Anal. Chem. 2004, 76, 1437-1444
Combination of LC/TOF-MS and LC/Ion Trap MS/MS for the Identification of Diphenhydramine in Sediment Samples Imma Ferrer,*,† Curt E. Heine,‡ and E. Michael Thurman†
U.S. Geological Survey, Denver, Colorado 80225, and Waters Corporation, 100 Cummings Center, Beverly, Massachusetts 01915
Diphenhydramine (Benadryl) is a popular over-thecounter antihistaminic medication used for the treatment of allergies. After consumption, excretion, and subsequent discharge from wastewater treatment plants, it is possible that diphenhydramine will be found in environmental sediments due to its hydrophobicity (log P ) 3.27). This work describes a methodology for the first unequivocal determination of diphenhydramine bound to environmental sediments. The drug is removed from the sediments by accelerated solvent extraction and then analyzed by liquid chromatography with a time-of-flight mass spectrometer and an ion trap mass spectrometer. This combination of techniques provided unequivocal identification and confirmation of diphenhydramine in two sediment samples. The accurate mass measurements of the protonated molecules were m/z 256.1703 and 256.1696 compared to the calculated mass of m/z 256.1701, resulting in errors of 0.8 and 2.3 ppm. This mass accuracy was sufficient to verify the elemental composition of diphenhydramine in each sample. Furthermore, accurate mass measurements of the primary fragment ion were obtained. This work is the first application of timeof-flight mass spectrometry for the identification of diphenhydramine and shows the accumulation of an over-thecounter medication in aquatic sediments at five different locations. Several studies in the past few years have established the presence of pharmaceuticals in the environment in both groundwater and surface water. Commonly used drugs excreted in urine or feces exist in municipal wastewaters in many forms and at sufficient concentrations for subsequent detection from environmental samples. Several publications report the presence of pharmaceutical compounds in water.1-5 Whereas water compart* To whom correspondence should be addressed. Current address: Department Hydrogeology and Analytical Chemistry, University of Almeria, 04120 Almeria, Spain. E-mail:
[email protected]. † U.S. Geological Survey. ‡ Waters Corp. (1) Ternes, T.; Bonerz, M.; Schmidt, T. J. Chromatogr., A 2001, 938, 175185. (2) Ternes, T.; Anderson, H.; Gilberg, D.; Bonerz, M. Anal. Chem. 2002, 74, 3498-3504. (3) Farre´, M.; Ferrer, I.; Ginebreda, A.; Figueras, M.; Olivella, L.; Tirapu, Ll.; Vilanova, M.; Barcelo´, D. J. Chromatogr., A 2001, 938, 187-197. 10.1021/ac034794m CCC: $27.50 Published on Web 02/03/2004
© 2004 American Chemical Society
ments have been widely studied and characterized for the presence of pharmaceuticals, solid matrixes, such as soil and sediments, have not been studied. Some hydrophobic compounds may accumulate in sediments, and the octanol-water partition coefficient (log P) values for pharmaceuticals provide a basis to predict the tendency of adsorption to soils or sediments. Diphenhydramine (2-diphenylmethoxy-N,N-dimethylethanamine) has been one of the most popular over-the-counter medications for the treatment of allergies for over 50 years. It is an antihistamine used to relieve allergic rhinitis, such as seasonal allergy and itching associated with uncomplicated allergic skin reactions. It is also used as a short-term sleep aid, to control coughs due to colds or allergy, and to treat motion sickness. Diphenhydramine is available in nonprescription products as the lone active ingredient and also in combination with other drugs. A typical capsule contains 25-50 mg of diphenhydramine for oral administration. The widespread use of diphenhydramine increases the likelihood of its occurrence in environmental compartments after excretion. The log P value for this compound is 3.27, indicating a high coefficient for partition into soils or sediments. It is also sufficiently water soluble to be transported to surface waters. The environmental relevance of this compound is associated with its potential for accumulation in bottom sediments and its possible toxicity to aquatic organisms that feed on these sediments. Moreover this compound could represent a model for other pharmaceuticals and their accumulation in solid matrixes. These are reasons for developing a sensitive extraction and confirmatory method for the identification and quantitation of pharmaceuticals, such as diphenhydramine, in sediment samples. Liquid chromatography/mass spectrometry with an ion trap mass analyzer (LC/ion trap MS) is a useful tool for qualitative analysis due to three important characteristics: high sensitivity for full spectral acquisitions, the efficient formation of fragment ions by collision-induced dissociation, which provide useful structural information, and the capability of conducting MSn experiments that establish the lineage and identity of individual product ions obtained from a selected precursor ion. Ion trap MS/ MS and MSn data have been used successfully for the identification of organic contaminants in complex environmental samples.6 (4) Golet, E. M.; Alder, A. C.; Giger, W. Environ. Sci. Technol. 2002, 36, 36453651. (5) Kolpin, D. W.; Furlong, E. T.; Meyer, M. T.; Thurman, E. M.; Zaugg, S. D.; Barber, L. B.; Buxton, H. T. Environ. Sci. Technol. 2002, 36, 1202-1211.
