August 2003 Page 1 of 14 319306-01 Introduction Solvent choices for electrospray ionization can be separated from those of column separation for systems utilizing a make-up solvent technique to provide required dilution. The following application note takes a look at a few common solvent systems and their eects on a set of pharmaceutical target compounds and related analytes. What was Investigated Mass-based fraction collection solvent selection for electrospray positive ionization mode was tested for a limited set of solvents against a small set of analytes to show examples of solvent choice on results. Often in prep analysis, unlike analytical work, compromises are made in ionization mode and HPLC conditions to suit a broad number of compounds. In this Application Note, we will look at some of these compromises in terms of ionization. Main Components: Gilson-Thermo Finnigan NEBULA™ Prep System, consisting of a Gilson 215 Liquid Handler/Autosampler/Fraction Collector, 322 H2 Prep Pump, 307 Make-up Pump, 155 UV/VIS Dual- wavelength Detector with 0.2-mm ow cell, and Thermo Finnigan MSQ Mass Spectrometer. Experimental—HPLC/LCMS Six solutions of analytes were prepared (10 mg/mL of each compound) in DMSO, which oers at least moderate separation for most analytes. All injection volumes were 500 µL on a 2-mL loop. Column: Metachem 21*50 ODS2. Flow rate: 20 mL/min.; gradient: 5% MEOH/95%Water + 0.1% HCOOH to 95% MEOH/5% Water +0.1%HCOOH in 10 minutes. Retention time estimates from screening analysis on a ZORBAX 4.6*50-mm column at 1.5 mL/min. using the same gradient. Make-up Flow Rate: 0.4 mL/min. Splitter: Standard LC-Packings/standard ACM 10–50; UV = 254nm. MS Conditions: Electrospray positive ionization mode. Capillary voltage: 3kV; cone voltage varied for experimental purposes. Capillary temperature (525°C) was determined to give optimal signal for all solvents. Scan range: 100–900 amu; scan rate: 0.8 sec. Cone stepping runs: 20V, 40V, 65V, and 95V at 1 sec. Make-up Solvent Considerations for Prep and Semi-Prep LC/MS Mass-Based Fraction Collection Materials & Methods Application Note 206 Tim Hegeman (Gilson, Inc.) Experiments: Collections performed for all solvents were tested for all mixes. Each system was run with four cone voltages to look at solvent inuence. Note: System is a solvent system identied by its organic constituent: MEOH, ACN, IPA, or MEOH/1,4 Dioxane. One representative mix from each system was chosen for the cone voltage experiment as indicated. August 2003 Page 2 of 14 319306-01 Make-up 1 Solvent A Solvent B Acid Methanol % Water % HCOOH % 100 0 0.10 90 10 0.10 75 25 0.10 X 50 50 0.10 Make-up 2 Solvent A Solvent B Acid Acetonitrile % Water % HCOOH % 100 0 0.10 90 10 0.10 75 25 0.10 X 50 50 0.10 Make-up 3 Solvent A Solvent B Acid IPA % Water % HCOOH % 100 0 0.10 90 10 0.10 75 25 0.10 X 50 50 0.10 Make-up 4 Solvent A Solvent B Solvent Acid 1,4 Dioxane % Methanol % Water % HCOOH % X 25 25 50 0.10 Make-up ow = 0.4 mL/min. HCOOH = Formic Acid % = v/v X = Mix chosen to represent system for cone voltage change This change was a sequential step of voltages at 20, 40, 65, and 95V Table 1: Experimental—Make-up Solvents Tested Photo 1: Gilson-Thermo Finnigan NEBULA™ Prep System August 2003 Page 3 of 14 319306-01 Figure 1: Single-Injection LC/MS System Figure 2: Make-up Splitter Allows Solvent Choices Dilution Factor= mass split(1000) * make-up ow/prep ow Dilution Factor = (1000 * 0.4 mL/min.)