Automated Large-Volume Injection for Direct Analysis of Alkylphenols and Ethoxylated Alkylphenols by Mixed-Mode Chromatography and Mass Spectrometry

Automated Large-Volume Injection for Direct Analysis of Alkylphenols and Ethoxylated Alkylphenols by Mixed-Mode Chromatography and Mass Spectrometry Leo Wang, Bill Schnute, Stacy Henday, Dionex Corporation, Sunnyvale, CA, USA INTRODUCTION Alkylphenyl polyethoxylates (APEOs) are among the most commonly used nonionic surfactants, with over 6, tons produced annually. APEO degradation products alkylphenols (APs) and their short chain ethoxylates (mainly mono- and diethoxylates, EO and EO2) have been detected and reported in treated wastewater, sewage sludge, and soil. Studies reveal that these substances have varying estrogenic potencies. 2 Currently, these compounds are on the US E s priority testing list for evaluation of environmental and health effects. Because APEOs with different EO units exhibit different estrogenic potencies, it is important to develop an analytical method to profile and monitor their presence in the environment. The development of a chromatographic method for separation of APs and APEOs remains a challenge due to their structural complexity. Current methods, including normal-phase liquid chromatography (NPLC) and reversed-phase liquid chromatography (RPLC) separate these compounds by the difference of alkyl groups or EO units. 9 The preferred detection technique for APEO analysis is mass spectrometry, due to its sensitivity and selectivity. Here, a mixed-mode chromatographic method was developed for separation of these compounds. On-line concentration and desalting was fully automated using large-volume injection ( ml) to increase sample throughput and improve sensitivity. Three mixed-mode columns with different selectivities were evaluated and the results compared, as well as MS and MS/MS instrumentation modes as detection techniques. Parameters affecting chromatographic separation and MS detection were explored and discussed, with each method individually evaluated with respect to linearity, detection limits, precision, and accuracy. EXPERIMENTAL Instrumentation HPLC: P68 Dual Ternary Pump ASI- Autosampler or AS-HV autosampler TCC- Thermostatted Column Compartment UVDU detector Mass Spectrometer: MSQ Plus single quadrupole mass spectrometer with cone wash TSQ Quantum Access triple quadrupole mass spectrometer Software: Chromeleon 6.8 SR6 Xcalibur 2. DCMS Link for Xcalibur (version 2.)* for MS/MS analysis *DCMS Link is a Chromeleon-based software module providing the interface for controlling Dionex chromatography instruments from different mass spectrometer software platforms. Sample Loading Conditions for AS-HV* Concentrator: Acclaim Guard (. mm, µm) Injection Volume: ml Loading Speed: ml/min * For ASI- autosampler, use loop injection and µl injection volume. Chromatographic Conditions Analytical Column: Acclaim Surfactant (2. mm, µm) Column Temperature: C Mobile Phase: A: Methanol (CH OH); B: Buffer ( mm sodium acetate (NaOAc) for MSQ Plus, mm ammonium acetate (NH OAc) for Quantum TSQ); C: water. Time (min) A% B% C% 7 2 9 7 2 9 8 9 9 7 2 Flow Rate:.2 ml/min

Mass Spectrometric Conditions MSQ Plus Interface: Electrospray Ionization (ESI) Detection Mode: Selected Ion Monitoring (SIM), see Table for scan details Nebulizer Gas: Nitrogen; 8 psi Probe Temperature: C Cone Wash: CH CN/H 2 O (/, v/v) at µl/min Quantum TSQ Interface: ESI Detection Mode: Selected Reaction Monitoring (SRM), see Table 2 for scan details Capillary Temperature: 27 C Needle Voltage: V Sheath Gas: units Auxiliary Gas: for the first segment, for the second segment Name Table. Scan Functions and Scan Events for MSQ Plus SIM Detection Adduct Mass m/z Span m/z SIM Group Time Range (min) Dwell Time (sec) Polarity Cone Voltage (V) [M+Na] + 27.2.. 8.. Pos. 6 [M+Na] + 7.2.. 8.. Pos. 6 SIM Group 2 [M+Na] + 287.2. 8. 2.. Pos. 6 NPEO2 [M+Na] +.2. 8. 2.. Pos. 6 SIM Group OP, tert-op [M+CH COO] 26.2.. 9.. Neg. 6 NP [M+CH COO] 279.2.. 9.. Neg. 6 Scan Segment ( ~ 2 min) 2 (2 ~ 2 min) Table 2. SRM Scan Functions for MS/MS Detection Analyte * Quantitative SRM ; ** Confirmative SRM. Polarity SRM Transitions Parent Ion (m/z) Product Ion (m/z) Collision Energy (V) (Q*) positive 268.2. (C**) positive 268.2 7. 7 (Q) positive 2.2 8. (C) positive 2.2 2. 22 (Q) positive 282.2 27. 7 (C) positive 282.2 8.2 NPEO2 (Q) positive 26.2 8. NPEO2 (C) positive 26.2 2. 22 Standard Preparation tert-op negative 2.. 2 -OP negative 2. 6. 2 -NP negative 29.2 6. 