Wireless Hazard Badges to Detect Nerve-Agent Simulants

Article abstract of DOI:10.1002/anie.201604431

Human exposure to hazardous chemicals can have adverse short- and long-term health effects. In this Communication, we have developed a single-use wearable hazard badge that dosimetrically detects diethylchlorophosphate (DCP), a model organophosphorous cholinesterase inhibitor simulant. Improved chemically actuated resonant devices (CARDs) are fabricated in a single step and unambiguously relate changes in chemiresistance to a wireless readout. To provide selective and readily manufacturable sensor elements for this platform, we developed an ionic-liquid-mediated single walled carbon nanotube based chemidosimetric scheme with DCP limits of detection of 28 ppb. As a practical demonstration, an 8 h workday time weighted average equivalent exposure of 10 ppb DCP effects an irreversible change in smartphone readout.

Full text of DOI:10.1002/anie.201604431

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201604431
German Edition: DOI: 10.1002/ange.201604431
Sensors
Wireless Hazard Badges to Detect Nerve-Agent Simulants
Rong Zhu+, Joseph M. Azzarelli+, and Timothy M. Swager*
Abstract: Human exposure to hazardous chemicals can have
adverse short- and long-term health effects. In this Communi-
cation, we have developed a single-use wearable hazard badge
that dosimetrically detects diethylchlorophosphate (DCP),
a model organophosphorous cholinesterase inhibitor simulant.
Improved chemically actuated resonant devices (CARDs) are
fabricated in a single step and unambiguously relate changes in
chemiresistance to a wireless readout. To provide selective and
readily manufacturable sensor elements for this platform, we
developed an ionic-liquid-mediated single walled carbon
nanotube based chemidosimetric scheme with DCP limits of
detection of 28 ppb. As a practical demonstration, an 8 h
Figure 1. a) A wearable, passive, disposable chemical dosimeter
hazard badge can be scanned periodically throughout the day. A fresh
workday time weighted average equivalent exposure of 10 ppb
DCP effects an irreversible change in smartphone readout.
tag would be peeled off from a pack at the beginning of each time
period. b) Quantification of the “dose” of the hazardous chemical
using a smartphone enables facile information collection to a central-
ized database, by means of cloud data transfer/storage. Standardized
decisions can be informed by predefined protective action criteria
(PAC) levels associated with equivalent exposure time weighted
average hazardous chemical concentration.
H
uman exposure to hazardous chemicals in the environ-
ment, in the workplace, or in military contexts remains
a critical human health concern. Furthermore, recent and
continued human exposure to hazardous chemicals in “every-
day” environments has prompted heightened interest in
chemical monitoring by citizens.[1,2] Most toxic gases are not
visually detectable and can be harmful at persistent exposures
below the human olfaction threshold. The ability to quantify
a chemical hazard dose in a temporally correlated fashion
would enable real-time personalized risk assessment (Fig-
ure 1a).
combustible or recyclable and therefore environmentally
friendly, and are smartphone-communicable (Figure 1b).
Passive radio frequency communication devices, such as
near field communication (NFC) tags, meet these require-
ments.[6]
Real-time, personalized situational awareness remains
impractical for all but the most specialized applications:
existing technological solutions are largely limited to colori-
metric reagent tests such as Drꢀger tubes,[3] various electronic
nose technologies,[4] and scaled-down spectroscopy-based
methods.[5] The principal limitations of these techniques are
either cost, reliability, sensitivity, ease-of-use, physical size,
power requirements, or all of the above. As a result, personal
chemical dosimeters have not been broadly implemented. To
address this need, we have developed a low-cost, user-
friendly, passive, reliable chemical hazard badge that collects
actionable health risk data that can be transmitted wirelessly
to cloud data storage.
We targeted diethyl chlorophosphate (DCP) in our
development of a prototype chemical hazard dosimeter.
DCP has been studied as a target analyte for chemical
sensors mainly as a chemical warfare nerve agent (NA)
simulant.[7] It is also a chemical analogue of cholinesterase
inhibiting organophosphate pesticides.[8] For chemical haz-
ards like DCP, time-weighted-average (TWA) permissible
exposure limits are the most relevant parameter in dictating
the necessity and type of protective actions to be taken.
