This journal is©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016 New J. Chem., 2016, 40, 9021--9024 | 9021

Cite this: NewJ.Chem., 2016,

40, 9021

A fluorescent probe for a lewisite simulant†

Doo-Hee Lee, Dong-Nam Lee and Jong-In Hong*

Herein, we report for the first time a highly sensitive fluorescent

probe for the detection of an organoarsenic blister agent simulant,

arsenic trichloride (AsCl3). This probe showed high selectivity to

AsCl3, in the presence of other metal ions, even at low concentra-

tions. Moreover, quantitative determination of AsCl3 in soil was

achieved with a detection limit of 61.2 lmol kg�1 that is sufficiently

lower than the reported LD50 of lewisite.

Chemical warfare agents (CWAs) are classified as weapons ofmass destruction,1 because of their potential to cause indis-criminate damage with severe psychological and physiologicaleffects. Although the military use of CWAs has been banned bythe Chemical Weapons Convention,2 its terroristic use is still athreat to the society.3,4 Accordingly, various detection methodsfor CWAs have been developed based on ion-mobility spectros-copy,5 mass spectrometry,6 electrochemistry,7 and other tech-niques.8–10 Of late, fluorescence-based detection methods arefound to be attractive due to their simplicity, high sensitivity,cost-effectiveness, and fast response, thereby complementing theprevious reports. Despite the recent advancements, fluorescence-based detection methods for CWAs have been focused on nerveagents11–13 and other sulfur/nitrogen mustard agents.14,15 In thiscontext, there is renewed interest in other categories of CWAssuch as organoarsenic compounds.

Although there are several methods for the detection ofarsenic compounds,16 new methods for fast and simple trackingare still desired for environmental monitoring.17 Lewisite is anorganoarsenic compound manufactured for use as a chemicalweapon, and it acts as a vesicant and a lung irritant. Herein, wereport for the first time a fluorescent probe which allows fora highly sensitive and selective detection of an organoarsenicblister agent, a lewisite simulant.

Lewisite (2-chloroethenylarsonous dichloride) has an arseniccore that is a thiophilic metalloid.18 Thiol-containing biomoleculessuch as cysteine, glutathione, and lipoic acid are known to stronglybind arsenic,19 resulting in impaired gluconeogenesis and oxidativephosphorylation.20 Lewisite exposure at low doses causes blistersand acute skin burns in humans. High-dose exposure can lead tosevere irreversible issues such as liver necrosis, renal failure,and even the lethal ‘‘Lewisite shock’’.21,22 During World War II,2,3-dimercaptopropanol (British Anti-Lewisite, BAL) was exploitedas an antidote for lewisite.23 BAL contains two thiol groups thatstrongly bind to the arsenic core, which attenuates the toxicity.Inspired by the thiophilic characteristics of arsenic species, wehave prepared a thiol containing 7-hydroxycoumarin for thefluorescence detection of a lewisite simulant, AsCl3 (Scheme 1).24

Probe 1 was synthesized in a four-step procedure with anoverall yield of 7.3% (Scheme S1, ESI†). Compound 6 was preparedby the Duff reaction,25 and was subjected to reductive aminationwith 7 to furnish compound 8 in 42% yield.26 Removal of thetrityl groups of 8 with a mixture of trifluoroacetic acid andtriisopropylsilane was performed to provide compound 1. Inaddition, probe 2 (cyclic disulfide of 1) was prepared by the airoxidation of 1 for easy handling, because dithiols are relativelyunstable and can spontaneously convert into the correspondingoxidized form.27

The performance of probe 1 in response to AsCl3 was con-firmed by mass spectrometry. Probe 2 was pretreated with tris-(2-carboxyethyl)phosphine (TCEP) to obtain its reduced form,

Scheme 1 Proposed mechanism for lewisite simulant detection.

Department of Chemistry, College of Natural Sciences, Seoul National University,

Seoul 08826, Republic of Korea. E-mail: jihong@snu.ac.kr

† Electronic supplementary information (ESI) available: Synthesis of probes 1 and 2,characterization, and detailed procedures. See DOI: 10.1039/c6nj01658h

Received (in Montpellier, France)26th May 2016,Accepted 21st September 2016

DOI: 10.1039/c6nj01658h

www.rsc.org/njc

NJC

LETTER

Publ

ishe

d on

22

Sept

embe

r 20

16. D

ownl

oade

d by

Seo

ul N

atio

nal U

nive

rsity

on

04/0

8/20

17 0

7:06

:22.