/20 mL/min. = 20 Low-Mount, Low-Pressure Valve (not shown installed in spring clamp) 215 Liquid Handler NC COMM NO 819 Injection Module to wast e to wast e Column Outlet Pump UV/VIS Detector Outlet Inlet Semi-Pre p Filter Assembly Make-up Pump low flow to high flow mass spectro collec t Flow Splitter Finnigan MSQ™ Inlet Inlet August 2003 Page 4 of 14 319306-01 Mix 1 in DMSO Compound RT RT MW* ion MH+ RT from screen Window 1 Niacinamide 0.6 1.5-2.5 122 123 2 4-Acetaminophen 2.9 2.3-3.5 151 152 3 Caff eine 4.9 4.7-5.4 194 195 4* Salicyclic acid acetate 5.8 5.6-6.3 180 none 5 Verapamil 6.7 6.5-7.5 454 455 6 N ifedaphine 7.9 7.7-8.1 346 347 * MW Molecular Weight * RT Retention time min. * MH+ Protonated ion 4* Salicyclic acid acetate not analyzed for OH NH O CH 3 2 N O NH 2 1 N N O O CH 3 CH 3 N N CH 3 3 O OH O C H 3 O 4 N + O - O N H CH 3 CH 3 O CH 3 O O O CH 3 6 O O CH 3 CH 3 CH 3 CH 3 N N CH 3 O O CH 3 CH 3 5 6 CH 3 CH 3 O S 10mg/mL DMSO Figure 3: Mix 1: 10 mg/mL DMSO Mix 2 in DMSO Compound RT RT MW* ion MH+ RT from screen Window 1 Pyridoxamine 0.4 0-1 168 169 2 Metropropolol 4.2 4.1-4.9 267 268 3 Quinine 5 4.5-5.5 324 325 4* Salicyclic acid acetate 6 5.5-6.5 138 none 5 C ortisone 7.2 7-7.5 360 361 6 Clofazimine 8.5 9 472 473 7 Triprolidine 4.8 4.5-5.5 278 279 * MW Molecular Weight * RT Retention time min. 4* Salicyclic acid acetate not analyzed for N OH NH 2 OH CH 3 CH 3 O O NH CH 3 CH 3 OH N O CH 3 N OH CH 2 O OH OH N N NH Cl N CH 3 CH 3 Cl CH 3 N N 7 1 2 3 4 5 6 O OH O CH 3 CH 3 O OH Figure 4: Mix 2: 10 mg/mL DMSO August 2003 Page 5 of 14 319306-01 Mix 3 in DMSO Compound RT RT MW* ion MH+ RT from screen Window 1 Pyridoxine 0.4 0-1 169 170 2 8 -Chlorotheophyline 4.4 4-4.9 214 215 3 Chlorpheniramine 5.2 5.0-6.0 274 275 4 Doxepin 6.4 5.8-7.0 279 280 5 Funarizine 7.9 7.7-8.2 404 405 * MW Molecular Weight * RT Retention time min. * MH+Protonated ion N CH 3 OH OH OH N N CH 3 CH 3 Cl N CH 3 O CH 3 N N F F 1 2 3 4 5 N N O O CH 3 CH 3 N H N Cl Double charged at high cone voltage (M2H+) Figure 5: Mix 3: 10 mg/mL DMSO Mix 4 in DMSO Compound RT RT MW* ion MH+ RT from screen Window 1 Pyridoxal 0.55 0-1 167 168 2 Primidone 4.8 4.5-5.5 218 219 3 3-Is obutyl-1-methylxantthene 6.1 6-6.7 222 223 4 Prednisone 7.2 7-7.6 358 359 5 D esoximetasone 8.7 8.5-9.0 376 377 * MW Molecular Weight * RT Retention time min. * MH+Protonated ion N CH 3 OH O OH NH N H O O CH 3 F OH O O CH 3 CH 3 OH CH 3 1 2 3 4 5 N N N H N O O CH 3 CH 3 CH 3 O OH O CH 3 CH 3 O OH Figure 6: Mix 4: 10 mg/mL DMSO August 2003 Page 6 of 14 319306-01 Mix 5 in DMSO Compound RT RT MW* ion MH+ RT from screen Window 1 Panthothenic Acid 2 1-2.9 219 220 2 Sulfamethazine 4.5 4.0-5.0 278 279 3 Estrone 5.1 4.8-5.5 270 271 4 Enalapril 6.4 6.2-6.8 376 377 * MW Molecular Weight * RT Retention time min. * MH+Protonated ion O OH NH O OH CH 3 CH 3 OH NH 2 S O O NH N N CH 3 CH 3 NH N O OCH 3 CH 3 O O O H 1 2 O CH 3 OH 3 4 Figure 7: Mix 5: 10 mg/mL DMSO Mix 6 in DMSO Compound RT RT MW* ion MH+ RT from screen Window 1 Bromopheniramine 4.9 4.1-5.8 318 319 2 Diltizem 6.6 6.4-7.4 414 415 3* Quininic Acid 7.9 7.7-8.1 203 none 3* Not analyzed for * MW Molecular Weight * RT Retention time min. N N CH 3 CH 3 Br 1 O CH 3 N S O O CH 3 O N CH 3 CH 3 2 CH 3 O OH O 3 Figure 8: Mix 6: 10 mg/mL DMSO Targets Which Showed a Signicant Solvent Eect Several of the target compounds studied were not inuenced by the solvents chosen; these are omitted from further discussion. These primarily consisted of compounds that have very good MH+ ionization sites, such as tertiary amines, which dwarfed all other inuences. August 2003 Page 7 