Analyst View Article Online
Published on 02 December 2013. Downloaded by UNIVERSITY OF OTAGO on 11/09/2014 12:42:22.
PAPER
Cite this: Analyst, 2014, 139, 807
View Journal | View Issue
Ultrasensitive trace analysis for 2,4,6trinitrotoluene using nano-dumbbell surface-enhanced Raman scattering hot spots† Zhinan Guo,‡ab Joonki Hwang,‡a Bing Zhao,b Jin Hyuk Chung,c Soo Gyeong Cho,c Sung-June Baekd and Jaebum Choo*a We report an ultra-sensitive surface-enhanced Raman scattering (SERS)-based detection system for 2,4,6trinitrotoluene (TNT) using nano-dumbbell structures formed by the electrostatic interaction between positively and negatively charged gold nanoparticles. First, Meisenheimer complexes were produced between TNT and L-cysteine on gold substrates, and 4-mercaptopyridine (4-MPY) labeled gold nanoparticles (positively charged) were allowed to interact with the Meisenheimer complexes through the electrostatic interaction between the negatively charged aromatic ring of the complex molecules and the positively charged nanoparticles. Then, negatively charged gold nanoparticles were added in order to form nano-dumbbells. As a result, many hot junctions were generated by the dumbbell
Received 12th October 2013 Accepted 1st December 2013
structures, and the SERS signals were greatly enhanced. Our experimental results demonstrate that the SERS-based assay system using nano-dumbbells provides an ultra-sensitive approach for the detection
DOI: 10.1039/c3an01931d www.rsc.org/analyst
of TNT explosives. It also shows strong potential for broad application in detecting various explosive materials used for military purposes.
Introduction The fast and sensitive identication of explosive materials at ultra-low concentration levels is a very important issue for homeland security and counter-terrorism applications.1,2 Among various explosive materials, 2,4,6-trinitrotoluene (TNT) is one of the most commonly used explosives in both military and terrorist activities.3,4 It is also known as a serious pollutant since people exposed to TNT over a prolonged period of time tend to experience anemia and abnormal liver function.5 Furthermore, some military testing grounds and wastewater that are contaminated with TNT are difficult and expensive to remedy.6,7 Therefore, it is very important to develop a highly sensitive detection system for monitoring trace amounts of TNT in the environment. Several analytical methods, including gas chromatography,8 high-performance liquid chromatography,9 ion mobility spectrometry,10 neutron activation analysis,11 a
Department of Bionano Engineering, Hanyang University, Ansan 426-791, South Korea. E-mail: [email protected]; Fax: +82-31-436-8188; Tel: +82-31-400-5201
b
State Key Laboratory of Supramolecular Structure and Materials, Jilin University, Changchun 130012, China
c Defense Advanced R&D Institute, Agency for Defense Development, Daejeon 305-152, South Korea d
Department of Electronics Engineering, Chonnam National University, Gwangju 500757, South Korea † Electronic supplementary 10.1039/c3an01931d
information
(ESI)
‡ Joint rst authors.
This journal is © The Royal Society of Chemistry 2014
available.
See
DOI:
chemical sensors,12 and biosensors13 have been employed until now. However, some of the problems associated with these analytical methods such as a long sample preparation time, poor detection limit, and lengthy measurement time, have made these detection systems less attractive. Recently, a surface-enhanced Raman scattering (SERS)based detection technique has been considered as a promising alternative for the sensitive detection of TNT.14–21 Since the solubility of TNT in water is very low, its inherent Raman signals are relatively weak. When the molecules are adsorbed on roughened metal surfaces, however, their SERS signals are greatly amplied due to the electromagnetic and chemical enhancement at the SERS active sites known as “hot junctions”. This enhancement effect shows promise in overcoming the low sensitivity problem inherent in conventional Raman spectroscopy. However, a couple of problems still need to be resolved in order to apply this detection capability as a practical application tool. One issue that needs to be addressed is the development of the nanostructures which can be used for highly sensitive and reproducible SERS measurements, and the other is the modication of TNT molecules in order to improve their metal surface affinity for stronger SERS signal enhancement. Until now, many different types of nanostructures, including polymer-encapsulated gold nanoparticles,14 silver-covered molybdate nanowires,15 silver-covered vanadate nano ribbons,16 alumina nanopore substrates with gold nanoparticle clusters,17 silver nanoparticle-covered carbon nanotube grids,18 gold nanoparticle-covered carbon nanotubes,19 silver-encapsulated
Analyst, 2014, 139, 807–812 | 807
View Article Online
Published on 02 December 2013. Downloaded by UNIVERSITY OF OTAGO on 11/09/2014 12:42:22.
Analyst
magnetic nanospheres,20 functionalized hybrid nickel nanostructures,21 silver nanotube arrays,22 and cysteine-modied silver nanoparticles23,24 have been developed for the selective and highly sensitive detection of TNT. In addition, there have been some efforts to chemically modify TNT molecules into SERS-sensitive active forms because TNT does not possess a chromophore in the visible region and does not strongly adsorb onto metal surfaces. Therefore, TNT was modied into SERS active species, such as TNT azo dye derivatives,25 TNT stilbene derivatives,26 TNT-p-aminobenzenethiol complexes,22 and Meisenheimer complexes,23,24,27 using various chemical methods. By these chemical modications, highly colored TNT derivatives could achieve a functionality that enables a strong interaction with the metal surface. Among the TNT complexes, the Meisenheimer complexes formed by the covalent addition of nucleophiles from amino compounds to an electron-decient aromatic ring of TNT are the most popularly used design for the TNT sensors. Herein, we will report on the design of conceptually new nano-dumbbell structures allowing for the highly sensitive SERS detection of TNT molecules at ultra-low concentration levels. These dumbbell structures are formed by the electrostatic interactions between positively charged gold nanoparticles (PC AuNPs) and negatively charged gold nanoparticles (NC AuNPs). In colloid-based SERS detection, signal enhancement is achieved from the hot junctions induced by particle aggregations. Thus, it is difficult to precisely control the uniform distribution of hot junctions. In our substrate-based nano-dumbbell system, however, hot junctions emerge from the gaps formed by the electrostatic interaction between PC AuNPs and NC AuNPs. Consequently, the appearance of many hot junctions, generated by the dumbbell structures, greatly enhances the SERS signals for the Meisenheimer complexes of the TNT molecules immobilized on gold substrates. The results of our research demonstrate that this SERS-based nano-sensing system using nano-dumbbell structures has strong potential for broad application in detecting various explosive materials.