Analytical Chemistry, Vol. 76, No. 5, March 1, 2004 1437
However, when appropriate standards are not available, uncertainty in the identity of a compound may exist even after analysis with an ion trap mass spectrometer. Furthermore, sediment extracts may contain isobaric interferences due to the presence of a complex mixture of organic compounds, and these interferences may confuse data interpretation. A modern orthogonal acceleration time-of-flight mass spectrometer (with a reflectron field) (TOF-MS)7 generates increased resolving power of signals on the m/z axis in comparison to an ion trap mass spectrometer. This resolving power benefits analyses involving complex environmental matrixes such as sediments, by separating isobaric interferences from the contaminant signals of interest. The improved resolution also facilitates the measurement of accurate masses within 5 ppm, which are accepted for the verification of elemental compositions. Elemental compositions of contaminants and their fragment ions clearly constitute higherorder identifications than those afforded by nominal mass measurements. TOF-MS has been applied to the identification of pesticides,8,9 surfactants,6 and some drugs.10-14 However, no environmental identifications of diphenhydramine have been reported by TOF-MS or any other LC/MS method. The combination of ion trap MS/MS for structural information and TOF-MS for accurate mass measurements and elemental compositions of ions represents a powerful analytical approach for the identification of organic compounds in complex matrixes, including potential unknowns in environmental samples. The work reported here demonstrates this approach for the analysis of diphenhydramine in sediment samples. EXPERIMENTAL SECTION Chemicals. Diphenhydramine and atrazine (surrogate standard) were obtained from Sigma-Aldrich Co. (St Louis, MO). Atrazine-d5 was purchased from Cambridge Isotopes (Cambridge, MA), and it was used as an internal standard. High-performance liquid chromatography (HPLC) grade acetonitrile, methanol, and water were purchased from Burdick and Jackson (Muskegon, MI). Ammonium formate and formic acid were obtained from Aldrich (Milwaukee, WI). Sampling. Sediment samples were collected downstream from different wastewater treatment plants and river sites near the plants. Grab sediment samples were collected from five locations in the United States. The sampled streams receive discharge from wastewater treatment plants. The sediment samples were collected from depositional zones in wadeable sections of the stream. Sediment samples were collected in glass jars (baked at 300 °C) (6) Ferrer, I., Thurman, E. M., Eds. Liquid Chromatography Mass Spectrometry/ Mass Spectrometry, MS/MS and Time-of-Flight MS: Analysis of Emerging Contaminants; American Chemical Society Symposium Series 850; American Chemical Society: Washington, DC, 2003; p 415. (7) Cotter, R. J. Anal. Chem. 1999, 71, 445A-451A. (8) Bobeldijk, I.; Vissers, J. P. C.; Kearney, G.; Major, H.; van Leerdam, J. A. J. Chromatogr., A 2001, 929, 63-74. (9) Thurman, E. M.; Ferrer, I.; Parry, R. J. Chromatogr., A 2002, 957, 3-9. (10) Zhang, H.; Heinig, K.; Henion, J. J. Mass Spectrom. 2000, 35, 423-431. (11) Zhang, N.; Fountain, S. T.; Bi, H.; Rossi, D. T., Anal. Chem. 2000, 72, 800806. (12) Sundstrom, I.; Hedeland, M.; Bondesson, U.; Andren, P. E. J. Mass Spectrom. 2002, 37, 414-420. (13) Eckers, C.; Haskins, N.; Langridge, J. Rapid Commun. Mass Spectrom. 1997, 11, 1916-1922. (14) Michelsen, P.; Karlsson, A. Rapid Commun. Mass Spectrom. 1999, 13, 21462150.
