ISSN 10619348, Journal of Analytical Chemistry, 2012, Vol. 67, No. 2, pp. 82–97. © Pleiades Publishing, Ltd., 2012. Original Russian Text © S.S. Aleksenko, 2012, published in Zhurnal Analiticheskoi Khimii, 2012, Vol. 67, No. 2, pp. 116–132.
REVIEWS
Liquid Chromatography with MassSpectrometric Detection for the Determination of Chemical Warfare Agents and Their Degradation Products S. S. Aleksenko Saratov Military Institute of Biological and Chemical Safety, pr. 50 Let Oktyabrya 5, Saratov, 410037 Russia Received November 24, 2010; in final form, May 16, 2011
Abstract—This review summarizes the results of highperformance liquid chromatography with massspec trometric detection (HPLC–MS) in the identification of chemical warfare agents and their degradation products, obtained in the last 13 years passed since the ratification of the Convention on the Prohibition of the Development, Production, Stockpiling, and Use of Chemical Weapons and on Their Destruction (the Convention). The conditions of the separation and detection of compounds in a variety of ionization tech niques (electrospray ionization, atmospheric pressure chemical ionization) and analyzers (timeofflight (TOF), tandem mass spectrometry, triple quadrupole, ion trap) are considered. The detection limits of the degradation products are given; the possibility of identifying compounds and the methods of sample prepa ration using contemporary methods of preconcentration (solvent microextraction) and separation (use of various adsorbents, molecularlyimprinted polymers) are described. The features of the HPLC–MS analysis of environmental samples (water, soil) and biological fluids (urine, blood serum) are discussed. The review is focused on the determination of the degradation products and derivatives of nerve agents, that is, alkylphos phonic acids. Keywords: highperformance liquid chromatography with mass spectrometric detection, chemical warfare agents, nerve agents, degradation products of chemical warfare agents, alkylphosphonic acids. DOI: 10.1134/S1061934812020025
from the aqueous phase into the organic phase, fol lowed by derivatization. In those cases, the application of HPLC is more perspective. A conventional spectro photometric detector is of little use in determining CWA and DP because of the lack of chromophoric groups in their molecules. The development of inter faces that allow the combination of HPLC with mass spectrometry (HPLC–MS) greatly expanded the pos sibilities of the method for the identification and quantification of these substances.
In connection with the ratification of the Conven tion and the beginning of the destruction of chemical weapons, the determination of chemical warfare agents (CWA) and their degradation products (DP) has become particularly important in the late 1990s. First of all, this is due to that the safety of the person nel and the public need to be ensured during storage, transportation, and destruction of chemical weapons. The latter suggests the implementation of the system for continuous monitoring of the destruction of CWA and the concentration of CWA and products of their degradation in the environment. The derivatives of alkylphosphonic acids (APAs) are among the main DP of organophosphorus nerve agents, while thiodiglycol and sulfur and oxygencontaining longchain diols are for blister agents (Fig. 1). In the last decade, a growing number of publications have been devoted to the identification of these compounds by gas chroma tography (GC), HPLC, and capillary electrophoresis. Gas chromatography coupled with mass spectromet ric detection remains one of the most reliable methods for determining agents in most cases due to its high sensitivity and selectivity and the presence of large libraries of mass spectra [1]. However, the determina tion of DP in water samples, water extracts from soil, or biological fluids requires the transfer of compounds
Achievements, constraints, and prospects of the development of HPLC–MS for the determination of CWA and DP were previously analyzed in several reviews [2–5]. Applied aspects of the method for determining various hazardous agents, including chemical warfare agents, for the period of 2002–2005 are presented in [2]. A place of HPLC among other chromatographic and electrophoretic methods for the determination of CWA and DP is considered [3–5]. We should also note the book, a chapter of which is devoted to the use of HPLC–MS for the determina tion of substances prohibited by the Convention [6]. However, similar works have not been published over the past 5 years. This review summarizes the results of the application of HPLC–MS for the determination of CWA and DP for 13 years have passed since the rat 82
LIQUID CHROMATOGRAPHY WITH MASSSPECTROMETRIC DETECTION
H3C GB
CH3 O O P F CH3
H3C I
CH3 O O P OH CH3
H3C
H3C VX
CH3
O H3C O P S CH3
CH3 III
O O P OH CH3
N CH3
S
Cl
HO
HD
CH3
H3C
O H3C P S HO EA 2192 Cl
O HO P OH CH3 MPA
CH3 CH3 O H3C CH3 O P OH II CH3
CH3 CH3 O H3C CH3 O P F GD CH3
83
S Thiodiglycol
N
CH3 CH3
OH
Fig. 1. Structure of chemical warfare agents and their destruction products: (GB) Sarin, Oisopropylmethylfluorophosphonate; (GD) Soman, Opinacolylmethylfluorophosphonate; (VX) OethylS(2diisopropylaminoethyl)methylthiophosphonate; (HD) sulfur mustard; (EA 2192) S(2diisopropylaminoethyl)methylthiophosphonate; (MPA) methylphosphonic acid; (I) Oisopropyl methylphosphonic acid; (II) Opinacolyl methylphosphonic acid; and (III) Ethyl methylphosphonic acid.
ification of the Convention, including the analysis of the conditions of chromatographic separation and detection, detection limits achieved, the possibility of identification of compounds, the means for sample preparation of certain samples (water, soil, biofluids), and the contemporary options for preconcentration and isolation of analytes. Determination of Chemical Warfare Agents Several papers were published demonstrating cap illary HPLC (column inner diameter, 0.32 mm) with mass spectrometric detection can principally be applied for the determination of relatively high con centrations of nerve agents in aqueous media (0.01– 0.1 mg/cm3) and soil (10 µg/g), which is several orders of magnitude higher than their maximum permissible concentrations (MPCs) [7, 8]. Tabun, cyclohexylme thylfluorophosphonate (Cyclosarin), Sarin, and Soman are separated for 18 min on a reversed phase column in a gradient elution mode with a mixture of solutions of 0.1% trifluoroacetic acid in water and in acetonitrile (Fig. 2) [7]. The mass spectra of ionization products consist mainly of ions [M + H]+ or adducts of type [M + H + CH3CN]+ and the protonated JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 67