2 IGEL CA-2 and IGEL CO-2 were purchased from Sigma- Aldrich.,, and NPEO2 were chromatographically fractionated in our laboratory using an Acclaim WAX- column. Concentrations of collected fractions were calculated based on relative UV peak areas. N-octylphenol, tert-octylphenol, and n-nonylphenol were purchased from Sigma-Aldrich and dissolved in CH CN to µg/ml. Structures are shown in Figure. C 8 H 7 OH 2 CH 2 OH Octylphenol monoethoxylate m.w. = 2. C 9 H 9 OCH 2 CH 2 OH Nonylphenol monoethoxylate m.w. = 26. C 8 H 7 Octylphenol diethoxylate m.w. = 29. C 9 H 9 (OH 2 CH 2 ) 2 OH Nonylphenol diethoxylate m.w. = 8. (OCH 2 CH 2 ) 2 OH H C(H 2 C) 6 H 2 C OH H C(H 2 C) 7 H 2 C OH -OP -Octylphenol m.w. = 26. -OP -Nonylphenol m.w. = 22. OH t-bu tert-op tert-octylphenol m.w. = 26. 289 Figure. Structures of studied analytes. 2 Automated Large-Volume Injection for Direct Analysis of Alkylphenols and Ethoxylated Alkylphenols by Mixed-Mode Chromatography and Mass Spectrometry

Calibration standards were prepared in mobile phase the following concentration levels: 2,,, 2,,, 2,, and ng/ml. Each level contains seven analytes:,,, NPEO2, -OP, tert-op, and -NP. RESULT AND DISCUSSION Chromatography Among the three column candidates, the Acclaim Surfactant and Acclaim Mixed-Mode WAX- possessed the capability of separating APEO EO short oligomers. To achieve the goal of total resolution of APEO EO oligomers and APEO groups, i.e., OPEOs and NPEOs, these columns were compared with respect to their group separation efficiency. As shown in Figure 2, the Acclaim Surfactant column demonstrated the ability to differentiate APEOs by the EO units as well as alkyl group difference. mau Columns: Acclaim Surfactant and Acclaim WAX- (2. mm, µm) Mobile Phases: For Surfactant: 8% CH CN % mm NH OAc ph.2 9% H 2 O For WAX-: % CH CN % mm NH OAc ph.2 % H 2 O NPEOs on WAX- OPEOs on WAX- Column Temp.: C Flow Rate:. ml/min Inj. Volume: µl Detection: UV @ 27 nm Samples: IGEL CA-2 IGEL CO-2 NPEOs on Surfactant OPEOs on Surfactant 6 9 2 Minutes Figure 2. Resolution comparison of WAX- and Surfactant columns for APEO groups. 289 Column temperature was a key parameter for fine tuning resolution and adjusting run time. Seven temperatures from 2 C were examined to determine optimum conditions (acetonitrile was used in the mobile phase). Results are shown in Table. Three resolutions (Rs) were measured to evaluate overall method resolution: Rs represents the resolution for EO oligomers; Rs represents the resolution for oligomer groups; Rs measures the separation of APEOs and AP. Oligomers with the same alkyl chain but different EO units were better resolved using low temperatures. Temperature ( C) Table. Effects of Column Operating Temperature Chromatography Time* (min) Resolution Resolution Resolution 2 9. 2.7 2.82 2. 2 7.8 2.6 2.92 2.2 6. 2..2 2..7 2..6.9. 2.26.8.7 2. 2.2.9.. 2.6.9. * Minimum time required to elute last peak (-NP). Mobile phase: 2% CH CN, % mm NH OAc, % H 2 O; flow rate:. ml/min Temperature dependence for the separation of APs and APEOs followed the same trend, with lower temperature yielding better resolution. However, separation of oligomer groups (OPEOs and NPEOs) showed the opposite trend: OPEOs and NPEOs were better resolved with elevated temperature. Column temperature was examined further using methanol as an organic modifier for MS detection and C was selected as the optimum condition to achieve total separation. A mobile phase with higher organic content is favorable for MS sensitivity. The methanol-containing mobile phase showed significantly improved MS response for APEO. Therefore, methanol was used for HPLC-MS analysis. Method Performance with Single Quadrupole MS Instrumentation Mass Spectrometry with SIM Detection When coupling MS detection to chromatography, ESI is the preferred ionization interface for APEO analysis. ESI showed better sensitivity and specificity for a wider range of oligomers. To improve detection limits using MS SIM detection,. mm sodium acetate was added to the mobile phase to form sodium adducts. Because sodium acetate is nonvolatile, salt condensation was observed after operating for a short period of time. To ensure the long term stability of MS detection, the optional cone wash feature was activated with a wash stream of µl/min methanol/water (/, v/v). Significant improvement of system stability was observed when employing the cone wash (.% decrease in intensity with cone wash compared with 6% decrease without cone wash after 6 runs.)

