offering a relatively high sample throughput and the scenario when sample amount is the limiting factor In the first case resolution and sensitivity is sacrificed for speed In the second one exact quantification conditions are established for the sake of sensitivity and speed and in the last case highest sensitivity is achieved at medium resolution and speed Whereas the first two scenarios can be covered with the same instrument for extremely sensitive analysis an instrument optimized for low flow rates and sample amounts has to be used Choosing the instrument which fits your purposes and setting the correct parameters in combination with high robustness and reliability of the Agilent 1100 Series modules allows to achieve a very high sample throughput in combination with qualitative and quantitative information of even very low sample amounts A summary of the detector settings MSD and DAD for the three diffe
20. ng the injection valve at high organic concentrations to clean the valve grooves DAD Signals Instrument 1 a 5 a as nom m Figure 5 DAD settings for ultrafast high throughput analysis For detection a DAD and an MSD can be used in combination or as stand alone detectors After having passed the DAD the flow is split to 1 mL min for an ESI source by a splitter For an APCI source an even lower split ration may be used The flow that is not directed to the MSD may be collected with a fraction collector but for this purpose an active splitter should be used to obtain good peak shapes This issue will be discussed in more detail in the section Opti mization of your system for quan tification purposes For the DAD the following settings in figure 5 are recommended to ensure fast sensitive and reliable detection After having chosen an appropriate detection wavelength e g 215 and 254 nm and a bandwidth of 10 nm to ensure fast detection in combination with adequate sensitivity an appropriate reference wavelength e g 360 nm in combination with a wide bandwidth e g 80 nm should be defined If you want to save spectra you should save all spectra in peak or apex slope and baseline spectra to get spectral information in combination without creating large file sizes To obtain good spectral resolution e g for library searches the step size should be set to 1 nm To obtain good chromatog
21. njection to injection times sensitivity and resolution have to be sacrificed Whereas identification of substances is achieved easily the resulting resolution often does not allow accurate quantification however an estimation of the pro portions can be done In many cases this is sufficient and allows to achieve the desired high sample throughput The Agilent 1100 Series high throughput LC MS sys tem shown in figure 2 is ideally suited for these analyses The sys tem consists of a degasser for online degassing the solvents two binary pumps a well plate autosampler equipped with an automation interface a well plate handler which is the plate storing device an internal or an external 10 port valve a thermostatted col umn compartment a diode array detector DAD a mass selective detector MSD which can be based on quadrupole or on ion trap technology and a data evaluation unit To reduce the dead volume of the system figure 3 as far as possible the mixer of the binary pump can be removed and capillaries with low internal dia meter 180 pm or below should be used Removing the mixer increases baseline noise somewhat but still leads to acceptable performance To reduce the dead volume further short Figure 2 The Agilent 1100 Series high throughput LC MS system 1 Data evaluation unit 2 Agilent 1100 Series LC system with degasser binary pumps well plate sampler thermostatted column compartment
22. o an internal diameter of 50 pm and 0 3 x 50 mm or even shorter capillary columns may be used These changes reduce the dead volume of the system to roughly 20 pL Depending on the information required the system can be equipped with a DAD and an MSD or an MSD only Depend ing on the requirements a low flow rate of 2 or 3 L min may be used for very sensitive analysis but this will result in quite long cycle times To keep the required time as low as possible fast sol vent exchange should be used at Timetable Total Lines 0 Canoni Figure 13 r Signals r Time Store Sample Bw Reference Bw Sees rene 7 AF fs fo fso oo 4 Stoptime no Limit min B m e54 fo feo foo nm Posttime ot min cm feso fso fso so nm pF feso fso feo feo 4 nm p Required Lamps er fso fo eo fo j am muv Evis 2 DP EOT oe Peakwidth Responsetime oO Range fiso to 400 inm ence Sm Step jao nm F Prerun fenm Threshold fi 000 mAU l Postrun Margin for negative Absorbance Help fioo mau DAD settings for the detection of lowest sample amounts 11 the end of the analysis but this cannot be combined with over lapped injection To reduce the analysis time a high flow rate of 20 L min can be used Due to the high flow rates the fast solvent exchange feature is not needed thus overlapped injections can be used to save time To ensure a complete delivery of