Ideal chemical hazard dosimeters enable the instanta-
neous assessment of a TWA exposure based on temporally
correlated chemical dose information. For instance, the US
Department of Energy has established “protective action
criteria” levels (PAC) for over 3000 hazardous substances,
including DCP.[9] Accordingly, hazard badges could be
designed to activate at dosage levels that induce mild,
transient health effects (PAC-1), irreversible or other serious
health effects that could inhibit the ability to further protect
oneself (PAC-2), or life threatening health effects (PAC-3).
Furthermore, personal exposure information could be
synchronized with a cloud database[6] to expose spatiotem-
poral trends[10] and enable emergency decision making by
a second party in the case of incapacitation of the hazard
badge wearer.
Our concept was to create wearable, battery-free, single-
use chemical hazard badges that cost less than $1, are
[*] Dr. R. Zhu,[+] J. M. Azzarelli,[+] Prof. Dr. T. M. Swager
Department of Chemistry and Institute for Soldier Nanotechnologies
Massachusetts Institute of Technology
Cambridge, MA (USA)
E-mail: tswager@mit.edu
[+] These authors contributed equally to this work. The author order was
determined by coin toss.
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under http://dx.doi.org/10.
1002/anie.201604431.
We have previously demonstrated that two-step conver-
sion of NFC tags into chemically actuated resonant devices
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
(CARDs) can enable semiquantitative, selective detection of
chemical gases with a smartphone.[11] However, ultra-trace
(less than part-per-million) sensing and dosimetric detection
remain challenging with our initial designs.[12] To address
these challenges we report herein key improvements to both
the circuit design as well as new single walled carbon
nanotube (SWCNT)-based chemiresistive dosimetric materi-
als.
In our first generation CARD platform (series-CARD, or
s-CARD), the chemiresistor (Rs) is incorporated in series with
the NFC integrated circuit (IC) in a two-step method. This
method involves the disruption and recompletion of the
circuit (Figure 2a). The raw chemical information that the
chemiresistor collects is converted and wirelessly transmitted,
mainly in the form of device resonant frequency (f0)
amplitude, or gain (in dB).
We hypothesized that incorporation of the chemiresistor
in parallel with the integrated circuit would serve to overcome
the drawbacks associated with s-CARDs. Such a modification
process produced a new type of CARD platform: parallel-
CARD, or p-CARD. For a proof of concept, parallel fixed
resistors ranging from 100 W to 100 kW were used to construct
a series of p-CARDs. Their gain readouts were measured. As
shown in Figure 2b (part (ii)), as Rs increased, the device
underwent a monotonic decrease in gain and proceeded from
nonresonant to resonant. The gain–logRs relationship was
linear from 1 kW through 100 kW (Figure 2b, inset in
part (ii)).
The importance of practicability in the fabrication proce-
dure of a new device structure should not be understated. In
this regard, the p-CARD design is advantageous; it does not
require disruption of the existing radio frequency ID (RFID)
circuit and can be fabricated in a single step. The p-CARD is
created by simple deposition of chemiresistive material
between the leads connecting the IC (Figure 2a). This
nondisruptive modification method not only results in more
consistent device performance but also makes p-CARDs
amenable to inkjet printing and roll-to-roll manufacturing
processes.[13]
With the p-CARD platform in hand, we turned our
attention to the development of a DCP-responsive dosimetric
SWCNT chemiresistor[14] based on the irreversible hydrolysis
of DCP.[15] To enhance the response by accelerating hydrol-
ysis, we targeted SWCNTs in ionic liquids.[16] In addition to
creating solution-phase reactivity at the SWCNT surface,
ionic liquids (IL) have been shown to partially debundle
SWCNTs when the two components are ground together in
the solid state or when a mixture of SWCNTs and an IL are
sonicated together in the presence of cosolvents.[17] Despite
these advantages, SWCNT/ILs are not an established chem-
iresistor platform.[18]
We initially tested the response of p-CARDs fabricated
with SWCNT/IL composites to DCP in nitrogen (N2). We
found that a combination of SWCNT and 1-butyl-3-methyl-
imidazolium chloride (BMIMCl) showed a good, irreversible
response. Previous work in our laboratory has shown that
small-molecule selectors incorporated into chemiresistor
formulations can selectively enhance the resistive response
to gas analytes.[19] By incorporating 2-(2-hydroxy-1,1,1,3,3,3-
hexafluoropropyl)-1-naphthol (HFIPN) as a hydrogen-bond-
ing chelator/catalyst[20] into the mixture, a 3.3-times improve-
ment in response to DCP was realized (Figure 3a).
