sources [6, 7]. Therefore, it remains necessary to identify new sources of halocarbons to gain a better understanding of the global halogen budget [2]. Marine environment have been recognized as one of the important natural sources of halocarbons. In the ocean, macroalgae and marine phytoplankton produce halocarbons (e.g., [8, 9, 10]). Recently, the bacterial production of CH3Cl, CH3Br [11], and CH3I [12] was also reported. However, microbial production of halocarbons remains highly uncertain and requires further study.
Several extraction procedures have been available for the measurement of VOCs in natural and biological samples (e.g., [13]). The static headspace (SHS) and headspace-solid phase microextraction (HS-SPME) methods have been widely utilized as a means of directly Introduction
Halocarbons are among the volatile organic compounds (VOCs) emitted from ecosystems, and they have been shown to affect a wide range of climate and atmospheric chemistries [1, 2]. Of par ticular interest are the halocarbons involved in the global halogen budget and the depletion of tropospheric and stratospheric ozone [3]. Long-lived halocarbons (methyl chloride, CH3Cl and methyl bromide, CH3Br) are abundant in the atmosphere and have been studied widely. However, natural sources of these long-lived halocarbons are insuf ficient to compensate for known sinks of these compounds [4, 5]. As for shor t-lived halocarbons (e.g., methyl iodide, CH3I; bromoform, CHBr3; and diiodomethane, CH2I2), there is few information about short-lived halocarbon
Halocarbons emitted from ecosystems play key roles in the degradation of the tropospheric and stratospheric ozone. In the estimation of the global budget, known sources of halocarbons have been considered insufficient with respect to the global sink. Therefore, it remains necessar y to identify new sources of halocarbons. To date, however, ver y few marine phytoplankton and bacteria have been studied. Research into halocarbon production from marine sources traditionally relies on purge-and-trap techniques, but these cannot be used to measure halocarbons in foam-forming samples (e.g., bacterial culture). An alternative method, an automated dynamic headspace (DHS) extraction technique with the gas chromatography − mass spectrometry (GC/MS) method, was applied to the simultaneous analysis of 15 halocarbons in aqueous and bacterial samples at pico- to nano- mol L− 1
range. The reproducibilities (n = 10) of the measurements were 24.7 % for CH3Cl, 16.3 % for CH3Br, 10.5 % for CH3I, and from 1.4 % to 8.8 % for the remaining halocarbons. Linear regression of the standard solution prepared with bacterial culture medium indicated that this method was successfully applied to the analysis of trace levels of halocarbons in bacterial culture. The DHS GC/MS method potentially provides a sensitive means for tracing numerous natural sources of halocarbons from environmental or biological samples.
Keywords: Dynamic headspace analysis (DHS), Gas chromatography-mass spectrometry (GC/MS), Halocarbon, Bacteria
Simultaneous Determination of 15 Halocarbons at Pico- to Nano - Mol
per Liter Levels in Water and Biological Samples Using Dynamic
Headspace Extraction and Gas Chromatography − Mass - Spectrometry
Gen TANIAI
＊, Hiroshi ODA
＊＊, Yuki YONEYAMA
＊, Minami ABE
＊, Takashi YAMAKOSHI
＊,
Keisuke AMBIRU
＊, Michiko KURIHARA
＊＊＊and Shinya HASHIMOTO
＊＊＊（Accepted November 14, 2012）
＊ Graduate School of Integrated Basic Sciences, Nihon University,
3-25Į40, Sakurajosui, Setagaya-ku, Tokyo 156-8550, Japan
＊＊ Institute for Environmental Sciences, University of Shizuoka,
52-1 Yada, Suruga-ku, Shizuoka 422-8526, Japan
＊＊＊ Department of Chemistry, College of Humanities & Sciences, Nihon
beverages [23]. Our group was the first to report the bacterial production of halocarbons using the DHS method [11]. However, there are no published data on the optimization of an analytical method for halocarbons at very low concentrations (pico- to nano- mol L− 1 range) in aqueous or biological samples that have high cell density.
In the present study, the optimization of the DHS extraction conditions combined with GC/MS was carried out to develop a sensitive quantitative analysis of 15 halocarbons, including highly volatile ones such as CH3Cl and CH3Br. Below, we discuss the development of the present analytical method and evaluate the applicability of DHS GC/MS to the analysis of halocarbons in bacterial cultures for the exploration of halocarbon sources.
2. Materials and methods 2 – 1. Chemicals
The mixed standard solution of CH3Cl (2000 Pg mL − 1) and CH3Br (2000 Pg mL− 1) was purchased from Supelco (Bellefonte, Pennsylvania, USA). The standards of C2H5Br (99 % pure), CH3I (99.5 % pure), C2H5I (99 % pure), (CH3)2CHI (99 % pure), CH2Br2 (99 % pure), CH3(CH2)2I (99 % pure), CHBrCl2 (98 % pure), CH2ClI (97 % pure), CHBr2Cl (98 % pure), CH2BrI (97 % pure, LOT 09211BJ), CHBr3 (99 % pure), CH2I2 (99 %), and (CD3)2CDI (isotopic purity 98 atom % D) were purchased from Sigma-Aldrich (Tokyo, Japan). CH2BrCl (98 % pure) and pesticide residue-grade solvents were obtained from Wako Pure Chemical Industries (Tokyo, Japan). These reagents were measuring VOCs in the headspaces of samples [14, 15].
