Simple Determination of Gaseous and Particulate ... - ACS Publications

Jun 4, 2018 - GF-CX-572 cartridge can be replaced with a new cartridge within 3 s. An overview of ... this claim. Few studies have reported analytical...
2 downloads 0 Views 2MB Size
This is an open access article published under a Creative Commons Non-Commercial No Derivative Works (CC-BY-NC-ND) Attribution License, which permits copying and redistribution of the article, and creation of adaptations, all for non-commercial purposes.

Article Cite This: Chem. Res. Toxicol. 2018, 31, 585−593

pubs.acs.org/crt

Simple Determination of Gaseous and Particulate Compounds Generated from Heated Tobacco Products Shigehisa Uchiyama,*,†,‡ Mayumi Noguchi,‡ Nao Takagi,‡ Hideki Hayashida,§ Yohei Inaba,† Hironao Ogura,§ and Naoki Kunugita† †

Department of Environmental Health, National Institute of Public Health, 2-3-6, Minami, Wako-shi, Saitama 351-0197, Japan Faculty of Engineering and §Graduate School of Engineering, Chiba University, 1-33 Yayoicho, Inage-ku, Chiba-shi, Chiba 263-8522, Japan

Downloaded via 185.101.71.40 on July 17, 2018 at 14:43:25 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.



S Supporting Information *

ABSTRACT: As a new form of cigarettes, heated tobacco products (HTPs) have been rapidly distributed worldwide. In this study, an improved method for analyzing gaseous and particulate compounds generated from HTPs is described. Smoke is collected using a GF-CX572 sorbent cartridge with 300 mg of carbon molecular sieves, that is, Carboxen 572 (CX572), and a 9 mm glass-fiber filter (GF). After collection, the CX572 particles from the cartridge are transferred along with the GF and deposited into a vial containing two phases of carbon disulfide and methanol. The CX572 particles settle into the lower carbon disulfide phase, while nonpolar compounds are desorbed. After the sample is allowed to stand, the solution is slowly stirred. The two-phase mixture of carbon disulfide and methanol is combined into a homogeneous solution. Polar compounds are then desorbed, while the desorbed nonpolar compounds remain in solution. For the analysis of carbonyl compounds, an enriched 2,4-dinitrophenylhydrazine solution is added to a portion of the combined solution for derivatization and subsequent high-performance liquid chromatography analysis. For the analysis of volatile organic compounds and water, a portion of the combined solution is analyzed by gas chromatography−mass spectrometry or equipped with a thermal conductivity detector. By applying the proposed GF-CX572 one-cartridge method to the analysis of the mainstream smoke generated from HTPs and traditional cigarettes, several chemical compounds are detected, and the chemical composition of smoke is revealed. The GF-CX572 one-cartridge method can analyze gaseous and particulate chemical compounds from the HTP smoke by utilizing not only the entire puff volume but also one puff volume because the GF-CX-572 cartridge can be replaced with a new cartridge within 3 s. An overview of the chemicals generated from HTPs is obtained in detail by one-puff volume sampling. In addition, the generated chemical compounds strongly depend on the temperature of tobacco leaves in HTPs.



INTRODUCTION Cigarette smoke contains more than 5000 chemical compounds, at least 50 of which are carcinogenic.1,2 It is associated with various pulmonary and cardiovascular disorders including emphysema, atherosclerosis, and cancer,3−7 which account for 28.6% of all cancer deaths.8 To reduce the health issues caused by cigarette smoke, a new cigarette product, that is, heated tobacco product (HTP), has been recently launched. Philip Morris International (PMI) has created an HTP called I-Quit-OrdinarySmoking (iQOS), which was initially launched in 2014 in Nagoya, Japan, and Milan, Italy.8 In addition, British American Tobacco (BAT) has created an HTP called “glo”, which was initially launched in 2016 in Sendai, Japan.9 Japan Tobacco (JT) subsequently released a hybrid product of HTPs and E-cigarettes called “Ploom TECH”, which was initially launched in 2016 in Fukuoka, Japan.10 iQOS and glo heat but do not burn the tobacco leaf for the inhalation of nicotine-containing smoke. With the use of Ploom TECH, nicotine-containing smoke is © 2018 American Chemical Society

inhaled, and an aerosol is generated by heating the E-liquid containing propylene glycol and glycerol passes through a capsule of granulated tobacco leaves. Currently, these HTP tobacco products are being distributed worldwide. HTP products are thought to release fewer chemical compounds because tobacco does not undergo combustion, leading to the reduction or elimination of compounds that are only produced at combustion temperatures. HTPs appear to be less detrimental than traditional cigarettes in terms of health-related effects; however, limited analytical data are available to support this claim. Few studies have reported analytical data for iQOS,11,12 but to the best of our knowledge, studies have not reported analytical data for glo and Ploom TECH products. In effect, sufficient research has not been carried out for providing a balanced view.9 Received: February 25, 2018 Published: June 4, 2018 585