Materials and methods Materials and reagents The TNT used in this study was provided by the Agency for Defense Development (ADD) in South Korea. It was in a ake form of high purity, which was in accordance with corresponding military specications. Gold(III) chloride trihydrate, sodium citrate, sodium borohydride, L-cysteine, 2-aminoethanethiol and 4-mercaptopyridine (4-MPY) were purchased from Sigma-Aldrich (St. Louis, MO, USA). All of the reagents were used as received without further purication. All of the aqueous solutions were prepared by using ultrapure deionized water (18 MU) obtained from a Milli-Q water purication system (Millipore Corporation, Billerica, MA, USA). Synthesis of positively and negatively charged gold nanoparticles For the formation of the nano-dumbbell structures on substrates, two different types of gold nanoparticles were fabricated. One was the positively charged gold nanoparticles
808 | Analyst, 2014, 139, 807–812
Paper
(PC AuNPs) and the other was the negatively charged gold nanoparticles (NC AuNPs). The PC AuNPs were prepared using the method previously reported by Niidome et al.28 Briey, 400 mL of 213 mM 2-aminoethanethiol was added to 40 mL of 1.42 mM HAuCl4. Aer stirring for 20 min at room temperature, 10 mL of 10 mM NaBH4 was added, and then the mixture was vigorously stirred for 10 min in the dark. At that point, the color of the solution changed from light yellow to red. Aer further mild stirring for 15 min, the sample was stored in the dark at 4 C. In order to prepare a Raman active probe, 4 103 M 4-MPY was added to PC AuNPs, and the mixture was allowed to react for 10 h while stirring. The excess 4-MPY in the solution was removed by centrifugation at 5000 rpm for 20 min. It was identied by the use of the UV/Vis absorption spectra that there was no aggregation of the PC AuNPs aer the entire modication procedure. The NC AuNPs were synthesized by the process developed by Frens.29 Immediately aer 100 mL of 1% HAuCl4 aqueous solution started to boil, 4 mL of 1% trisodium salt dehydrate (Na3 citrate) solution was added. Aer 15 min boiling, the mixed solution was cooled down to room temperature. The diameter of the gold nanoparticles was controlled by adjusting the amount of dropping solution. Formation of Meisenheimer complexes on a gold wafer for highly sensitive SERS measurements A gold-patterned wafer was used for the highly sensitive SERS detection. Initially, a glass substrate was cleaned using surfacecleaning solutions SPM (H2SO4 : H2O2 ¼ 4 : 1) and SC-1 (NH4OH : H2O2 : DI water ¼ 1 : 1 : 5). Titanium metal was sputtered onto this clean glass slide, followed by gold deposition. Then the entire glass substrate was cut into small pieces (3 mm 3 mm). The gold wafer was placed in 20 mL of 1 mM L-cysteine solution for 6 hours. Then the surface was thoroughly washed with deionized water, and dried under a stream of high purity nitrogen gas. Here, L-cysteine was adsorbed on gold substrates by thiol bonds and it forms self-assembled monolayers on the surface.30 Next, the L-cysteine modied gold wafer was immersed in 3 mL of different concentrations of a TNT (1.0 pM–10 mM range) solution (1 : 1 acetone–water) and incubated for 24 h. The surface was then washed with a 1 : 1 acetone– water solution and dried with nitrogen gas. At this point in the process, Meisenheimer complexes were formed on the gold substrates by the chemical interaction of L-cysteine and TNT. This substrate was immersed into a 4-MPY modied PC AuNP solution for 24 h. The 4-MPY modied PC AuNPs were assembled on the Meisenheimer complexes through the electrostatic interaction between the PC AuNPs and the negatively charged Meisenheimer complexes. For the construction of the nanodumbbell hot spots, this PC AuNP-modied gold substrate was put into a freshly prepared NC AuNP solution for 24 h. Finally, this nano-dumbbell-immobilized gold substrate was washed with deionized water and then dried with nitrogen gas. Nanoparticle characterization and SERS measurements A Cary 100 spectrophotometer (Varian, USA) was used to acquire the UV/Vis absorption spectra. High magnication
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 02 December 2013. Downloaded by UNIVERSITY OF OTAGO on 11/09/2014 12:42:22.
Paper
Analyst
Fig. 1 shows the sequential steps for the highly sensitive detection of TNT using nano-dumbbell SERS hot spots. First of all, the gold substrate was modied with L-cysteine solution (Fig. 1(b)). The cysteine molecules were immobilized on the gold substrate through an Au–S covalent bond. Aer the modication of the gold substrate, the cysteine-modied gold substrate was immersed in different concentrations of TNT solution for 24 hours. At this point in the process, TNT molecules could be immobilized on the gold substrate through the chemical interaction with L-cysteine. The Meisenheimer complexes shown in Fig. 1(c) are s-complexes formed by the covalent addition of nucleophiles to a ring carbon atom of the electron-decient aromatic rings. Since TNT is typically electron-decient due to the strong electron-withdrawing effect of three nitro groups, TNT is capable of forming Meisenheimer complexes in the solution phase.24 In addition, Raman reporter (4-MPY)-labeled PC AuNPs were dropped onto the Meisenheimer complexes on the gold substrate as shown in Fig. 1(d). The PC AuNPs could interact with the complexes
through the electrostatic interactions between the negatively charged aromatic ring and the positively charged particle surfaces. Finally, the NC AuNPs were added onto the surface in order to create nano-dumbbell SERS hot spots (Fig. 1(e)). The SERS signals of the 4-MPY were greatly enhanced by the hot spot junctions formed through the electrostatic interaction between the two different types of gold nanoparticles. In order to prepare the SERS active nanoprobes, various concentrations of Raman reporter molecules (4-MPY) were added to the PC AuNPs, and the mixture was allowed to react for 10 h while stirring. Fig. 2 shows the UV/Vis spectra of the gold nanoparticles for various concentrations of 4-MPY in the 2.0 103 to 7.0 103 M range. When the concentration of 4-MPY was 5.0 103 M, its color changed from red to dark purple (Fig. S1†), and a new absorption band appeared at 750 nm. This result indicates that the SERS-active nanoprobes started to aggregate at this concentration. On the basis of our measurements, the optimal 4-MPY concentration was estimated to be 4.0 103 M. For the stable immobilization of the TNT molecules on the gold substrate, the surface was treated with L-cysteine. Due to the very strong chemical binding between the Au and S atoms and the self-assembled monolayer (SAM) formation of L-cysteine, the entire surface could be fully covered by the 30 L-cysteine molecules. For quantitative analysis of TNT, the L-cysteine functionalized gold substrate was immersed into various concentrations of TNT in a 1 : 1 acetone–water solution. During this process, the TNT molecules could be immobilized on the gold substrate through the formation of Meisenheimer complexes. According to previous reports, 98% of TNT forms Meisenheimer complexes through the chemical reaction with L-cysteine. For higher concentration of TNT, most of the L-cysteine molecules can react with TNT molecules, and the entire surface of the gold substrate will be covered with newly formed Meisenheimer complexes. On the other hand, the coverage percentage will be decreased for lower concentrations of TNT.
