The benzil-cyanide reaction and its application to the development of a selective cyanide anion indicator

Article abstract of DOI:10.1021/ja8039025

The benzil-cyanide reaction is a cyanide-specific reaction that has been exploited to produce a colorimetric indicator for this toxic anion. This was done by producing a π-extended analogue of benzil, 7, which is soluble in a 70:30 (v/v) mixture of methanol-water. In this medium, dilute solutions of 7 are yellow but produce colorless products when exposed to low concentrations of cyanide anion (≥1.7 μM; added as an aqueous NaCN solution), but no other common anions (e.g., OH, F , N₃ , benzoate , and H₂PO ₄^-). On the basis of these observations and supporting mechanistic analyses, it is concluded that the modified benzil system 7 is a promising cyanide anion indicator that is attractive in terms of its selectivity, ease-of-use, water compatibility, and the low, naked-eye discernible cyanide detection limit it provides.

Full text of DOI:10.1021/ja8039025

Published on Web 08/13/2008
The Benzil-Cyanide Reaction and Its Application to the
Development of a Selective Cyanide Anion Indicator
Dong-Gyu Cho, Jong Hoon Kim, and Jonathan L. Sessler*
Department of Chemistry and Biochemistry and Institute for Cellular and Molecular Biology,
1 UniVersity Station A5300, The UniVersity of Texas at Austin, Austin, Texas 78712-0165
Received May 24, 2008; E-mail: sessler@mail.utexas.edu
Abstract: The benzil-cyanide reaction is a cyanide-specific reaction that has been exploited to produce
a colorimetric indicator for this toxic anion. This was done by producing a π-extended analogue of benzil,
7, which is soluble in a 70:30 (v/v) mixture of methanol-water. In this medium, dilute solutions of 7 are
yellow but produce colorless products when exposed to low concentrations of cyanide anion (g1.7 µM;
added as an aqueous NaCN solution), but no other common anions (e.g., OH-, F-, N3-, benzoate-, and
H2PO4-). On the basis of these observations and supporting mechanistic analyses, it is concluded that the
modified benzil system 7 is a promising cyanide anion indicator that is attractive in terms of its selectivity,
ease-of-use, water compatibility, and the low, naked-eye discernible cyanide detection limit it provides.
Introduction
cently, we carried out a mechanistic study of the benzil
rearrangement reaction in organic solvents and used this analysis
Cyanide binds to the ferric form of cytochrome-c and inhibits
the mitochondrial electron-transport chain.1 It is thus highly toxic
to living creatures. Nonetheless, the use of cyanide salts remains
widespread, particularly in gold mining, electroplating, and
metallurgy.2 Despite safeguards and increasing levels of moni-
toring and control, accidental releases of cyanide into the
environment do occur. This can lead to disastrous consequences,
as underscored by the large cyanide spill that took place in
Romania in 2000, an event that was considered to be the worst
environmental disaster in Europe since Chernobyl.3 There is
thus a well-appreciated need for cyanide-selective receptors,
sensors, and indicators. Indeed, considerable effort within the
anion recognition community has been devoted to preparing
such species.4 However, many of the cyanide anion receptors
reported to date have relied on hydrogen-bonding motifs and,
as a consequence, have generally displayed poor selectivities
relative to other anions.4 To overcome this limitation, reaction-
based receptors, rationally designed cyanide anion indicators,