However, neither method is sensitive enough to detect VOCs at the environmental concentration level (pmol L− 1 level). To measure trace levels of VOCs, the purge-and-trap (P&T) method is generally preferred over the headspace extraction methods. The P&T method has been used to assess halocarbons in liquid samples such as fresh water [16], brackish water [17], seawater [18, 19, 20], and phytoplankton cultures [21]. Unfortunately, the P&T method concentrates VOCs on the trap by bubbling an iner t gas (e.g., ultrapure helium) through liquid samples, so this method cannot be used to measure VOCs in foam-forming samples (e.g., bacterial culture) or in solid samples (e.g., sediment).
An alternative method, called the dynamic headspace (DHS) extraction method, is an extraction technique concentrating VOCs on the trap tube by passing a stream of an inert gas into the headspace above the sample (Fig. 1). The DHS extraction method depends on the continuous regeneration of an extracting gas to force volatiles from a sample into the headspace gas phase by sweeping the headspace. Compared to the SHS or HS-SPME method, the DHS method may be a more sensitive means of detecting volatile compounds, and it may have great potential for researching volatile biogenic trace gases in a wide variety of samples (e.g., seawater, cultured microorganisms, and sediment). The DHS method has generally been used for the analysis of volatiles in several kinds of samples, such as fermented foods [22] and
Absorbent
Volatiles
Sample
Exhaust
Inert gas inlet
a) Concentration
b) Desorption
GC
Inert gas inlet
Helium flow
Helium flow
MS
before each analysis. The dynamic headspace system was used to purge the gas phase above each sample (10 mL) with ultrapure helium (>99.9999 % pure, Air Liquide Kogyo Gas Ltd.; Tokyo, Japan), and halocarbons with boiling points between − 24 ℃ (CH3Cl) and 181 ℃ (CH2I2) were pre-concentrated in the trap column of a glass tube (60 mm, 4 mm ID, 6 mm OD) containing about 60 mg of Tenax TA, which was maintained at room temperature. The halocarbons were released from the trap column by heating it at 200℃ with a thermal desorption unit (TDU, Gerstel K. K.; Tokyo, Japan), and the desorbed halocarbons were cryofocusing on a liner tube trap of Tenax TA held at − 150 ℃ with a cooled injection system (CIS 4, Gerstel K. K.; Tokyo, Japan). After cryofocusing, a liner tube was rapidly heated up to 200℃ and halocarbons were introduced into the capillary column (DB-624, length, 20 m; inner diameter, 0.18 mm; and film thickness, 1 Pm; Agilent Technologies; Tokyo, Japan). Halocarbons were measured with a gas chromatograph (GC, 6890N, Agilent Technologies, Inc.; Wilmington, North Carolina, USA) mass spectrometer (MS, 5975C, Agilent). Analysis of halocarbons was performed by selected ion monitoring to attain high sensitivity. The analytical conditions for the DHS GC/MS and the target ions are given in Tables 1 used without further purification.
The primary standards were prepared gravimetrically and dissolved in ultrapure methanol. Aqueous working standards were prepared by diluting suitable aliquots of primary standards with ultrapure water (Milli-Q water; Nihon Millipore, Tokyo, Japan) or culture medium (used for the microbial experiments). Calibrations using aqueous working standards were carried out approximately once per week. A liquid standard of (CD3)2CDI was added as an internal standard to each sample at a final concentration of approximately 1 nmol L− 1 to monitor GC/MS sensitivity drift. Instr umental blanks were measured prior to sample measurements.
2 – 2. Instrumentation
Sample extraction and introduction were fully automated using a Gerstel MPS-2 autosampler configured for the auto-DHS injection, which was operated with Gerstel Maestro software Ver. 1.4.8.14. The halocarbons in aqueous samples contained in glass vials sealed with silicone/PTFE septa (Gerstel K.K.; Tokyo, Japan) were measured using an automated dynamic headspace extraction system (DHS, Gerstel K. K.; Tokyo, Japan). The baking of an adsorbent trap (230℃，30 min) was carried out
Table 1 Experimental conditions for the dynamic headspace extraction system and GC/MS Examined
range
Optimized condition Dynamic headspace extraction system
Thermostatting Thermostatting time (min) Agitation speed (rpm)
1 – 4 350 – 800
2 500
Trap column Trapping material Tenax TA
Trap temperature (℃ ) 20
Purge volume (mL) 75 – 150 100
Purge flow rate (mL min− 1) 7.5 – 20 10
Desorption temperature (℃ ) 200
Cryo-focus Trapping material Tenax TA
Trap temperature (℃ ) − 150
Desorption temperature (℃ ) 200
GC/MS
Capillary column DB-624 (20 m × 0.18 mm × 0.1 Pm)
Carrier gas He
Carrier gas flow rate 1.0 mL min− 1 Oven temperature Initial 40 ºC for 4 min
programmed to 100 ºC at 10 ºC min− 1 programmed to 200 ºC at 20 ºC min− 1 Ionization mode EI (electron ionization, 70 eV)
with the composite working standards at least at four concentrations each that spanned the range of trace gas concentrations in the bacterial cultures. The detection limit was defined as three times the standard deviation of the control samples ( 3V). To evaluate the influence of the matrix effect, the standard solutions were measured with DHS GC/MS, which enabled us to individually ascertain calibration cur ves in ultrapure water or in a bacterial medium matrix.
2 – 5. Application for the analysis of halocarbons in bacterial culture
Bacterial culture samples were obtained from batch cultures of Erythrobacter longus JCM6170T and
Alteromo-nas macleodii JCM20772T(from Riken Bioresource Center, Saitama, Japan; Japan Collection of Microorganisms [JCM]). Each strain was preincubated in test tubes. The cells were then collected by centrifugation (12,000 rpm; 3 min), washed three times with fresh medium, inoculated in 10 mL of marine broth 2216 (Difco Laboratories; Detroit, Michigan, USA) with potassium iodide (KI; final concentration, 1 Pmol L− 1 for Erythrobacter longus, 1 mmol L− 1 for Alteromonas macleodii) in 20-mL glass vials sealed with silicone/PTFE septa, and incubated in the dark at 25 ℃ in the Sanyo MIR-253 incubator (Sanyo, and 2, respectively.