View Article OnlineView Journal | View Issue

9022 | New J. Chem., 2016, 40, 9021--9024 This journal is©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016

probe 1.28 The mass spectrum obtained upon the additionof TCEP to probe 2 showed a peak at m/z 312.0 (calcd for[C14H18O3NS2]+ 312.07), which indicated the presence of probe1 (Fig. S7, ESI†). After AsCl3 addition, a new peak appeared atm/z 383.9, indicating the formation of complex 1–As (calcd for[C14H15O3NS2As]+ 383.97), in a molar ratio of 1 : 1 (Fig. S8, ESI†).

The photophysical properties of probe 2 in aqueous solution(10 mM) upon the addition of AsCl3 were monitored by UV-Visand fluorescence spectroscopy in the presence of TCEP (12 mM).Earlier reports suggest that TCEP (12 mM) reduces probe 2(10 mM) rapidly and stoichiometrically to generate probe 1(10 mM).28 Probe 1 shows an absorption maximum centeredat 320 nm. The addition of AsCl3 induces a 7 nm hypsochromicshift, showing the isosbestic point at 332 nm (Fig. 1a). Theintrinsic fluorescence of probe 1 (lmax = 445 nm, FF = 0.62)29

linearly decreased upon the addition of AsCl3 (0–10 mM) and wascompletely quenched by 10 equiv. of AsCl3 (Fig. 1b). However,there were no significant responses upon the addition ofaqueous solution of various metal ions, Hg2+, Ag+, Cu2+, Fe2+,Zn2+, Pb2+, and Cd2+ (10 equiv. each), demonstrating selectivesensing behaviour toward AsCl3. It is noteworthy that probe 1showed a selective response for AsCl3 over other thiophilicmetal ions such as Hg2+, Cu2+, and Ag+, as these are the majorcompetitors with As3+ in real field samples. Subsequent addi-tion of AsCl3 to each metal ion solution resulted in a quenchingfactor (F0/F) of 16.5–19.7, which was comparable with the valueobtained for probe 1 in AsCl3 solution (F0/F = 18.3). Thus, theresults obtained can be considered to indicate a competitiveresponse toward AsCl3, even in the presence of other metal ions(Fig. 2a). Furthermore, the quenching response of probe 1towards AsCl3 allows for the highly sensitive naked-eye detec-tion, while there were no such noticeable changes with othermetal ions (Fig. 2b). The limit of detection (LOD) estimated by the3s/slope was determined to be 1.34 � 10�6 M (1.34 mmol kg�1),showing a linear range upon addition of 0–10 mM AsCl3, with anR2 value of 0.981 (Fig. S2, ESI†). The LOD value demonstratesthe sufficient sensitivity of probe 1 for practical applications,in accordance with the reported LD50 of lewisite (30 mg kg�1,145 mmol kg�1) (Table S1, ESI†).30 It is interesting to note thatthe fluorescence response toward AsCl3 is visible over a broadpH range of 4–9 (Fig. S1, ESI†), and hence can be used for the

detection of CWAs in battle fields under weakly acidic condi-tions (pH 4.8–5.8).31

From the observed results, we propose a fluorescence quench-ing mechanism of 1 toward AsCl3. Thiophilic AsCl3 approachesthe dithiol group of probe 1, and the dithiol and hydroxyl groupssimultaneously react with a nearby AsCl3 to form a coordinationcomplex, releasing 3 equiv. of HCl. The bound As3+ triggersintersystem crossing of singlet excited electrons by the heavyatom effect, quenching the fluorescence of the probe.32 However,HCl, a byproduct of AsCl3 detection, did not have a significantinfluence on the fluorescence change (Fig. S3, ESI†).

Interestingly, probe 2 in the absence of TCEP showed similarresults to probe 1 in regard to AsCl3 recognition (Fig. S5, ESI†).The 1,2,5-dithiazepane group on probe 2 would facilitate proximalarrangement of AsCl3 as a result of the thiophilic characteristics.7-Hydroxycoumarin, a control fluorophore lacking any sulfur-containing functional group, shows negligible fluorescencechanges upon treatment with AsCl3, which demonstratesthat 1,2,5-dithiazepane plays an active role in AsCl3 detection(Fig. S6, ESI†).