Fig. 1 Schematic illustration of SERS-based TNT detection using the electrically charged nano-dumbbells: (a) gold substrate, (b) L-cysteinemodified gold substrate, (c) formation of Meisenheimer complexes between L-cysteine and TNT, (d) immobilization of positively charged gold nanoparticles on the gold substrate, and (e) formation of nanodumbbell structures.
Fig. 2 UV/Vis absorption spectra of 4-MPY labeled gold nanoparticle colloids in different 4-MPY concentrations: (a) 2 103 M, (b) 3 103 M, (c) 4 103 M, (d) 5 103 M, (e) 6 103 M, and (f) 7 103 M.
transmission electron micrograph (TEM) images were taken using a JEOL JEM 2100F instrument at an accelerating voltage of 200 kV. Scanning electron microscope (SEM) images were taken using a TESCAN (MIRA3) instrument at an accelerating voltage of 20 kV. Dynamic light scattering (DLS) data of the NPs were obtained using a Nano-ZS90 (Malvern). The SERS signals of each of the gold substrates treated with different TNT concentrations were measured using a Renishaw 2000 Raman microscope system (Renishaw, UK.). A Melles Griot He–Ne laser operating at l ¼ 632.8 nm was used as the excitation source with a laser power of 20 mW. The Rayleigh line was removed from the collected Raman scattering data using a holographic notch lter located in the collection path. The Raman scattering data were collected using a charge-coupled device (CCD) camera at a spectral resolution of 4 cm1.
Results and discussion
This journal is © The Royal Society of Chemistry 2014
Analyst, 2014, 139, 807–812 | 809
View Article Online
Published on 02 December 2013. Downloaded by UNIVERSITY OF OTAGO on 11/09/2014 12:42:22.
Analyst
Fig. 3 Comparison of magnified SEM images for the 109 M TNT solution: (a) single nanoparticles and (b) nano-dumbbells structures, and their corresponding animation images for (c) single nanoparticles and (d) nano-dumbbell structures.
Fig. 4 Comparison of the SERS spectra for the 109 M TNT solution: (a) single nanoparticles and (b) nano-dumbbell structures.
Paper
For SERS measurements, 4-MPY-labeled PC AuNPs were dropped onto the gold surface. At this point, these particles could be immobilized on the surface through the electrostatic interactions between the negatively charged aromatic ring of the Meisenheimer complexes and the positively charged nanoparticles. Consequently, the number of PC AuNPs that could be immobilized on the surface was dependent on the number of Meisenheimer complexes. Several SEM micrographs of the nanoparticle assemblies on the gold substrate were taken in order to understand their immobilization properties. Fig. S2† shows the changes in the gold nanoparticle densities for the different loading conditions. Fig. S2(a)† shows the SEM image of the gold substrate treated with L-cysteine. Fig. S2(b)† displays the SEM image of the gold substrate treated with L-cysteine and 4-MPY-labeled PC AuNPs. In this case, very few nanoparticles were adsorbed onto the surface. Fig. 3(a) shows the magnied SEM image (original image: Fig. S2(c)†) of the gold substrate treated with L-cysteine, 109 M TNT, and 4-MPY-labeled PC AuNPs. The nanoparticles were immobilized through the electrostatic interaction between the nanoparticles and the Meisenheimer complexes, and they were also well dispersed on the substrate. In the next step, NC AuNPs were dropped onto the PC AuNP-immobilized gold substrate in order to create nanodumbbell structures for the SERS signal enhancement. Fig. 3(b) displays their magnied SEM image for a randomly chosen area from Fig. S3(d).† As expected, many of the nano-dumbbell structures, formed by the electrostatic interactions between the PC AuNPs and NC AuNPs, could be identied in this gure. Fig. 3(c) and (d) show their animation images. Fig. 4 shows that the SERS signals for the nano-dumbbell structures are at least 10 times stronger than those for the PC AuNPs only. This result indicates that the very large hot spot junctions emerging from the dumbbell edges between the positively charged nanoparticles and the negatively charged particles make an important contribution to the signal enhancement. Therefore, it is obvious that a stronger Raman enhancement can be achieved when nano-dumbbell structures are used as articial hot spot generators. In addition, Raman
Fig. 5 (a) SERS spectra for decreasing concentrations of TNT: (i) 107 M, (ii) 108 M, (iii) 109 M, (iv) 1010 M, (v) 1011 M, (vi) 1012 M, and (vii) blank (without TNT) as well as (b) the corresponding calibration curve of the SERS signal at 1577 cm1 as a function of the TNT concentration. The error bars indicate standard deviations from five measurements at different spots.
810 | Analyst, 2014, 139, 807–812
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 02 December 2013. Downloaded by UNIVERSITY OF OTAGO on 11/09/2014 12:42:22.
Paper
reporter molecules (4-MPYs in this work) displayed a stronger enhancement than other molecules (TNT and L-cysteine) due to their effective electromagnetic enhancement and the charge transfer effects between the particles and the reporter molecules. Fig. 5(a) displays the SERS spectra of the nano-dumbbell structures for various concentrations of TNT. The concentration of TNT was varied from 107 M to 1012 M and the SERS peak of the 4-MPY centered at 1577 cm1 was used for the quantitative evaluation of TNT. In the absence of TNT, a weak SERS signal was observed. This observation indicates that a small amount of PC AuNPs remained on the gold substrate by non-specic binding even though most of the nanoparticles were removed from the substrate by the washing process. When the TNT solution was added, the SERS signals gradually increased with an increase in the concentration of TNT. The corresponding calibration curve is displayed in Fig. 5(b), where the error bars indicate the standard deviations from ve measurements at different spots. A good linear response was achieved at the given concentration range. This result indicates that our dumbbell structure-based SERS detection technique can be used for highly accurate quantitative evaluation of TNT. The limit of detection (LOD) was determined to be 1012 M with ve standard deviations from the background information. Our experimental data demonstrate that this technique is an excellent analytical tool for the highly sensitive detection of TNT.