2 – 3. Optimization of DHS parameters
Among the possible variables that might exert some ef fect on halocarbon recover y, four DHS parameters (thermostatting time, agitator speed, purge volume, and purge flow rate) were optimized. To assess the effects of sample thermostatting time, agitation speed, purge volume, and purge flow rate on the DHS extraction efficiency, 10.0 mL of aqueous standard solutions were used. The concentrations for the 15 halocarbons in the aqueous standard solution were as follows: CH3Cl, 1000 pmol L−1 ; CH 3Br, 40 pmol L −1 ; C 2H5Br, 20 pmol L −1 ; CH 3I, 50 pmol L− 1 ; C 2H5I, 9 pmol L − 1 ; CH 2BrCl, 9 pmol L − 1 ; (CH3)2CHI, 7 pmol L− 1 ; (CD3)2CDI, 7 pmol L− 1 ; CH2Br2 , 10 pmol L− 1 ; CH
3 (CH2)2I, 6 pmol L
− 1 ; CHBrCl
2 , 8 pmol L− 1 ; CH
2ClI, 9 pmol L− 1 ; CHBr2Cl, 8 pmol L− 1 ; CH2BrI, 10 pmol L− 1 ; CHBr
3 , 9 pmol L
− 1 ; and CH
2I2 , 9 pmol L − 1. The examined ranges for thermostatting time, agitation speed, purge volume, and purge flow rate are listed in
Table 1.
2 – 4. Calibration and assessment of DHS
For the calibration plot of the standard solutions, each 10 mL of ultrapure water or culture medium was spiked
Table 2 Retention time (RT), target ions, linearity range, and correlation coefficient ( R2
) for the target compounds Compound RT a (min) Target ionsb (m / z) Linearity range ( pmol L− 1 ) Correlation coefficient ( R2 ) CH3Cl 2.0 50 52 3,000 – 300,000 0.961 CH3Br 2.3 94 96 100 – 10,000 0.989 C2H5Br 3.1 108 110 50 – 5,000 0.968 CH3I 3.1 142 127 70 – 7,000 0.984 C2H5I 4.8 156 127 9 – 900 0.995 CH2BrCl 5.1 130 128 30 – 3,000 0.998 (CH3)2CHI 6.1 170 127 7 – 700 0.990 (CD3)2CDI 6.0 177 127 7 – 700 0.994 CH2Br2 7.1 174 93 10 – 1,000 0.999 CH3 (CH2)2I 7.2 170 127 6 – 600 0.993 CHBrCl2 7.3 83 85 8 – 800 0.998 CH2ClI 7.7 176 178 9 – 900 0.999 CHBr2Cl 9.4 129 127 80 – 800 0.999 CH2BrI 9.6 222 220 10 – 1,000 0.999 CHBr3 11.1 173 171 9 – 900 0.999 CH2I2 11.6 141 127 9 – 900 0.998 a
RT: Experimental conditions were as in Table 1.
b
Tokyo, Japan). After incubation of the cells for several days, the concentration of halocarbons in the culture samples of marine bacteria in 20-mL vials were analyzed using DHS GC/MS. Bacterial growth was monitored by measuring optical density at 600 nm (OD600) during the culture period using a WPA CO7500 colorimeter (Biochrom, Cambridge, UK). After the measurement of halocarbons in cultured samples (10 mL), the cap of the sealed vial was removed, and optical density was measured with a 10-mm cell. For each measurement of halocarbons (and optical density) at 0, 1, 2, 3, 4, 8, and 10 d, separate vials of cultured samples were prepared.
3. Results
3 – 1. Optimization of the conditions for DHS extraction
To optimize the conditions for the DHS extraction procedure, we investigated the effects of four parameters: sample thermostatting time, agitation speed, purge volume, and purge flow rate. The peak area grew substantially as the time of thermostatting increased to around 2 or 3 min for CH3Cl and CH3Br (Fig. 2a). Further increases in the thermostatting time led to decreases in the peak areas of CH3Cl and CH3Br. At thermostatting times of longer than 3 min, CH3I, C2H5Br, and C2H5I were found to lead to increases in the peak areas (Fig. 2b). As for the other halocarbons, no significant changes were observed in the peak area by increasing the thermostatting time from 1 to 4 min (Fig. 2c–e). To avoid the loss of CH3Cl and CH3Br, a thermostatting time of 2 min was chosen as the optimum condition. As regards the agitation speed during thermostatting and purging, the maximum peak area was obser ved at 500 rpm for all analytes (Fig. 3a–e). Thus, to achieve effective extraction, an agitation speed of 500 rpm was selected as the optimum condition.
