|Year : 2019 | Volume
| Issue : 3 | Page : 130-135
Fe3O4 nanoparticle-modified multi-walled carbon nanotubes for the determination of synthetic pyrethroids in human blood
Cui Yanhua1, Shen Weidong1, Yang Fan1, Pan Yaoyue1, Zhang Suling2, Yao Weixuan3
1 Institute of Criminal Science and Technology, Huzhou Public Security Bureau, Huzhou, China
2 College of Materials and Environmental Engineering, Hangzhou Dianzi University, Hangzhou, China
3 Department of Criminal Science and Technology, Zhejiang Police College, Hangzhou, China
|Date of Web Publication||18-Sep-2019|
Huzhou Public Security Judicial Appraisal Center, Huzhou 313000
Source of Support: None, Conflict of Interest: None
In this study, a sensitive and rapid magnetic solid-phase microextraction (MSPE) procedure based on Fe3O4 nanoparticle-modified, multi-walled carbon nanotubes combined with a gas chromatography/electron capture detector (GC/ECD) was developed to quantify pyrethroids in human blood samples. This study compared liquid–liquid extraction and solid-phase extraction methods and found that the MSPE-GC/ECD method had many advantages, such as high recovery rate, high extraction efficiency, high detection sensitivity, low limit of detection, and simple operation. It can meet the actual requirements for the determination of synthetic pyrethroids in human blood.
Keywords: Carbon nanotube, Fe3O4, nanoparticles, synthetic pyrethroids
|How to cite this article:|
Yanhua C, Weidong S, Fan Y, Yaoyue P, Suling Z, Weixuan Y. Fe3O4 nanoparticle-modified multi-walled carbon nanotubes for the determination of synthetic pyrethroids in human blood. J Forensic Sci Med 2019;5:130-5
|How to cite this URL:|
Yanhua C, Weidong S, Fan Y, Yaoyue P, Suling Z, Weixuan Y. Fe3O4 nanoparticle-modified multi-walled carbon nanotubes for the determination of synthetic pyrethroids in human blood. J Forensic Sci Med [serial online] 2019 [cited 2021 May 15];5:130-5. Available from: https://www.jfsmonline.com/text.asp?2019/5/3/130/267153
| Introduction|| |
Pyrethroid insecticides are a broad spectrum of insecticides that control a variety of pests. Pyrethroid is a contact pesticide with pesticidal effects 10–100-times stronger than those of organic chlorine and organic phosphorus. In the recent decades, pyrethroid has been widely used in the cultivation of fruits, vegetables, cotton, and other plants. Such insecticides can induce neurotoxicity in insects and mammals. In humans, a variety of reversible symptoms, such as paresthesia, irritation of the skin and mucosa, headache, dizziness, and nausea, are reported following pyrethroid exposure. Analysis of pyrethroids in human blood is challenged by its low detectable levels and interference by the various proteins, enzymes, hormones, and electrolytes present in the blood. Therefore, it is important to develop analytical methods that allow accurate determination of pyrethroids in the blood.
The main analytical methods used to detect pyrethroids are gas chromatography (GC),,,,, GC/mass spectrometry (MS),, high-performance liquid chromatography,,, and liquid chromatography-MS,,,, which are preceded by various sample preparation techniques, such as liquid–liquid extraction (LLE),,, solid-phase extraction (SPE),,, and solid-phase microextraction (SPME)., However, conventional methods, such as LLE and SPE, require a long time and can easily lead to clogging in the adsorption column and other issues. The literature reports several extraction methods that do not require solvents, such as SPME and liquid-phase microextraction (LPME). SPME has several disadvantages; it is not cost-effective and carry-forward effect is common with SPME, and recovery by LPME is not ideal. In general, the pretreatment method in the analysis of pyrethroid pesticides in the blood is ubiquitous but time-consuming and laborious.
MSPE, a new alternative mode of SPE that uses a sorbent phase with magnetic susceptibility, facilitates the separation of solid materials from a solution by means of an external magnetic field, and thus, has the advantages of easy operation and short extraction time for sample pretreatment. Moreover, for environmental or biological samples, the MSPE mode eliminates the possibility of column clogging.