The irreversible response and significant enhancement
associated with the BMIMCl/HFIPN-based chemiresistor was
consistent with observed hydrolysis kinetics of DCP in
solution (Scheme 1). We found that at room temperature,
DCP undergoes only minor hydrolysis after stirring in CD3CN
for 10 min, even in the presence of excess water (8 equiv) as
monitored by 31P NMR spectroscopy (Scheme 1, condi-
tions ii: for DCP d ꢀ 4.8 ppm). In contrast, when DCP was
added to a mixture of HFIPN and BMIMCl (a minimal
amount of CD3CN was added to obtain a liquid mixture) in
the absence of any additional water, instantaneous hydrolysis
occurred with the trace water present under an ambient
atmosphere (conditions i). Specifically, a significant portion
Figure 2. a) Single-step nondisruptive conversion of a commercial
NFC tag to a p-CARD and comparison to our first-generation design.
To create an s-CARD, a hole is punched removing some of the
aluminum lead material and disrupting the circuit, which is recon-
nected by the chemiresistor. To create a p-CARD, the material is
deposited on the aluminum leads connecting the IC. Note that the
p-CARD photograph has opposite contrast (aluminum metal is bright)
to reveal the deposited chemiresistor. b) Resonance-frequency traces
for s-CARD (i) and p-CARD (ii) with varying Rs. Respective gain values
measured for each Rs (insets in (i) and (ii)).
Although the s-CARD proved successful in selectively
detecting chemically diverse analytes at parts-per-million
(ppm) levels, we noticed that its circuit structure could
introduce physical constraints to the response magnitude
corresponding to a certain change in the sensor resistance
(DGain/DRs). We have systematically examined the gain
readout of a series of s-CARDs with fixed resistors (Rs) in
place of chemiresistors. These resistors ranged from 100 W to
100 kW, which encompassed the typical dynamic range of our
CNT-based chemiresistors (1 kW to 100 kW). As shown in
Figure 2b (part (i)), the device resonated in all cases, and
a non-monotonic change in the gain readout was observed as
Rs increased (Figure 2b, inset in part (i)). This resulted in
a minimal gain difference observed between Rs = 1 kW and
10 kW. More importantly, the non-monotonicity leads to
ambiguous results when the device is operating within a large
dynamic gain range.
2
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
Figure 3. a) When exposed to DCP (1 ppm) for 50 s, p-CARDs fabri-
cated with HFIPN (blue bar, left) had a 3.3-times larger magnitude of
response than those fabricated without HFIPN (red bar, right).
b) Magnitude of response (DGain) as a function of device “age”. Blue
squares represent different devices. The structures of the components
of the optimized system are shown (inset). c) SEM image of a typical
SWCNT-wrapped microcrystal. Scale bar=1 mm.
Figure 4. p-CARD DCP dosimeter performance with varying DCP
concentrations: a) Saturation plots. DCP exposure started at 100 s.
The results are the averages of multiple individual devices tested
(X=5, except for [DCP]=28 ppb (inset) where X=3). Shaded areas
indicate standard deviations. b) Relative resistance changes per
minute. The upper inset shows the expanded portion for
[DCP]=28 ppb. The lower inset shows the exposure-dose-normalized
response determined for p-CARDs at each concentration tested.
Scheme 1. 31P NMR spectroscopy kinetics study of DCP hydrolysis.
fabricated a series of p-CARDs that incorporated the
BMIMCl/HFIPN/SWCNT-based chemiresistor. The individ-
ual device was exposed to a nitrogen flow for 100 s followed
by DCP vapor until its gain readout reached saturation. The
results are summarized in Figure 4a.
was hydrolyzed within a few seconds, indicated by the
emerging signal for diethylphosphoric acid at d ꢀ ꢁ0.5 ppm.
Nearly full conversion was observed within 10 min.