have been developed recently; these include oxazines, such as
1,5 cationic borane derivatives, for example, 2,6 and acridinium
salts such as 3.7 Unfortunately, none of these is ideal. In many
cases, the mechanistic basis for reaction is not fully established.
More importantly, there are practical limitations to their use.
For instance, the cationic borane receptor 2, while very
promising in that reaction with cyanide takes place in water,
fails to produce a color change, a shortcoming that restricts its
utility. Likewise, the oxazine-based indicators required specific
biphasic conditions, while the acridinium salts required an
elevated reaction temperature. There thus remains a need for
yet-improved reaction-based cyanide indicator systems. Re-
to produce a benzil-based indicator, 4, which could be used to
detect the presence of cyanide anion in the absence of water.8
Further analysis of the underlying chemistry has now led us to
produce a modified benzil indicator, 7, which undergoes a
different cyanide-driven reaction, the so-called benzil-cyanide
reaction, in aqueous media. As detailed below, this new system
can be used to detect the presence of cyanide via a naked-eye
discernible color change in 70:30 (v/v) methanol-water at a
limit of detection below the maximum permissible level for
drinking water set by the World Health Organization (1.9 µM).9
(4) (a) Kim, Y.-H.; Hong, J.-I. Chem. Commun. 2002, 512–513. (b)
Anzenbacher, P., Jr.; Tyson, D. S.; Jurs´ıkova´, K.; Castellano, F. N.
J. Am. Chem. Soc. 2002, 124, 6232–6233. (c) Ros-Lis, J. V.; Mart´ınez-
Ma´n˜ez, R.; Soto, J. Chem. Commun. 2002, 2248–2249. (d) Chow,
C.-F.; Lam, M. H. W.; Wong, W.-Y. Inorg. Chem. 2004, 43, 8387–
8393. (e) Badugu, R.; Lakowicz, J. R.; Geddes, C. D. Anal. Biochem.
2004, 327, 82–90. (f) Garc´ıa, F.; Garc´ıa, J. M.; Garc´ıa-Acosta, B.;
Mart´ınez-Ma´n˜ez, R.; Sanceno´n, F.; Soto, J. Chem. Commun. 2005,
2790–2792. (g) Badugu, R.; Lakowicz, J. R.; Geddes, C. D. J. Am.
Chem. Soc. 2005, 127, 3635–3641. (h) Liu, H.; Shao, X.-B.; Jia, M.-
X.; Jiang, X.-K.; Li, Z.-T.; Chen, G.-J. Tetrahedron 2005, 61, 8095–
8100. (i) Badugu, R.; Lakowicz, J. R.; Geddes, C. D. J. Am. Chem.
Soc. 2005, 127, 3635–3641. (j) Badugu, R.; Lakowicz, J. R.; Geddes,
C. D. Dyes Pigm. 2005, 64, 49–55. (k) Jin, W. J.; Ferna´ndez-Arguelles,
M. T.; Costa-Ferna´ndez, J. M.; Pereiro, R.; Sanz-Medel, A. Chem.
Commun. 2005, 883–885. (l) Chung, Y. M.; Raman, B.; Kim, D.-S.;
Ahn, K. H. Chem. Commun. 2006, 186–188. (m) Chen, C.-L.; Chen,
Y.-H.; Chen, C.-Y.; Sun, S.-S. Org. Lett. 2006, 8, 5053–5056. (n)
Lee, K.-S.; Kim, H.-J.; Kim, G.-H.; Shin, I.; Hong, J.-I. Org. Lett.
2008, 10, 49–51. (o) Ekmekci, Z.; Yilmaz, M. D.; Akkaya, E. U. Org.
Lett. 2008, 10, 461–464.
(5) (a) Tomasulo, M.; Raymo, F. M. Org. Lett. 2005, 7, 4633–4636. (b)
Tomasulo, M.; Sortino, S.; White, A. J. P.; Raymo, F. M. J. Org.
Chem. 2006, 71, 744–753.
(1) Young, C.; Tidwell, L.; Anderson, C. Cyanide: Social, Industrial, and
Economic Aspects; Minerals, Metals, and Materials Society: Warren-
dale, 2001.
(2) Bhattacharya, P. K. Metal Ions in Biochemistry; Alpha Science
International Ltd.: Harrow, U.K., 2005.
(6) Hudnall, T. W.; Gabbai, F. P. J. Am. Chem. Soc. 2007, 129, 11978–
11986.
(7) Yang, Y.-K.; Tae, J. Org. Lett. 2006, 8, 5721–5723.
(8) Sessler, J. L.; Cho, D.-G. Org. Lett. 2008, 10, 73–75.
(9) Guidelines for Drinking-Water Quality; World Health Organization:
Geneva, 1996.
(3) Koenig, R. Science 2000, 287, 1737–1738.
9
10.1021/ja8039025 CCC: $40.75
2008 American Chemical Society
J. AM. CHEM. SOC. 2008, 130, 12163–12167 12163