Multi-walled carbon nanotubes (MWNTs),,,, are comprised of a layer of carbon atoms (graphite sheets) rolled into a seamless, hollow tube at a certain angle. Carbon nanotubes are another allotrope of carbon discovered after C60. Their radial size is small; the outer diameter of the tube is in the nanometer range, whereas the inner diameter of the tube is smaller than that (some are only approximately 1 nm). Their length is generally in the micrometer range, and their length-to-diameter ratio is very large. These nanotubes have excellent mechanical, electrical, and chemical properties. Specifically, they present outstanding mechanical strength, heat resistance, resistance to acid and alkali corrosion, and a specific surface area. Therefore, MWNTs are widely used in mechanical, electronic, optical, magnetic, chemical, and biological applications. The surface of MWNTs with multiple π electrons is very hydrophobic and can easily cause grouping, which greatly limits the use of MWNTs. Therefore, in the present study, MWNTs were combined with hydroxyl and carboxyl hydrophilic groups under acidic conditions to enhance their polarity and hydrophilicity. The Fe3O4/acidic MWNTs (AMWNTs) material has the advantages of large surface area and faster diffusion. Fe3O4/AMWNTs show great potential as adsorbents for various organic compounds, such as phthalates, aconitic acids, and (fluoro) quinolones. In the extraction process, even a small amount of Fe3O4/AMWNTs can enhance the separation effect within a short equilibration time.
Pyrethroid insecticides usually contain cyano or halogen groups; therefore, they are strongly electronegative. GC/electron capture detector (ECD) analysis is superior to other chromatographic methods. Thus, the present study aimed to develop a simple and sensitive method to determine pyrethroid content in blood samples by MSPE-GC/ECD analysis using AMWNTs as an adsorbent. Fe3O4/AMWNTs synthesized using low toxicity, low cost, and easy availability of Fe3O4 magnetic materials and acidified MWNTs were shown. To achieve optimal performance, the conditions of the proposed MSPE procedure were systematically studied. The method was fully validated and successfully applied to detect pyrethroids in whole blood samples using MSPE-GC/ECD.
| Experimental Methods|| |
Chemicals and reagents
Ferric chloride (FeCl3•6 H2O; 99%), ferrous chloride (FeCl2•4 H2O; 99%), and ammonium hydroxide (NH3•4 H2O) were purchased from ANPEL Laboratory Technologies, Inc. (Shanghai, China). MWNTs (outside diameter, 10–20 nm; purity >97%) were purchased from Shenzhen Nanometer Gang Co., Ltd. (Shenzhen, China). Deionized water was obtained from a Millipore Milli-Q system used in the study. The organic solvents used (methanol, dichloromethane, acetone, ethyl acetate, chloroform, and n-hexane) were obtained from Tedia. The other solvents were of analytical grade. Allethrin, bifenthrin, fenpropathrin, cypermethrin, fenvalerate, deltamethrin, lambda-cyhalothrin, permethrin, and cyfluthrin (each at a concentration of 1 mg/mL in n-hexane) [Figure 1] were purchased from ANPEL.
We used the following instruments in our study: extraction instrument (IKA MS 3, Germany), pure water meter (Milli-Q Advantage A10, USA), nitrogen blowing instrument (ANPEL), and water bath (ANPEL).
A Varian 3800 GC/ECD was used for pyrethroid analysis. The chromatograph was equipped with a CP Sil-8 capillary column (30 m length × 0.25 mm i.d. × 0.25 mm film thickness). Injections were made in the split mode with a ratio of 20:1. The column temperature program was as follows: initialization at 150°C, held for 2 min, increased at 20°C/min–280°C, and then, held for 20 min. The injector port and ECD were heated to 260°C and 300°C, respectively.
Synthesis of Fe3O4/O-multi-walled carbon nanotubes
First, owing to the hydrophobic property of MWNTs, purification and oxidization procedures were conducted according to a previously reported method. Briefly, 1 g of MWNTs was dispersed in concentrated HNO3(60 mL, 2.2 mol/L) in an ultrasonic bath for 10 min and then stirred for 4 h at 60°C. The obtained MWNTs were filtered and thoroughly rinsed with purified water several times until the effluent was pH neutral. Finally, the collected solid was dried 8 h at 80°C. This procedure removed impurities from the MWNTs and introduced some acidic groups (i.e., the carboxyl and hydroxyl groups) onto the surface (referred to as AMWNTs) to improve dispersion in aqueous solution.