1H NMR experiments corroborate the role of 1,2,5-dithiazepanein AsCl3 recognition. Upon the addition of increasing amountsof AsCl3 to the solution of compound 3, the aromatic peaks(6.80–7.24 ppm) gradually shifted to the downfield region, whilethe 1H resonance from the 7-hydroxyl group at 10.4 ppm dis-appeared. The aliphatic proton resonances on 1,2,5-dithiazepanealso shifted from 2.93 and 3.27 ppm to 3.24 and 3.85 ppmaccompanied by broadening of the peaks. These observationssuggest that As3+ is located around the 1,2,5-dithiazepane moietyof compound 3, binding at the phenolic oxygen (Fig. S4, ESI†).

Fig. 1 UV-Vis absorbance (a) and fluorescence spectra (b) of 1 (10 mM)upon addition of AsCl3 (0–10 equiv.) in the co-presence of TCEP (12 mM) inwater. The insets of the panels show the absorption (labs = 360 nm) (a) andemission (lem = 445 nm) (b) intensity of 1 (10 mM) decreasing upon additionof AsCl3 (0–10 equiv.). The fluorescence emission was monitored byexcitation at 370 nm.

Fig. 2 (a) Fluorescence intensity changes of 1 (10 mM) upon addition ofvarious metal ions (10 equiv.) in water (blue bar) and competitive responseupon subsequent addition of AsCl3 (10 equiv.) into each metal ion solution(red bar). The fluorescence intensity was obtained by excitation at 370 nm.(b) Photograph under UV hand lamp irradiation at 365 nm showing fluores-cence quenching upon addition of AsCl3.

Letter NJC

Publ

ishe

d on

22

Sept

embe

r 20

16. D

ownl

oade

d by

Seo

ul N

atio

nal U

nive

rsity

on

04/0

8/20

17 0

7:06

:22.

View Article Online

This journal is©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016 New J. Chem., 2016, 40, 9021--9024 | 9023

Also, another plausible pathway would be as follows: As3+ couldbe first positioned around the 1,2,5-dithiazepane moiety andreduce disulfide to dithiol,33 giving the As-ligated complex afterbinding of the dithiol with As3+.34

Since lewisite, upon release, has deleterious effects on humanhealth and the environment, its detection from samples such assoil is therefore very important.35 To demonstrate the in-fieldutility of probe 2, quenching experiments were conducted usingAsCl3-spiked soil after extraction with water.36 The amount ofspiked AsCl3 was quantitatively analyzed in the range of 0–3.5 mMthrough fluorescence. Most importantly, the LOD, estimated to be6.12 � 10�5 M (61.2 mmol kg�1), was sufficiently low to detectdoses below the LD50 in soil.30 This demonstrates the utility of thepresent probes for actual field tests. The fluorescence intensityof probe 2 decreased upon increasing the amount of the spikedAsCl3, and the corresponding intensity change could be detectedby the naked eye (Fig. 3).

In summary, we have developed for the first time a fluorescentprobe for the lewisite mimic, AsCl3. The dithiol or 1,2,5-dithiazepane group on probe 1 or 2 drives proximal arrangementof the thiophilic AsCl3, thereby resulting in binding of theprobe with As3+ and fluorescence quenching owing to the heavyatom effect. We have also demonstrated the selective and highlysensitive sensing behaviour of the probes toward AsCl3. Moreover,quantitative determination of AsCl3 in soil has been performedat low concentrations, suggesting the potential utility of our

system in real-world applications. The studies herein constitutea fluorescence-based detection for a class of CWAs which havepreviously been conducted only through methods requiringheavy equipment.

Experimental sectionSynthesis of compound 8

To a solution of compound 6 (1.99 g, 10 mmol) in methanol/dichloromethane (100 mL/50 mL), compound 7 (6.22 g, 10 mmol)was added, and then a small amount of acetic acid was added.Sodium cyanoborohydride (0.63 g, 10 mmol) was added drop-wise to the resulting ice-cooled solution under stirring. After themixture was stirred for three days at room temperature, it wasacidified by adding conc. HCl and the resulting solution wasevaporated almost to dryness under reduced pressure. Theresidue was dissolved in a saturated aqueous solution of Na2CO3

and extracted with dichloromethane.26 Combined fractions weredried over anhydrous Na2SO4, and evaporated under reducedpressure to give an amber-colored oil. The residue was furtherpurified on a silica-gel column with hexane and ethyl acetateto give compound 8 in 42.0% yield. 1H NMR (CDCl3, 300 MHz)d = 2.13 (t, 4H), 2.26 (t, 4H), 3.61 (s, 2H), 6.17 (d, 1H), 6.70 (d, 1H),7.20 (m, 31H), 7.62 (d, 1H).