23. ominent ion as quantifier ion quantification TF enabled range possible quantifier ion To 50 due to larger ensure the analysis qualifier ion peak width you may fragment ions and add other intense and specific m z as qualifier ions Identification of low gt 0 1 FS disabled lt 2000 Da 0 3 Da Use the most 100 est sample amounts DR disabled extended scan prominent ion as quantifier ion TF enabled range possible quantifier ion For due to larger peak width maximum sensitivity use only one SIM ion at a time and separate fragmentor values for each ion 1 Data reconstruction can be used for singly charged ions only Table 5 Generic settings as starting conditions for DAD and MSD Comparison of the DAD and MSD settings for ultra high throughput analysis for high throughput analysis in combination with quantification and for highest detection sensitivity during high throughput analysis 15 Detector setup For known separations DAD Application Identification in combination with minimal analysis time Peakwidth Narrowest peak width at inflection points Signal Wavelength SW and Signal Bandwidth BW SW Choose an appropriate wavelength for each compound BW Choose a narrow bandwidth to allow rough quantification in case of coeluting compounds Reference Wavelength RW and Reference Bandwidth RB RW Choose a reference wavelength outside but also as close as possible to the
24. orks for singly charged ions The time filter box activates that the specified peak width is used for the Gaussian filter used within the ChemStation This should always be activated assuming that the correct peak width was entered To reduce the file size condensed data storage may be activated but if detailed isotopic spectral infor mation is required it is better to Figure 6 MSD settings for ultrafast high throughput analysis activate full data storage to store all data acquired To obtain fast quadrupolar scan cycles in combi nation with high sensitivity the scan range should be narrowed as much as possible and a relatively large step size for example 0 25 Da should be used but the stepsize used depends on the spec tral resolution needed To reduce injection to injection times as much as possible the outlet capil lary of the DAD should be con nected to the ESI source directly rather than to the valve inside the MSD Using an ion trap as detec tor allows to perform automated MS MS experiments In contrast to the single quad this allows to select a single parent ion and perform MS MS experiments thus generating a daughter ion spectrum for this compound only even when other substances coelute To maintain chromato graphic resolution the number of precursors should be set to a low value and the threshold should be set to a relative high value As con centration is not an issue accumu lation time and space c
25. r ion in combination with qualifier ions thus leading to an accurate and sensitive quantification in combi nation with secure identification Figure 12 shows the separation of the substances previously shown in figure 7 analyzed for quantitative purposes The separation is much better allowing quantification of the substances however a longer analysis time is necessary In table 4 precisions of the measurements are shown When using two columns in parallel for quantitative analysis it must be considered that methods have to be validated either on both columns together thus leading to a slightly reduced precision of the analysis or separately for each column However then it has to be controlled which sample is injected onto which column This can easi ly be achieved with Agilent Chem Station and Agilent ChemStore 3 Highest sensitivity in combination with fast analyses The system used for high through put analysis of lowest sample amounts is similar to the one described above only miniaturization of the flow path was achieved to increase the sensitivity For this purpose two capillary pumps are used instead of the binary pumps a micro well plate autosampler IDAD Signals Instrument 1 equipped with an 8 pL or alterna tively a 40 pL sample loop instead of the standard well plate autosampler and a micro valve instead of the normal 10 port valve Additionally all tubing material diameters should be adjusted t
26. r setup for high throughput analysis To achieve shortest cycle times two columns can be used in parallel This allows to analyze the sample on one column while the other one is regenerated The control of the columns is done by a valve switching to the next position after having finished the first analysis This reduces the cycle time of the amount necessary for column regeneration You only have to ensure that the gradient has already reached the detector and is followed by starting conditions before switching the valve to the next column For sample storage and delivery the well plate autosampler in combination with the well plate handler is used All different kinds of vial and well plates can be used with this system and you are also able to define your own ones Using 384 well plates the well platehandler has a sample capaci ty of up to 30 720 samples thus allowing the analysis of thou sands of samples per day in com bination with the high robustness of the complete system and fully unattended analysis with very short cycle times To achieve fast analysis automatic delay volume reduction and overlapped injec tion should be enabled Even dur ing these very short cycle times you still have enough time to per form a 10 s needle wash with an appropriate solvent prior to injec tion to ensure a low carry over of the well plate sampler figure 4 The carry over of sticky com pounds can be reduced further by switchi
