Gas phase detection of chemical warfare agents CWAs with portable Raman

https://doi.org/10.1016/j.jhazmat.2019.121279Get rights and content

Highlights

  • Charge and energy transfer in support of plasmonic nanostructures vital in SERS.

  • Low-cost graphite quenches fluorescence and avoids photo-thermal decomposition.

  • SERS Fingerprint of simulant of CWA with Ag plates on graphite and laser λ = 785 nm.

  • Interactions of DMMP with Ag calculated by DFT matching Raman spectrum.

  • 2.5ppmV gas of simulant CWAs detected with portable Raman equipment.

Abstract

The development of SERS substrates for chemical detection of specific analytes requires appropriate selection of plasmonic metal and the surface where it is deposited. Here we deposited Ag nanoplates on three substrates: i) conventional SiO2/Si wafer, ii) stainless steel mesh and iii) graphite foils. The SERS enhancement of the signal was studied for Rhodamine 6 G (R6 G) as common liquid phase probe molecule. We conducted a comprehensive study with λ = 532, 633 and 785 nm on all the substrates. The best substrate was investigated, at the optimum laser 785 nm, for gas phase detection of dimethyl methyl phosphonate (DMMP), simulant of the G-series nerve agents, at a concentration of 2.5 ppmV (14 mg/m3). The spectral fingerprint was clearly observed; with variations on the relative intensities of SERS Raman bands compared to bulk DMMP in liquid phase reflects the DMMP-Ag interactions. These interactions were simulated by Density Functional Theory (DFT) calculations and the simulated spectra matched with the experimental one. Finally, we were detected the characteristics DMMP fingerprint with hand-held portable equipment. These results open the way for the application of SERS technique on real scenarios where robust, light-weight, miniaturized and simple to use and cost-effective tools are required by first responders.

Introduction

The nerve gas attacks with sarin in Matsumoto (1994) and Tokyo (1995) and in the armed conflict in Syria (Goutha 2013 and Khan Shaykhun 2017) are recent and representative examples of the present world threat spectrum. Chemical Warfare Agents (CWAs), initially developed in the First World War, include vesicant and blister agents such as phosgene (PD) and mustard gas (HD) and nerve agents, as sarin (GB) and venomous agent X (VX). The nerve agents are highly toxic due to the irreversible binding with the nerve sites responsible for acetylcholinesterase breakdown, a necessary process in neurotransmission. Sarin gas presents the highest toxicity with a IDLH (Immediately Dangerous to Life or Health) of 0.1 mg m−3. These CWAs are able to spread out in the atmosphere, creating a toxic scenario in few seconds. The early detection and identification of these agents in gas phase is essential for the safety of first responders and for the efficient evacuation of threatened public spaces.

Surface Enhanced Raman Spectroscopy (SERS) is a technique capable of label-free ultrasensitive vibrational “fingerprinting”, and as such it is recognized as highly interesting method for explosive and chemical threat detection (Hakonen et al., 2015). The technique has also become suitable for on-site detection thanks to the development of portable equipment with adequate spectral resolution (ranging from 7 to 12 cm−1) and low weight (330 to 2450 g) at reasonable prices, ca. 10 thousand euros without spectral library. However there are still several challenges to be overcome before this technique can be implemented as “off-the-shelf” solution for CWA detection.

SERS mainly relies on the enhancement of the Raman signal due to the plasmon resonance of a metallic nanostructure. This effect only occurs when the target molecule is in close contact to the metal, and for this reason an interaction i.e. physical or chemical adsorption is required between the molecule and the metal. The magnitude of the enhancement depends on several parameters including composition, size, morphology, topology, surface distribution, and dielectric environment of the metallic nanostructure on substrate surface. Many efforts have been devoted in SERS field for the development of optimum SERS substrates ranging from fancy shapes of nanoparticles with sharp edges to ordered nanostructures whose the shape and spacing was optimized with simulation and then on fabricated in cleanroom. Other key factors for the highly sensitive detection of molecules include the cross-section of the molecule, its stability, interactions with the SERS substrate and its solid, liquid or gaseous state.