Synthesis of compound 2

Compound 8 (3.34 g, 4.2 mmol) was deprotected by treatmentwith dichloromethane : trifluoroacetic acid : triisopropylsilane(50 : 47.5 : 2.5, v/v/v, 400 mL) for 1 hour. Deprotection solutionwas evaporated under reduced pressure and residual trifluoro-acetic acid was removed by co-evaporation with cyclohexane(3 � 100 mL) and dried in vacuo. The residue was extracted withdichloromethane and dried over anhydrous Na2SO4 and evapo-rated under reduced pressure to give an amber-coloured oil.The residue was air-oxidized in methanol/water with Na2CO3

prior to purification. The resulting residue was further purifiedon a silica-gel column with hexane and dichloromethane to givea white solid in 33% yield. 1H NMR (CDCl3, 300 MHz) d = 2.95(t, 4H), 3.30 (t, 4H), 4.35 (s, 2H), 6.20 (d, 1H), 6.81 (d, 1H), 7.31(d, 1H), 7.62 (d, 1H). 13C NMR (CDCl3, 75 MHz) d = 162.7, 161.0,153.0, 144.3, 128.4, 113.9, 111.7, 111.5, 108.3, 56.6, 53.3, 38.0.HRMS (TOF, MeOH): calculated for C14H16O3NS2

+, [M + H+]+

310.0572, found 310.0578.

Fluorescence & UV-Vis experiments

Probe 2 was dissolved in tetrahydrofuran to afford a stock solu-tion (10 mM), which was then diluted to 10 mM in distilled water.After in situ reduction of 2 with TCEP (12 mM), analyte was addedto the solution, and the photophysical properties were measuredin real time.

Acknowledgements

This work was supported by the NRF grant (No.2015R1A2A1A15055347) funded by the MSIP.

Fig. 3 (a) Fluorescence changes of 2 (10 mM) in response to the spikedconcentration of AsCl3 (0–3.5 mM) in soil. The spiked AsCl3 was extractedinto the aqueous layer, followed by mixing with probe 2, and the fluores-cence intensity of 2 was monitored by excitation at 370 nm. The insetshows the decreasing fluorescence intensity (lem = 445 nm) of 2 (10 mM)in response to the spiked AsCl3. (b) Photograph under UV hand lampirradiation at 365 nm shows fluorescence quenching of 2 in response tothe spiked concentration of AsCl3 (0–3.5 mM).

NJC Letter

Publ

ishe

d on

22

Sept

embe

r 20

16. D

ownl

oade

d by

Seo

ul N

atio

nal U

nive

rsity

on

04/0

8/20

17 0

7:06

:22.

View Article Online

9024 | New J. Chem., 2016, 40, 9021--9024 This journal is©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016

Notes and references

1 Government Printing Office, National strategies to combatweapons of mass destruction, The White House, Washington,DC, 2002.

2 States that have neither signed nor acceded to the ChemicalWeapons Convention: Angola, Egypt, North Korea, South Sudan;see: https://www.opcw.org/about-opcw/non-member-states,accessed 5/2015.

3 H. Morita, N. Yanagisawa, T. Nakajima, M. Shimizu,H. Hirabayashi, H. Okudera, M. Nohara, Y. Midorikawaand S. Mimura, Lancet, 1995, 346, 290–293.

4 L. A. McCauley, G. Rischitelli, W. E. Lambert, M. Lasarev,D. L. Sticker and P. S. Spencer, Int. J. Occup. Environ. Health,2001, 7, 79–89.

5 K. Tuovinen, H. Paakkanen and O. Hanninen, Anal. Chim.Acta, 2001, 440, 151–159.

6 W. E. Steiner, S. J. Klopsch, W. A. English, B. H. Clowers andH. H. Hill, Anal. Chem., 2005, 77, 4792–4799.

7 Y. Lin, F. Lu and J. Wang, J. Electroanal., 2004, 16, 145–149.8 K. Kim, O. G. Tsay, D. A. Atwood and D. G. Churchill, Chem.