Fe3O4 nanoparticles were attached to the surface of AMWNTs by a chemical coprecipitation method with minor adjustments. Briefly, FeCl3•6 H2O (2.35 g), FeCl2•4 H2O (0.86 g), and AMWNTs (1.0 g) were mixed in 100 mL of deionized water using an ultrasonic bath for 10 min, followed by stirring at 80°C under a nitrogen atmosphere. Next, a 5% ammonia solution (30 mL) was added dropwise, and the mixture was stirred vigorously for 30 min. After cooling the mixture to a temperature range in degree Celsius (°C), the black precipitates were isolated from the supernatant by magnetic decantation and then washed with deionized water several times until the pH of the washing solution approached neutral. The dried product was identified as Fe3O4/AMWNTs.
Drug-free whole blood specimens from 6 de-identified individuals were provided by Huzhou central blood station. For each individual, 100 mL whole blood specimen was collected. In this study, all the tests were replicated with the six blood specimens from different individuals (each individual's specimen for three times). This research project was approved by the local institutional review board in accordance with ethics guidelines.
For the analysis of whole blood samples, an aliquot (1 mL) of blood was transferred into a 5-mL centrifuge tube, followed by the addition of 20 μL of pyrethroid insecticides (10 μg/mL). The spiked concentration of each pyrethroid insecticide was 200 ng/mL. The spiked blood was then dissolved in 3 mL of acetone, and the mixture was shaken for 30 min and centrifuged at 4000 rpm for 15 min. The liquid sample and precipitate were collected separately. Next, 1 mL of acetone was added to the residue, and this step was repeated twice. Combined with the previous acetone solution, the mixture was concentrated under nitrogen to remove acetone and then diluted to 4 mL with deionized water, after which 20 mg of the adsorbent was added to the sample solution, followed by the MSPE procedure under the optimal conditions. An aliquot of organic solvent (2 mL) was added as an eluent, and the solution was subjected to ultrasonication for 10 min. The eluted organic solvent was finally concentrated to 200 μL. An aliquot (1 μL) of the eluate was injected into the GC/ECD for analysis.
| Results and Discussion|| |
Optimization of the extraction process
To achieve the highest pyrethroid extraction efficiency, the adsorption and extraction processes, which may affect the extraction efficiency of Fe3O4/AMWNTs, were studied. The four parameters studied included the salting-out effect, solution pH, solvent desorption, and contact time. This study sought to optimize these factors and find the best adsorption and extraction conditions.
An increase in ionic strength of the solution directly decreases the solubility of pyrethroids, which is favorable for the adsorption of analytes by Fe3O4/AMWNTs. However, an increase in ionic strength also leads to an increase in the viscosity of the solution, which is unfavorable for the adsorption of analytes by Fe3O4/AMWNTs. In this experiment, NaCl was used as the salting-out agent, and the effect of a salting-out agent on extraction efficiency was investigated by adding NaCl at concentrations of 0%–30% (w/v) [Figure 1]. Addition of NaCl at 10% resulted in high peak areas, which slightly decreased after further addition of NaCl. This finding indicated that the increase in salt concentration reduced the solubility of analytes in the sample solution and enhanced their adsorption onto the surface of the sorbent. However, excessive salt concentration led to low signal strength because the viscosity of the aqueous solution increased, thus reducing the molecular mass transfer rate, resulting in low extraction efficiency.
Effect of solution pH
The pH of the sample solution will affect the dissociation equilibrium of the analytes. The nine pyrethroids used in this study had two chiral centers on their cyclopropyl ring, connected to a cyano group with an α-carbon atom, thereby yielding four diastereoisomers. Pyrethroid insecticides easily isomerize in polar solvents due to alpha-proton exchange, which causes them to be stable in acidic solutions but decomposed in alkaline media. At higher pH values, these structures may be positively charged, but for MWNTs, the negative carboxyl group may cause complete adsorption of the positively charged pyrethroid. In addition, pyrethroid insecticides are converted into a neutral form under weak acidic conditions, which promotes the formation of hydrogen bonds, resulting in π–π interactions between pyrethroids and AMWNTs due to the polar interactions. Thus, the adsorption of pyrethroids by Fe3O4/AMWNTs was improved. In this study, the adsorption effects of Fe3O4/AMWNTs at pH values of 2, 4, 6, and 9 were evaluated [Figure 2]. According to the chromatographic response, the adsorption of pyrethroids to Fe3O4/AMWNTs was best at pH 4.0.