DOI: 10.1021/acs.chemrestox.8b00024 Chem. Res. Toxicol. 2018, 31, 585−593

Article

Chemical Research in Toxicology

An LX20 linear 20-port piston-type smoking machine (Heinrich Borgwaldt GmbH, Hamburg, Germany) was used to collect mainstream cigarette smoke in accordance with the Health Canada Intense (HCI) regime20 and the International Organization for Standardization (ISO) regime.21 In the HCI regime, mainstream smoke constituents were collected at a puff volume of 55 mL, puff duration of 2 s, and puff interval of 30 s, with 100% blocking of the filter ventilation holes using a Mylar adhesive tape. In the ISO regime, mainstream smoke constituents were collected at a puff volume of 35 mL, puff duration of 2 s, and puff interval of 60 s, with no blocking of the filter ventilation holes. Reagents. CX572 (20/45 mesh) was purchased from Sigma-Aldrich Inc. (St. Louis, MO, USA). Glass fiber prefilters (GF, AP2504700) were purchased from Merck Millipore Ltd. (Darmstadt, Germany). A standard 1,3-butadiene solution (2.0 mg/mL in methanol) was purchased from AccuStandard Inc. (New Haven, CT, USA). The water used for HPLC and sample preparation was deionized and purified using a Milli-Q Water System equipped with a UV lamp (Millipore, Bedford, MA, USA). Benzene-d6 (99.95%), isoprene (95.0%), acrylonitrile (97%), benzene (99.7%), toluene (99.7%), and carbon disulfide (99.0%) were purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). Acetonitrile (HPLC grade, > 99.9%), ethanol (>99.5), methanol (anhydrous, 99.8%), phosphoric acid (85% solution in water), (−)-nicotine (≥99%), isoquinoline (97%), and ammonium acetate (99.999%) were purchased from Sigma-Aldrich Inc. (St. Louis, MO, USA). DNPH hydrochloride (>98%) was purchased from Tokyo Kasei Co. Ltd. (Tokyo, Japan). All carbonyl DNPH derivatives for HPLC analysis were synthesized according to previous studies.22,23 HTPs and Reference Cigarettes. Three HTP cigarette brands were used. For iQOS (PMI Inc., Desales NW, Washington, DC. USA), “regular”, “balanced regular”, “mint”, and “menthol” tobacco sticks called HeatSticks were used. For glo (BAT, Southampton, UK), “bright tobacco”, “fresh mix”, and “intensely fresh” tobacco sticks called neostiks were used. For Ploom TECH (JT Inc., Tokyo, Japan), “Mevius Legular”, “Cooler Green”, and “Cooler Purple” liquid capsules called Tobacco caps were used. Figure S1 in the Supporting Information shows the composition of these HTPs. Conventional cigarettes 3R4F and 1R5F from the University of Kentucky (Lexington, KY, USA) and CM6 from the Cooperation Center for Scientific Research Relative to Tobacco (CORESTA, Paris, France) were used. Test tobacco sticks and conventional cigarettes were used for measurement after being placed at 22 °C and 60% humidity for 2 days. Preparation of the Sorbent Cartridge (CX572 Cartridge) and GF Filter. Three-hundred milligrams of CX572 particles was weighed into a glass tube and conditioned using a tube conditioner (TC-20, Markes Int. Ltd., Mid Glamorgan, UK) at 380 °C for 5 h under a purified nitrogen flow at 50 mL/min. After cooling to room temperature, carbon adsorbents were packed into an empty polyethylene SPE tube (3 mL, Supelco Inc., Bellefonte, PA, USA) with end frits. The GF was washed with 50 mL of acetonitrile and dried under vacuum to 5−8 MPa at room temperature. After drying, the washed GF was cut to a φ of 9 mm. Figure 1 shows the CX572 cartridge and GF filter. Preparation of Concentrated DNPH Solution. Phosphoric acid (10 mL) and DNPH hydrochloride (1 g) were added to a 50 mL volumetric flask and diluted to 50 mL with acetonitrile. This mixture solution was then continuously stirred with a magnetic stirrer until a clear solution was obtained. The solution was then stored in a refrigerator at 4 °C until use. Preparation of Internal Standards for GC/MS and GC/TCD Analysis. An internal standard mixed solution (ISMS) containing ethanol, benzene-d6, and isoquinoline was used for GC/MS and GC/TCD. Ethanol (10 g), benzene-d6 (0.1 g), and isoquinoline (0.5 g) were added into a 40 mL volumetric flask and diluted with methanol. The resulting ISMS contained 250 mg/mL ethanol (IS1), 2.5 mg/mL benzene-d6 (IS2), and 12.5 mg/mL isoquinoline (IS3). Collection of Cigarette Smoke Using Sorbent Cartridge. Figure 1 shows the outline of the cigarette smoke sample collection method using the sorbent cartridge. The end-cap from the CX572 cartridge was removed, and one or two pieces of GF cut to a φ of 9 mm was set. Ordinarily, a single GF is used for smoke collection;