1438
Analytical Chemistry, Vol. 76, No. 5, March 1, 2004
with Teflon-lined lids, chilled, and either directly transported to the laboratory or shipped (chilled) to the laboratory by overnight express. Samples were held at 4 °C until extraction. Sediment Fortification Experiments. The recoveries (percentage of standard added to sample that is recovered during extraction and cleanup) and reproducibility (relative standard deviation for triplicate analysis) of the extraction method were determined by sediment fortification. Previously freeze-dried sediments from Evergreen Lake, CO, were fortified with diphenhydramine at 100 µg/kg and atrazine (used as a surrogate standard) at 500 µg/kg and then incubated for at least 24 h to allow time for the analyte to interact with the matrix of the sediment and thus approximate real conditions. These sediment samples were previously used for validating a gas chromatography/mass spectrometry (GC/MS) method for semivolatile compounds, such as polyaromatic hydrocarbons, and the ambient concentrations of other contaminants did not interfere in the method developed in this work. The diphenhydramine fortification solution (10 µg/mL) was prepared in methanol, and it was also used for quantitation of the analyte recovered after extraction. A dilution of this standard solution was made, and it was analyzed by ion trap. The areas for diphenhydramine in the standard solution and the areas from the sample extracts were compared in order to calculate the recoveries of extraction. Unfortified sediment samples also were analyzed for interferences and for determining whether ambient concentrations of diphenhydramine were present. No traces of diphenhydramine were found in the unfortified sediment samples. Accelerated Solvent Extration (ASE) Extraction. An automated Dionex-ASE 200 system (Dionex Co., Sunnyville, CA) was used for the sediment extractions. Ten grams of wet sediment was packed into an 11-mL stainless steel ASE vessel. The packed vessels were sealed at both ends with circular cellulose filters of 2.1-cm diameter (Whatman, Springfield Mill, Maidstone, Kent, U.K.). Optimized conditions included the use of a 7:3 (by volume) mixture of acetonitrile and water as extraction solvent. The temperature of extraction was 120 °C, and the pressure was 1500 psi. A 6-min preheat of the sediment/solvent mixture preceded the extractions. The sediment was subjected to three 5-min static extraction cycles. At the end of each extraction, nitrogen gas was used to expel the extract into glass collection vials (60-s purge). The total volume of extract was ∼18 mL. Atrazine-d5 was added to the final volume at a concentration of 0.05 µg/mL. Afterward, 1 mL of the extract was passed through a syringe filter for cleanup. The acetonitrile content was evaporated with a gentle stream of nitrogen and reconstituted with mobile phase, and aliquots of this solution were subsequently injected on the two LC/MS systems for analysis. Ion Trap MS/MS. Liquid chromatography/electrospray/ion trap mass spectrometry (LC/ESI-MS), in positive ionization, fullscan operation, was used to separate and identify diphenhydramine. The analyte was separated from matrix interferences using an HPLC (series 1100, Agilent Technologies, Palo Alto, CA) equipped with a reversed-phase C18 analytical column (Phenomenex RP18, Torrance, CA) of 250 mm by 3 mm and 5-µm particle diameter. Column temperature was maintained at 25 °C. Mobile phase A was acetonitrile, and mobile phase B consisted of 10 mM ammonium formate buffer. A linear gradient progressed from 15%