Conclusions In the present study, we developed an ultra-sensitive SERSbased TNT detection method using the formation of nanodumbbells between positively and negatively charged gold nanoparticles. Meisenheimer complexes between TNT and cysteine were produced on gold substrates and then 4-MPY labeled PC AuNPs were allowed to interact with the Meisenheimer complexes through the electrostatic interaction between the negatively charged aromatic ring of the complex molecules and the positively charged nanoparticles. For stronger SERS signal enhancement, NC AuNPs were added onto the PC AuNPs immobilized on the gold substrates. As a result, nano-dumbbell structures were formed through the electrostatic interaction between the positively and negatively charged nanoparticles. These dumbbell structures were identied using SEM images and their SERS signals were 10 times stronger than those for the PC AuNPs alone due to the appearance of many hot spots. Our experimental results demonstrate that TNT could be detected accurately at a 1.0 pM level. In particular, the SERS-based detection using nanodumbbell structures provides a promising method for the highly sensitive detection of other chemical analytes with unprecedented advantages.
Acknowledgements This work was supported by the Agency for Defense Development (ADD) in South Korea. This work was also partially
This journal is © The Royal Society of Chemistry 2014
Analyst
supported by the National Research Foundation of Korea (grant numbers R11-2008-0061852 and K20904000004-11A050000410).
Notes and references 1 J. S. Caygill, F. Davis and S. P. J. Higson, Talanta, 2011, 88, 14–29. 2 R. S. Golightly, W. E. Doering and M. J. Natan, ACS Nano, 2009, 3, 2859–2869. 3 K. D. Smith, B. R. McCord, W. A. MacCrehan, K. Mount and W. F. Rowe, J. Forensic Sci., 1999, 44, 789–794. 4 P. T. Charles, B. M. Dingle, S. Ven Bergen, P. R. Gauger, C. H. Patterson and A. W. Kusterbeck, Field Anal. Chem. Technol., 2001, 5, 272–280. 5 J. Hawari, S. Beaudet, A. Halasz, S. Thiboutot and G. Ampleman, Appl. Microbiol. Biotechnol., 2000, 54, 605– 618. 6 J. B. Belden, G. R. Lotufo, C. K. Chambliss, J. C. Fisher, D. R. Johnson, R. E. Boyd and J. G. Sims, Environ. Pollut., 2011, 159, 1363–1368. 7 M. T. Montgomery, R. B. Coffin, T. J. Boyd, J. P. Smith, S. E. Walker and C. L. Osbern, Environ. Pollut., 2011, 159, 3673–3680. 8 M. Nambayah and T. I. Quickenden, Talanta, 2004, 63, 461– 467. 9 E. L. Izake, Forensic Sci. Int., 2010, 202, 1–8. 10 T. M. Swager, Acc. Chem. Res., 1998, 31, 201–207. 11 D. T. McQuade, A. E. Pullen and T. M. Swager, Chem. Rev., 2000, 100, 2537–2574. 12 D. W. Kimmel, G. LeBlanc, M. E. Meschievtz and D. E. Cliffel, Anal. Chem., 2012, 84, 685–707. 13 M. K. Habib, Biosens. Bioelectron., 2007, 23, 1–18. 14 K. Qian, H. Liu, L. Yang and J. Liu, Analyst, 2012, 137, 4644– 4646. 15 L. Yang, L. Ma, G. Chen, J. Liu and J. Q. Tian, Chem.–Eur. J., 2010, 16, 12683–12693. 16 M. W. Shao, L. Lu, H. Wang, S. Wang, M. L. Zhang, D. D. D. Ma and S. T. Lee, Chem. Commun., 2008, 2310– 2312. 17 H. Ko, S. Chang and V. V. Tsukruk, ACS Nano, 2009, 3, 181– 188. 18 Y. Sun, K. Liu, J. Miao, Z. Wang, B. Tian, L. Zhang, Q. Li, S. Fan and K. Jiang, Nano Lett., 2010, 10, 1747–1753. 19 T. Demeritte, R. Kanchanapally, Z. Fan, A. K. Singh, D. Senapati, M. Dubey, E. Zakar and P. C. Ray, Analyst, 2012, 137, 5041–5045. 20 Q. An, P. Zhang, J. M. Li, W. F. Ma, J. Guo, J. Hu and C. C. Wang, Nanoscale, 2012, 4, 5210–5216. 21 P. R. Sajanlal and T. Pradeep, Nanoscale, 2012, 4, 5210– 5216. 22 H. Zhou, Z. Zhang, C. Jiang, G. Guan, K. Zhang, Q. Mei, R. Liu and S. Wang, Anal. Chem., 2012, 84, 685–707. 23 Z. Xu, J. Hao, W. Braida, D. Strickland, F. Li and X. Meng, Langmuir, 2011, 27, 13773–13779. 24 S. S. Dasary, A. K. Singh, D. Senapati, H. Yu and P. C. Ray, J. Am. Chem. Soc., 2009, 131, 13806–13812.
Analyst, 2014, 139, 807–812 | 811
View Article Online
Analyst
28 T. Niidome, K. Nakashima, H. Takahashi and Y. Niidome, Chem. Commun., 2004, 1978–1979. 29 G. Frens, Nature, Phys. Sci., 1973, 241, 20–22. 30 W. Yang, J. J. Gooding and D. B. Hibbert, J. Electroanal. Chem., 2001, 516, 10–16.
Published on 02 December 2013. Downloaded by UNIVERSITY OF OTAGO on 11/09/2014 12:42:22.
25 C. J. McHugh, R. Keir, D. Graham and W. E. Smith, Chem. Commun., 2002, 580–581. 26 C. J. McHugh, A. R. Kennedy, W. E. Smith and D. Graham, Analyst, 2007, 132, 986–988. 27 X. Zhou, H. Liu, L. Yang and J. Liu, Analyst, 2013, 138, 1858–1864.