It is crucial to identify and quantify volatile organic compounds (VOCs), carbonyls, tars, and other chemical compounds in cigarette smoke and HTP vapors and to evaluate the effect of smoking on human health. Currently, three conventional methods are available to collect nicotine, tar, VOCs, and carbonyls from mainstream cigarette smoke. For analyzing VOCs, mainstream cigarette smoke is passed through a Cambridge filter pad (CFP); the vapor phase is cryogenically trapped in an impinger containing 10 mL of methanol and cooled to a temperature of less than −70 °C in a dry ice/isopropanol bath.13,14 Then the impinger solution is subjected to gas chromatography−mass spectrometry (GC/MS) analysis for quantification. For analyzing carbonyls, carbonyls are collected by the passage of mainstream smoke into an impinger containing 80 or 35 mL of a 2,4-dinitrophenylhydrazine (DNPH) solution,15,16 and the chemical composition is quantitatively determined by high-performance liquid chromatography (HPLC). For analyzing nicotine, mainstream cigarette smoke is passed through a CFP. After elution with 2-propanol, a portion of the eluate is subjected to GC/ flame ionization detection for quantification. These conventional methods require multiple, sometimes large, sampling devices such as an impinger with a cryogenic bath. Previously, a simple method for the simultaneous collection of nicotine, tar, VOCs, and carbonyls from mainstream cigarette smoke, using a CFP for particulate compounds and a sorbent cartridge packed with carbon molecular sieves, that is, Carboxen 572 (CX572), to isolate gaseous compounds without a traditional impinger, has been reported by our group.17,18 This sorbent cartridge method has been applied to traditional cigarette and E-cigarettes.19 As the analysis of the compounds generated from HTP is thought to be difficult because of low compound concentrations, an even simpler and more sensitive extraction method was developed using a small single-sorbent cartridge comprising a glass-fiber filter pad and a CX572 cartridge, which can be applied for the analysis of smoke from HTPs and traditional cigarettes.



MATERIALS AND METHODS

Apparatus. A Prominence LC-20 HPLC system (Shimadzu, Kyoto, Japan) was used with two LC-20AD pumps, an SIL-20AC autosampler, and an SPD M20A photodiode array detector. An Ascentis RP-Amide analytical column, with a 3 μm particle size and an i.d. of 150 mm × 3 mm i.d. (Supelco Inc., Bellefonte, PA, USA), was used. Solutions A and B of the mobile phase mixture comprised acetonitrile/ water (50:50 v/v) containing 10 mmol/L ammonium acetate and acetonitrile/water (80:20 v/v), respectively. HPLC elution was carried out using 100% A for 5 min, followed by a linear gradient from 100% A to 100% B in 50 min, and then maintained for 10 min. The flow rate of the mobile phase was 0.8 mL/min. The column temperature was 30 °C, and the injection volume was 10 μL. Detection was carried out at maximum wavelengths from 300 to 500 nm. A QP 2010 Plus GC/MS system (Shimadzu, Kyoto, Japan) was used with a fused-silica column (InertCap AQUATIC-2 60 m × 0.25 mm i.d., d = 1.4 μm, GL Sciences, Tokyo, Japan) and programmed from 40 °C (held for 6 min) to 250 °C at 6 °C/min, using the carrier gas He at 1 mL/min and EIMS detection of 70 eV operated in the full-scan mode from m/z 40−400 and SIM mode. Table S1 in the Supporting Information summarizes the quantified and verified ions detected for each compound in the SIM mode. The injection volume was 1 μL (split injection, split ratio 10:1), the septum purge was 5 mL/min, and the injector temperature was 240 °C. A QP 2010 GC-TCD system (Shimadzu, Kyoto, Japan) was used with a packed column (Porapack Q 2 m × 3 mm i.d., 80−100 mesh deactivated stainless, GL Sciences, Tokyo, Japan) and operated at an oven temperature of 170 °C with the carrier gas He at 30 mL/min. The injection volume was 2 μL, the injector temperature was 250 °C, and the detector temperature was 250 °C. 586