27. raphic peak shapes the Agilent ChemStation tries to acquire at least 20 spectra during a chromatographic peak Therefore the expected chromatographic peak width at half height has to be entered in the DAD screen Due to the high flow rates the expected peak width is very low A typical value for the narrowest peak is 0 01 min For the slit width a good combination of spectral resolution and sensitivity is achieved at 4 nm Whereas for normal analysis an autobalance is performed prior to each run this is skipped to further reduce the cycle time As the detector drift is very low for quali tative analysis only it is sufficient to do an autobalance only once before starting an analysis This saves additional 5 s per run For the MSD the settings in figure 6 are recommended to ensure sensi tive and reliable detection In the MSD screen a peak width of 0 01 min should also be entered Here fast scan data reconstruction and time filter should be enabled to ensure optimal performance The fast scan option optimizes the scan speed of the quadrupole There fore some resolution and sensitiv ity are sacrificed Data reconstruc tion uses a moving average filter to reconstruct the spectra from the recorded data With this math ematical operation the resolution sacrificed with the fast scan option is gained back thus resulting in a comparable spectral resolution to a normal scan however the data reconstruction algorithm only w
28. rent scenarios is shown in tables 5 and 6 Generic settings as starting conditions DAD Signal Wavelength Reference Application Peakwidth SW and Signal Wavelength RW Slit Step Store Auto Bandwidth BW and Reference balance Bandwidth RB Identification in gt 0 01 SW RW 4nm 1nm Allin peak off combination with 205 215 254 280 nm 360 nm minimal analysis BW RB time 10 nm 80 nm Identification and gt 0 05 SW RW 8nm 1nm Apex Slope on most accurate 205 215 254 280 nm 360 nm and quantification BW RB Baseline 10 nm 80 nm Identification of gt 0 1 SW RW 16 nm 4nm Allor every on lowest sample 205 215 254 280 nm 360 nm second amounts BW RB 30 nm 80 nm To reduce the high data amount the step size is set to 4 nm still leading to good identification results 14 MSD Fast Scan FS Application Peakwidth Data Reconstruction Scan Range Scan Step Size SIM Relative Dwell DR time for SIM Time Filter TF Identification in gt 0 01 FS enabled lt 600 Da 0 25 Da Use the most 50 combination with DR enabled for a lower step prominent ion as quantifier ion minimal analysis TF enabled size you have to quantifier ion To 50 time decrease the scan ensure the analysis qualifier ion range further you may fragment ions and add other intense and specific m z as qualifier ions Identification and gt 0 05 FS enabled lt 2000 Da 0 2 Da Use the most 50 most accurate DR enabled extended scan pr
29. st accurate quantification Narrowest peak width at inflection points Activate FS DR and TF to sacrifice some sensitivity for fast scan cycles in combination with high spectral resolution and acceptable sensitivity A narrow scan range is recom mended to ensure increased spectral resolution and sensitivity For quantification extracted ion chromatograms have to be gener ated The lower the step size the higher the resolution but less analysis time can be spent per single step thus resulting in a decreased sensitivity and vice versa The fewer ions being moni tored the more analysis time is spent for a single ion thus resulting in higher sensi tivity These signals can be integrated directly for quantification The higher the dwell time per centage the more analysis time is spent on this ion resulting in higher sensitivity Identification of low est sample amounts Typical peak width at inflec tion points Activate TF only and deactivate FS and DR to obtain highest sensitivity Narrow the scan range as much as possible to gain as much time per single quadrupole step resulting in highest sensitivity The higher the step size the less steps have to be made per quadrupole scan cycle thus more time can be spent per single step resulting in increased sensitivity The less ions the more analysis time is spent for a single ion thus resulting in higher sensi tivity Usin
30. zine 3 Sulfadimethoxine 4 500 ng each Gradient 0 5 min from 5 95 ACN Flow 4 mL min Column Zorbax STM SB C18 column 4 6 x 15 mm particle size 1 8 ym Normal Analysis Sample Sulfamethizole 1 Sulfamethazine 2 Sulfachloropyridazine 3 Sulfadimethoxine 4 500 ng each Gradient 15 min from 5 95 ACN Flow 1 mL min Column Extend C18 column 4 6 x 50 mm 3 5 um particle size 100 7791 311 1 80 60 1 2 4 40 20 271 0 312 1 0 200 400 600 800 200 400 600 800 200 400 600 800 Mass charge m z Absorbance mAU 600 12 DAD 254 nm 400 3 4 200 0 0 0 1 0 2 0 3 0 4 0 5 0 6 TIC To 4 1200000 a MSD 800000 400000 0 T T T 0 0 1 0 2 0 3 0 4 0 5 0 6 a Time min Figure 7 High throughput analysis of 4 sulfonamides 54 nm mAU 500 gt x x 254 nm mAU 400 N x 500 gt 400 300 300 0 5 min gradient 200 200 7 100 7 100 a 0 a 0 02 04 06 0 8 1 0 A Time min 0 1 Time min 254 nm mAU 15 min gradient 0 2 4 6 8 10 12 14 Time min Figure 8 Comparison of the UV chromatograms at 254 nm of a 0 5 min gradient and a 15 min gradient Absorbance mAU Time min Figure 9 UV chromatogram 280 nm of 8 Caffeine injections acquired into a single data file Injector Draw sample 20 sec needle wash Inject Wait 0 7 min Valve bypass Draw sample 20 sec needle wash Needle into seat Valve mainpass Eject sampl
Download Pdf Manuals
Related Search
Agilent Optimizing the Agilent 1100 Series high throughput LC/MS system