Paper
812 | Analyst, 2014, 139, 807–812
This journal is © The Royal Society of Chemistry 2014
Published on 02 December 2013. Downloaded by UNIVERSITY OF OTAGO on 11/09/2014 12:42:22.
PAPER
Cite this: Analyst, 2014, 139, 807
View Journal | View Issue
Ultrasensitive trace analysis for 2,4,6trinitrotoluene using nano-dumbbell surface-enhanced Raman scattering hot spots† Zhinan Guo,‡ab Joonki Hwang,‡a Bing Zhao,b Jin Hyuk Chung,c Soo Gyeong Cho,c Sung-June Baekd and Jaebum Choo*a We report an ultra-sensitive surface-enhanced Raman scattering (SERS)-based detection system for 2,4,6trinitrotoluene (TNT) using nano-dumbbell structures formed by the electrostatic interaction between positively and negatively charged gold nanoparticles. First, Meisenheimer complexes were produced between TNT and L-cysteine on gold substrates, and 4-mercaptopyridine (4-MPY) labeled gold nanoparticles (positively charged) were allowed to interact with the Meisenheimer complexes through the electrostatic interaction between the negatively charged aromatic ring of the complex molecules and the positively charged nanoparticles. Then, negatively charged gold nanoparticles were added in order to form nano-dumbbells. As a result, many hot junctions were generated by the dumbbell
Received 12th October 2013 Accepted 1st December 2013
structures, and the SERS signals were greatly enhanced. Our experimental results demonstrate that the SERS-based assay system using nano-dumbbells provides an ultra-sensitive approach for the detection
DOI: 10.1039/c3an01931d www.rsc.org/analyst
of TNT explosives. It also shows strong potential for broad application in detecting various explosive materials used for military purposes.
Introduction The fast and sensitive identication of explosive materials at ultra-low concentration levels is a very important issue for homeland security and counter-terrorism applications.1,2 Among various explosive materials, 2,4,6-trinitrotoluene (TNT) is one of the most commonly used explosives in both military and terrorist activities.3,4 It is also known as a serious pollutant since people exposed to TNT over a prolonged period of time tend to experience anemia and abnormal liver function.5 Furthermore, some military testing grounds and wastewater that are contaminated with TNT are difficult and expensive to remedy.6,7 Therefore, it is very important to develop a highly sensitive detection system for monitoring trace amounts of TNT in the environment. Several analytical methods, including gas chromatography,8 high-performance liquid chromatography,9 ion mobility spectrometry,10 neutron activation analysis,11 a
Department of Bionano Engineering, Hanyang University, Ansan 426-791, South Korea. E-mail: [email protected]; Fax: +82-31-436-8188; Tel: +82-31-400-5201
b
State Key Laboratory of Supramolecular Structure and Materials, Jilin University, Changchun 130012, China
c Defense Advanced R&D Institute, Agency for Defense Development, Daejeon 305-152, South Korea d
Department of Electronics Engineering, Chonnam National University, Gwangju 500757, South Korea † Electronic supplementary 10.1039/c3an01931d
information
(ESI)
‡ Joint rst authors.
This journal is © The Royal Society of Chemistry 2014
available.
See
DOI:
chemical sensors,12 and biosensors13 have been employed until now. However, some of the problems associated with these analytical methods such as a long sample preparation time, poor detection limit, and lengthy measurement time, have made these detection systems less attractive. Recently, a surface-enhanced Raman scattering (SERS)based detection technique has been considered as a promising alternative for the sensitive detection of TNT.14–21 Since the solubility of TNT in water is very low, its inherent Raman signals are relatively weak. When the molecules are adsorbed on roughened metal surfaces, however, their SERS signals are greatly amplied due to the electromagnetic and chemical enhancement at the SERS active sites known as “hot junctions”. This enhancement effect shows promise in overcoming the low sensitivity problem inherent in conventional Raman spectroscopy. However, a couple of problems still need to be resolved in order to apply this detection capability as a practical application tool. One issue that needs to be addressed is the development of the nanostructures which can be used for highly sensitive and reproducible SERS measurements, and the other is the modication of TNT molecules in order to improve their metal surface affinity for stronger SERS signal enhancement. Until now, many different types of nanostructures, including polymer-encapsulated gold nanoparticles,14 silver-covered molybdate nanowires,15 silver-covered vanadate nano ribbons,16 alumina nanopore substrates with gold nanoparticle clusters,17 silver nanoparticle-covered carbon nanotube grids,18 gold nanoparticle-covered carbon nanotubes,19 silver-encapsulated
Analyst, 2014, 139, 807–812 | 807
View Article Online
Published on 02 December 2013. Downloaded by UNIVERSITY OF OTAGO on 11/09/2014 12:42:22.
Analyst
magnetic nanospheres,20 functionalized hybrid nickel nanostructures,21 silver nanotube arrays,22 and cysteine-modied silver nanoparticles23,24 have been developed for the selective and highly sensitive detection of TNT. In addition, there have been some efforts to chemically modify TNT molecules into SERS-sensitive active forms because TNT does not possess a chromophore in the visible region and does not strongly adsorb onto metal surfaces. Therefore, TNT was modied into SERS active species, such as TNT azo dye derivatives,25 TNT stilbene derivatives,26 TNT-p-aminobenzenethiol complexes,22 and Meisenheimer complexes,23,24,27 using various chemical methods. By these chemical modications, highly colored TNT derivatives could achieve a functionality that enables a strong interaction with the metal surface. Among the TNT complexes, the Meisenheimer complexes formed by the covalent addition of nucleophiles from amino compounds to an electron-decient aromatic ring of TNT are the most popularly used design for the TNT sensors. Herein, we will report on the design of conceptually new nano-dumbbell structures allowing for the highly sensitive SERS detection of TNT molecules at ultra-low concentration levels. These dumbbell structures are formed by the electrostatic interactions between positively charged gold nanoparticles (PC AuNPs) and negatively charged gold nanoparticles (NC AuNPs). In colloid-based SERS detection, signal enhancement is achieved from the hot junctions induced by particle aggregations. Thus, it is difficult to precisely control the uniform distribution of hot junctions. In our substrate-based nano-dumbbell system, however, hot junctions emerge from the gaps formed by the electrostatic interaction between PC AuNPs and NC AuNPs. Consequently, the appearance of many hot junctions, generated by the dumbbell structures, greatly enhances the SERS signals for the Meisenheimer complexes of the TNT molecules immobilized on gold substrates. The results of our research demonstrate that this SERS-based nano-sensing system using nano-dumbbell structures has strong potential for broad application in detecting various explosive materials.