A (initial conditions) to 100% A in 40 min, after which the mobilephase composition was maintained at 100% A for 5 min. The flow rate was 0.6 mL/min, and 50 µL of the sediment extracts was injected. This HPLC system was connected to an ion trap mass spectrometer (Esquire LC, Bruker Daltonics, Billerica, MA) equipped with an ESI probe. The operating parameters of the MS system were optimized in full-scan mode (m/z range, 50-400) by flow injection analysis of the analyte. The maximum ion accumulation time was 200 ms. The internal pressure in the ion trap was of 1.2 × 10-5 mbar. Quantitation of the sediment samples was performed with an external standard curve by direct injection of diphenhydramine at concentrations ranging from 0.01 to 1 µg/ mL (y ) 4 × 106x - 8473) and corrected for percent recovery. Limits of detection for diphenhydramine were 100 pg on column. LC/TOF-MS. An Agilent series 1100 HPLC was used for the LC/TOF-MS analyses. The standard and extracted sediment samples were eluted from a MetaChem MetaSil AQ C18 column with dimensions of 2 × 150 mm, at a flow rate of 0.5 mL/min. Mobile phases A and B were water with 0.1% formic acid and acetonitrile with 0.1% formic acid, respectively. The chromatographic method held the initial mobile-phase composition (10% B) constant for 8 min, followed by a linear gradient to 100% B at 25 min; 15 µL of the extracts was injected. An orthogonal acceleration TOF mass spectrometer, the LCT (Waters Corp., Manchester, U.K.), was used in positive ion mode with electrospray ionization. This instrument is equipped with a 3.6-GHz timeto-digital converter and a reflectron that generated a resolving power of 5900 at m/z 556 (fwhm definition). Leucine enkephalin ([M + H]+ ) 556.2771) was added postcolumn as a lock mass to compensate for drift of the external calibration. The instrument was operated with a cone voltage of 22 V. Accurate mass data were processed (centroiding of continuum data with lock mass and digital deadtime correction) using Masslynx 3.5 software. Exact masses corresponding to particular elemental compositions were also calculated by Masslynx 3.5. LC/TOF-MS analyses were used for identification and confirmation rather than quantitation as will be shown later. RESULTS AND DISCUSSION ASE Extraction and Recoveries. An optimization of the recoveries of extraction for diphenhydramine from sediments was carried out with different solvent compositions and different temperature and pressure values. Solvent, temperature, and pressure are the three most important variables that can affect the extraction efficiency by ASE. The results obtained for the recoveries using different solvent mixtures are shown in Table 1. The highest recoveries were obtained using an acetonitrile/water mixture (70:30). The average extraction recovery for a diphenhydramine standard from sediments using this solvent composition was 75% with a relative standard deviation of 8%. We also investigated previously the effects of changing the temperature and pressure on extraction efficiency. Temperatures were varied between 75 and 200 °C at a constant pressure of 1500 psi. An increase in temperature (above 75 °C) led to higher extraction recoveries. Above 120 °C, no substantial differences were observed. Temperaturesrather than pressure or timesis expected to have a greater effect on solubility and mass transfer for the performance of ASE. The increased solubility of water in organic solvents at elevated temperatures increases the contact
Table 1. ASE Extraction Recoveries for Diphenhydramine from Sediments Using Different Solvents and Final Conditions Used in This Study extraction solvent
recovery of extraction (%) and RSDa
ACN/water (60:40) MeOH/water (60:40) acetone/water (60:40) MeOH/acetone/water (60:20:20) ACN/water (70:30)
59 ((5) 17 ((9) 54 ((7) 33 ((7) 75 ((8)
Final Conditions sediment 10-15 g cell size 11 mL solvent 70% ACN/30% H2O temperature 120 °C pressure 1500 psi cycles of extraction 3 a
RSD, relative standard deviation.