DOI: 10.1021/acs.chemrestox.8b00024 Chem. Res. Toxicol. 2018, 31, 585−593

Article

Chemical Research in Toxicology

Figure 1. Schematic of the mainstream smoke collection system using a GF-CX572 cartridge. however, when high amounts of total particulate matter were generated in the HCI regimen, two GF pieces were used. Single GF was used for all HTPs in this study. Elution of GF-CX572 Cartridge and Analysis. Figure 2 shows the flowchart of the analytical procedure, and Figure 3 shows the

were desorbed. After allowing the sample to stand for 5 min, the solution was slowly stirred. The two-phase solution of carbon disulfide and methanol was dissolved into a single phase, polar compounds were desorbed, and nonpolar compounds remained in solution. The solution was then stirred in a rotary shaker at 120 cycles per min for 20 min. For analyzing carbonyls, 0.5 or 1 mL of the combined solution was transferred to a 5 mL volumetric flask, and 0.2 mL of the enriched DNPH solution was added. After 10 min, the solution was diluted to 5 mL with ethanol and subjected to HPLC analysis. For analyzing VOCs and water, 1 mL of the combined solution was transferred to a 1.5 mL autosampler vial, ISMS (10 μL for ISO or 20 μL for HCI) was added, and the solution was subjected to GC/MS and GC/TCD analysis under the conditions described in the Apparatus section. The total gaseous and particulate matter (TGPM) was determined by the comparison of the mass of the trapped smoke before and after smoking runs.



RESULTS AND DISCUSSION Chromatographic Profiles of Target Compounds by GC/MS, GC/TCD, and HPLC. Several VOCs are detected in the eluates from the mainstream smoke of HTPs and traditional cigarettes using GC/MS. Figure S2 in the Supporting Information shows a TIC chromatogram recorded in the scan mode obtained from iQOS-Menthol with the HCI smoking regimen as reference. VOCs from 1,3-butadiene to nicotine are completely separated. Several carbonyl compounds are detected in eluates from the mainstream smoke of HTPs and traditional cigarettes using HPLC. Figure S3 in the Supporting Information shows an HPLC chromatogram with photodiode array detection at maximum wavelengths obtained from iQOS-Menthol with the HCI smoking regimen. DNPH derivatives from formaldehyde to 2-nonenal are detected and completely separated. Water in the eluate from mainstream smoke was analyzed by GC/TCD. Figure S4 in the Supporting Information shows a GC/TCD chromatogram. The peak of water is detected just in front of the methanol peak. Water and internal standard peaks are completely separated from solvent peaks including those of methanol and carbon disulfide. Limit of Detection, Limit of Quantitation, and Reproducibility. VOCs and carbonyl compounds. Under optimum analytical conditions, the equations for the calibration curve, linear range, limit of detection (LOD), and limit of quantitation (LOQ) are determined for the analytes in a sample matrix (Table S1 in the Supporting Information). LODs defined on the basis of the signal-to-noise ratios of 3 were calculated as 0.76−17 μg/L, except for water. The LOQ, which is calculated using a signal-to-noise ratio of 10, was 2.5−58 μg/mL.

Figure 2. Flowchart of the analytical procedure for the simultaneous determination of gaseous and particulate chemical compounds from mainstream cigarette smoke.