dimers, which can serve to confirm the molecular weights of unknown compounds (Fig. 2). Microcolumn HPLC was used to analyze soil, water, and snow samples, containing, in addition to CWA, their DP [9–11]. In addition to the environ mental samples, the methods for determining sarin, soman, cyclosarin, and tabun at a level of 4–20 µg/g in the office stuff such as paper, fabrics, synthetic carpet, and wallpaper were developed (Table 1) [12]. To over come the strong masking effect exerted by the matrix components, a combination of HPLC with tandem mass spectrometry (HPLC–MS–MS) was used to obtain chromatograms for selected ions (for example, to separate the signal of sarin [M + H]+, the spectrum was scanned at m/z 141). Prior to determination, CWA and DP were isolated from soil or office stuff usually with water [8, 12]. The degree of extraction of sarin and soman varied depending on the type of soil from 20 to 90% [8] and from 50 to 85% for office stuff [12]. Cyclosarin and soman are well adsorbed with wall wallpaper painted with latex paints: the degree of recovery was only 25% [12]. The detection limits of agents by HPLC–MS are usually much higher than their MPCs. For example, the MPC of sarin for water reservoirs is 5 × 10–5 mg/dm3, while for the soil, 2 × 10–4 mg/kg, which determines No. 2
2012
84
ALEKSENKO 3
Intensity, % 100
(а)
2 4
80 1 60 40 20
0
8
4
12
16 t, min
(b) +
100
99
MH [MH + CH3CN]+ 141 182 M2H+ 281 1 159
0 [MH + CH3CN]+ MH+ 204 163
100
2 M2H+ 325
276 0 [MH + CH3CN]+
M2H+ 361
222
100 MH+ 181
140 99
3 279
0 100
M2H+ 365
[MH + CH3CN]+
140 85
281
224 181
99 126
4
238
0 100
150
200
250
300
350
400 m/z
Fig. 2. (a) Chromatogram of a water sample containing organophosphorus nerve agents and (b) mass spectra of chemical warfare agents [7]. Capillary column (150 mm × 0.32 mm), Zorbax C18 SB (5 mm), gradient elution from 0.1% CF3COOH solution in water (A) to 0.1% СF3COOH solution in a 95 : 5 mixture of СН3СN⎯water (B) with the concentration of B increasing from 1 to 75% for 30 min; the mobile phase flow rate, 5 µL/min; mass spectrometric detector with electrospray ionization: (1) sarin, (2) tabun, (3) cyclosarin, and (4) soman (concentration 0.1 mg/cm3). JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 67
No. 2
2012
LIQUID CHROMATOGRAPHY WITH MASSSPECTROMETRIC DETECTION
85
Table 1. Sample preparation and HPLC–MS determination of chemical warfare agents and their destruction products
Compounds
Sample
Stationary Sample phase, particle Eluent (flow rate, Detector µL/min) preparation size (column size, mm)
Sarin, soman, tabun, cyclosarin
Water
–
Zorbax C18 SB, 5 μm (150 × 0.32)
Gradient elu ESI, tion: from 0.1% tandem CF3COOH solu tion in water to 0.1% CF3COOH solution in a 95 : 5 CH3CN– water (5)
Sarin, soman, isopropyl MPA and pinacolyl MPA
Soil
Ultrasonic water ex traction
As above
As above (16)
Sarin, soman, tabun, cyclosarin
Office Ultrasonic stuff: car water ex pet, paper, traction wallpaper, fabric
As above
Linear range, ng/cm3
Detection limit, ng/cm3
Refer ence
10–100
–
[7]
ESI 10 μg/g (Zshape), timeof flight
–
[8]
ESI 4–20 μg/g – Gradient elu tion: from 0.1% (Zshape), CF3COOH solu tandem tion in water to CH3CN (5)
[12]
Ethyl MPA (I), Plasma, isopropyl MPA water (II), and pinaco lyl MPA (III)
Derivatiza CAPCELL PAK UG C18 tion with pbro (150 × 1.5) mophena cyl bromide (BPB). Precipita tion of se rum pro teins with CH3CN, derivatiza tion with BPB, tran sition to CH2Cl2, adsorption on a Bond Elut SI car tridge, elu tion (CH2Cl2– MeOH), evaporation
Isocratic elution: FAB, tan 1—(I); 5–100 (I, II) [15] (55 : 45 vol) dem 0.5—(II); 10–100 5 mM 5—(III) (III) CH3COONH4– CH3CN contain ing 0.1% solution of glycerol (100)
Soil MPA, EPA, PPA; isopropyl MPA, pinacolyl MPA, nhexyl MPA and npentyl MPA; PPA iso propyl ester
Extraction Spherisorb S5 ODS2 with (150 × 2.0) CH3OH, extraction with water, evaporation
Gradient elu ESI, tan tion: from 1 mM dem tributylamine so lution in water to 1 mM tributyl amine solution in CH3OH (300)
JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 67
No. 2
2012
500 (es – ters) 10000 (alkyl phosphon ic acids)
[16]
86
ALEKSENKO
Table 1. (Contd.)
Compounds
Sample
Stationary Sample phase, particle Eluent (flow Detector preparation size (column rate, µL/min) size, mm)
Isocratic elu ESI triple tion: quadru (86 : 14 vol) pole CH3CN– 20 mM CH3COONH4 (500–1000)
0.075 (I), 0.16 (II), 0.24 (III), 0.03 (IV), 0.05 (V)
(2(methyl)pro Urine pyl MPA (I), ethyl MPA (II), isopropyl MPA (III), pinacolyl MPA (IV), and cyclohexyl MPA (V)
Evapora Atlantis tion to dry HILIC, 3 μm ness, redis (50 × 2.1) solution, re covery on a Strata Si1 cartridge
Sulfur and oxa Soil containing diols (DP of mustard)
Ultrasonic water ex traction
Gradient elu ESI 80 µg/g Zorbax C18SB, 5 μm tion: from 0.1% (Zfrom), (150 × 0.32) CF3COOH so timeof lution in water flight; to 0.1% ESI, tan CF3COOH so dem lution in (95 : 5) CH3CN–wa ter (10)
MPA; ethyl Standard MPA, npropyl solutions MPA, isopropyl MPA, isobutyl MPA, cyclohex yl MPA, and pi nacolyl MPA; EPA; methyl, ethyl, npropyl, and isopropyl esters of EPA; PPA; methyl and sec butyl esters of PPA
–
Hichrom RPB C8/C18, 5 μm (250 × 2.1)
Gradient elu ESI, tion: from 0.1% APCI HCOOH solu tion in water to 0.1% HCOOH solution in MeOH (200)
MPA; ethyl Water, soil MPA, isopropyl MPA, isobutyl MPA, pinacolyl MPA, and cy clohexyl MPA
Water: membrane filtration Soil: ultra sonic water extraction
Zorbax C18 SB, 5 μm (150 × 2.1), tcol = 25°C
Gradient elu tion: from 10 mM HCOONH4 solution in wa ter to 10 mM HCOONH4 so lution in CH3OH (200)
ESI (Zform), timeof flight
Linear range, ng/cm3
Detection limit, ng/cm3
Refer ence
1–200
[17]
–
[18]
–
[19]
1–8 µg/g – (soil), 700– 2000 (water)
[20]
<50
JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 67
No. 2
2012
LIQUID CHROMATOGRAPHY WITH MASSSPECTROMETRIC DETECTION
87
Table 1. (Contd.)
Compounds
VX
Sample
Blood
Stationary Sample phase, particle Eluent (flow Detector preparation size (column rate, µL/min) size, mm)
Detection Linear range, Refer limit, ng/cm3 ence ng/cm3
Microdialy Luna C18(2), Isocratic elu ESI, triple – tion: quadru sis 3 µm (80 : 20 vol) pole (150 × 2.0) 0.05% CF3COOH solution in wa ter–0.05% CF3COOH solution in CH3CN (200)
Methyl MPA (I), Water isopropyl MPA (II), isobutyl MPA (III), pen tyl MPA (IV), pinacolyl MPA (V), hexyl MPA (VI), and heptyl MPA (VII) MPA, isoPPA pentyl ester (VIII)
Hollow fi ber liquid phase mi croextrac tion, 1 h
Ethyl MPA, iso Urine propyl MPA, isobutyl MPA, pinacolyl MPA, and cyclohexyl MPA
Isotope di Xterra MS, lution, solid 3.5 µm (150 × 2.1) phase ex traction on an Isolute PEAX car tridge, evaporation
MPA (I), ethyl Beverag – MPA (II) and es isopropyl MPA (III), EPA (IV), PPA (V)
Hypercarb (100 × 2.1)
PRPX100, 5 µm (100 × 4.6) PRPX100, 5 µm (150 × 2.1)
JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 67
0.002–1
[21]
Gradient elu ESI, ion tion: from trap 0.01 M HCOONH4 solution in wa ter to 0.01 M HCOONH4 solution in CH3OH (200)
100 (I), 5000–30000 2 (II, III), 0.1 (IV–VIII)
[22]
Gradient elu ESI, tion: from tandem 0.5% HCOOH solution in wa ter to CH3OH (200), T = 40°C
≤0.5
[23]
Gradient elu tion 1: from 0.5% HCOOH solution in wa ter (pH ~ 2.3) containing 5 vol % of CH3OH to 0.3 M HCOONH4 (pH ~ 2.3) containing 22 vol % of CH3OH (1000). Gradient elu tion 2: 10 mM (NH4)2CO3 (pH ~ 8.5)— 50 mM (NH4)2CO3 (pH ~ 8.5) (500)
18.3*, 7,4 (I), 20.4*, 11.2 (II) 19.5*, 28,6 (III) 19.9*, 7.4 (IV) 22.9*, 14.5 (V)
No. 2
2012
ESI, in ductively coupled plasma
1–200
[24]
88
ALEKSENKO
Table 1. (Contd.)