Materials and methods Materials and reagents The TNT used in this study was provided by the Agency for Defense Development (ADD) in South Korea. It was in a ake form of high purity, which was in accordance with corresponding military specications. Gold(III) chloride trihydrate, sodium citrate, sodium borohydride, L-cysteine, 2-aminoethanethiol and 4-mercaptopyridine (4-MPY) were purchased from Sigma-Aldrich (St. Louis, MO, USA). All of the reagents were used as received without further purication. All of the aqueous solutions were prepared by using ultrapure deionized water (18 MU) obtained from a Milli-Q water purication system (Millipore Corporation, Billerica, MA, USA). Synthesis of positively and negatively charged gold nanoparticles For the formation of the nano-dumbbell structures on substrates, two different types of gold nanoparticles were fabricated. One was the positively charged gold nanoparticles
808 | Analyst, 2014, 139, 807–812
Paper
(PC AuNPs) and the other was the negatively charged gold nanoparticles (NC AuNPs). The PC AuNPs were prepared using the method previously reported by Niidome et al.28 Briey, 400 mL of 213 mM 2-aminoethanethiol was added to 40 mL of 1.42 mM HAuCl4. Aer stirring for 20 min at room temperature, 10 mL of 10 mM NaBH4 was added, and then the mixture was vigorously stirred for 10 min in the dark. At that point, the color of the solution changed from light yellow to red. Aer further mild stirring for 15 min, the sample was stored in the dark at 4 C. In order to prepare a Raman active probe, 4 103 M 4-MPY was added to PC AuNPs, and the mixture was allowed to react for 10 h while stirring. The excess 4-MPY in the solution was removed by centrifugation at 5000 rpm for 20 min. It was identied by the use of the UV/Vis absorption spectra that there was no aggregation of the PC AuNPs aer the entire modication procedure. The NC AuNPs were synthesized by the process developed by Frens.29 Immediately aer 100 mL of 1% HAuCl4 aqueous solution started to boil, 4 mL of 1% trisodium salt dehydrate (Na3 citrate) solution was added. Aer 15 min boiling, the mixed solution was cooled down to room temperature. The diameter of the gold nanoparticles was controlled by adjusting the amount of dropping solution. Formation of Meisenheimer complexes on a gold wafer for highly sensitive SERS measurements A gold-patterned wafer was used for the highly sensitive SERS detection. Initially, a glass substrate was cleaned using surfacecleaning solutions SPM (H2SO4 : H2O2 ¼ 4 : 1) and SC-1 (NH4OH : H2O2 : DI water ¼ 1 : 1 : 5). Titanium metal was sputtered onto this clean glass slide, followed by gold deposition. Then the entire glass substrate was cut into small pieces (3 mm 3 mm). The gold wafer was placed in 20 mL of 1 mM L-cysteine solution for 6 hours. Then the surface was thoroughly washed with deionized water, and dried under a stream of high purity nitrogen gas. Here, L-cysteine was adsorbed on gold substrates by thiol bonds and it forms self-assembled monolayers on the surface.30 Next, the L-cysteine modied gold wafer was immersed in 3 mL of different concentrations of a TNT (1.0 pM–10 mM range) solution (1 : 1 acetone–water) and incubated for 24 h. The surface was then washed with a 1 : 1 acetone– water solution and dried with nitrogen gas. At this point in the process, Meisenheimer complexes were formed on the gold substrates by the chemical interaction of L-cysteine and TNT. This substrate was immersed into a 4-MPY modied PC AuNP solution for 24 h. The 4-MPY modied PC AuNPs were assembled on the Meisenheimer complexes through the electrostatic interaction between the PC AuNPs and the negatively charged Meisenheimer complexes. For the construction of the nanodumbbell hot spots, this PC AuNP-modied gold substrate was put into a freshly prepared NC AuNP solution for 24 h. Finally, this nano-dumbbell-immobilized gold substrate was washed with deionized water and then dried with nitrogen gas. Nanoparticle characterization and SERS measurements A Cary 100 spectrophotometer (Varian, USA) was used to acquire the UV/Vis absorption spectra. High magnication
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 02 December 2013. Downloaded by UNIVERSITY OF OTAGO on 11/09/2014 12:42:22.
Paper
Analyst
Fig. 1 shows the sequential steps for the highly sensitive detection of TNT using nano-dumbbell SERS hot spots. First of all, the gold substrate was modied with L-cysteine solution (Fig. 1(b)). The cysteine molecules were immobilized on the gold substrate through an Au–S covalent bond. Aer the modication of the gold substrate, the cysteine-modied gold substrate was immersed in different concentrations of TNT solution for 24 hours. At this point in the process, TNT molecules could be immobilized on the gold substrate through the chemical interaction with L-cysteine. The Meisenheimer complexes shown in Fig. 1(c) are s-complexes formed by the covalent addition of nucleophiles to a ring carbon atom of the electron-decient aromatic rings. Since TNT is typically electron-decient due to the strong electron-withdrawing effect of three nitro groups, TNT is capable of forming Meisenheimer complexes in the solution phase.24 In addition, Raman reporter (4-MPY)-labeled PC AuNPs were dropped onto the Meisenheimer complexes on the gold substrate as shown in Fig. 1(d). The PC AuNPs could interact with the complexes
through the electrostatic interactions between the negatively charged aromatic ring and the positively charged particle surfaces. Finally, the NC AuNPs were added onto the surface in order to create nano-dumbbell SERS hot spots (Fig. 1(e)). The SERS signals of the 4-MPY were greatly enhanced by the hot spot junctions formed through the electrostatic interaction between the two different types of gold nanoparticles. In order to prepare the SERS active nanoprobes, various concentrations of Raman reporter molecules (4-MPY) were added to the PC AuNPs, and the mixture was allowed to react for 10 h while stirring. Fig. 2 shows the UV/Vis spectra of the gold nanoparticles for various concentrations of 4-MPY in the 2.0 103 to 7.0 103 M range. When the concentration of 4-MPY was 5.0 103 M, its color changed from red to dark purple (Fig. S1†), and a new absorption band appeared at 750 nm. This result indicates that the SERS-active nanoprobes started to aggregate at this concentration. On the basis of our measurements, the optimal 4-MPY concentration was estimated to be 4.0 103 M. For the stable immobilization of the TNT molecules on the gold substrate, the surface was treated with L-cysteine. Due to the very strong chemical binding between the Au and S atoms and the self-assembled monolayer (SAM) formation of L-cysteine, the entire surface could be fully covered by the 30 L-cysteine molecules. For quantitative analysis of TNT, the L-cysteine functionalized gold substrate was immersed into various concentrations of TNT in a 1 : 1 acetone–water solution. During this process, the TNT molecules could be immobilized on the gold substrate through the formation of Meisenheimer complexes. According to previous reports, 98% of TNT forms Meisenheimer complexes through the chemical reaction with L-cysteine. For higher concentration of TNT, most of the L-cysteine molecules can react with TNT molecules, and the entire surface of the gold substrate will be covered with newly formed Meisenheimer complexes. On the other hand, the coverage percentage will be decreased for lower concentrations of TNT.