As regards the purge volume (Fig. 4a–e), larger purge volumes (>100 mL) reduced the peak areas of extremely volatile halocarbons (e.g., CH3Cl, boiling point: − 24 ℃ ; CH3Br, boiling point: 3.5 ℃ ) (Fig. 4a), whereas increases in purge volume tended to enhance the peak areas of other halocarbons, especially CHBr3 (boiling point: 149 ℃ ) and CH2I2 (boiling point: 181℃ ) (Fig. 4d–e). Purge volumes lower than 100 mL appeared to be insufficient to extract
Fig. 2 Ef fect of thermostatting time on the extraction efficiency of 15 halocarbons. Agitation speed: 500 rpm, purge volume: 100 mL, purge flow rate of helium: 10 mL min− 1. Concentrations of spiked halocarbons were as follows. CH3Cl, 1000 pmol L− 1 ; CH3Br, 40 pmol L− 1 ; C
2H5Br, 20 pmol L− 1 ; CH3I, 50 pmol L− 1 ; C2H5I, 9 pmol L− 1 ; CH2BrCl, 9 pmol L− 1 ; (CH3)2CHI, 7 pmol L−1 ; (CD
3)2CDI, 7 pmol L−1 ; CH2Br2 , 10 pmol L− 1 ; CH
3 (CH2)2I, 6 pmol L− 1 ; CHBrCl2 , 8 pmol L− 1 ; CH2ClI, 9 pmol L− 1; CHBr2Cl, 8 pmol L− 1 ; CH2BrI, 10 pmol L− 1 ; CHBr
3 , 9 pmol L− 1 ; and CH2I2 , 9 pmol L− 1 0 4,000 8,000 12,000 0 1,000 2,000 3,000 4,000 CH3Cl CH3Br a P eak ar ea of CH 3 Cl Peak ar ea of CH 3Br 0 10,000 20,000 30,000 40,000 50,000 3,000 4,000 5,000 6,000 7,000 8,000 b C2H5Br CH3I C2H5I Peak ar ea of CH 3 I Peak ar ea of C 2H 5I and C 2H 5Br 0 3,000 6,000 9,000 d CH2BrCl CH2Br2 CHBrCl2 CHBr2Cl CHBr3 Peak ar ea 0 1 2 3 4 5 0 500 1,000 1,500 2,000 2,500 e CH2ClI CH2BrI CH2I2
Thermostatting time (min)
P e ak ar ea 0 1,000 2,000 3,000 c (CH3)2CHI (CD3)2CDI CH3(CH2)2I P eak ar ea
)
)LJ
Fig. 3 Effect of agitation speed on the extraction efficiency of 15 halocarbons. Thermostatting time: 2 min, purge volume: 100 mL, purge flow rate of helium: 10 mL min− 1. Concentrations of spiked halocarbons were as follows. CH3Cl, 1000 pmol L− 1 ; CH3Br, 40 pmol L− 1 ; C2H5Br, 20 pmol L− 1 ; CH
3I, 50 pmol L− 1 ; C2H5I, 9 pmol L− 1 ; CH2BrCl, 9 pmol L− 1; (CH3)2CHI, 7 pmol L− 1 ; (CD3)2CDI, 7 pmol L− 1 ; CH2Br2, 10 pmol L− 1 ; CH3 (CH2)2I, 6 pmol L− 1 ; CHBrCl2 , 8 pmol L− 1 ; CH2ClI, 9 pmol L− 1 ; CHBr
2Cl, 8 pmol L− 1; CH2BrI, 10 pmol L− 1 ; CHBr3 , 9 pmol L− 1 ; and CH2I2 , 9 pmol L− 1
Fig. 4 Effect of purge volume on the extraction efficiency of 15 halocarbons. Thermostatting time: 2 min, agitation speed: 500 rpm, purge flow rate of helium: 10 mL min− 1. Concentrations of spiked halocarbons were as follows. CH3Cl, 1000 pmol L− 1 ; CH3Br, 40 pmol L− 1 ; C2H5Br, 20 pmol L− 1 ; CH
3I, 50 pmol L− 1 ; C2H5I, 9 pmol L− 1 ; CH2BrCl, 9 pmol L− 1 ; (CH3)2CHI, 7 pmol L− 1 ; (CD3)2CDI, 7 pmol L− 1 ; CH2Br2, 10 pmol L− 1 ; CH3 (CH2)2I, 6 pmol L− 1 ; CHBrCl2 , 8 pmol L− 1 ; CH2ClI, 9 pmol L− 1 ; CHBr
2Cl, 8 pmol L− 1 ; CH2BrI, 10 pmol L− 1 ; CHBr3 , 9 pmol L− 1 ; and CH2I2 , 9 pmol L− 1
0 5,000 10,000 15,000 0 1,000 2,000 3,000 CH3Cl CH3Br a P eak ar ea of C H3 Cl Peak ar ea of C H 3Br 0 10,000 20,000 30,000 40,000 50,000 0 2,000 4,000 6,000 8,000 b C2H5Br CH3I C2H5I P eak ar ea of C H3 I P eak ar ea of C 2H 5I an d C 2H 5Br 0 3,000 6,000 9,000 d CH2BrCl CH2Br2 CHBrCl2 CHBr2Cl CHBr3 P eak ar ea 200 400 600 800 1000 0 500 1,000 1,500 2,000 e CH2ClI CH2BrI CH2I2 Agitation speed (rpm) P eak ar ea 0 1,000 2,000 3,000 c (CH3)2CHI CH3(CH2)2I (CD3)2CDI P eak ar ea
)L
0 5,000 10,000 15,000 20,000 0 1,000 2,000 3,000 4,000 5,000 CH3Cl CH3Br a Peak ar ea of C H3 Cl Peak ar ea of C H 3Br 0 10,000 20,000 30,000 0 2,000 4,000 6,000 8,000 10,000 C2H5Br CH3I C2H5I b P e ak ar ea of C H3 I Peak ar ea of C 2H 5I a nd C 2H 5Br 0 3,000 6,000 9,000 d CH2BrCl CH2Br2 CHBrCl2 CHBr2Cl CHBr3 Peak ar ea 75 100 125 150 0 1,000 2,000 3,000 CH2ClI CH2BrI CH2I2 e Purge volume (ml) P e ak ar ea 0 1,000 2,000 3,000 c (CH3)2CHI (CD3)2CDI CH3(CH2)2I Peak ar ea)L
)LJ
compounds with high boiling points from the samples, whereas purge volumes higher than 100 mL appeared to increase the loss of low-boiling-point analytes from the trap column. For the simultaneous analysis of 15 halocarbons, the following experiments were performed with a purging volume of 100 mL helium. The peak areas of very volatile halocarbons, such as CH3Cl and CH3Br, were decreased by increasing the purge flow rates of helium from 7.5 to 20 mL min−1 (Fig. 5a–c). On the other hand, the purge flow rate had no significant effect on the peak areas of other compounds, such as CHBr3 and CH2I2 (Fig. 5d–e). Since the lowest flow rate (7.5 mL min− 1) appeared to reduce the precision of the analyses of CH3Cl and CH3Br, the following experiments were conducted using a purge flow rate of 10 mL min− 1.