Linearity and Method Detection Limits The linearity of this method was evaluated by three replicate injections of calibration standards. The method detection limit (MDL) for each analyte was calculated by the equation: MDL = t 99% S (n=7) Where t is Student s t at 99% confidence intervals (t 99%, n=7 =.) and S is the standard deviation. The standard deviation was obtained from seven replicate injections of a calibration standard at ppb for OPEOs and ppb for NPEO2 and APs (a calibration standard at ppb was used to obtain the standard deviation for due to its low response). The results for linearity and MDLs are summarized in Table. When using adducts (sodium adducts for APEOs and acetate adducts for phenols) as quantitative ions, linear responses were observed in the range from low ppb to 2 ppb with correlation coefficients greater than.99. The calculated MDLs vary from 7. () to 7.6 ppb (). Precision and Accuracy Method precision and accuracy were calculated from seven replicated injections of standards at different concentrations based on their MS response. Precision is shown by RSD from the replicate analyses, and accuracy was calculated by (observed amount)/(specified amount) %. Method precision for each analyte is less than % with the highest bias of 9.2% for (Table ). Accuracy measurements were within % bias for each analyte except for which was 8% higher than the specified amount. Table. Linearity, Precision, Accuracy, and Method Detection Limits Using MS SIM Analyte MS Adduct Linearity From To R 2 Precision (RSD, n = 7) Accuracy (%) MDL (pg) [M+Na] + 2.999 9.2 96.9 29 a [M+Na] +.999.8 9.2 82 a [M+Na] + 2 2.996 2. 8 6 b NPEO2 [M+Na] + 2.99 2.2 96.9 2 c tert-op [M+CH COO] 2.998. 9. 222 c -OP [M+CH COO] 2.99. 97.2 282 c -NP [M+CH COO] 2.997.6 22 c a Calculated from 7 replicate injections of a ppb standard. b Calculated from 7 replicate injections of a ppb standard. c Calculated from 7 replicate injections of a ppb standard. Method Performance with Triple Quadrupole MS/MS Instrumentation Mass Spectrometry with SRM Detection An ESI interface and methanol-containing mobile phase were also used for MS/MS SRM detection. NH OAc was used to buffer the mobile phase instead of NaOAc because analyte-sodium adducts did not show strong fragment ions with SRM conditions. Two SRM transitions were used for quantification and confirmation. The SRM scan details are shown in Table 2. Calibration Range and Method Detection Limits Coefficient of determination (R 2 ) measured greater than.99 for each analyte from low calibration levels (2 ppb) to 2 ppb. MDLs were calculated by the same means as for MS SIM detection, and details are shown in Table. Lower detection limits can be achieved using MS/MS SRM compared with MS SIM detection. The SRM chromatograms of APs and APEOs are shown in Figure with the injection amount of 2 ng of each analyte. LC Conditions Column: Acclaim Surfactant (2., µm) Column Temp.: C Flow Rate:.2 ml/min Injection: 2. ng in µl Mobile Phase: A: CH OH; B: NH OAc; C: H 2 O Relative Abundance Gradient: Time (min) A B C 7 7 2 7 2 9 7 2 9 8 9 2 7 2 RT: 6.7 S/N 77 RMS MS Conditions MS Detector: TSQ Quantum Access Ionization: Electrospray Ionization (ESI) Detection Mode: Polarity Switching SRM Needle Voltage: V Capillary Temp.: 27 C Sheath Gas: Arb. Auxiliary Gas: / Arb. (segment / segment 2) Ion Sweep Gas: See Table 2 for SRM Scan Segments and Events NL: 6.89E +SRM: 268 A RT:.6 S/N 77 RMS NL: 2.9E6 +SRM: 2 8 B NL: 2.8E RT: 7.97 S/N 9 RMS +SRM: 282 27 C NL:.2E6 RT: 7. S/N 79 RMS NPEO2 +SRM: 26 8 D NL:.8E2 RT: 8.77 S/N 27 RMS tert-octylphenol SRM: 2 E NL: 7.E RT:.77 S/N 2 RMS -Octylphenol SRM: 2 6 F NL: 8.87E RT: 6.67 S/N 298 RMS -Nonylphenol SRM: 29 6 G 2 Minutes Figure. SRM Chromatograms of alkylphenols and alkylphenol ethoxylates. Reconstructed SRM chromatograms A G show APs and APEOs with 2 ng injection. Automated Large-Volume Injection for Direct Analysis of Alkylphenols and Ethoxylated Alkylphenols by Mixed-Mode Chromatography and Mass Spectrometry