Figure 3. Two-phase elution of chemical compounds from the GF-CX572 cartridge. outline of a two-phase elution of chemical compounds from the GF-CX572 cartridge. A 1 mL portion of carbon disulfide was added into an 8 mL vial with a PTFE septum, and 4 mL of methanol was slowly added into the vial. The solution in the vial was maintained in two phases (bottom phase = carbon disulfide). After the cigarette smoke was collected, the CX572 particles with GF and frits were transferred from the cartridge into the vial. CX572 particles settled into the bottom carbon disulfide phase in which nonpolar compounds 587

DOI: 10.1021/acs.chemrestox.8b00024 Chem. Res. Toxicol. 2018, 31, 585−593

Article

Chemical Research in Toxicology

Therefore, the nicotine generation rates are calculated as 23% for iQOS; 30% for glo; and 3.5% for Ploom TECH. The nicotine generation rate from Ploom TECH is extremely low, possibly because the tobacco leaf of Ploom TECH is not heated, and nicotine is generated by passing propylene glycol mist into the tobacco leaf. Figure 4 shows the major chemical compounds generated from HTPs and traditional cigarettes for comparison. Chemical compounds are listed in the descending order of the amount in the traditional cigarette. In a traditional cigarette, the most abundant chemical compounds generated are nicotine, followed by acetaldehyde, glycerol, isoprene, and acetone. In HTP, the most abundant chemical compounds generated are glycerol, followed by menthol, nicotine, propylene glycol, and acetol. Traditional cigarettes and HBNs use tobacco leaves. The difference in compound generation is thought to be related to the different temperatures required to heat (HTP) compared to combust (traditional cigarette) tobacco leaves. Generation of Chemical Compounds on Each Puff Smoking and Temperature of Tobacco Leaf. The GF-CX572 one-cartridge method can analyze gaseous and particulate compounds from the mainstream smoke by utilizing not only one entire cigarette, but also from one-puff volume because of its high sensitivity and simple operation. The GF-CX-572 cartridge installed in the smoking machine can be replaced with a new cartridge at 5 s within a puff interval of HCI (28 s). Sampling of each puff was continuously performed until 20 puffs were sampled for one tobacco stick. The smoking protocol was in accordance with the HCI regime with a puff number of 20 and an operating time of 600 s. Heating times are 360 s after preheating of 20 s for iQOS and 180 s min after preheating of 50 s for glo. At this time, the temperature of tobacco leaves, at 1-s intervals using a TC-08 thermocouple data logger (Pico Technology, Cambridgeshire, UK), was measured. A thermocouple was placed in the center of the tobacco leaves in the HTPs, and temperature changes were recorded. Figure 5 shows the temperature change on the tobacco leaf and generation of chemical compounds (such as TGPM, nicotine, acetol) with respect to iQOS, glo, and Ploom TECH. Figure S6 in the Supporting Information shows the changes in cumulative amounts of nicotine from mainstream HTP smoke with puff number as reference. The cumulative amounts of nicotine generated from iQOS and Ploom TECH from 12 puffs are similar to the data listed in Table 1. For iQOS and glo, high amounts of TGPM including water are generated in the initial puffs and rapidly decrease (Figure 5). For iQOS, the tobacco leaf is heated by inserting an electronically controlled heating blade (length 13.5 mm, width 5 mm, and thickness 0.35 mm), and after heating, the temperature of the tobacco leaf rapidly decreases. For glo, the tobacco leaf is heated by begirded using a heating block, and after heating, the temperature of the tobacco leaf slowly decreases. For iQOS and glo, temperature in the tobacco leaf decreased intermittently with each puff. For Ploom TECH, the heating program is basically different from those of iQOS and glo. The heater is switched on only for smoking and generates propylene glycol mist. Nicotine is generated by passing propylene glycol mist into the tobacco leaf. Therefore, the temperature of tobacco leaf and TGPM including nicotine remain substantially constant. The mean temperatures for the heating times of the tobacco leaves in iQOS, glo, and Ploom TECH were 210, 170, and 23 °C, respectively. For traditional cigarettes, the temperature increased with each puff, with a mean combustion temperature of 460 °C (Figure S5).