Upon a more thorough investigation, we observed that
there was a maturing process with the SWCNT/BMIMCl
system, accompanied by two coinciding observations: 1) the
p-CARD baseline (gain) drift was effectively zero and 2) the
magnitude of the response to DCP increased substantially as
a function of time, reaching a maximum after about 18 h
(Figure 3b). Visual inspection and optical microscopy con-
firm time-dependent crystallization of BMIMCl at the surface
of p-CARD (Figure S7). Scanning electron microscopy
(SEM) revealed the formation of SWCNT-wrapped micro-
crystal structures (Figure 3c). We hypothesized that such
structures could increase SWCNT surface area and thus lead
to an enhanced response. This is consistent with our
observation that p-CARDs fabricated with 1-butyl-3-methyl-
imidazolium type ILs that are liquids at room temperature
(anion = hexafluorophosphate, bromide, or iodide) did not
demonstrate this behavior. Recent reports suggest solid-
phase ILs hold promise as a tunable class of materials for
a broad array of applications.[21]
Consistent magnitudes of saturation response were
observed (DGain ꢀ ꢁ1.9 dB, or DR/R0 ꢀ 1400%) when the
DCP concentration was greater than 100 parts-per-billion
(ppb), whereas a slightly diminished overall change in gain
was obtained with 28 ppb DCP (DGain ꢀ ꢁ1.5 dB; see Fig-
ure 4a, inset). From the linear region of the saturation plots,
we were able to extract the relative sensor resistance change
per minute as a benchmark for detection sensitivity evalua-
tion (Figure 4b). It was found that Rs increased by over 60%
after exposure to 2.4 ppm DCP for only 1 min. This high
sensitivity and fast response kinetics allowed the detection of
DCP at a concentration as low as 28 ppb and on a practical
time scale (0.4%minꢁ1).
A true test of a dosimeter is based on the response
behavior across all combinations of time and concentration.
Ideally the response of a dosimeter is proportional to the
exposure dose, the product of analyte concentration, and
exposure time ([analyte]·Dt). It was determined across three
orders of concentration magnitude that the exposure-dose-
normalized response (gain) of all p-CARDs tested fell within
a relatively narrow range (Figure 4b, lower inset), indicating
With the optimized system in hand, we next evaluated its
performance toward various concentrations of DCP. We
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
concentration-independent dosimetry of our optimized
system. By taking the average exposure-dose-normalized
response, theATT="tpl=0mm tpr=0mm tptxpt=4.8mm
tptxt=5.2mm tptxb=4.2mm" exposure-dose-dependent rela-
tionship to the change in gain of a p-CARD can be
empirically derived as:
Z
t2
DGain ¼ A
½DCPꢂdt
ð1Þ
t1
Where A = ꢁ(2.5 ꢃ 1.2) ꢁ 10ꢁ4 dBppbꢁ1 minꢁ1 and t is mea-
sured in minutes. This key relationship thus enables unam-
biguous equivalent exposure assessments based on the
relative change in gain. Furthermore, when combined with
knowledge of the power threshold of binary p-CARDs,
preprogrammed p-CARDs that switch on and off after
passing a predefined equivalent exposure threshold can be
easily and predictably fabricated.
Figure 6. Selectivity of p-CARD DCP dosimeter towards interferents.
Change in p-CARD gain as a function of a) exposure time (min) and
b) exposure dose (min ppm).
Next, the NA-simulant-triggered smartphone binary
switch of this system was demonstrated wherein a p-CARD
was repeatedly exposed to DCP vapor (2 ppm) for 50 s
followed by 170 s N2 (Figure 5). The device was designed to
switch on when it had exceeded a PAC-1 threshold. Initially,
the p-CARD was below the smartphoneꢂs on/off threshold at
gain = ꢁ0.225 dB, and thus was unreadable (“off” state,
indicated by green dots). The corresponding frequency-gain
plot of this device (t = 0 s) was nearly flat, demonstrating poor
device resonance within the working frequency region of the
NFC reader (see Figure S9 in the Supporting Information).