Fig. 1 Schematic illustration of SERS-based TNT detection using the electrically charged nano-dumbbells: (a) gold substrate, (b) L-cysteinemodified gold substrate, (c) formation of Meisenheimer complexes between L-cysteine and TNT, (d) immobilization of positively charged gold nanoparticles on the gold substrate, and (e) formation of nanodumbbell structures.
Fig. 2 UV/Vis absorption spectra of 4-MPY labeled gold nanoparticle colloids in different 4-MPY concentrations: (a) 2 103 M, (b) 3 103 M, (c) 4 103 M, (d) 5 103 M, (e) 6 103 M, and (f) 7 103 M.
transmission electron micrograph (TEM) images were taken using a JEOL JEM 2100F instrument at an accelerating voltage of 200 kV. Scanning electron microscope (SEM) images were taken using a TESCAN (MIRA3) instrument at an accelerating voltage of 20 kV. Dynamic light scattering (DLS) data of the NPs were obtained using a Nano-ZS90 (Malvern). The SERS signals of each of the gold substrates treated with different TNT concentrations were measured using a Renishaw 2000 Raman microscope system (Renishaw, UK.). A Melles Griot He–Ne laser operating at l ¼ 632.8 nm was used as the excitation source with a laser power of 20 mW. The Rayleigh line was removed from the collected Raman scattering data using a holographic notch lter located in the collection path. The Raman scattering data were collected using a charge-coupled device (CCD) camera at a spectral resolution of 4 cm1.
Results and discussion
This journal is © The Royal Society of Chemistry 2014
Analyst, 2014, 139, 807–812 | 809
View Article Online
Published on 02 December 2013. Downloaded by UNIVERSITY OF OTAGO on 11/09/2014 12:42:22.
Analyst
Fig. 3 Comparison of magnified SEM images for the 109 M TNT solution: (a) single nanoparticles and (b) nano-dumbbells structures, and their corresponding animation images for (c) single nanoparticles and (d) nano-dumbbell structures.
Fig. 4 Comparison of the SERS spectra for the 109 M TNT solution: (a) single nanoparticles and (b) nano-dumbbell structures.
Paper
For SERS measurements, 4-MPY-labeled PC AuNPs were dropped onto the gold surface. At this point, these particles could be immobilized on the surface through the electrostatic interactions between the negatively charged aromatic ring of the Meisenheimer complexes and the positively charged nanoparticles. Consequently, the number of PC AuNPs that could be immobilized on the surface was dependent on the number of Meisenheimer complexes. Several SEM micrographs of the nanoparticle assemblies on the gold substrate were taken in order to understand their immobilization properties. Fig. S2† shows the changes in the gold nanoparticle densities for the different loading conditions. Fig. S2(a)† shows the SEM image of the gold substrate treated with L-cysteine. Fig. S2(b)† displays the SEM image of the gold substrate treated with L-cysteine and 4-MPY-labeled PC AuNPs. In this case, very few nanoparticles were adsorbed onto the surface. Fig. 3(a) shows the magnied SEM image (original image: Fig. S2(c)†) of the gold substrate treated with L-cysteine, 109 M TNT, and 4-MPY-labeled PC AuNPs. The nanoparticles were immobilized through the electrostatic interaction between the nanoparticles and the Meisenheimer complexes, and they were also well dispersed on the substrate. In the next step, NC AuNPs were dropped onto the PC AuNP-immobilized gold substrate in order to create nanodumbbell structures for the SERS signal enhancement. Fig. 3(b) displays their magnied SEM image for a randomly chosen area from Fig. S3(d).† As expected, many of the nano-dumbbell structures, formed by the electrostatic interactions between the PC AuNPs and NC AuNPs, could be identied in this gure. Fig. 3(c) and (d) show their animation images. Fig. 4 shows that the SERS signals for the nano-dumbbell structures are at least 10 times stronger than those for the PC AuNPs only. This result indicates that the very large hot spot junctions emerging from the dumbbell edges between the positively charged nanoparticles and the negatively charged particles make an important contribution to the signal enhancement. Therefore, it is obvious that a stronger Raman enhancement can be achieved when nano-dumbbell structures are used as articial hot spot generators. In addition, Raman
Fig. 5 (a) SERS spectra for decreasing concentrations of TNT: (i) 107 M, (ii) 108 M, (iii) 109 M, (iv) 1010 M, (v) 1011 M, (vi) 1012 M, and (vii) blank (without TNT) as well as (b) the corresponding calibration curve of the SERS signal at 1577 cm1 as a function of the TNT concentration. The error bars indicate standard deviations from five measurements at different spots.
810 | Analyst, 2014, 139, 807–812
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 02 December 2013. Downloaded by UNIVERSITY OF OTAGO on 11/09/2014 12:42:22.
Paper
reporter molecules (4-MPYs in this work) displayed a stronger enhancement than other molecules (TNT and L-cysteine) due to their effective electromagnetic enhancement and the charge transfer effects between the particles and the reporter molecules. Fig. 5(a) displays the SERS spectra of the nano-dumbbell structures for various concentrations of TNT. The concentration of TNT was varied from 107 M to 1012 M and the SERS peak of the 4-MPY centered at 1577 cm1 was used for the quantitative evaluation of TNT. In the absence of TNT, a weak SERS signal was observed. This observation indicates that a small amount of PC AuNPs remained on the gold substrate by non-specic binding even though most of the nanoparticles were removed from the substrate by the washing process. When the TNT solution was added, the SERS signals gradually increased with an increase in the concentration of TNT. The corresponding calibration curve is displayed in Fig. 5(b), where the error bars indicate the standard deviations from ve measurements at different spots. A good linear response was achieved at the given concentration range. This result indicates that our dumbbell structure-based SERS detection technique can be used for highly accurate quantitative evaluation of TNT. The limit of detection (LOD) was determined to be 1012 M with ve standard deviations from the background information. Our experimental data demonstrate that this technique is an excellent analytical tool for the highly sensitive detection of TNT.