The optimum conditions for the DHS method are summarized in Table 1. The total duration of analysis (DHS extraction and GC/MS analysis) was approximately 5 hours for 10 samples.
3 – 2. Linearity, detection limits, and precision
Linearity was studied over two orders of magnitude. The linear range and the correlation coefficient R2 relative to each compound are shown in Table 2. The calibration curves indicated good correlations with R2 ≥ 0.990, except in the cases of CH3Cl (0.961), CH3Br (0.989), C2H5Br (0.968), and CH3I (0.984). The limit of detection (LOD) was calculated as 3V, where V is the standard deviation of 10 consecutive measurements of the standard solutions (Table 3). The relative standard deviation (RSD) of the results of the DHS GC/MS measurements were 24.7 % for CH3Cl, 16.3 % for CH3Br, 10.5 % for CH3I, and from 1.4 % to 8.8 % for the other halocarbons (Table 3).
3 – 3. Effects of bacterial medium matrices
To estimate the influence of the matrix of bacterial culture medium, standard solutions were separately prepared in ultrapure water and bacterial medium. The calibration cur ves showed excellent linearity for halocarbons at pico- to nano- mol L− 1 level in bacterial medium (R2 ≥ 0.987, except in the cases of CH
3Cl (0.960), CH3Br (0.895), C2H5Br (0.906), and CH3I (0.937)) as in ultrapure water (Table 2). These results indicated that the DHS analysis using the calibration cur ve of each matrix is suitable for the analysis of trace levels of Fig. 5 Effect of purge flow rate of helium on the extraction
efficiency of 15 halocarbons. Thermostatting time: 2 min, agitation speed: 500 rpm, purge volume: 100 mL. Concentrations of spiked halocarbons were as follows. CH3Cl, 1000 pmol L− 1 ; CH3Br, 40 pmol L− 1 ; C2H5Br, 20 pmol L− 1 ; CH3I, 50 pmol L− 1 ; C2H5I, 9 pmol L− 1 ; CH
2BrCl, 9 pmol L− 1 ; (CH3)2CHI, 7 pmol L− 1 ; (CD
3)2CDI, 7 pmol L− 1 ; CH2Br2, 10 pmol L− 1 ; CH3 (CH2)2I, 6 pmol L− 1 ; CHBrCl2, 8 pmol L− 1 ; CH2ClI, 9 pmol L− 1 ; CHBr2Cl, 8 pmol L− 1 ; CH2BrI, 10 pmol L− 1 ; CHBr
3 , 9 pmol L− 1 ; and CH2I2 , 9 pmol L− 1 0 5,000 10,000 15,000 0 1,000 2,000 3,000 4,000 CH3Cl CH3Br a P eak ar ea of CH 3 Cl Peak ar ea of CH 3Br 0 10,000 20,000 30,000 40,000 50,000 0 2,000 4,000 6,000 8,000 C2H5Br CH3I C2H5I b P eak ar ea of CH 3 I P eak ar ea of C 2H 5I and C 2H 5Br 0 2,000 4,000 6,000 8,000 10,000 CHBrCl2 CHBr2Cl CHBr3 CH2BrCl CH2Br2 d Peak ar ea 0 1,000 2,000 3,000 (CH3)2CHI (CD3)2CDI CH3(CH2)2I c P eak ar ea 5 10 15 20 0 500 1,000 1,500 2,000 CH2ClI CH2BrI CH2I2 e
Purge flow rate (ml min-1)
Peak
ar
ea
)
halocarbons in aqueous and culture bacterial samples. Since the slopes of the calibration curves were different between ultrapure water and bacterial medium, the halocarbons in the bacterial samples were quantified based on the calibration curves established by an analysis of halocarbon standards in the bacterial medium to correct for the matrix effects. Figure 6 shows typical ion chromatograms obtained from the analysis of the bacterial culture medium spiked with halocarbon standards. As shown in Fig. 6, the separation of every peak was sufficient for quantitative analysis, and there were no disturbances due to any other peaks from the medium.
3 – 4. Application of DHS GC/MS method to analysis of halocarbons in bacterial culture samples
The DHS GC/MS method was used to quantify halocarbons in the cultures of Erythrobacter longus (D- Proteobacteria) and
Alteromonas macleodii (J- Proteobacteria). The time courses of CH3Cl, CH3Br, and CH3I concentrations in the cultures of
E. longus are shown in Fig. 7. No other halocarbons (i.e.,
C2H5Br, C2H5I, CH2BrCl, (CH3)2CHI, CH2Br2, CH3 (CH2)2I,
CHBrCl2, CH2ClI, CHBr2Cl, CH2BrI, CHBr3, or CH2I2) were detected in the cultures. The OD600 time courses of the cultured samples indicated that E. longus increased during the incubation period (0.54 at 10 d). These results indicated that the concentrations of CH3Cl, CH3Br, and CH3I increased in the culture as they grew. Alteromonas macreodii incubated in culture medium with 1 mmol L− 1 KI, produced CH
3Cl, CH3Br, and CH3I up to 1.58×104±0.25×104, 6.41×103±0.91 × 103, 1.37 × 104
± 0.33 × 104 nmol L− 1, respectively (n = 3). The DHS GC/MS system can analyze about 1,000 bacterial samples within 6 months without being dismantled for cleaning.