of the solvent with the analytes in wet sediment, and they should be extracted more efficiently than at room temperature. Also, another effect of increasing extraction temperature is to overcome the energy barrier of desorption. The pressure applied to the cell is necessary to keep the solvents in a liquid state when temperatures greater than their boiling points at 1 atm are used. Secondarily, pressure enhances permeation of solvent into the sample matrix. The effect of pressure on analyte recoveries was investigated by extracting the fortified sediment samples at pressures ranging from 1000 and 2000 psi at a constant temperature of 120 °C. The purpose of this test was to determine whether elevated pressures improved solvent diffusivity into the matrix pores, where the analyte might be sequestered. In this case, no substantial differences were observed in the extraction recoveries for diphenhydramine within this pressure range. Table 1 also shows the final conditions chosen for the diphenhydramine extractions from sediments. LC Ion Trap Standard Analyses. Figure 1a shows the m/z 256 mass chromatogram obtained from an extract of a sediment fortified with diphenhydramine at 100 µg/kg. The m/z 256 signal corresponds to the protonated diphenhydramine molecule. This figure also shows mass spectral data acquired during the indicated elution of diphenhydramine. Upon protonation, diphenhydramine undergoes a facile fragmentation pathway to an extremely stable secondary benzylic carbocation with an m/z value of 167. It is important to note the pronounced tendency to form this fragment ion as indicated by the low relative intensity of the [M + H]+ signal (only 5%) in the mass spectrum, despite the low fragmentor voltage of 70 V. Many pharmaceutical compounds do not fragment significantly under these mild conditions. Figure 1b shows the MS/MS data obtained from the m/z 256 precursor, utilizing a 1.2-V amplitude for fragmentation in the ion trap. The fragmentation to m/z 167 was the only one observed, and it corresponds to the cleavage of the ether linkage (see Figure 1b) and charge transfer to form the even-electron secondary benzylic carbocation with an elemental composition of C13H11. Figure 1c shows the chromatogram corresponding to the analysis of an unfortified sediment sample (blank); as we can see in this figure, no signal at m/z 167 is observed; therefore, no presence of diphenhydramine was observed. Analytical Chemistry, Vol. 76, No. 5, March 1, 2004
1439
Figure 1. (a) m/z 256 mass chromatogram from a sediment fortified with diphenhydramine analyzed by ion trap LC/ESI-MS in positive ion mode. The corresponding full-scan mass spectrum is also shown (total quantity of analyte injected ∼3 ng). (b) MS/MS product ion spectrum from m/z 256 precursor and (c) m/z 167 extracted ion chromatogram of an unfortified (blank) sediment sample.
LC/TOF-MS Standard Analyses. LC/MS data were also acquired from a diphenhydramine standard using time-of-flight mass spectrometry. Figure 2a presents the mass chromatogram at m/z 256, and Figure 2b shows a background-subtracted mass spectrum at the elution time of diphenhydramine (15.5 min). As observed in the ion trap data, the peak at m/z 167 dominates the mass spectrum even at a gentle cone voltage of 22 V. For most pharmaceutical compounds, this cone voltage would not impart enough energy to the [M + H]+ ions (through collisions with N2 molecules in the source region) to stimulate pronounced fragmentation. As discussed above, this observation confirms the facile 1440 Analytical Chemistry, Vol. 76, No. 5, March 1, 2004
nature of the fragmentation to the ion at m/z 167. Although both instruments readily generate the m/z 167 fragment, the Z-spray source of the LCT apparently transfers less energy to the diphenhydramine [M + H]+ ions as indicated by the higher intensity of the [M + H]+ signal relative to the fragment ion signal (∼30%). The incorporation of the reflectron field in the TOF mass analyzer design increases resolving power in comparison to an ion trap or quadrupole system. This improved resolution enhances the selectivity of the data by separating isobaric interferences. Figure 3a contrasts the narrow peaks associated with the TOF
Figure 2. (a) m/z 256 mass chromatogram for diphenhydramine standard by LC/TOF-MS, and (b) the corresponding background-subtracted mass spectrum. Total quantity of analyte injected, 1 ng. Table 2. Accurate Mass Measurements for the Diphenhydramine Standard by LC/TOF-MS
m/z 256 m/z 167
Figure 3. (a) Time-of-flight mass spectrum from m/z 166 to 169 and (b) ion trap mass spectrum from m/z 166 to 169 for the diphenhydramine standard.
instrument with those obtained from the ion trap (Figure 3b) for the m/z 167 fragment ion signal. Table 2 contains accurate mass measurements for the protonated diphenhydramine standard and the abundant benzylic fragment ion. This table also contains calculated exact masses corresponding to the elemental compositions of the protonated molecule (C17H22NO) and the fragment (C13H11). The agreement (within 5 ppm) demonstrated between the measured and calculated masses serves to verify the proposed elemental compositions and represents a higher order identification of this compound than that based on the nominal mass assignments available from the ion trap mass spectrometer.
measured mass
calculated exact mass
mDa error
ppm error
formula
256.1700 167.0859
256.1701 167.0861
-0.1 -0.2
-0.4 -1.3
C17H22NO C13H11
Identification of Diphenhydramine in Sediment Samples. The two LC/MS methods were evaluated with sediment samples suspected to contain environmentally relevant concentrations of diphenhydramine. Sediment properties were not fully characterized, but the clay fraction (defined as that portion of the sediments