of 14 319306-01 MNa + MHDM SO + (+23) (+79) 1 Niacinamide none low cone 2 4-acetaminophen none low cone 3 C affeine none low cone 4 Quinine none low cone 5 Pyridoxal none low cone 6 Primidone none low cone 7 3-Isobutylxanthine none low cone 8 Panthothenic A cid All none 9 Sulfamethazine high cone none 10 Nifedapine (not shown) high cone none 11 Cortisone high cone All 12 Desoximetasone none All 13 Estrone none low cone 14 Prednisone none low cone low cone = present at 20 and 40 volts at >10% MH+ high cone present at high cone = 65 and 95 volts; > 10% of MH+ All = adduct present at > 10% MH+ at all cone voltages System test make-up solvent 50% MEOH + 0.1% HCOOH none = < 10% MH+ Makeup flow 0.4mL/minute N N O O CH 3 CH 3 N N CH 3 N O NH 2 OH NH O CH 3 N O CH 3 N OH CH 2 N CH 3 OH O OH NH N H O O CH 3 O OH NH O OH CH 3 CH 3 OH NH 2 S O O NH N N CH 3 CH 3 N N N H N O O CH 3 CH 3 CH 3 Figure 9: Results (Make-up 1) 50/50 Methanol/Water + 0.1% HCOOH MNa + MHDM SO + (+23) (+79) 1 Niacinamide none low cone 2 4-acetaminophen none low cone 3 C affeine none low cone 4 Quinine none low cone 5 P yridoxal none low cone 6 Primidone none low cone 7 3-Isobutyixanthine none low cone 8 Panthothenic A cid All none 9 Sulfamethazine high cone none 10 Nifedaphine (not shown) high cone none 11 Co rtisone high cone All 12 Desoximetasone none All 13 Estrone none low cone 14 Prednisone none low cone low cone = present at 20 and 40 volts at >10% MH+ high cone present at high cone = 65 and 95 volts; > 10% of MH+ All = adduct present at > 10% MH+ at all cone voltages System test make-up solvent 50% Water 50% MEOH + 0.1% HCOOH none = <10% MH+ F OH O O CH 3 CH 3 OH CH 3 O CH 3 OH O OH O CH 3 CH 3 O OH O OH O CH 3 CH 3 O OH The following is a brief look at the dierent solvent systems and the compounds found to be most aected by the solvent choice. These tend to be molecules with weaker basicity (i.e., exposed oxygens). Adducts seen: MH+=(MW+1); MHACN+=(MW+42)+; MHIPA+=(MW+61)+; and MHDMSO+=(MW+79)+ Figure 10: Results (Make-up 1) 50/50 Methanol/Water + 0.1% HCOOH August 2003 Page 8 of 14 319306-01 Figures 11 & 12: Results (Make-up 1) 50/50 Methanol/Water + 0.1% HCOOH Compounds having acidic functionalities (e.g., phenols and carboxylic acids) tend to form Na+adducts, and the relative intensities tend to increase with cone voltage. DMSO adduct is generally fairly weak and can be driven o with increased cone voltage. Results (Make-up 1) 50/50 Methanol/Water + 0.1% HCOOH This system does not tend to add to the adduct problem via an intrinsic solvent adduct. This can result in a more simplied and consistent spectra and collection. Na + (M+23) is a problem for all analytes that have the ability to form these persistent adducts. Other adducts, such as the DMSO, adduct (M+1+78) can form if the species is available from the injection solvent (i.e., can be removed by increased cone voltage). The chief argument against this system may be solubility of certain classes of compounds. PrednisoneMH+(359) and D MSO (MH +78)+ adduct (438) at 20V and 65V shows that higher voltage drives off this adduct (MW 358, MH+ 359) 20V 65V 20V 65V Panthothenic Acid MW 219, MH+ = 220 ; Na+ adduct (242) shows that unlike DMSO adduct (298)- relative abundance of N a+ adduct increases with cone voltage. (note DMSO adduct at 20V) 157 common DMSO solvent ion 2(DMSO)H+ from sample solvent MNa + MHACN+ MHDMSO+ (+23) (+42) (+79) 1 Niacinamide None All Low Cone 2 4-acetaminophen None All Low Cone 3 Caeine None All None 4 Quinine None None