As the device was iteratively exposed to DCP: 1) a
consistent irreversible gain readout decrease (R increase) was
observed, as showcased by the staircase-shape plot; 2) the
resonance amplitude of the device increased, as indicated by
the increasing depth of the minima in the frequency–gain
plot; 3) After 3 cycles, corresponding to a PAC-1 TWA of
10 ppb, the deviceꢂs resonance amplitude exceeded the read-
ability threshold and became readable by the smartphone
(state “on”, indicated by red dots).[22]
When it comes to in-field chemical hazard dosimetry,
chemiresistor selectivity to the target analyte over commonly
encountered interferent chemicals is critical for minimizing
the occurrence of false alarms.[23] Toward this end, a series of
potential interferent chemicals were evaluated. Overall, the
responses (DGain/Dt) of our p-CARD DCP dosimeter
toward NA simulants (DCP and diisopropyl fluorophos-
phates (DFP)) at ppb or low-ppm level were at least one order
of magnitude larger than those resulted from the highly
concentrated vapors (about 100–1000 ppm) of the interfer-
ents studied (Figure 6a). This good selectivity can be better
characterized by the exposure-dose-normalized responses
(Figure 6b). Among the interferents tested, ketone, ester, or
hydroxy functional group containing compounds afforded
a detectable sensor resistance increase, presumably through
hydrogen-bonding. Hydrocarbon/halogenated hydrocarbons
were essentially inert to the sensor.[24] Dimethyl methyl-
phosphonate (DMMP), a hydrogen-bond-accepting NA sim-
ulant that does not react covalently under these conditions,
elicited a much smaller response than DCP and DFP at
comparable concentrations, which is consistent with the
irreversible hydrolysis proposed for the response to DCP
and DFP.[25]
In conclusion, we have developed a highly sensitive and
selective disposable wireless dosimetric chemical hazard
badge that can reliably detect NA simulants down to
28 ppb. This was enabled by 1) the invention of a new
wirelessly addressable sensor platform, p-CARD, derived
from commercial NFC tags in a single step, and 2) the
identification of a SWCNT/IL-based chemiresistor that
selectively responds to the target analyte through instanta-
neous irreversible reactions. This badge allows the quantifi-
cation of chemical hazard dose in a temporally correlated
fashion and transmission of that information wirelessly, which
has been demonstrated by the smartphone readability switch
upon repeated exposures. The consistent dosimetric behavior
of this system enables real-time hazard assessment that is
relevant to widely employed chemical hazard regulation
standards.
Figure 5. p-CARD DCP PAC-1 dosimeter. A p-CARD’s gain (left axis,
blue data points) was measured while it was iteratively exposed to
DCP (2 ppm in N2; shaded bars). The equivalent exposure TWA (right
axis, blue line) was calculated, and the p-CARD was read by a smart-
phone once per cycle. Below the PAC-1 threshold, the p-CARD is
unreadable (green dots). Above the PAC-1 threshold, the p-CARD
becomes readable (red dots).
4
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
656; b) K. Kordas, T. Mustonen, G. Toth, H. Jantunen, M.
Acknowledgements
Lajunen, C. Soldano, S. Talapatra, S. Kar, R. Vajtai, P. M.
Ajayan, Small 2006, 2, 1021 – 1025; c) R. A. Potyrailo, A. Burns,
C. Surman, D. J. Lee, E. McGinniss, Analyst 2012, 137, 2777 –
2781; d) W. Lee, H. Koo, J. Sun, J. Noh, K.-S. Kwon, C. Yeom, Y.
Choi, K. Chen, A. Javey, G. Cho, Sci. Rep. 2015, 5, 17707; e) M.
Jung, J. Kim, J. Noh, N. Lim, C. Lim, G. Lee, J. Kim, H. Kang, K.
Jung, A. D. Leonard, et al., IEEE Trans. Electron Devices 2010,
57, 571 – 580.
The authors would like to acknowledge the financial support
of the Chemical and Biological Technologies Department at
the Defense Threat Reduction Agency (DTRA-CB) via
Grant BA12PHM123 in the “Dynamic Multifunctional Mate-
rials for a Second Skin D[MS]2” program, and the Army
Research Office through the Institute for Soldier Nano-
technologies. We thank Marshall Craft (MIT) for building our
VNA trace analysis LabView program and assisting in 3D
printing of a gas sensing enclosure, Dr. Steven Kooi for
assisting in 3D printing of a gas sensing enclosure, Dr. Fei Yan
(MIT) for discussion, and Vera Schroeder (MIT) for 3D
graphic design.
[14] J. F. Fennell, S. F. Liu, J. M. Azzarelli, J. G. Weis, S. Rochat, K. A.
Mirica, J. B. Ravnsbæk, T. M. Swager, Angew. Chem. Int. Ed.
2015, 54, 1266 – 1269; Angew. Chem. 2015, 127, 1282 – 1285.