4. Discussion
4 – 1. DHS GC/MS method
The DHS GC/MS method was evaluated by comparison with established methods: P&T GC/MS and solid-phase microextraction (SPME) GC/MS (Table 3). Table 3 shows that the DHS method was applicable to the simultaneous analysis of 15 halocarbons at pico- to nano-mol L− 1 levels. The sensitivity of the P&T method is more sensitive than the Table 3 Relative standard deviation (RSD) and limit of detection (LOD) using DHS GC/MS, LOD
using purge-and-trap (P&T) GC/MS, and LOD using solid-phase microextraction (SPME) GC/MS DHSa LOD using P&Tc (pmol L− 1 ) LOD using SPMEd (pmol L− 1 ) RSD ( % ) LODb ( pmol L− 1 ) CH3Cl 24.7 (1000)e 762 1.2 ̶f CH3Br 16.3 (40) 19.1 0.14 ̶f C2H5Br 5.6 (20) 4.2 ̶ f ̶f CH3I 10.5 (50) 17.9 0.03 ̶ f C2H5I 6.5 ( 9 ) 1.1 ̶ f ̶f CH2BrCl 4.7 ( 9 ) 2.4 0.02 ̶ f (CH3)2CHI 8.5 ( 7 ) 1.3 ̶f ̶f (CD3)2CDI 8.8 ( 7 ) 1.4 ̶f ̶f CH2Br2 2.9 (10) 2.0 0.05 ̶f CH3 (CH2)2I 8.1 ( 6 ) 1.3 ̶f ̶f CHBrCl2 1.4 ( 8 ) 1.5 0.02 104 CH2ClI 5.1 ( 9 ) 1.7 0.02 ̶ f CHBr2Cl 2.1 ( 8 ) 1.9 0.02 14.4 CH2BrI 2.9 (10) 1.2 ̶ f ̶f CHBr3 4.1 ( 9 ) 2.1 0.03 31.7 CH2I2 2.3 ( 9 ) 2.4 0.12 ̶f a This study
b LOD was calculated as 3V, where V is the standard deviation of 10 consecutive measurements of the standard solutions c
Kurihara et al. 2010
d
Allard et al. 2012
e
Values in brackets are the compound concentrations in pmol L− 1. f
DHS method. Unfortunately, however, the P&T method concentrates VOCs on the trap by bubbling an inert gas (e.g., ultrapure helium) through liquid samples, so this method cannot be used to measure VOCs in foam-forming samples (e.g., bacterial culture) or in solid samples (e.g., sediment). Whereas the SPME method can be used to measure VOCs in various samples, including bacterial cultures and sediment, the SPME method [15] is
less sensitive than the DHS method (Table 3) for the halocarbon measurement.
It was reported that Tenax TA has low water adsorption capacity, which is an impor tant proper ty for the measurement of trace gases in aqueous samples [24]. Furthermore, Tenax TA exhibits high thermal stability, relatively low water retention, and a low bleed rate [25]. To avoid or at least reduce the effects due to water matrix Fig. 6 Ion chromatograms of 15 halocarbon species in a bacterial medium sample using DHS GC/MS
analysis. Single ion chromatogram collected in selected-ion monitoring mode. Concentrations of spiked halocarbons were as follows. CH3Cl : 7,000 pmol L− 1; CH3Br : 7,000 pmol L− 1 ; C2H5Br : 500 pmol L− 1 ; CH
3I, C2H5I, CH2BrCl : 900 pmol L− 1 ; (CD3)2CDI, (CH3)2CHI : 700 pmol L− 1; CH2Br2 : 1,200 pmol L− 1 ; CH
3(CH2)2I : 600 pmol L− 1 ; CHBrCl2 : 800 pmol L− 1; CH2ClI : 900 pmol L− 1 ; CHBr2Cl : 800 pmol L− 1 ; CH2BrI : 1,000 pmol L− 1 ; CHBr3 , CH2I2 : 900 pmol L− 1
Abundance 1200 Abundance Abundance Abundance Abundance Abundance Abundance Abundance 600 0 400 200 200 100 800 400 0 160 80 120 60 60 80 120 60 100 200 300 100 200 300 100 50 600 300 0 120 60 800 400 0 180 120 60 2 4 6 8 10 2 4 6 8 10 Time (min) Time (min) a) P] 50 b) P] 94 c) P] 108 d) P] 142 e) P] 156 f) P] 130 g) P] 177 h) P] 170 i) P] 174 j) P] 83 k) P] 176 l) P] 129 m) P] 222 n) P] 173 o) P] 141 CH3Cl CH3Br C2H5Br CH3I C2H5I CH2BrCl (CD3)2CDI (CH3)2CHI CH3(CH2)2I CH2Br2 CHBrCl2 CH2ClI CHBr2Cl CH2BrI CHBr3 CH2I2
Fig. 5 Taniai et al
)LJ 7DQLDL HW DO
Alteromonas macreodii (J- Proteobacteria) produce halocarbons in the culture. Previous research has indicated that Erythrobacter species produced CH3Cl, CH3Br, and CH3I [11], similar to the results of halocarbon production in the present study. Even though the detected halocarbon concentrations were as low, i.e., at sub-nmol L− 1 levels, this DHS GC/MS method confirmed the production of halocarbons in the bacterial samples. In a previous study on bacteria belonging to J - Proteobacteria,
Alteromonas macleodii was reported to produce CH3I [12] and Pseudomonas species were repor ted to produce CH3Cl, CH3Br, and CH3I [11] in the culture experiment. These results suggested that not only Erythrobacter longus
(D- Proteobacteria) but also Alteromonas macreodii
(J - Proteobacteria) would produce halocarbons in the marine environment.