Rev., 2011, 111, 5345–5403.9 J. P. Walker and S. A. Asher, Anal. Chem., 2005, 77, 1596–1600.

10 Y. Yang, H.-F. Ji and T. Thundat, J. Am. Chem. Soc., 2003,125, 1124–1125.

11 T. J. Dale and J. Rebek, Jr., J. Am. Chem. Soc., 2006, 128,4500–4501.

12 Y. J. Jang, O. G. Tsay, D. P. Murale, J. A. Jeong, A. Segev andD. G. Churchill, Chem. Commun., 2014, 50, 7531–7534.

13 A. Barba-Bon, A. M. Costero, S. Gil, F. Sancenon andR. Martınez-Manez, Chem. Commun., 2014, 50, 13289–13291.

14 V. Kumar and E. V. Anslyn, J. Am. Chem. Soc., 2013, 135,6338–6344.

15 R. C. Knighton, M. R. Sambrook, J. C. Vincent, S. A. Smith,C. J. Serpell, J. Cookson, M. S. Vickers and P. D. Beer, Chem.Commun., 2013, 49, 2293–2295.

16 M. Leermakers, W. Baeyens, M. De Gieter, B. Smedts, C. Meert,H. C. De Bisschop, R. Morabito and Ph. Quevauviller, TrendsAnal. Chem., 2006, 25, 1–10.

17 The currently accepted upper limit for arsenic in water is10 ppb; United States Environmental Protection AgencyPublication. EPA816-F-01-004, http://www.epa.gov/safewater/arsenic/compliance.html.

18 M. Baglan and S. Atılgan, Chem. Commun., 2013, 49, 5325–5327.19 S. D. Tuorinsky, Medical aspects of chemical warfare, Department

of the Army, Office of the Surgeon General, Borden Institute,US, 2008, pp. 259–332.

20 S. J. S. Flora, J. Biomed. Ther. Sci., 2014, 1, 48–64.21 J. A. Romano, B. J. Lukey and H. Salem, Chemical warfare

agents: chemistry, pharmacology, toxicology, and therapeutics,CRC Press, Boca Raton, 2nd edn, 2008.

22 C. Bismuth, S. W. Borron, F. J. Baud and P. Barriot, Toxicol.Lett., 2004, 149, 11–18.

23 J. A. Vilensky and K. Redman, Spectrochim. Acta, Part A,2003, 41, 378–383.

24 V. C. Ezeh and T. C. Harrop, Inorg. Chem., 2012, 51,1213–1215.

25 K.-S. Lee, H.-J. Kim, G.-H. Kim, I. Shin and J.-I. Hong,Org. Lett., 2008, 10, 49–51.

26 Y. Shimazaki, S. Huth, S. Hirota and O. Yamauchi, Bull.Chem. Soc. Jpn., 2000, 73, 1187–1195.

27 N. Ollivier, J. Dheur, R. Mhidia, A. Blanpain and O. Melnyk,Org. Lett., 2010, 12, 5238–5241.

28 J. C. Han and G. Y. Han, Anal. Biochem., 1994, 220, 5–10.29 M. Fischer and J. Georges, Chem. Phys. Lett., 1996, 260, 115–118.30 S. L. Hoenig, Compendium of Chemical Warfare Agents,

Springer, New York, NY, 2007, pp. 1–46.31 T. Bausinger, E. Bonnaire and J. Preuss, Sci. Total Environ.,

2007, 382, 259–271.32 K. Rurack, Spectrochim. Acta, Part A, 2001, 57, 2161–2195.33 T. R. Kulp, S. E. Hoeft, M. Asao, M. T. Madigan, J. T. Hollibaugh,

J. C. Fisher, J. F. Stolz, C. W. Culbertson, L. G. Miller andR. S. Oremland, Science, 2008, 321, 967–970.

34 We thank an anonymous referee for valuable comments ona plausible mechanism.

35 G. Certini, R. Scalenghe and W. I. Woods, Earth-Sci. Rev.,2013, 127, 1–15.

36 V. Kumar and E. V. Anslyn, Chem. Sci., 2013, 4, 4292–4297.

Letter NJC

Publ

ishe

d on

22

Sept

embe

r 20

16. D

ownl

oade

d by

Seo

ul N

atio

nal U

nive

rsity

on

04/0

8/20

17 0

7:06

:22.

View Article Online