Compounds
Sample
Thiodiglycol, Soil bis(2hydroxy ethylthio)meth ane, 1,2bis(hy droxyethyl thio)ethane, 1,3bis(hyd roxyethyl thio)propane, 1,4bis(hyd roxyethyl thio)butane
Stationary Sample phase, particle Eluent (flow preparation size (column rate, µL/min) size, mm) Double ul LiChrosorb trasonic wa RP18.5 µm ter extrac (150 × 0.32); tion
Detector
Linear range, ng/cm3
Detection limit, ng/cm3
Isocratic elu ESI, triple – tion: (80 : 20 quadru vol) 0.2 vol % pole HCOOH solu tion in water– CH3OH
–
Refer ence [25]
Notes: MPA, methylphosphonic acid; EPA, ethylphosphonic acid; PPA, propylphosphonic acid; ESI, electrospray ionization; FAB, fast atom bombardment; APCI, atmospheric pressure chemical ionization. * The first value is obtained in gradient mode 1, and the second value, in gradient mode 2.
the need for preconcentration of the samples. At the same time, the ability to determine in one sample both CWA and their degradation products without derivati zation of the latter is an advantage of HPLC over GC. According to the rules of the Organization for the Pro hibition of Chemical Weapons, the results of the deter minations of CWA should be confirmed by at least two methods; therefore, in a number of papers, HPLC– MS or HPLC–MS–MS were used in a combination with several other methods, including GC and NMR [13, 14]. Determination of Decomposition Products of Chemical Warfare Agents Conditions of the chromatographic separation of DP are summarized in Table 1; it is seen that volatile acetate, formate, and trifluoroacetate water–metha nol or water–acetonitrile solutions are used as mobile phases. Compounds that differ little in polarity, are separated in an isocratic mode, for example, meth ylphosphonic acid (MPA) ethyl, isopropyl, and pina colyl esters are separated for 5 min [15]. Reversed phase columns, in which hydrophobic interactions are mainly implemented, are not always suitable for the separation of relatively small polar molecules, such as MPA and its ethyl ester because of their small reten tion. The introduction of an ionpairing reagent (tributylamine) into the mobile phase allowed the sep aration of eight alkylphosphonic acids and their esters, including propyl MPA and isopropyl MPA [16]. The application of mathematical methods ensures rapid optimization of several parameters of separation simultaneously, taking into account their mutual effect on each other, as demonstrated in the ionpair version of HPLC–MS with electrospray ionization [16]. To
improve the separation of phosphonic acids in HPLC, a column was proposed with an sorbent based on porous graphitic carbon [26], which has stronger hydrophobic properties and higher stability compared with the С18 sorbent [27]. It was shown that hydro philic compounds were well retained on this stationary phase due to the specific electronic properties of the planar graphitic surface and delocalized πelectron density. This phase is also used for the separation of isomers [27–29]. Thus, isopropylphosphonic and pro pylphosphonic acids, tertbutylphosphonic and butylphosphonic acids, and a number of esters of eth ylphosphonic acid and MPA were determined for less than 15 min in eluting with 0.1% trifluoroacetic acid and acetonitrile in the isocratic mode. The alkyl esters of MPA are separated on a column with highpurity silica gel because of hydrophilic interactions with the eluent flow rate varying from 500 to 1000 µL/min, which enabled the determination of five DP in urine on a column 50 mm in length for less than 2 min (Fig. 3) with low values of detection limits (Table 1) with the use of isotopelabeled compounds as internal standards [17]. Along with standard columns 2 mm in diameter, capillary columns (inner diameter 0.32 mm) are suc cessfully used to determine the degradation products of organophosphorus nerve agents and various derivatives of sulfur mustard. Their advantage is that the eluent con sumption is an order of magnitude lower (5–10 µL/min) compared with conventional HPLC columns (100– 300 µL/min), which provides better compatibility with mass spectrometric detection because the flow rate needs not to be decreased [7, 8, 18, 30]. It is known that the presence of salts in the eluent (for example, ammonium formate) affects the efficiency of the elec trospray ionization of compounds, contributing to the
JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 67
No. 2
2012
LIQUID CHROMATOGRAPHY WITH MASSSPECTROMETRIC DETECTION Intensity 1.6 × 104
4
89
(а) 5
1.2 × 104 8 × 103
1 2 3
4 × 103
0 8 × 103
(b)
4 × 103
0 2.4 × 104
1
(c) 4
2 1.6 × 104
5
3 8 × 103
0
0.5
1.0
1.5
2.0
2.5
3.0
3.5 t, min
Fig. 3. Chromatograms of the extracts of destruction products from (a) synthetic urine with DP concentration 1 ng/cm3, (b) urine without DP, and (c) patient’s urine with DP concentration 2 ng/cm3 [17]. MPA esters: (1) pinacolyl, (2) (2(methyl)propyl, (3) cyclohexyl, (4) isopropyl, and (5) ethyl.