Table. Calibration Range and Coefficient of Determination Using MS/MS SRM Analyte SRM Calibration Range NPEO2 From To R 2 Fitting Function Weighting 268 2.9972 Linear /X 268 7 2.996 Linear /X 2 8 2 2.9976 Quadratic /X 2 2 2 2.9976 Quadratic /X 282 27 2.996 Linear /X 282 8 2.996 Linear /X 26 8 2 2.9977 Quadratic /X 26 2 2 2.9976 Quadratic /X tert-op 2 2.997 Linear /X -OP 2 6 2.999 Linear /X -NP 29 6 2.998 Linear /X Precision and Accuracy Method precision and accuracy were measured at three levels:, 2, and 2 ppb. Calculations were performed by the same means as for MSQ SIM detection, and results are shown in Table 6. Accuracy varies from 86% to %, and RSDs range from.% to 9.%. Analyte Table 6. Precision and Accuracy Using MS/MS SRM Detection NPEO2 tert-op -OP -NP Level Mean Accuracy (%) RSD ND N/A N/A 2 2. 2 899 9..2 8.9 89. 2.7 2 96 98.. 2 76 87.8.9 ND N/A N/A 2 2 2 6.9 2 82 92.6.2 9.88 98.8 9. 2 87 9. 2. 2 82 9.6 2.2 ND N/A N/A 2 22 6 8.9 2 2.2.6 ND N/A N/A 2 72 86. 8.8 2 978 98.9. ND N/A N/A 2 2 2..9 2 26.8.6 MDL (pg) 27 2 8.6 6. 62 27 Automation of Large Volume Injection for Direct Analysis of Water Samples To improve the method detection limits and to reduce intensive sample preparationin the laboratory, an automated large-volume injection method was employed for direct analysis of these analytes using the AS-HV autosampler. The system schematic is shown in Figure. The water sample is loaded into a ml sample loop by the AS-HV then delivered by the second system pump unit through a concentrator column (Acclaim,.6 mm, µm) using DI water as the mobile phase. The analytes trapped on the concentrator column were back-eluted and separated on the analytical column (Acclaim Surfactant, 2. mm, µm), and detected by UV and MS detectors. Automation was achieved through Chromeleon software. Detection limits were significantly improved by using large-volume injections. Direct analysis of water samples for APs and APEOs can be performed using this method. Method specificity was evaluated by analyzing ml treated DI water (organic residues were removed by passing DI water through an analytical column (.6 mm, µm) as a system blank. No measurable peaks were observed at the specific retention times for target analytes except for NPEO2 (2% peak area of NPEO2 at ppt). Carryover was examined by analyzing a system blank sample after injection of a calibration standard at the highest concentration, i.e., 2 ppt. No carryover peaks were observed except for NPEO2 whose peak area was 6% of NPEO2 in a blank sample. Therefore, carryover was deemed to be negligible when using large-volume injection for direct analysis. The coefficient of determinations (R 2 ) achieved were greater than.99 for each analyte from low levels (: ppt; : ppt; : 2 ppt; NPEO2: ppt; t-op: ppt; -OP: 2 ppt; -NP: ppt) to 2 ppt. MDLs were measured by the statistical method mentioned in prior sections with n 6. MDLs for APEOs were measured at low ppt levels (:. ppt; : 6.79 ppt; : 8.2 ppt; NPEO2:. ppt), tert-op at 78 ppt and -OP at ppt. (MDL for -NP was not measured due to low response.)