Effect of extraction time
MSPE is an equilibration technique that extends adsorption time to help analyze the interactions between particles until they reach equilibrium. In this study, to achieve maximum desorption of the target compound, the effects of different desorption time (5–20 min) were studied [Figure 3]. The adsorption of the nine pyrethroids by Fe3O4/AMWNTs did not change significantly after 10 min. The results showed that the peak areas of the nine pyrethroid insecticides did not change significantly within 10–20 min. To save time cost, an analytical time of 10 min was selected.
Choice of desorption solvent
Pyrethroid insecticides usually contain cyano or halogen groups; thus, they are strongly electronegative. Strongly polar organic solvents have better extraction effects. Ethyl acetate, dichloromethane, ethyl acetate/n-hexane (1:1, v/v), and ethyl acetate/chloroform (2:1, v/v) were used as elution solvents. The experimental results [Figure 4] showed that for the nine pyrethroids, the extraction efficiencies of four desorption solvents could be ranked from best to worst as follows: ethyl acetate, ethyl acetate/n-hexane (1:1, v/v), dichloromethane, and ethyl acetate/chloroform (2:1, v/v). Thus, ethyl acetate/chloroform (2:1, v/v) was selected as the extraction solvent. Chloroform alone is not conducive to allow separation of magnetic nanomaterials. Therefore, a good separation effect can be achieved by adding ethyl acetate to chloroform.
The optimal extraction conditions were as follows:first, 1 mL of blood was transferred to a 5-mL centrifuge tube. Next, 3 mL of acetone was added, and the mixture was oscillated for 30 min and centrifuged for 15 min at 4000 rpm. After removing the supernatant, 1 mL of acetone was added to the residue and the process was repeated twice. The mixture was combined with the previous acetone solution and then concentrated to remove the acetone. Next, 0.2 mL of the solution was diluted to 5 mL in a pH 4.0 buffer solution, and 5 g NaCl (final concentration of 10%) was added to the aqueous solution. Afterward, 20 mg of the absorbent was added to the solution and mixed using a digital vortex at 3000 rpm for 10 min. The sample solution was separated from the adsorbent by applying an external magnetic field. Next, the target magnetic nanoparticles were mixed with 2 mL of methyl acetate/chloroform (2:1, v/v) solution. The mixture was eluted for 10 min, allowing separation of organic solvent, and then concentrated to 200 L.
Validation of quantification
The ability of Fe3O4/AMWNTs to adsorb pyrethroids in the complex samples was quantitatively examined. The linear range and correlation coefficient of the nine pyrethroids were determined by a combination of MSPE-GC/ECD [Table 1]. The responses of the nine pyrethroids were linear in the range of 0.05–5 μg/mL, with an R2 value higher than 0.9908.
The proposed GC method was established to determine pyrethroids in the blood after adsorption by Fe3O4/AMWNTs. No pyrethroid pesticide was found in normal human blood samples. [Figure 5] shows the chromatograms of blood samples spiked with pyrethroid standards and those of unspiked blood samples. All nine pyrethroids showed a good chromatographic response after Fe3O4/AMWNTs extraction. Owing to the complexity of the blood sample matrix, proteins in the blood may cause significant matrix effects during biological sample pretreatment. Thus, a deproteinization step was added, and the recovery was measured by adding a spiked sample. Blood samples were collected from six healthy volunteers; three tests were conducted for each blood sample. Twenty microliters of pyrethroid (10 μg/mL) was added to each blood sample (1 mL) to allow good equilibration. The concentration of each pyrethroid insecticide was 200 ng/mL. The recovery rate of each blood sample was determined [Table 2]. The average recovery was 77.5%–88.4% and relative standard deviation (RSD) <7.0% [Table 3].