formation of ions of the analytes. However, for the determination of MPA alkyl esters, it is enough to use a 2% aqueous solution of formic acid in a gradient elu tion mode with increasing concentration of acetoni trile to 80 (vol) % [31]. Methods of ionization and detection of degradation products. For the determination of five DP at a level of 20 µg/L, HPLC–MS was implemented for the first time in 1988 in the version with thermospray ioniza tion [32]. For the determination of alkylphosphonic acids and their esters, electrospray ionization is cur rently the most common [8, 16, 18, 19, 22, 23, 25, 31]. Atmospheric pressure chemical ionization is used much less frequently [19, 33]. All these ionization methods are mild, because they generate simple JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 67
molecular ions of CWA derivatives such as [M – H]–, [M + H]+, [M + NH4]+, etc. without fragmentation of the molecules, which makes unambiguous identifica tion of small molecules difficult and does not offer any advantage over GC having extensive databases for mass spectrometric identification. In this case, HPLC–MS could be more preferable when rapid ini tial screening for the presence of DP in water samples and soil extracts is required with minimal, compared to GC, sample preparation. In HPLC–MS–MS, the obtained mass spectra of compounds are more informative for the identifica tion of DP [12, 16–18, 20, 21]. This version of detec tion in HPLC for the analysis of DP was used for the No. 2
2012
90
ALEKSENKO
first time in 2004: the derivatives of mustard (thiodi glycol and nine sulfurcontaining diols, including five components that failed to be identified by HPLC– MS) were determined in aqueous extracts of soil with a quadrupole and a timeofflight massspectrometric detectors [18]. The comparison of the methods of ionization with the formation of positively and negatively charged ions by the example of various degradation products of organophosphorus and blister agents showed that the latter method was more preferable for the detection of MPA alkyl esters [19, 20, 33, 34]. In contrast to the spectra obtained in positive ionization mode, compli cated by the presence of multiple adducts formed with solvents (methanol, acetonitrile), ions of salts (Na+), and dimers [2M + H]+ and trimers [3M + H]+, “clean” mass spectra with a dominant molecular ion of deprotonated molecule [M – H]– are generated upon negative ionization mode. In the comparison of different methods of ioniza tion in HPLC–MS, including electron beam ioniza tion, electrospray ionization, and chemical ioniza tion, it was shown by the example of a series of alky lphosphonates and analogues of thiodiglycol that the majority of compounds in the first case did not yield molecular ions; however, fragment ions contained in the mass spectrum were sufficient to identify each component [35]. Despite the fact that electrospray ionization generates mainly molecular ions and that various adducts and dimers can be used for additional confirmation of molecular weight of compounds, it is desirable to use tandem mass spectrometry for unam biguous identification. In addition, adducts may lead to the suppression of ionization, thereby reducing the sensitivity of detection. It should be noted that the mass spectra obtained in all three modes of ionization differ from each other and can be used for additional confirmation of the identification [35]. Some com pounds, such as mustard and relatively nonpolar DP 1,4thioxane and 1,4dithiane, are not ionize in elec trospray and, therefore, cannot be determined by HPLC–MS; in this case, GC is used [11]. To determine three MPA alkyl esters in the form of the derivatives with nbromophenacylbromide (BPB), fast atom bombardment ionization was used [15]. The optimal use of positive ionization, giving almost three times more intense ions [M + H]+ at m/z 323 and 337 for MPA ethyl and isopropyl esters, respectively, com pared with [M – BPB]–. However, derivatization in HPLC is still rare; only the HPLC–MS determination of EA 2192 (the destruction product of VX, Fig. 1) is described in the form of methyl ester obtained by the postcolumn reaction with trimethylphenylammo nium hydroxide [36]. Among analyzers, beside the standard quadrupole, other instruments such as triple quadrupole [17, 21, 23], ion trap [22], and timeofflight [8, 18, 20] ana lyzers are used. The latter enabled the identification of
six MPA alkyl esters with the determination of the masses of ions [M – H]– and fragments in the mode of tandem mass spectrometry with an accuracy of Δm/z = 0.01 [20]. The combination of HPLC–MS and induc tively coupled plasma mass spectrometry for the anal ysis of different beverages allowed the identification of ten APAs and APA alkyl esters with the detection lim its 7.4–88.1 ng/cm3 [24]. Since all compounds of interest in this case were detected by 31P, an important condition is the chromatographic separation of all alkylphosphonates and other phosphoruscontaining matrix components. Determination of structure. The fragmentation of MPA alkyl esters in HPLC–MS–MS, probably, pro ceeds with the destruction of the carbon–oxygen bond with the separation of the alkyl radical and the forma tion of the [M ⎯ H] ⎯ ions of MPA at m/z 95 in the neg ative ion ionization mode or the [M + H]+ ions of MPA at m/z 97 in the positive ion ionization mode. Moreover, the delocalization of the phosphorus–oxy gen bonds, according to [23], is due to the hydrogen atom migrating from the terminal methyl group of the splitted alkyl radical that is transformed into alkene: O O P O R CH3
•–
– O O P OH + Alkene, CH3
which was confirmed by HPLC–MS–MS of Dlabeled MPA ethyl ester (Fig. 4). Some examples of the characteristic ion adducts, fragments, and molec ular ions of alkyl esters of APAs are presented in Table 2. It is noted that when tributylamine is used as an ionpairing reagent in the HPLC separation of alkylphosphonates, molecular radical anions are gen erated in the electrospray ionization instead of molec ular ions [M – H]– [16]. In the threestage mass spec trometry, the formation of an ion at m/z 79, corre sponding to [PO3]– is observed. The fragmentation of longchain diols in HPLC–MS–MS with the deter mination of m/z of the molecular ion and fragments with an accuracy of 0.01 Da using a timeofflight detector enabled the identification of compounds with different positions of the atoms of sulfur and oxygen, which are the destruction products of mustard, that were not previously identified in HPLC–MS with electrospray ionization (Fig. 5) [18]. During the identification and quantification of DP after the enzymatic hydrolysis of organophosphorus nerve agents, one should take into account the possi ble presence of a significant amount of adducts formed by the components of buffer solutions and the hydrol ysis products of sarin, soman, and cyclosarin [37]. One of these components is tris(hydroxymethyl)ami nomethane used in biochemical and biomedical research. As shown by HPLC–MS–MS, in the study of organophosphorus nerve agents, it is better to use, for example, 2(Nmorpholino)ethanesulfonic acid or
JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 67
No. 2
2012
LIQUID CHROMATOGRAPHY WITH MASSSPECTROMETRIC DETECTION Intensity 2.0 e4
91
95 (а)
123
0 2.1 e5
96
(b)
128
95 0 85
95
105
115
125
135 m/z
Fig. 4. Mass spectra of (a) MPA ethyl ester C3H9O3P (the destruction product of VX) and (b) isotopelabeled reference material С3Н4D5О3Р, obtained by tandem mass spectrometry [23].