Most of the SERS applications in the open literature refer to the detection in liquid phase and only few reports for gas phase detection at concentrations below 10 ppmV could be found. Van Duyne and coworkers reported the detection of 8 ppmV of benzene thiol on a SERS substrate consisting of well-ordered silica nanospheres coated with a 200 nm thick Ag film (Biggs et al., 2009). A recyclable surface-enhanced Raman scattering (SERS) substrate, consisting of Ag nanorods coated with an ultrathin HfO2 shell (Ag NRs@HfO2), was able to detect, after 40 min exposure, concentrations in the gas phase as low as 20 ppb of the molecule 2-Naphthalenethiol (2-NAT). Then substrate was easily regenerated by heating at 250 °C for 25 s (Ma et al., 2016). The SERS detection of volatile organic compounds (VOCs) biomarkers in human breath, such as acetone in the case of diabetes or hydrogen cyanide in patients with cystic fibrosis, is gaining of importance as non-invasive tool in primary screening diagnosis. The group of Boisen detected 5ppmV of hydrogen cyanide in gas phase (Lauridsen et al., 2015) and later on P. aeruginosa cultures (Lauridsen et al., 2017). The term of plasmonic nose was recently coined for a SERS substrate that consists on a core-shell structure of Ag nanocubes encapsulated in porous metal organic framework MOF, in particular ZIF-8 (Koh et al., 2018). The VOCs adsorbed on the porous layer formed a 3D confinement space for the molecules near the plasmonic structure. In this system 200 ppmV of toluene were detected and the detection limit could be lowered to 50 ppb for 2-nitrotoluene, a molecule with higher cross section. Another recent example of low concentration in gas phase was the detection of 10 ppmV of benzene. In this case the strategy for decreasing the detection limit involved deep cooling of the SERS substrate down to -80 °C, to promote adsorption and condensation of the molecules (Oh et al., 2018). Recently, a 3D porous substrate successfully detected for the first time 0.1 ppmV of NO2 in the gas phase with a hand-held Raman detector (Kim et al., 2018). The high surface area employed consisted of a 3D multilayer structure made of Au coated-Ag nanowires (AgNWs) forming lots of random hot spots in the cross points of the fibers. In our laboratory, the detection of 625 ppbV of DMMP, a surrogate molecule of the G-series nerve agents which are of particular concern due to its extreme toxicity and persistence, was recently reported using a substrate consisting of self-assembled AuNPs coated with a citrate layer that acted as an effective trap for the target molecules (Lafuente et al., 2018a). The above results illustrate the importance of substrate selection for gas phase using SERS. It must provide high electromagnetic enhancement (EF), but also chemical affinity towards the target molecule while avoiding band interference with the functional groups responsible for this affinity.

In this work we selected Ag nanoplates as the plasmonic material to be used in DMMP detection in view of four reasons: i) the optical properties of Ag (real and imaginary parts of the dielectric function), that makes this metal the main candidate to get interesting optical effects, including plasmon resonances (Le Ru and Etchegoin, 2009); ii) the specific interaction of DMMP with Ag through the Psingle bondO bond, as already reported in the first study of DMMP detection in roughened Ag electrodes (Taranenko et al., 1996); iii) the electromagnetic field enhancement at the tips and edges of the nanoplates; and finally, iv) the coincidence of the plasmon resonance band of Ag nanoplates with the excitation laser wavelength used for the Raman measurements, 785 nm. We have compared the response of the synthesized Ag nanoplates on three different substrates, SiO2/Si, stainless steel mesh and graphite foil. The SERS response was analyzed using Rhodamine (R6 G) as probe molecule with lasers of λ = 532, 633 and 785 nm. The same substrates were also studied for the detection of 2.5 ppmV of DMMP in gas phase with a benchtop and portable Raman equipments. The interactions of the DMMP molecule with the Ag surface were simulated using Density Functional Theory (DFT) calculations carried out by Gaussian09 quantum chemistry program (Frisch et al., 2013) to explain the relative intensities variation of the characteristic Raman vibrational modes.