[15] a) K. Kim, O. G. Tsay, D. A. Atwood, D. G. Churchill, Chem.
Rev. 2011, 111, 5345 – 5403; b) S.-Y. Moon, Y. Liu, J. T. Hupp,
O. K. Farha, Angew. Chem. Int. Ed. 2015, 54, 6795 – 6799; Angew.
Chem. 2015, 127, 6899 – 6903; c) E. Lꢄpez-Maya, C. Montoro,
L. M. Rodrꢃguez-Albelo, S. D. A. Cervantes, A. A. Lozano-
Pꢅrez, J. L. Cenꢃs, E. Barea, J. A. R. Navarro, Angew. Chem.
Int. Ed. 2015, 54, 6790 – 6794; Angew. Chem. 2015, 127, 6894 –
6898; d) H. Sohn, S. Lꢅtant, M. J. Sailor, W. C. Trogler, J. Am.
Chem. Soc. 2000, 122, 5399 – 5400.
Keywords: carbon nanotubes · dosimeter · ionic liquids ·
nerve agents · sensors
[16] a) T. Welton, Chem. Rev. 1999, 99, 2071 – 2083; b) P. Pavez, D.
Millꢆn, C. Cocq, J. G. Santos, F. Nome, New J. Chem. 2015, 39,
1953 – 1959; c) P. Pavez, D. Millꢆn, C. J. I. Morales, E. A. Castro,
C. Lꢄpez A., J. G. Santos, J. Org. Chem. 2013, 78, 9670 – 9676.
[17] a) T. Fukushima, A. Kosaka, Y. Ishimura, T. Yamamoto, T.
Takigawa, N. Ishii, T. Aida, Science 2003, 300, 2072 – 2074; b) T.
Fukushima, T. Aida, Chem. Eur. J. 2007, 13, 5048 – 5058; c) S.
Zhang, Q, Zhang, Y. Zhang, Z. Chen, M. Watanabe, Y. Deng,
Prog. Mater. Sci. 2016, 77, 80 – 124; d) M. Tunckol, J. Durand, P.
Serp, Carbon 2012, 50, 4303 – 4334.
[18] For a graphene–IL-based quartz crystal microbalance gas sensor,
see: a) Q. Ji, I. Honma, S.-M. Paek, M. Akada, J. P. Hill, A. Vinu,
K. Ariga, Angew. Chem. Int. Ed. 2010, 49, 9737 – 9739; Angew.
Chem. 2010, 122, 9931 – 9933; A SWCNT– IL hybrid has been
used as electrode material for voltammetric sensing: b) S. Fan, F.
Xiao, L. Liu, F. Zhao, B. Zeng, Sens. Actuators B 2008, 132, 34 –
39.
[19] K. A. Mirica, J. M. Azzarelli, J. G. Weis, J. M. Schnorr, T. M.
Swager, Proc. Natl. Acad. Sci. USA 2013, 110, E3265 – E3270.
[20] a) F. Wang, T. M. Swager, J. Am. Chem. Soc. 2011, 133, 11181 –
11193; b) J.-P. Bꢅguꢅ, D. Bonnet-Delpon, B. Crousse, Synlett
2004, 18 – 29.
[21] I. M. Warner, B. El-zahab, N. Siraj, Anal. Chem. 2014, 86, 7184 –
7191.
[22] While we demonstrated the binary switch with the commercial
NFC tags that have a threshold around ꢁ0.2 dB, we also devised
a simple method to customize this parameter, which is described
in the Supporting Information.
[23] Y. Seto, M. Kanamori-Kataoka, K. Tsuge, I. Ohsawa, K.
Matsushita, H. Sekiguchi, T. Itoi, K. Iura, Y. Sano, S. Yamashiro,
Sens. Actuators B 2005, 108, 193 – 197.
[24] We noted that ammonia vapor led to sensor resistance decrease,
which is likely due to a deprotonation process.
[25] p-CARDs did not respond significantly to low concentrations of
water vapor (< 10% relative humidity) and the response to DCP
was not significantly affected under such a level of humidity.
However, a continuous flow (0.5–1.5 Lminꢁ1) of moisture vapor
of elevated humidity (30% to 100% of the saturation humidity)
could compromise the sensor. For details see the Supporting
Information.