AXP Sample Sample 2 Wash solvent CH OH/H 2 O (/) Figure. System schematic. /8 - /6 Adapter Eluent Trap 6 Pump Left - Off 2 2 6 Peristaltic Pump /8 - /6 Adapter Eluent Trap 6 Pump Left - DI H 2 O 2 /8 - /6 Adapter Eluent Trap 6 2 Peristaltic Pump Peristaltic Pump Pump Left - Off 2 6 2 6 Surfactant Surfactant Surfactant Step UV Pump Right Eluent In Step 2 UV Pump Right Eluent In Step UV Pump Right Eluent In MS MS MS 2897 REFERENCES. Guenther, K.; Heinke, V.; Thiele, B.; Kleist, E.; Prast, H.; Raecker, T. Environ. Sci. Technol. 22, 6, 676 68. 2. Loos, R.; Hanke, G.; Umlauf, G.; Eisenreich, S. J. Chemosphere 27, 66, 69 699.. Rice, C. P.; Schmitz-Afonso, I.; Loyo-Rosales, J. E.; Link, E.; Thoma, R.; Fay, L.; Altfater, D.; Camp, M. J. Environ. Sci. Technol. 2, 7, 77 7.. Nimrod, A. C.; Benson, W. H. Crit. Rev. Toxicol. 996, 26, 6.. Shang, D. Y.; Ikonomou, M. G.; Macdonald, R. W. J. Chromatogr. A 999, 89, 67 82. 6. Ferguson, P. L.; Iden, C. R.; Brownawell, B. J. Anal. Chem. 2, 72, 22. 7. Ferguson, P. L.; Iden, C. R.; Brownawell, B. J. J. Chromatogr. A 2, 98, 79 9. 8. Loyo-Rosales, J. E.; Schmitz-Afonso, I.; Rice, C. P.; Torrents, A. Anal. Chem. 2, 7, 8 87. 9. Andreu, V.; Ferrer, E.; Rubio, J. L.; Font, G.; Pico, Y. Sci. Total Environ. 27, 78, 2 29.. Wang, J.; Henday, S.; Schnute, W. Profiling Analysis of the Degradation Products of Alkylphenol Polyethoxylates by LC-MS Using an Acclaim Surfactant Column with Mass Spectrometric Detection. Presented at the 6th ASMS Conference, 28.. Petrovic, M.; Diaz, A.; Ventura, F.; Barcelo, D. Anal. Chem. 2, 7, 886 89. CONCLUSION This study produced methods for profiling degradation products of ethoxylated alkylphenols in water samples. Using a large-volume injection, automated direct analysis of APs and APEOs was demonstrated without labor-intensive sample preparation. Suppliers Sigma-Aldrich (St. Louis, MO, USA). Total separation of seven target analytes was achieved and selective, sensitive detection was achieved using mass spectrometric detection. Detection limits and specificity were improved over MS SIM using MS/MS SRM detection, and additional confirmation information can be obtained by introducing a second SRM transition. DCMS LINK is a trademark, and Acclaim and Chromeleon are registered trademarks of Dionex Corporation. MSQ Plus is a trademark and TSQ Quantum Access and Xcaliber are registered trademarks of Thermo Fisher Scientific, Inc. Passion. Power. Productivity. Dionex Corporation 228 Titan Way P.O. Box 6 Sunnyvale, CA North America U.S./Canada (87) 29-7 South America Europe Austria () 66 2 Benelux () 2 68 9768 (2) 29 Denmark () 6 6 9 9 France () 9 Germany (9) 626 99 Ireland () 6 6 Italy (9) 2 62 267 Sweden (6) 8 7 8 6988-6 Automated Large-Volume Brazil () 7 Injection Switzerland for () Direct 62 2 9966 Analysis United Kingdom of () Alkylphenols 276 69722 and (8) 77-7 Ethoxylated Alkylphenols by Mixed-Mode Chromatography and Mass Spectrometry www.dionex.com Asia Pacific Australia (6) 2 92 2 China (82) 228 282 India (9) 22 276 27 Japan (8) 6 688 2 Korea (82) 2 26 28 Singapore (6) 6289 9 Taiwan (886) 2 87 66 LPN 22- /9 29 Dionex Corporation