3(Nmorpholino)propanesulfonic acid instead of amino or hydroxylcontaining buffer solutions, espe cially at high concentrations of organophosphorus agents or in the analysis proceeding over a long period of time [37]. The detection limits of the destruction products of CWA in different samples for APAs and their alkyl esters are presented in Table 1. It is found that for the esters of APAs, the detection limits (0.5 µg/cm3) are lower than those values characterizing APAs (10 µg/cm3) due to a lower efficiency of ionization of the latter [16]. The lowest detection limits—30 to 240 pg/cm3 (Table 1)—were obtained for MPA alkyl esters in urine with a combination of solidphase extraction JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 67
and preconcentration by evaporation and with the use of isotope dilution in HPLC–MS–MS [17]. Some what higher values of the detection limits (0.5 ng/cm3) were obtained in the determination of four alkylphos phonates in urine by HPLC–MS–MS with isotope dilution [23]. In determining ethyl, isopropyl, and pinacolyl esters of MPA in the form of derivatives of BPB in the total ion current mode, the detection limits were 1–5 ng/cm3 for the samples of river water and blood plasma, and in the selected ion monitoring mode, 0.5–5 ng/cm3 [15]. The preconcentration by hollow fiber liquid phase microextraction (LPME) enabled the obtain ing of the detection limits of dialkylphosphonic acids No. 2
2012
92
ALEKSENKO
Table 2. Signals in the mass spectra of some destruction products of CWA in HPLC–MS and in HPLC–MS–MS with electrospray ionization [22] Compound Methyl ester MPA Ethyl ester MPA Isopropyl ester MPA Isobutyl ester MPA Pentyl ester MPA Pinacolyl ester MPA Isopropylphosphonic acid pentyl ester
Mr
m/z [M + H]+
m/z [M + NH4]+
110 124 138 152 166 180 194
111 125 138 153 167 181 195
128 142 156 170 184 198 212
at 0.1–2 ng/cm3, with the exception of MPA methyl ester (100 ng/cm3), in the selected ion detection mode, whereas in the total ion current mode, these values increased by 5–200 times [22]. To compare the detection limits of MPA alkyl esters by gas chromatog raphy–mass spectrometry (GC–MS) after the extrac tion from water using hollow fiber LPME, preconcen tration, and derivatization with alkyl bromide were 0.1–0.75 µg/cm3 [38], and in the form of derivatives with N(tertbutyldimethylsilyl)Nmethyltrifluoro acetamide by GC–MS, 0.01⎯0.54 µg/dm3 (the total
m/z of fragments in MS–MS 111 [M + H]+ 97 [125C2H4]+ 97 [139C3H6]+ 97 [153C4H8]+ 97 [167C5H10]+ 97 [181C6H12]+ 125 [195C5H10]+
ion current mode) [39]. The ionpair solidphase extraction of the esters of APAs (methyl, ethyl, propyl, butyl) from water, followed by their methylation and determination by GC–MS, yielded the detection limits of 0.1–0.2 µg/cm3 and 10–15 ng/cm3 in the total ion current mode and in the selected ion mode, respectively [40]. The use of preconcentration and derivatization reduces the detection limits in HPLC–MS method by several orders of magnitude compared with the deter mination performed without these steps. However, in this case, the HPLC method has no advantages over Ions
m/z, Da
H]+
100
HO
(а)
243.056 225.044 197.016 165.043 137.011 105.038
241 OH O S MH+ (b) 161 259 137 199213
[M + H]+ [M + HH2O]+ [M + HSC2H4]+ [M + HHOC2H4SH]+ [M + HHOC2H4SC2H4OH]+
259.049 241.039 199.048 181.037 137.010
[M + H]+ [M + HH2O]+ [M + HHOC2H4SC2H4OH]+ [M + HHOC2H4SC2H4OC2H5]+
287.081 269.070 165.042 137.008
[M + H]+ [M + HH2O]+ [M + HHOC2H4SC2H4OH]+ [M + HHOC2H4SC2H4OC2H5]+ [M + HHOC2H4SC2H4OC2H4SH]+ [M + HHOC2H4SC2H4OC2H4SC2H4OH]+
331.107 313.097 209.068 181.037 149.063 105.038
S
S
105 137 165
0 100
S
[M + [M + HH2O]+ [M + HHOC2H5]+ [M + HHOC2H4SH]+ [M + HHOC2H4SC2H5]+ [M + HHOC2H4SC2H4OH]+
HO
S
105 0 100 HO
S
225 OH
197
MH+ 243
S
S
O
137
105 0
S
165
269 (c) MH+ OH 287
(d) 313 100 HO
S
O
105
149 181 209
100
150
0 50
S
O
200
S
250
MH+ 331
OH
300
350 m/z
Fig. 5. Mass spectra of longchain diols—mustard derivatives—in aqueous soil extracts, obtained by HPLC–MS–MS with elec trospray ionization and their interpretation [18]: (a) 3,6,9trithia1,11undecanediol, (b) 6oxa3,9,10trithia1,12dode canediol, (c) 6oxa3,9,12trithia1,14tetradecanediol, and (d) 6,12dioxa3,9,15trithia1,17heptadecanediol. JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 67
No. 2
2012
LIQUID CHROMATOGRAPHY WITH MASSSPECTROMETRIC DETECTION
GC in that polar DP is determined only in the form of volatile derivatives. These examples show the compa rability of the determination of CWA derivatives by GC and HPLC with a certain superiority of the latter in the values of detection limits. Application of other detectors. In the interlabora tory experiments on the analysis of various samples of unknown composition with the aim to identify CWA and DP of them, organized every year for the national laboratories by the Organization for the Prohibition of Chemical Weapons, microcolumn HPLC (column inner diameter 75 µm) was applied with a flame pho tometric detector (FPD) specific for sulfur and phos phorus [25, 41–44]. The main DP of mustard were found using the sulfur channel: bis(2hydroxoet hyl)sulfide and bis(2hydroxoethylthio)alkanes were determined in soil with a detection limit of 1 µg/cm3 and separation efficiency (1.5–6.0) × 105 theoretical plates per meter with the peak compression in the presence of displacers [25, 41]. The flame photometric detector tuned to the phosphorus mode was also used to deter mine APAs and their esters (Fig. 6) [43], whose detec tion limits were 1–800 ng/cm3 [42], which satisfies the requirements of the Organization for the Prohibition of Chemical Weapons (detection of 1 µg/cm3 of sub stances and below). A lightscattering detector for determining APAs and the esters of MPA and ethylphosphonic acid is known to be used, enabling more effective optimiza tion of the separation conditions (0.1% solutions of HCOOH, CH3COOH, CF3COOH, C3F7COOH, and C4F9COOH were studied as mobile phases) with sub sequent transfer of these conditions to an HPLC–MS system [26]. The use of a spectrophotometric detector for the determination of DP is limited due to a lack of chromophoric groups in the majority of compounds, which leads to a low sensitivity of the determination and yields lowinformative absorption spectra for the identification of DP. In this regard, the alkyl esters of MPA were proposed to be determined with a UV detector as derivatives of BPB [45], or with a fluores cence detector in the form of n(9anthroyloxy)phen acyl derivatives [46]. Sample preparation, methods of preconcentration. Water and soil samples. For the analysis of water sam ples containing DP at the level of several µg/cm3, the membrane filtration is enough [20]. The degradation products of organophosphorus nerve agents or mus tards in soil samples with the concentration at a level of µg/g are extracted with water under sonication fol lowed by centrifugation [8, 42]. In the analysis of soil samples with unknown composition of APAs and their esters, a sequential extraction with methanol and water was performed for the recovery of a wide range of substances; then, the extracts were combined and evaporated by a factor of 20 [16]. The determination of the esters of APAs in the samples at the level of fractions of µg/cm3 and lower JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 67
93
FPD signal (mV) 2 1 3 4 5
0
5
10
15
6
7
20
8
25 t, min
Fig. 6. Chromatogram of the solution of a mixture of phos phonic acid reference materials [42]; column (185 mm × 0.32 mm), PRPX100. Gradient elution from 0.13 M HCOOH (0–20.5 min) to (30 : 70 vol) 0.3 M НСООNH4–CH3OH (20.5–28.5 min); the mobile phase flow rate, 8 µL/min; sample volume, 20 µL; FPD: (1) MPA, (2) EPA, (3) isopropylphosphonic acid, (4) pro pylphosphonic acid, (5) Ethyl MPA, (6) Isopropyl ester, (7) Isopropyl EPA, and (8) Pinacolyl MPA.