[1] a) D. Vallero Fundamentals of Air Pollution, 5th ed., Elsevier,
Waltham, 2014, pp. 139 – 378; b) S. Steinle, S. Reis, C. E. Sabel,
Sci. Total Environ. 2013, 443, 184 – 193; c) A. Vale, T. C. Marrs, P.
Rice, Medicine 2012, 91, 106 – 108; d) B. Erickson, Chem. Eng.
News 2016, 94, 30 – 34.
[2] R. P. Pohanish, Sittigꢀs Handbook of Toxic and Hazardous
Chemicals and Carcinogens, 6th ed., Elsevier Science, Burling-
ton, 2011.
[3] Drꢁger-Tubes & CMS-Handbook Soil, Water, and Air Investiga-
tions as Well as Technical Gas Analysis, 16th ed., Drꢀger Safety
AG & Co., Lubeck, 2001.
[4] A. D. Wilson, M. Baietto, Sensors 2009, 9, 5099 – 5148.
[5] R. Sferopoulos, Human Protection and Performance Division,
Commonw. Aust., Victoria, 2008, DSTO-GD-0570.
[6] V. Coskun, B. Ozdenizci, K. Ok, Wirel. Pers. Commun. 2013, 71,
2259 – 2294.
[7] For selected recent examples, see: a) X. Zhou, Y. Zeng, C. Liyan,
X. Wu, J. Yoon, Angew. Chem. Int. Ed. 2016, 55, 4729 – 4733;
Angew. Chem. 2016, 128, 4807 – 4811; b) C. Belger, J. G. Weis, E.
Egap, T. M. Swager, Macromolecules 2015, 48, 7990 – 7994; c) Z.
Lei, Y. Yang, J. Am. Chem. Soc. 2014, 136, 6594 – 6597; d) H. J.
Kim, J. H. Lee, H. Lee, J. H. Lee, J. H. Lee, J. H. Jung, J. S. Kim,
Adv. Funct. Mater. 2011, 21, 4035 – 4040; e) S.-W. Zhang, T. M.
Swager, J. Am. Chem. Soc. 2003, 125, 3420 – 3421.
[8] Toxicology of Organophosphate & Carbamate Compounds,
(Ed.: R. C. Gupta), Elsevier, Burlington, 2006, pp. 103 – 160;
pp. 209 – 217; pp. 599 – 655.
[9] a) Temporary Emergency Exposure Limits for Chemicals: Meth-
ods and Practice, U. S. Dept. Energ., Washington, D.C., 2008,
DOE-HDBK-1046-2008; b) Protective Action Criteria Rev. 28A
Table 2, U.S. Dept. Energ. Subcomittee on Consequence Assess-
ment and Protective Actions (SCAPA).
[10] N. Cressie, C. K. Wilke in Statistics for Spatio-Temporal Data,
Wiley, Hoboken, 2011, pp. 243 – 356.
[11] J. M. Azzarelli, K. A. Mirica, J. B. Ravnsbæk, T. M. Swager,
Proc. Natl. Acad. Sci. USA 2014, 111, 18162 – 18166.
[12] a) S. Goswami, S. Das, K. Aich, RSC Adv. 2015, 5, 28996 – 29001;
b) A. M. Costero, M. Parra, S. Gil, R. Gotor, R. Martꢃnez-
MaÇez, F. Sancenꢄn, S. Royo, Eur. J. Org. Chem. 2012, 4937 –
4946.
[13] a) L. Yang, R. Zhang, D. Staiculescu, C. P. Wong, M. M.
Tentzeris, IEEE Antennas Wirel. Propag. Lett. 2009, 8, 653 –
Received: May 10, 2016
Published online: && &&, &&&&
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Sensors
Invisible dangers: A single-use, wirelessly
addressable, highly sensitive, and selec-
tive chemical hazard badge that dosi-
metrically detects diethyl chlorophos-
phate, a nerve agent simulant, down to
28 ppb has been developed. The device
comprises a wireless sensor platform and
a chemidosimeter system incorporating
single walled carbon nanotubes and ionic
liquids. Exposure to 10 ppb DCP effects
an irreversible switch in smartphone
readout.
R. Zhu, J. M. Azzarelli,
T. M. Swager*
&&&& — &&&&
Wireless Hazard Badges to Detect Nerve-
Agent Simulants
6
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!