None 5 Pyridoxal None All Low Cone 6 Primidone None All Low Cone 7 3-Isobutyixanthine none All Low Cone 8 Panthothenic Acid <10%++ None None 9 Sulfamethazine None None None 10 Nifedaphine (not shown) None None None 11 Cortisone None All None 12 Desoximetasone None All None 13 Estrone None None None 14 Prednisone None All None Low Cone = present at 20V and 40V at >10% MH+ System test make-up solvent = 50% water/50% ACETN with 0.1 HCOOH ow 0.4 mL/min. All = >10% of MH+ for all cone voltages measure (20, 40, 65, 95V) None = <10% for all cone voltages August 2003 Page 9 of 14 319306-01 Table 2: Results (Make-up 2) 50/50 Acetonitrile/Water + 0.1% HCOOH Caffeine (MH+ 195) Acetonitrile adduct(M+ACE + H)+ (236) shows that in many cases this adduct is not reduced by increased cone voltage. Here it is the base peak at both voltages. Primidone (MH+ 219) Acetonitrile adduct(M+ACETN + H)+ (260) shows that in many cases this adduct is still present at increased cone voltage. Here it is the base peak at both voltages. 20V 20V 65V 65V Figures 13 & 14: Results (Make-up 2) 50/50 Acetonitrile/Water + 0.1% HCOOH August 2003 Page 10 of 14 319306-01 Acetonitrile adducts for many compounds form resilient adducts that resist cone voltage fragmentation. These include compounds such as steroids and those with exposed carbonyls. This adduct can often form the base peak of the spectrum. (Acetonitrile adduct MW + 42 from (MH+Acet)+) The absence of Na+ may be due either to the absence of this metal ion or to the blockage by this solvent. MNa + MHACN+ MHDMSO+ (+23) (+61) (+79) 1 Niacinamide None None None 2 4-acetaminophen None None Low Cone 3 Caeine None None Low Cone 4 Quinine None None None 5 Pyridoxal None All Low Cone 6 Primidone None All Low Cone 7 3-Isobutyixanthine None All Low Cone 8 Panthothenic Acid All None None 9 Sulfamethazine None None None 10 Nifedaphine (not shown) All None None 11 Cortisone None All Low Cone 12 Desoximetasone None All None 13 Estrone None None None 14 Prednisone None All None Low Cone = present at 20V and 40V at >10% MH+ System test make-up solvent = 50% water 50%/50% IPA with 0.1% HCOOH ow 0.4 mL/min. IPA = isopropanol All = >10% of MH+ for all cone coltages measured (20, 40, 65, 95V) None = <10% of MH+ Table 3: Results (Make-up 1) 50/50 IPA/Water + 0.1% HCOOH Cortisone MW 360 MH+ 361 shows IPA adduct (+60) at 421 amu for M=H+IPA +. This is a prominent ion at both 20 and 65 V 20V 65V Figure 15: Results (Make-up 1) 50/50 IPA/Water + 0.1% HCOOH IPA can form adducts—particularly with steroids—that are not easily cleaved with high cone voltage. (Adduct at MW + 61 from (MH+IPA)+ ) Na+ adduction was present similar to the MEOH/Water system. [...]... methanol make-up solvent without the detrimental adduct affect of acetonitrile make-up mixes for some compounds If a solvent is used for solubility or compatibility reasons which are found or speculated to produce additional adducts—the ions for all the target adducts should be added to the collection method If using MS to check fractions for hits, keep in mind that the solvent was the mobile phase for. .. sequentially at one second each Conclusion Mass-based fraction collection with target compounds at the 1 mg and higher level usually will require a dilution-type of splitter for the mass spectrometer interface This offers the user some choice in the solvent employed as the dilution solvent This choice can significantly influence the ionization for LC/MS target mass ions Solvents such as 1,4 Dioxane can increase... is too