|Figure 5: Chromatographic map of blank blood and the addition of standard substances 1. Allethrin 2. Bifenthrin 3. Fenpropathrin 4. lambda-cyhalothrin 5. Permethrin 6. Cyfluthrin 7. Cypermethrin 8. Fenvalerate 9. Deltamethrin|
Click here to view
Comparison with other reported methods
This study also compared the MSPE method with LLE and SPE methods. The results showed that the limits of detection (LODs) of the MSPE method were 2.9–18.6 ng/mL and its recoveries ranged from 77.5% to 88.4%. In comparison, the LODs of SPE were between 3.6 and 25.5 ng/mL, and its recovery was in the range of 60.0%–88.6%. The LODs of LLE were between 3.2 and 20.8 ng/mL, and its recovery was between 62.7% and 89.5%. The experimental MSPE method discussed in this study had advantages over SPE and LLE methods, such as operational simplicity, low volumes of organic solvents, low sample requirement, low cost, and high sensitivity.
| Conclusions|| |
This study presented an analytical method using MSPE-GC/ECD to detect pyrethroids in the blood. The method used Fe3O4/AMWNTs as an adsorbent, and the adsorption conditions were optimized to improve the analytical efficiency of the method. The LODs for this method were 2.9–18.6 ng/mL, which showed its superior sensitivity to those of the methods presented in other reports. Our method had a linearity range of 0.05–5 μg/mL, with an R2 higher than 0.9908. Sample recoveries after removal of blood proteins were 77.5%–88.4%. The method exhibited high precision and accuracy (<7.0% RSD). All the test were replicated with six whole blood samples from different individuals, and consisted results were achieved. In contrast to the existing methods of pyrethroid insecticide detection in the blood, our method used magnetic nanomaterials to facilitate connection between a laparoscopy and a GC, which was shown to be convenient, efficient, sensitive, and environmentally friendly. We believe that this method can be used in the field of criminal technology.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Huang K, Li H. Quantitative analysis of pyrethroid insecticides in water samples by using SPE-GC/MS. Forensic Sci Technol 2001;1:30-1.
Zhang XH, Yan LS, Tang ML, Luo FZ, Ma WQ. Determination of four pyrethroid in water by gas chromatography. J Wuhan Institute Technol 2011;11:35-7.
Wan YW, Xiao W, Huang HW. Simultaneous determination of residues of pyrethroid insecticides in aquatic products by gas chromatograohy. Food Machinery 2010;26:68-70.
He H. Rapid determination of organochlorine, pyrethroid and organophosphorus pesticide residues in food by gas chromatography. J Food Saf Quality 2016;03:1063-8.
Ma Z. Determination of five pyrethroid in potable water by gas chromatography. Fujian Anal Testing 2007;16:61-4.
Liu HB, Zhao XF, Shi GY, Shao QS, Li MV. Determination of five pyrethroid pesticides residues in Anoectochilus roxburghii
by cloud point extraction-back extraction and GC-MS. Zhongguo Zhong Yao Za Zhi 2014;39:2859-62.
Shao H. Determination of the residue of the organic phosphorus and pyrethroid pesticide in onion by GC-MS. J Anhui Agric Sci 2011;39:3327-9.
Liu ZA, Gong QY, Xie ZH. Determination of pyrethroid multi-residues in vegetables by high-performance liquid chromatography. J Fuzhou Univ (Natl Sci Ed) 2008;36:122-5.
Weng S, You Y. SPE-HPLC determination of pyrethroid pesticide residues in Platycodon grandiflorum
. Chin J Pharm Anal 2011;31:1818-23.
Xiao-Li MU, Jiang LM, Wen DU. Simultaneous determination of pyrethroids insecticides in tobacco by LC-MS/MS method. Agrochemicals 2009;48:365-7.
Jin YE, Da-Sheng LU, Wen YM, Wang GQ. Determination of pyrethroid pesticide metabolites in human urine by liquid chromatography-mass spectrometry. Shanghai J Prev Med 2012;24:289-92.
Qin L, Wei J, Chen CJ, Liao GH, Duan TT. Advances in analysis technology of pyrethroids insecticides residues in Chinese herbal medicine. Guizhou Agric Sci 2014;42:71-5.
Yu WS, Wang XG, Yu HW, Rao XS. Determination of 33 kinds of Pesticide residues in food by liquid chromatography-mass spectrometry. Occup Health 2011;27:512-4.
Li X, Zhao M, Li S, Chen H, Shen J. Rapid determination of pyrethroids in tomatoes using gas chromatography combined with dispersive liquid-liquid microextraction. Se Pu 2012;30:926-30.