Section snippets

Materials

Silver nitrate (AgNO3, 99.9999%), sodium citrate tribasic dehydrate (>99%), hydrogen peroxide solution (H2O2, 30% wt), potassium bromide (KBr, >99%), sodium borohydride (NaBH4, >99%), poly(diallyl dimethylammonium (PDDA, 20% wt), rhodamine 6 G (99%) and dimethyl methylphosphonate (DMMP, 97%) were purchased from Sigma-Aldrich. All solutions were prepared in distilled water. Acetone and isopropyl alcohol were also purchased from Sigma-Aldrich. DMMP vapours were generated using a calibrated

Characterization of SERS substrates

The synthesis of the Ag nPlates was reproducible and in all the cases the UV–vis spectrum (Fig. 2a) displayed the main broad band at around 700 nm, assigned to the in-plane dipole plasmon resonance. The other two LSPR bands at 460 nm and 331 nm are assigned to the in-plane quadrupole and out-of-plane quadrupole plasmon resonances of triangular nanoplates, respectively (Hou et al., 2015). We observed that during storage at room temperature and in the darkness, the peak intensity progressively

Conclusions

The effect of different substrates (SiO2/Si, stainless steel mesh and graphite foil), where plasmonic silver nanoplates were deposited, on the SERS signal, with three lasers (λ = 532, 633 and 785 nm) has been evaluated with two molecules R6G in liquid and DMMP, 2.5ppmV in gas phase. It has been demonstrated that graphite is the best choice because of efficient quenching of the fluorescent signal (λ = 532 and 633 nm), via charge transfer or energy transfer. This energy transfer is also important

Declaration of Competing Interest

Authors declare that there are no conflicts of interest.

Acknowledgements

Authors are grateful for financial support from the European Union’s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement Nr. 823895; MICINN, Spain (CTQ2013-49068-C2-1-R and CTQ2016-79419-R); Gobierno de Aragón (T57-17R p), Feder 2014-2020 “Construyendo Europa desde Aragón”) and CIBER-BBN (initiative funded by the VI National R&D&i Plan 2008–2011, Iniciativa Ingenio 2010, Consolider Program, CIBER Actions and financed by the Instituto de Salud Carlos

References (25)

  • A. Hakonen et al.

    Explosive and chemical threat detection by surface-enhanced Raman scattering: a review

    Anal. Chim. Acta

    (2015)
  • R.K. Lauridsen et al.

    Towards quantitative SERS detection of hydrogen cyanide at ppb level for human breath analysis

    Sens. Biosensing Res.

    (2015)
  • F.J. Beck et al.

    Resonant SPP modes supported by discrete metal nanoparticles on high-index substrates

    Opt. Express

    (2011)
  • A.D. Becke

    Density-functional thermochemistry. III. The role of exact exchange

    J. Chem. Phys.

    (1993)
  • K.B. Biggs et al.

    Surface-enhanced raman spectroscopy of benzenethiol adsorbed from the gas phase onto silver film over nanosphere surfaces: determination of the sticking probability and detection limit time

    J. Phys. Chem. A

    (2009)
  • J.C.S. Costa et al.

    Surface-enhanced Raman spectroscopy studies of organophosphorous model molecules and pesticides

    Phys. Chem. Chem. Phys.

    (2012)
  • A.J. Frank et al.

    Synthesis of silver nanoprisms with variable size and investigation of their optical properties: a first-year undergraduate experiment exploring plasmonic nanoparticles

    J. Chem. Educ.

    (2010)
  • M.J. Frisch et al.

    Gaussian 09

    (2013)
  • S.K. Ghosh et al.

    Solvent and ligand effects on the localized surface plasmon resonance (LSPR) of gold colloids

    J. Phys. Chem. B

    (2004)
  • H. Hou et al.

    Graphene oxide-supported Ag nanoplates as LSPR tunable and reproducible substrates for SERS applications with optimized sensitivity

    ACS Appl. Mater. Interfaces

    (2015)
  • X. Jiang et al.

    Thiol-frozen shape evolution of triangular silver nanoplates

    Langmuir

    (2007)
  • S. Kim et al.

    Highly sensitive and on-site NO 2 SERS sensors operated under ambient conditions

    Analyst

    (2018)
  • Cited by (0)

    View full text