It was repor ted that biogenic marine aggregates, which were formed by phytoplankton-bacteria microbial communities, yield iodocarbons (CH3I, C2H5I, CH3CHICH3, CH3CH2CH2I) [26]. During a large aggregation event in the northern Adriatic, Alteromonadaceae (J- Proteobacteria) dominated in mucilaginous aggregates [27]. The results of the present study, taken together with the production of iodocarbons by marine aggregates, showed that aquatic bacterial communities could be new sources of methyl halides in marine environments. The present DHS GC/ MS method would be applicable to the evaluation of halocarbon production from marine microbial communities such as marine aggregates.
5. Conclusion
The DHS GC/MS method was optimized for the simultaneous determination of 15 halocarbons. This method can be used to quantify halocarbons at pico- to nano- mol L− 1 levels in aqueous samples. Linear regression analysis of the standard solution prepared with bacterial culture medium indicated that this method was successfully applied to the analysis of trace levels of halocarbons in bacterial culture. This DHS GC/MS method was applied to the quantitative determination of sub-nmol L− 1 levels of halocarbons for several days in bacterial culture. Besides having sufficient sensitivity, the DHS GC/MS system can analyze about 1,000 bacterial samples without being dismantled for cleaning. The present DHS GC/MS method will be reliable for and bleeding, Tenax TA was selected and used in the
present study as a solid sorbent for the stable measurement of trace halocarbons.
4 – 2. Production of halocarbons in bacterial culture samples
The results shown in Table 3 and Fig. 6 indicate that the DHS method is applicable to the determination of pico- to nano- mol L− 1 levels of halocarbons in bacterial medium as an alter native to the P&T method (for example, foam-forming samples when using the P&T method). The results shown in this study also indicate that both Erythrobacter longus (D- Proteobacteria) and
0
2000
4000
6000
8000
a
CH
3Cl
(
p
m
o
l L
-1)
0
200
400
600
b
CH
3Br
(
p
mo
l L
-1)
0
5
10
15
0
50
100
150
200
c
Incubation time (d)
CH
3I (
p
m
o
l L
-1)
Fig. 7 Time courses of CH3Cl, CH3Br, and CH3I concentrations in a culture of Erythrobacter longus. The circles and the er ror bars indicate the mean and the standard deviation, respectively (n = 3).
Scientific Research (C) (20510013) and a Grant-in-Aid for Scientific Research (B) (23310010) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan.
examinations of halocarbon production or degradation in marine microbial communities such as marine aggregates, sediment, and symbiotic cultures.
Acknowledgments
This study was supported in part by a Grant-in-Aid for
1) Ballschmiter, K. 2003. Pattern and sources of naturally produced organohalogens in the marine environment: biogenic formation of organohalogens. Chemosphere 52: 313-324. doi: 10.1016/S0045-6535(03)00211-X
2) O'Doherty S. J. & L. J. Carpenter, 2007. Halogenated volatile organic compounds. In Koppmann, R. (ed), Volatile Organic Compounds in the Atmosphere. Blackwell, Oxford: 173-220.
3) World Meteorological Organization, 2011. Scientific Assessment of Ozone Depletion: 2010. World Meteorological Organization Global Ozone Research and Monitoring Project; Geneva.
4) Butler, J. H. 2000. Better budgets for methyl halides? Nature 403: 260-261.
5) Schäfer, H., L. G. Miller, R. S. Oremland & J. C. Murrell, 2007. Bacterial Cycling of Methyl Halides. Advances in Applied Microbiology 61: 307-346. doi: 10.1016/S0065-2164(06)61009-5
6) Kurihara, M.K., M. Kimura, Y. Iwamoto, Y. Narita, A. Ooki, Y.-J. Eum, A. Tsuda, K. Suzuki, Y. Tani, Y. Yokouchi, M. Uematsu & S. Hashimoto, 2010. Distributions of short-lived iodocarbons and biogenic trace gases in the open ocean and atmosphere in the western Nor th Pacific. Marine Chemistr y 118: 156- 170. doi: 10.1016/j. marchem.2009.12.001
7) Saiz-Lopez, A., J. M. C. Plane, A. R. Baker, L. J. Carpenter, R. von Glasow, J. C. Gómez Martín, G. McFiggans & R. W. Saunders, 2012. Atmospheric Chemistr y of Iodine. Chemical Reviews 112: 1773-1804. doi: 10.1021/cr200029u 8) Laturnus, F., 1996. Volatile halocarbons released from
Arctic macroalgae. Marine Chemistr y 55: 359-366. doi:10.1016/S0304-4203(97)89401-7
9) Scarratt, M. G. & R. M. Moore, 1996. Production of methyl chloride and methyl bromide in laboratory cultures of marine phytoplankton. Marine Chemistry 54: 263-272. doi:10.1016/0304-4203(96)00036-9
10) Colomb, A., N. Yassaa, J. Williams, I. Peeken & K. Lochte, 2008. Screening volatile organic compounds (VOCs) emissions from five marine phytoplankton species by head space gas chromatography/mass spectrometry (HS-GC/MS). Journal of Environmental Monitoring 10: 325-330. doi: 10.1039/b715312k