low for a MEOH/Water system alone Results for Areas of the M+H+1 Ions for the Target Compounds The table shows different percentages of solvents and the measured areas, which are a crude guide to solvent choice results Areas are considered estimates only due to fact that issues such as linearity of response are not accessed at these overload levels Compounds strongly affected by solvent choice... Sulfamethazine 10 Nifedaphine (not shown) Low Cone = present at 20V and 40V at >10% MH+ High cone present at high cone = 65V and 95V; >10% of MH+ System test make-up solvent = 50% water/25% methanol/25% 1,4 dioxane DX = 1,4 Dioxane None = 10% of MH+ for all cone coltages measured (20, 40, 65, 95V) Tabel 4: Results (Make-up 4) 25/25/50 MEOH/1,4 Dioxane/Water + 0.1% HCOOH 20V 65V Cortisone... but low ion at both 20 and 65 V Note the DMSO adduct at 439 that lost at 65V Figure 16: Results (Make-up 4) 25/25/50 MEOH/1,4 Dioxane/Water + 0.1% HCOOH August 2003 Page 11 of 14 319306-01 1,4 Dioxane can form adducts with some compounds, but not to the degree of other solvents The Na+ adduct was present similar to the MEOH/water system 1,4 Dioxane may provide a good alternative solvent choice if the...Results (Make-up 3) 50/50 IPA/Water + 0.1% HCOOH This system does not tend to add to the adduct problem via an intrinsic solvent adduct to the degree that acetonitrile-based systems do IPA does form an adduct with many of the same compounds affected by acetonitrile Na + (M+23) is a problem for all analytes that have the ability to form these persistent adducts Other adducts,... x = compounds found to be significantly influenced by solvent choices gp = group (solution containing the compound listed) DX = 1,4 Dioxane at 25%, MEOH at 25%, and water at 50%, HCOOH 0.1% MEOH = Methanol IPA = Isopropyl Alcohol 1,4DX = 1,4 Dioxane ACETN = Acetonitrile * = V/VVolume/Volume Table 5: Measured Areas of Analytes M+H Signal for All Solvents Tested (mix 1, 2, 3) August 2003 Page 12 of 14... Compounds found to be significantly influenced by solvent choices gp = group (solution containing the compound listed) DX = 1,4 Dioxane at 25%, MEOH at 25%, and Water at 50%, HCOOH 0.1% MEOH = Methanol IPA = Isopropyl Alcohol 1,4DX = 1,4 Dioxane ACETN = Acetonitrile * = V/V Volume/Volume *10E6 = Area * 10E6 Table 6: Measured Areas of Analytes M+H Signal for All Solvents Tested (mix 4, 5, 6) Cortisone 50/50... affected by acetonitrile Na + (M+23) is a problem for all analytes that have the ability to form these persistent adducts Other adducts, such as the DMSO adduct (M+1+78), can form if the species is available from the injection solvent This system offers many of the same advantages as a methanol system, but can increase the solubility range MNa + MHDX+ MHDMSO+ (+23) (+89) (+79) 1 Niacinamide None None... accessed at these overload levels Compounds strongly affected by solvent choice would be expected to show the greatest fluctuations in areas as a result of such properties as poor ionization or adduct formation shifting the target mass MH+ to MHACN+, MHDMSP+, Mna+, etc MEOH MEOH MEOH MEOH ACETN ACETN ACETN ACETN IPA IPA DX H2O H2O H2O H2O H2O H2O H2O H2O H2O H2O DX 100/0* 75/25* 50/50* 90/10* 50/50* . Cone stepping runs: 20V, 40V, 65V, and 95V at 1 sec. Make-up Solvent Considerations for Prep and Semi -Prep LC/MS Mass-Based Fraction Collection Materials. Collections performed for all solvents were tested for all mixes. Each system was run with four cone voltages to look at solvent inuence. Note: System is a solvent