Conclusions In the present study, we developed an ultra-sensitive SERSbased TNT detection method using the formation of nanodumbbells between positively and negatively charged gold nanoparticles. Meisenheimer complexes between TNT and cysteine were produced on gold substrates and then 4-MPY labeled PC AuNPs were allowed to interact with the Meisenheimer complexes through the electrostatic interaction between the negatively charged aromatic ring of the complex molecules and the positively charged nanoparticles. For stronger SERS signal enhancement, NC AuNPs were added onto the PC AuNPs immobilized on the gold substrates. As a result, nano-dumbbell structures were formed through the electrostatic interaction between the positively and negatively charged nanoparticles. These dumbbell structures were identied using SEM images and their SERS signals were 10 times stronger than those for the PC AuNPs alone due to the appearance of many hot spots. Our experimental results demonstrate that TNT could be detected accurately at a 1.0 pM level. In particular, the SERS-based detection using nanodumbbell structures provides a promising method for the highly sensitive detection of other chemical analytes with unprecedented advantages.
Acknowledgements This work was supported by the Agency for Defense Development (ADD) in South Korea. This work was also partially
This journal is © The Royal Society of Chemistry 2014
Analyst
supported by the National Research Foundation of Korea (grant numbers R11-2008-0061852 and K20904000004-11A050000410).
Notes and references 1 J. S. Caygill, F. Davis and S. P. J. Higson, Talanta, 2011, 88, 14–29. 2 R. S. Golightly, W. E. Doering and M. J. Natan, ACS Nano, 2009, 3, 2859–2869. 3 K. D. Smith, B. R. McCord, W. A. MacCrehan, K. Mount and W. F. Rowe, J. Forensic Sci., 1999, 44, 789–794. 4 P. T. Charles, B. M. Dingle, S. Ven Bergen, P. R. Gauger, C. H. Patterson and A. W. Kusterbeck, Field Anal. Chem. Technol., 2001, 5, 272–280. 5 J. Hawari, S. Beaudet, A. Halasz, S. Thiboutot and G. Ampleman, Appl. Microbiol. Biotechnol., 2000, 54, 605– 618. 6 J. B. Belden, G. R. Lotufo, C. K. Chambliss, J. C. Fisher, D. R. Johnson, R. E. Boyd and J. G. Sims, Environ. Pollut., 2011, 159, 1363–1368. 7 M. T. Montgomery, R. B. Coffin, T. J. Boyd, J. P. Smith, S. E. Walker and C. L. Osbern, Environ. Pollut., 2011, 159, 3673–3680. 8 M. Nambayah and T. I. Quickenden, Talanta, 2004, 63, 461– 467. 9 E. L. Izake, Forensic Sci. Int., 2010, 202, 1–8. 10 T. M. Swager, Acc. Chem. Res., 1998, 31, 201–207. 11 D. T. McQuade, A. E. Pullen and T. M. Swager, Chem. Rev., 2000, 100, 2537–2574. 12 D. W. Kimmel, G. LeBlanc, M. E. Meschievtz and D. E. Cliffel, Anal. Chem., 2012, 84, 685–707. 13 M. K. Habib, Biosens. Bioelectron., 2007, 23, 1–18. 14 K. Qian, H. Liu, L. Yang and J. Liu, Analyst, 2012, 137, 4644– 4646. 15 L. Yang, L. Ma, G. Chen, J. Liu and J. Q. Tian, Chem.–Eur. J., 2010, 16, 12683–12693. 16 M. W. Shao, L. Lu, H. Wang, S. Wang, M. L. Zhang, D. D. D. Ma and S. T. Lee, Chem. Commun., 2008, 2310– 2312. 17 H. Ko, S. Chang and V. V. Tsukruk, ACS Nano, 2009, 3, 181– 188. 18 Y. Sun, K. Liu, J. Miao, Z. Wang, B. Tian, L. Zhang, Q. Li, S. Fan and K. Jiang, Nano Lett., 2010, 10, 1747–1753. 19 T. Demeritte, R. Kanchanapally, Z. Fan, A. K. Singh, D. Senapati, M. Dubey, E. Zakar and P. C. Ray, Analyst, 2012, 137, 5041–5045. 20 Q. An, P. Zhang, J. M. Li, W. F. Ma, J. Guo, J. Hu and C. C. Wang, Nanoscale, 2012, 4, 5210–5216. 21 P. R. Sajanlal and T. Pradeep, Nanoscale, 2012, 4, 5210– 5216. 22 H. Zhou, Z. Zhang, C. Jiang, G. Guan, K. Zhang, Q. Mei, R. Liu and S. Wang, Anal. Chem., 2012, 84, 685–707. 23 Z. Xu, J. Hao, W. Braida, D. Strickland, F. Li and X. Meng, Langmuir, 2011, 27, 13773–13779. 24 S. S. Dasary, A. K. Singh, D. Senapati, H. Yu and P. C. Ray, J. Am. Chem. Soc., 2009, 131, 13806–13812.
Analyst, 2014, 139, 807–812 | 811
View Article Online
Analyst
28 T. Niidome, K. Nakashima, H. Takahashi and Y. Niidome, Chem. Commun., 2004, 1978–1979. 29 G. Frens, Nature, Phys. Sci., 1973, 241, 20–22. 30 W. Yang, J. J. Gooding and D. B. Hibbert, J. Electroanal. Chem., 2001, 516, 10–16.
Published on 02 December 2013. Downloaded by UNIVERSITY OF OTAGO on 11/09/2014 12:42:22.
25 C. J. McHugh, R. Keir, D. Graham and W. E. Smith, Chem. Commun., 2002, 580–581. 26 C. J. McHugh, A. R. Kennedy, W. E. Smith and D. Graham, Analyst, 2007, 132, 986–988. 27 X. Zhou, H. Liu, L. Yang and J. Liu, Analyst, 2013, 138, 1858–1864.
Paper
812 | Analyst, 2014, 139, 807–812
This journal is © The Royal Society of Chemistry 2014