11) Fujimori, T., Y. Yoneyama, G. Taniai, M. Kurihara, H. Tamegai & S. Hashimoto, 2012. Methyl halide production by cultures of marine proteobacteria Erythrobacter and Pseudomonas and isolated bacteria from brackish water. Limnology and Oceanography 57: 154-162. doi: 10.4319/
lo.2012.57.1.0154
12) Amachi, S., Y. Kamagata, T. Kanagawa & Y. Muramatsu, 2001. Bacteria Mediate Methylation of Iodine in Marine and Terrestrial Environments. Applied and Environmental Microbiology 67: 2718-2722. doi: 10.1128/ AEM.67.6.2718-2722.2001
13) Jakubowska, N., B. Zygmunt, 㶀. Polkowska, B. Zabiegała & J. Namie㶄nik, 2009. Sample pr eparation for gas chromatographic determination of halogenated volatile organic compounds in environmental and biological samples. Journal of Chromatography A 1216: 422-441. doi: 10.1016/j.chroma.2008.08.092
14) Safarova, V. I., S. V. Sapelnikova, E. V. Djazhenko, G. I. Teplova, G. F. Shajdulina & F. Kh. Kudasheva, 2004. Gas chromatography-mass spectrometry with headspace for the analysis of volatile organic compounds in waste water. Journal of Chromatography B 800: 325-330. doi: 10.1016/ j.jchromb.2003.10.070
15) Allard, S., J. W. A. Charrois, C. A. Joll & A. Heitz, 2012. Simultaneous analysis of 10 trihalomethanes at nanogram per liter levels in water using solid-phase microextraction and gas chromatography mass-spectrometry. Journal of Chr omatography A 1238 : 15 - 21. doi:10.1016/j. chroma.2012.03.020
16) Huang, W., X. Bu, L. Nguyen, R. H. Gammon & J. L. Bullister, 2000. Production and consumption of methyl halides in a freshwater lake. Limnology and Oceanography 45: 1537-1545.
17) Christof, O., R. Seifert & W. Michaelis, 2002. Volatile halogenated organic compounds in European estuaries. Biogeochemistry 59: 143-160.
18) Hashimoto, S., T. Tanaka, N. Yamashita & T. Maeda, 2001. An automated purge and trap gas chromatography- mass spectrometry system for the sensitive shipboard analysis of volatile organic compounds in seawater. Journal of Separation Science 24: 97-103.
19) Hashimoto, S., S. Toda, K. Suzuki, S. Kato, Y. Narita, M. K. Kurihara, Y. Akatsuka, H. Oda, T. Nagai, I. Nagao, I. Kudo & M. Uematsu, 2009. Production and air-sea flux of halomethanes in the western subarctic Pacific in relation to phytoplankton pigment concentrations during the iron fertilization experiment (SEEDS II). Deep Sea Research Part II 56: 2928-2935. doi: 10.1016/j.dsr2.2009.07.003 20) Kurihara, M., M. Iseda, T. Ioriya, N. Horimoto, J. Kanda,
T. Ishimar u, Y. Yamaguchi & S. Hashimoto, 2012. Brominated methane compounds and isoprene in surface seawater of Sagami Bay: Concentrations, fluxes,
of the solid adsorbents Tenax TA, Tenax GR, Carbotrap, Carbotrap C, Carbosieve SIII, and Carboxen 569 and water management techniques for the atmospheric sampling of volatile organic trace gases Analytical Chemistr y 67: 4380-4386.
25) Pillonel, L., J. O. Bosset & R. Tabacchi, 2002. Rapid Preconcentration and Enrichment Techniques for the Analysis of Food Volatile. A Review. Lebensmittel-Wissenschaft und-Technologie 35: 1-14. doi: 10.1006/ fstl.2001.0804
26) Hughes, C., G. Malin, C. M. Turley, B. J. Keely, P. D. Nightingale & P. S. Liss, 2008. The production of volatile iodocarbons by biogenic marine aggregates. Limnology a n d O c e a n o g r a p h y 5 3 : 8 6 7 - 8 7 2 . d o i : 1 0 . 4 3 1 9 / lo.2008.53.2.0867
27) Blažina, M., M. Najdek, D. Fuks & N. Smodlaka, 2011. Microbial communities in northern Adriatic mucilaginous aggregates: insight into the early phase of aggregate formation Hydrobiologia 658: 213-220, doi: 10.1007/ s10750-010-0474-6
and relationships with phytoplankton assemblages. Marine Chemistr y 134- 135: 71-79. doi: 10.1016/j. marchem.2012.04.001
21) Hughes, C., D. J. Franklin & G. Malin, 2011. Iodomethane production by two impor tant marine cyanobacteria: Prochlorococcus marinus (CCMP 2389) and Synechococcus sp. (CCMP 2370) . Marine Chemistr y 125: 19-25. doi: 10.1016/j.marchem.2011.01.007
22) Manzini, S., C. Durante, C. Baschieri, M. Cocchi, S. Sighinolfi, S. Totaro & A. Marchetti, 2011. Optimization of a Dynamic Headspace - Thermal Desorption - Gas Chromatography/Mass Spectrometry procedure for the determination of furfurals in vinegars. Talanta 85: 863-869. doi: 10.1016/j.talanta.2011.04.018
23) Ochiai, N, K. Sasamoto, A. Hoffmann & K. Okanoya, 2012. F u l l e v a p o r a t i o n d y n a m i c h e a d s p a c e a n d g a s chromatography-mass spectrometry for uniform enrichment of odor compounds in aqueous samples. Journal of Chr omatography A 1240: 59- 68. doi: 10.1016/j. chroma.2012.03.097