Zheng X. Determination of pyrethroid pesticide residues in vegetable and fruit by microwave-assisted liquid-liquid microextraction and gas chromatography-mass spectrometry. Chin J Anal Chem 2004;32:225-8.
Ma XW, Lun XW, Zheng X, Bi H, Hou XH. Analysis of pyrethroid insecticides in water samples by dispersive liquid-liquid microextraction based on solidification of floating organic droplet. Agrochemicals 2013;52:110-3.
Liu YB, Jia LH, Niu SY. Rapid determination of 44 organochlorine and pyrethroid residues in vegetables and fruits by solid phase extraction-gas chromatography-mass spectrometry. Chin J Anal Chem 2005;33:290.
Xu Y, Wang ZQ, Qiao YQ, Wang YP, Wang DM. Determination of multi-component pesticide residues in fruits and vegetables by automatic solid phase extraction and gas chromatography tandem mass spectrometry. Pract Prev Med 2017;24:117-9.
Wu Y, Jiang B, Zhang XB, Zhao W, Pan ZL, Li LL. Determination of 31 insecticides residual in cereals by SPE- multidimensional gas chromatography-tandem mass spectrum. Environ Chem 2016;1:225-8.
Zhu J, Yang X, Feng JF, Shi ZX, Zhao YF, Zhou YM. Headspace solid-phase microextraction combined with gas chromatography mass spectrometry for determination of organochlorine and pyrethroid insecticides in milk. Chin J Food Hyg 2007;19:289-93.
Yuan N, Yu B, Zhang M, Zeng J, Chen X. Simultaneous determination of residues of organochlorine and pyrethroid pesticides in tea by microwave assisted extraction solid phase microextraction-gas chromatography. Se Pu 2006;24:636-40.
He S, Huang Z, Zhu Y. Recent advances in preparation of novel chromatographic stationary phases with carbon nanotubes. Se Pu 2013;31:1146-53.
Zheng CH, Meng FG, Mo XR, Zhao DY, Yang MM. The application of electrospunpolymer nanofibers in determination of five pyrethroid insecticides by gas chromatography. Chin J Anal Lab 2010;29:80-4.
Li CX, Tan W, Mei LY, Hu XX, Wang HB, Yang M. GC Determination of residual amount of fenpropathrin in vegetable with MWCNTs-solid phase extraction. Phys Test Chem Anal 2013;49:709-12.
Liu XL, Guo ZW, Huang LL, Feng GJ, Wang YH, Zeng DQ. Detection of 23 insecticides in vegetables by multi-walled carbon nanotubes dispersive solid phase extraction combined with gas chromatography. J Southern Agric 2016;47:401-6.
Asensio-Ramos M, Hernández-Borges J, Ravelo-Pérez LM, Rodríguez-Delgado MA. Simultaneous determination of seven pesticides in waters using multi-walled carbon nanotube SPE and NACE. Electrophoresis 2008;29:4412-21.
Wang Y, Zhi XX, Zhang LJ. Determination of polychlorinated biphenyls and organochlorine insecticides in water by GC with multiwalled carbon nanotubes as a solid-phase extraction adsorbent. J Instrumental Anal 2008;27:493-6.
Luo YB, Yu QW, Yuan BF, Feng YQ. Fast microextraction of phthalate acid esters from beverage, environmental water and perfume samples by magnetic multi-walled carbon nanotubes. Talanta 2012;90:123-31.
Zhang HF, Shi YP. Preparation of fe3O4 nanoparticle enclosure hydroxylated multi-walled carbon nanotubes for the determination of aconitines in human serum samples. Anal Chim Acta 2012;724:54-60.
Morales-Cid G, Fekete A, Simonet BM, Lehmann R, Cárdenas S, Zhang X, et al
. In situ
synthesis of magnetic multiwalled carbon nanotube composites for the clean-up of (fluoro) quinolones from human plasma prior to ultrahigh pressure liquid chromatography analysis. Anal Chem 2010;82:2743-52.
Du Z, Zhang S, Zhou C, Liu M, Li G. L-histidine functionalized multi-walled carbon nanotubes for on-line affinity separation and purification of immunoglobulin G in serum. Talanta 2012;99:40-9.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]
[Table 1], [Table 2], [Table 3]