and with high concentrations of interfering matrix components by HPLC–MS–MS requires careful sample preparation. The application of solidphase extraction with С18 sorbent for this purpose is limited because of poor retention of polar compounds. The potential of hydrophilic–hydrophobic polymer adsor bents modified with polar groups, for example, Oasis HLB on the basis of Nvinylpyrrolidone and divinyl benzene, which well retain acidic, basic, and neutral compounds, providing a high degree of their elution, were evaluated for the isolation and preconcentration of DP [47, 48]. However, it is difficult to optimize the conditions for these adsorbents for the adsorption and elution of the most hydrophobic derivative of CWA, MPA pinacolyl ether (the degradation product of soman), and the most hydrophilic derivative, that is, MPA ethyl ester (the degradation product of VX). Thus, the use of Oasis HLB to isolate the DP of orga nophosphorus nerve agents in urine did not ensure a complete elution of the most polar esters of APAs [49]. At the same time, similar hydrophilic–hydrophobic adsorbents were more effective in comparison with cartridges filled with C18 sorbents in the extraction of CWA from water [50]. To implement the predomi nantly hydrophobic retention mechanism for these sorbents, analytes are adsorbed from a sample acidi fied to pH 1 and are eluted with acetonitrile [51]. The use of molecularlyimprinted polymers, which selectively separates the analyte components from the matrix, greatly increases the selectivity of determina tion with the suppression of the matrix effect [52–55]. To isolate the alkyl esters of APAs, polymers were syn thesized with molecular fingerprints based on meth acrylic acid (a monomer) and trimethylolpropane tri methacrylate in acetonitrile or ethylene glycol No. 2
2012
94
ALEKSENKO
dimethacrylate in dichloromethane (a reagents for crosslinking), using pinacolyl MPA as a template for the formation of cavities of a certain size in the poly mer [51, 56, 57]. The disadvantage of this method for recovery of the target component, increasing the dura tion of the procedure, introducing an additional error in the determination, and increasing the possibility of partial loss of the components, is the need to transfer the analyte from water into the organic medium. To eliminate evaporation and redissolving of the sample, it is proposed to use a combination of solidphase extraction on a polymeric sorbent Oasis HLB and polymers having molecular memory, which enables the transfer of analytes into the organic phase during elution of DP from the sorbent Oasis HLB, the effi cient and selective preconcentration of the water sam ple, and the removal of interfering components on two sorbents successively [51]. Samples of biological fluids. In the sample prepara tion of biological fluids (urine, blood plasma, or blood), solidphase extraction (mostly, anion exchange) is usually used to remove interfering com ponents and to preconcentrate the analyte DP prior to the HPLC determination [17, 23, 31, 58]. Thus, the urine sample containing ethyl, isopropyl, pinacolyl, cyclohexyl, and isobutyl esters of MPA was preacidi fied with acetic or hydrochloric acid and diluted with water; interfering ions were separated by solidphase extraction on an anionexchange cartridge; and the analytes were eluted with methanol acidified with for mic acid [23, 31]. Along with the alkyl esters of MPA, other acidic components, occurring in urine and interfering with the determination, were extracted by the anionexchange sorbent; therefore, the adsorption of ethyl, isopropyl, pinacolyl, cyclohexyl, and 2(methyl)propyl esters of MPA on polar sorbents based on silica, for example, Strata Si1 (Phenome nex), was studied for the extraction from urine sample preliminary transferred into an organic solvent. Polar sorbents showed better recovery rates with respect to the phases containing the С2, С1, and CNgroups [17]. The automation of solidphase extraction signifi cantly reduces the duration of sample preparation, ensuring the processing of 96 samples in 20 min, whereas the sample preparation of 12 samples takes 40 min when manual operations are applied. At the same time, the detection error decreases, and the degree of extraction of compounds reaches 62–87% [23]. Further precon centration of the samples involves the evaporation either in a stream of nitrogen at 70°С [17, 23] or under vacuum [31]. Samples of blood plasma are prepared in a similar way, after precipitating the proteins by adding perchlo ric acid and separating them by centrifugation [31]. A combination of solidphase extraction and HPLC in real time is described for the determination of ethyl, isopropyl, and pinacolyl esters of MPA in blood plasma samples and water samples after deriva tization with BPB [15]. However, despite the fact that
the developed procedure has low detection limits (at a level of several ng/cm3), it is durable and rather time consuming, which hinders its use for mass analysis. In addition to the reaction with BPB at 80°С for 30 min, the sample preparation includes multiple evaporation before and after derivatization, preliminary purification of derivatives on a column, and transfer of the sample into the aqueous phase before entering into the system (Table 1) [15]. The recent trend toward miniaturization of sample preparation is observed, for example, the use of LPME as an alternative to liquid extraction [59–62]. One of the options—threephase hollow fiber liquid–liquid– liquid microextraction—was used for the preconcen tration of nine esters of APAs by a factor of 11–135 and for the determination them in water [22]. The method is based on successive extraction of the analyte from the donor phase (water sample) with pH < рКа of APA into the organic phase (1octanol) immobilized in the pores of the fiber, followed by the reextraction into the acceptor phase (pH 14) located in the microsyringe immersed the fiber. In this case, all three phases are not mixed with each other. The duration of extraction (60 minutes) is limited by the rate of diffu sion of the least hydrophobic substances from the aqueous phase into the organic phase, which have the lowest coefficient of preconcentration (11 for methyl MPA) compared with more hydrophobic molecules (the coefficients of preconcentration of pinacolyl and pentyl esters of MPA are 85 and 114, respectively). The conditions of LPME were optimized using facto rial experimental design [22]. Analysis of samples. Environmental samples. In the majority of procedures, DP were determined by the spiking procedure. In these cases, DP (the esters of APAs, thiodiglycol) were introduced into water or soil to simulate the effects of contamination [8, 18, 22], and the samples were kept for some time (1 h) [18, 42]. Less commonly, original CWA, such as organophos phorus nerve agents, were used for “contamination” by adding them in the blood plasma [31], water, or soil samples followed by soaking samples prior to the determination of the resulting hydrolysis products, that is, the alkyl esters of APAs [20]. Thus, 0.7– 1.6 µg/cm3 of isobutyl, ethyl, isopropyl, cyclohexyl, and pinacolyl esters of APA was found 5 days after the introduction of 2 µg/cm3 of various agents into the water using HPLC with a timeofflight mass spectro metric detector [20]. In the introducing of CWA in soil (8 µg/g), from 1.1 to 6.3 µg/g of these compounds was found after 10 h, including detectable amounts of MPA, which could not be determined in water [20]. It is shown that the use of water to extract the degrada tion products, the degree of extraction from soil for MPA was only 36%, while for its pinacolyl ester, 90% [20]. It is also shows that DP is extracted by water with the degree of extraction of more than 80% (isopropyl
JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 67
No. 2
2012
LIQUID CHROMATOGRAPHY WITH MASSSPECTROMETRIC DETECTION
and pinacolyl esters of MPA and bis(2hydroxyeth ylthio)alkanes) [8, 42]. There are only few works on the determination of arseniccontaining CWA derivatives by HPLC. The determination of phenyl derivatives (phenylarsonic acid, phenylarsinoxide, diphenylarsinic acid) by HPLC–MS with inductively coupled plasma in ground water with dominating concentration of diphenylarsinic acid (up to 2.1 mg/L) is described in [63]. To avoid losses of arsenic due to its precipitation because of the high concentration of iron(II) (up to 23 mg/L), after the contact of water with the atmo sphere, the sample was preserved by adding phospho ric acid. Biological fluids. The problems of revealing the fact of intoxication with a toxic agent and the identifica tion of the type of the agent by biomarkers in blood plasma and urine in the retrospective analysis, the determination of the adducts of organophosphorus agents and sulfur mustard with biotargets, the assess ing of the localization of toxic substances in proteins, including both transport and enzymatic, by HPLC– MS are considered in review [64] and partially covered in review [5]. To confirm the effects of nerve agents on a man, a procedure was developed for determining butyrylcholinesterase (BHE) adducts with sarin, soman, and VX in blood plasma by HPLC–MS with electrospray ionization [65]. Basing on the identifica tion of peptide fragments and adducts BHE–sarin by tandem mass spectrometry, it was suggested that the addition of agents occurred through serine residue in the BHE molecule, which was confirmed by the data of [66], where the site of attachment of diisopropylflu orophosphate was identified as a serine residue. The application of HPLC–MS–MS with isotope dilution enabled an express identification of the metabolites of sarin, soman, cyclosarin, and VX in urine with a per formance of 120 tests for 24 h and the detection limits below 200 fg [23]. The procedure was developed for quantifying alkyl esters of APAs in urine with the detection limits of 0.8–6 pg, ensuring the analysis of 288 samples per day, including sample preparation using automated solidphase extraction and determi nation by HPLC–MS; the duration of one analysis was less than 5 min [17]. It should be noted that in a retrospective analysis of biological fluids, the time interval between the impact of a toxic agent and subsequent sampling plays an important role in the selection of a biomarker for the identification of the agent. For example, sarin and soman can be found in unaltered form in plasma only for a few hours after exposure, whereas VX is resistant to changes in the body for about 40 h [67]. For the pur pose of positive identification of CWA, one should be aware that the products of hydrolytic decomposition are removed from the plasma within 1–2 days [68]; however, they may be determined in urine within two weeks [49], and the reactivation of the original toxic agent from the BHE adduct can occur within three JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 67
95 7
Intensity 10 × e4
(а)
8 × e4
6
6 × e4
3 4 5
4 × e4 2 × e4
1
0
4
2
8
12 7
10 × e4
8
(b)
8 × e4
6
6 × e4
3 4
4 × e4 2 × e4 0
1 2 4
8
12
16 t, min
Fig. 7. Chromatograms of the extracts of (a) blood plasma and (b) urine with the addition of reference solutions of destruction products [31]; column Hypercarb 100 mm × 2.1 mm, 30°С. Gradient elution with 2 vol % HCOOH solution (A) and СН3СN (B) in the mode: 100% A (1 min, 150 µL/min), increase for 1 min to 20% B + 80% A (5 min, 175 µL/min) increase for 1 min to 80% B + 20% A (6 min), and return to 100% A (6 min). MPA esters (concentration 100 ng/cm3): (1) ethyl, (2) isopropyl, (3) isobutyl, (4) butyl, (6) cyclohexyl, (7) pinacolyl, and (8) 2ethyl hexyl (internal standard); (5) phenylphosphonic acid (internal standard).