Linearity was examined over a concentration range of 5− 2000 μg/L for VOCs and 0.1−1000 μmol/L for carbonyl compounds, with excellent coefficients of determination greater than 0.9966, except for glycerol and propylene glycol. The reproducibility of the analytical method, expressed as the relative standard deviation (RSD), was estimated from data of 10 blank one-pot solutions spiked with a 10-μL VOC mixture standard solution (500 mg/L) and carbonyl-DNPhydrazone mixture standard solution (20 mmol/L). Table S1 summarizes the RSD values, which range from 1.9% to 5.1% (carbonyls) and from 0.23% to 4.4% (VOCs), indicating good reproducibility of the method. Water. Water blank in the sample solution is always detected because the methanol reagent contains water. Therefore, LOD and LOQ were determined on the basis of three times the standard deviation of the sample blank (10 reference blank solutions). Measurement of Mainstream Smoke Generated from HTP and Traditional Cigarette Using GF-CX572 Cartridge Followed by Two-Phase Elution. Mainstream smoke of two HTP brands (i.e., iQOS, glo, and Ploom TECH respectively) and traditional cigarettes (CM6, 3R4F, and 1R5F) were analyzed using the GF-CX572 cartridge followed by the two-phase desorption method. iQOS can perform for 6 min or 15 puffs per stick, while glo can perform for 3 min per stick. Ploom TECH can perform until 100 s or 50 puffs (JT’s recommendation). Therefore, the smoking protocol of iQOS and Ploom TECH is in accordance with the HCI (12 puffs) or ISO (6 puffs) regimen. However, for the smoking protocol of glo, the puff interval of the HCI regimen was modified to 16 s for the HCI regimen and 38 s for the ISO regimen to ensure that the same puff number is obtained for glo and iQOS. Tables 1 and 2 summarize the measured values of chemical compounds from mainstream smoke based on the HCI and ISO smoking protocols, respectively. For conventional cigarette, almost all of the measured values by the GF-CX572 method are similar to previously reported data.13,16,24 The amounts of some chemical compounds generated from the iQOS mainstream smoke listed in Table 1 have been previously reported12,25 and the approximate values are shown in this table. Overall, chemical compounds generated from HTP are fewer than those generated from traditional cigarettes, except water, propylene glycol, glycerol, and acetol. For iQOS and glo, the most abundant compound is water, which accounts for 75−85% of TGPM. In contrast, water generated from traditional cigarettes accounts for 17−27% of TGPM. Mainstream HTP smoke contains higher amounts of propylene glycol (HCI, 240−850 μg/stick) compared to traditional cigarettes (HCI, 11−28 μg/stick). In addition, HTP also generates higher amounts of acetol (HCI, 140−260 μg/stick) compared with traditional cigarettes (HCI, 50−110 μg/stick). Except Ploom TECH, propylene glycol in HTP sticks may be oxidized by heating, affording acetol and methylglyoxal (Scheme 1).19,26 When HTP sticks designated “Mint”. “Menthol” (iQOS), and “Fresh” (glo) were used, the highest amounts of menthol (HCI, 2000−2700 μg/stick) and 2-nonenal (HCI, 6.5−74 μg/stick) were detected from mainstream smoke. From the analysis of the tobacco leaf from the HTP sticks by the conventional method,27 nicotine content was 5.2 mg/stick for iQOS; 1.7 mg/stick for glo; and 6.5 mg/stick for Ploom TECH. However, the content of nicotine generated from HTPs by the HCI regimen is roughly 1200 μg/stick for iQOS; 510 μg/stick for glo; and 230 μg/stick for Ploom TECH. 588

DOI: 10.1021/acs.chemrestox.8b00024 Chem. Res. Toxicol. 2018, 31, 585−593

589

2.1 ± 0.32 9.7 ± 2.0 39 ± 12 5000 ± 790 2000 ± 240 1200 ± 170 6.0 ± 0.86 230 ± 9.9 48 ± 3.2 8.3 ± 0.86 17 ± 1.2 3.8 ± 0.31 11 ± 0.69 22 ± 0.38 2.4 ± 0.43 14 ± 1.3 5.4 ± 0.24 5.4 ± 1.0 6.9 ± 0.44 74 ± 1.6 38 ± 2.3 45 ± 2.7

12 0.21 ± 0.07 2.8 ± 0.80 0.14 ± 0.03 75 ± 15 0.89 ± 0.07 2.0 ± 0.28 260 ± 75 370 ± 150

mint

0.33 ± 0.02 5.2 ± 0.56 100 ± 7.0 5000 ± 390 6.8 ± 2.4 570 ± 66 10 ± 1.5 240 ± 4.8 26 ± 0.6 5.5 ± 0.12 15 ± 0.65 18 ± 0.34 15 ± 0.18 28 ± 0.63 6.0 ± 0.23 12 ± 0.34 6.5 ± 0.24 37 ± 2.2 13 ± 0.1