weeks after the impact [68]. In the case of the deriva tives of organophosphorus agents, adducts with cho linesterase or with plasma proteins (albumin) and hydrolysis products (alkylmethylphosphonic acids) may occur. To determine the metabolites of CWA, it is preferable to use urine, because they are rapidly removed from the blood and their concentration in urine is higher. For example, the butyl, cyclohexyl, isopropyl, isobutyl, pinacolyl, and ethyl esters of MPA were determined in urine at the concentrations of 45– 150 ng/cm3 (Fig. 7) [31]. HPLC–MS with atmo spheric pressure chemical ionization enables a reliable detection of the metabolites (isopropyl, isobutyl, and pinacolyl esters of MPA) in the blood plasma samples of rats treated with nonlethal doses of organophospho rus agents, 48–72 h after intoxication with the detec tion limits of 0.6–4 ng/cm3 [69]. No. 2
2012
96
ALEKSENKO
*** Thus, HPLC enables the determination of the deg radation products of CWA with the detection limits at a level of tens of pg/cm3. The introduction of contem porary options of sample preparation, including liq uidphase microextraction, ensures the preconcentra tion of DP by more than 100 times, significantly reducing the detection limits. The use of tandem mass spectrometry to HPLC allows one to receive informa tive spectra and to identify degradation products with high reliability, even without their prior separation. A short time of determination and partial automation of sample preparation make possible, even now, to pro cess hundreds of tests per day. A lack of comprehensive libraries of mass spectra for HPLC–MS because of difficulties in accounting for all possible ions and adducts, generated in a variety of mobile phases and under different conditions of ionization, is a certain limitation of the method. However, HPLC–MS is increasingly used in determining the decomposition products of CWA along with GC–MS and is capable of solving the problem of identification and quantifi cation of substances controlled by the Convention. REFERENCES 1. Rybal’chenko, I.V., Khlebnikova, N.S., Savel’eva, E.I., Radilov, A.S., and Rembovskii, V.R., Ros. Khim. Zh., 2005, vol. 49, no. 2, p. 26. 2. Wood, M., Laloup, M., Samyn, N., Fernandez, M.M.R., Bruijn, E.A., Maes, R.A.A., and Boeck, G.D., J. Chro matogr., A, 2006, vol. 1130, p. 3. 3. Kientz, Ch.E., J. Chromatogr., A, 1998, vol. 814, p. 1. 4. Hooijschuur, E.W.J., Kientz, Ch.E., and Brink man, U.A.Th., J. Chromatogr., A, 2002, vol. 982, p. 177. 5. Rybal’chenko, I.V., Ros. Khim. Zh., 2007, vol. 51, no. 2, p. 101. 6. Black, R.M. and Read, R.W., In: Chemical weapons convention chemical analysis. Ed. M. Mesilaakso. Chichester, UK: Jon Wiley&Sons Ltd, 2005. p. 476. 7. D’Agostino, P.A., Hancock, J.R., and Provost, L.R., J. Chromatogr., A, 1999, vol. 840, p. 289. 8. D’Agostino, P.A., Hancock, J.R., and Provost, L.R., J. Chromatogr., A, 2001, vol. 912, p. 291. 9. D’Agostino, P.A., Hancock, J.R., and Provost, L.R., J. Chromatogr., A, 1999, vol. 837, p. 93. 10. D’Agostino, P.A., Chenier, C.L., and Hancock, J.R., J. Chromatogr., A, 2002, vol. 950, p. 149. 11. D’Agostino, P.A., Hancock, J.R., and Chenier, C.L., Eur. J. Mass Spectrom., 2003, vol. 9, p. 609. 12. D’Agostino, P.A., Hancock, J.R., Chenier, C.L., and Lepage, C.R.J., J. Chromatogr., A, 2006, vol. 1110, p. 86. 13. Brickhouse, M.D., Creasy, W.R., Williams, B.R., Mor rissey, K.M., O’Connor, R.J., and Durst, H.D., J. Chromatogr., A, 2000, vol. 883, p. 185. 14. Creasy, W.R., Stuff, J.R., Williams, B., Morrissey, K., Mays, J., Duevel, R., and Durst, H.D., J. Chromatogr., A, 1997, vol. 774, p. 253.
15. Katagi, M., Tatsuno, M., Nishikawa, M., and Tsuchi hashi, H., J. Chromatogr., A, 1999, vol. 833, p. 169. 16. Tak, V., Kanaujia, P.K., Pardasani, D., Kumar, R., Srivastava, R.K., Gupta, A.K., and Dubey, D.K., J. Chromatogr., A, 2007, vol. 1161, p. 198. 17. Mawhinney, D.B., Hamelin, E.I., Fraser, R., Silva, S.S., Pavlopoulos, A.J., and Kobelski, R.J., J. Chromatogr., B, 2007, vol. 852, p. 235. 18. D’Agostino, P.A., Hancock, J.R., and Chenier, C.L., J. Chromatogr., A, 2004, vol. 1058, p. 97. 19. Black, R.M. and Read, R.W., J. Chromatogr., A, 1998, vol. 794, p. 233. 20. Liu, Q., Hu, X., and Xie, J., Anal. Chim. Acta, 2004, vol. 512, p. 93. 21. Stubbs, S.J. and Read, R.W., J. Chromatogr., B, 2010, vol. 878, p. 1253. 22. Tak, V., Pardasani, D., Kanaujia, P.K., and Dubey, D.K., J. Chromatogr., A, 2009, vol. 1216, p. 4319. 23. Ciner, F.L., McCord, C.E., Plunkett, Jr., R.W., Martin, M.F., and Croley, T.R., J. Chromatogr., B, 2007, vol. 846, p. 42. 24. Kubachka, K.M., Richardson, D.D., Heitkem per, D.T., and Caruso, J.A., J. Chromatogr., A, 2008, vol. 1202, p. 124. 25. Hooijschuur, E.W.J., Kientz, Ch.E., Hulst, A.G., and Brinkman, U.A.Th., Anal. Chem., 2000, vol. 72, p. 1199. 26. Mercier, J.P., Morin, Ph., Dreux, M., and Tambute, A., J. Chromatogr., A, 1999, vol. 849, p. 197. 27. Ross, P. and Knox, J.H., Adv. Chromatogr., 1997, vol. 37, p. 121. 28. Elfakir, C., Chaimbault, P., and Dreux, M., J. Chro matogr., A, 1998, vol. 826, p. 193. 29. Petritis, K.N., Chaimbault, P., Elfakir, C., and Dreux, M., J. Chromatogr., A, 1999, vol. 833, p. 147. 30. D’Agostino, P.A., Provost, L.R., and Hancock, J.R., J. Chromatogr., A, 1998, vol. 808, p. 177. 31. Koller, M., Becker, Ch., Thiermann, H., and Worek, F., J. Chromatogr., B, 2010, vol. 878, p. 1226. 32. Wils, E.R.J. and Hulst, A.G., J. Chromatogr., 1988, vol. 454, p. 261. 33. Black, R.M. and Read, R.W., J. Chromatogr., A, 1997, vol. 759, p. 79. 34. Read, R.W. and Black, R.M., J. Chromatogr., A, 1999, vol. 862, p. 169. 35. Smith, J.R. and Shih, M.L., J. Appl. Toxicol., 2001, vol. 21, p. 27. 36. Creasy, W.R., J. Am. Soc. Mass Spectrom., 1999, vol. 10, p. 440. 37. Gab, J., John, H., Melzer, M., and Blum, M.M., J. Chromatogr., B, 2010, vol. 878, p. 1382. 38. Pardasani, D., Kanaujia, P.K., Gupta, A.K., Tak, V., Shrivastava, R.K., and Dubey, D.K., J. Chromatogr., A, 2007, vol. 1141, p. 151. 39. Lee, H.S.N., Sng, M.T., Basheer, Ch., and Lee, H.K., J. Chromatogr., A, 2007, vol. 1148, p. 8. 40. Saradhi, U.V.R.V., Prabhakar, S., Reddy, T.J., and Murty, M.R.V.S., J. Chromatogr., A, 2007, vol. 1157, p. 391.
JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 67
No. 2
2012
LIQUID CHROMATOGRAPHY WITH MASSSPECTROMETRIC DETECTION 41. Hooijschuur, E.W.J., Kientz, Ch.E., and Brink man, U.A.Th., J. Chromatogr., A, 1999, vol. 849, p. 433. 42. Hooijschuur, E.W.J., Kientz, Ch.E., and Brinkman, U.A.Th., J. Chromatogr., A, 2001, vol. 928, p. 187. 43. Hooijschuur, E.W.J., Kientz, Ch.E., Hulst, A.G., and Brinkman, U.A.Th., J Chromatogr. A, 2001, vol. 907, p. 165. 44. Hooijschuur, E.W.J., Hulst, A.G., De Jong, A.L., De Reuver, L.P., Van Krimpen, S.H., Van Baar, B.L.M., Wils, E.R.J., and Kientz, Ch.E., Trends Anal. Chem., 2002, vol. 21, p. 116. 45. Bossle, P.C., Martin, J.J., Sarver, E.W., and Sommer, H.Z., J. Chromatogr., 1983, vol. 267, p. 209. 46. Roach, M.C., Ungar, L.W., Zare, R.N., Reiner, L.M., Pompliano, D.L., and Frost, J.W., Anal. Chem., 1987, vol. 59, p. 1056. 47. Cheng, Y.F., Phillips, D.J., Neue, U., and Bean, L., J. Liq. Chromatogr., 1997, vol. 20, p. 2461. 48. Fontanals, N., Galia, M., Cormack, P.A.G., Marce, R.M., Sherrington, D.C., and Borrull, F., J. Chromatogr., A, 2005, vol. 1075, p. 51. 49. Riches, J., Morton, I., Read, R.W., and Black, R.M., J. Chromatogr., B, 2005, vol. 816, p. 251. 50. Kanaujia, P.K., Pardasania, D., Gupta, A.K., and Dubey, D.K., J. Chromatogr., A, 2007, vol. 1139, p. 185. 51. Moullec, S.L., Truong, L., Montauban, C., Begos, A., Pichon, V., and Bellier, B., J. Chromatogr., A, 2007, vol. 1139, p. 171. 52. Pichon, V. and ChapuisHugon, F., Anal. Chim. Acta, 2008, vol. 622, p. 48. 53. Pichon, V., J. Chromatogr., A, 2007, vol. 1152, p. 41. 54. Turiel, E. and MartinEsteban, A., Anal. Chim. Acta, 2010, vol. 668, p. 87.
JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 67
97
55. Baggiani, C., Anfossi, L., and Giovannoli, C., Anal. Chim. Acta, 2007, vol. 591, p. 29. 56. Moullec, S.L., Begos, A., Pichon, V., and Bellier, B., J. Chromatogr., A, 2006, vol. 1108, p. 7. 57. ZiHui, M. and Qin, L., Anal. Chim. Acta, 2001, vol. 435, p. 121. 58. Kataoka, M. and Seto, Y., J. Chromatogr., B, 2003, vol. 795, p. 123. 59. Lee, J., Lee, H.K., Rasmussen, K.E., and Pedersen Bjergaard, S., Anal. Chim. Acta, 2008, vol. 624, p. 253. 60. PedersenBjergaard, S. and Rasmussen, K.E., J. Chro matogr., A, 2008, vol. 1184, p. 132. 61. Rasmussen, K.E. and PedersenBjergaard, S., Trends Anal. Chem., 2004, vol. 23, p. 1. 62. Psillakis, E. and Kalogerakis, N., Trends Anal. Chem., 2003, vol. 22, p. 565. 63. Daus, B., Mattusch, J., Wennrich, R., and Weiss, H., Talanta, 2008, vol. 75, p. 376. 64. Podol’skaya, E.P. and Babakov, V.N., Nauch. Pri borostr., 2008, vol. 18, no. 4, p. 5. 65. Tsuge, K. and Seto, Y., J. Chromatogr., B, 2006, vol. 838, p. 21. 66. Tsuge, K. and Seto, Y., J. Chromatogr., B, 2002, vol. 776, p. 79. 67. Van der Schans, M.J., Lander, B.J., Van der Wiel, H., Langenberg, J.P., and Benschop, H.P., Toxicol. Appl. Pharmacol., 2003, vol. 191, p. 48. 68. Noort, D., Benschop, H.P., and Black, R.M., Toxicol. Appl. Pharmacol., 2002, vol. 184, p. 116. 69. Rodin, I.A., Braun, A.V., Anan’eva, I.A., Shpigun, O.A., Savel’eva, E.I., Rybal’chenko, I.V., Bolotov, S.L., and Rodchenkov, G.M., Massspektrometria, 2011, vol. 8, no. 1, p. 45.
No. 2
2012