A R T I C L E S
Cho et al.
soluble in MeOH and, perhaps, MeOH-water mixtures. It was
prepared in 38% yield by subjecting bis(4-ethynylphenyl)e-
thanedione 511 and 4-iodo-N,N-bis(2-methoxyethyl)benzenamine
612 to Sonogashira coupling. The unexpected low yield is
thought to reflect the fact that intermediate 5 is consumed more
rapidly than 6 during the reaction. Consistent with this latter
supposition was the finding that the optimized yield of 38%
(based on 6) was only obtained when 1 extra equiv of 5 was
used.
Results and Discussion
The ability of 7 to act as a reaction-based cyanide anion sensor
was monitored by UV-vis spectroscopy, as shown in Figures
1 and 2. In the absence of an added anion, the absorption
maximum of 7 (2.30 × 10-5 M in 80:20 (v/v) MeOH-water)
appears at 410 nm. After the addition of 1 equiv of sodium
cyanide and base (10 µM, 1 N aqueous NaOH) to 2 mL of this
aqueous methanolic solution of 7, a large bathochromic shift
was observed (∆λmax ) 43 nm), with all spectral changes being
complete within 10 min with occasional shaking (Figure 1a).
The blue shift in the absorption maximum produced as the result
of this treatment is reflected in a change in the color of the
solution from yellow to colorless (Figure 1b), a feature that
allows for facile qualitative and semiquantitative analysis via
simple naked-eye inspection.13
Our recent studies of 4 confirmed that in the absence of water
and in the presence of an organic-soluble cyanide anion source,
specifically tetrabutylammonium cyanide (TBACN), a benzil
rearrangement took place; this produced cyanobenzoyl benzoate
in decent yields in aprotic solvents via intermediate C (Schemes
1 and 2).8 This same intermediate is implicated in the venerable
benzil-cyanide reaction,10 kinetic studies of which were
reported in 1958.10a Cleavage of benzil derivatives according
to this latter reaction produces the corresponding benzaldehyde
derivatives and benzoate esters in MeOH. Only a catalytic
amount of cyanide is required to trigger this reaction, and this
can be provided by the use of simple CN- salts that have been
rendered miscible in MeOH via the addition of water. The
benzil-cyanide reaction also leads to bond cleavage and thus
an obvious change in the π-conjugation pathway of the starting
benzil derivative. We therefore considered that the benzil-cyanide
reaction would prove superior to the benzyl rearrangement-based
process we used previously for cyanide anion detection.8 In
particular, we thought it might allow detection of cyanide in
the presence of water, provided a benzil-like system could be
produced that would give rise to a detectible color change
following cyanide-mediated bond cleavage.
In the absence of added base, the color changes (and
presumably the underlying benzil-cyanide reaction) were not
complete within 1 h under what were otherwise identical
conditions. This observation is consistent with the intuitively
reasonable conclusion that intermediate C is subject to more
rapid solvolysis in a basic medium. However, it is important to
appreciate that even in the absence of base, the reaction produces
the same final products albeit more slowly. The specific nature
of these products will be discussed in context of Figures 3 and
4 below.
To exploit the benzil-cyanide reaction for the purposes of
cyanide indicator development, it is necessary to produce a
π-extended benzil derivative that is soluble in MeOH. Unfor-
tunately, the previous indicator (4) was designed to be soluble
in organic media and proved, as expected, to be insoluble in
MeOH. We therefore focused on synthesizing a benzil deriva-
tive, the putative reaction-based indicator 7, which was expected
Further studies revealed that, in the case of 7, the rate of the
cyanide-promoted reaction is independent of water concentration
up to a limit of 30% water v/v, with the solution becoming
turbid once a limit of 50% water in MeOH is reached. Using a
dilute solution of 7 (7.20 × 10-6 M) in 70:30 (v/v) MeOH-water
allows a limit of detection (LoD) of less than 1.7 µM to be
Scheme 1. Proposed Mechanisms of the Benzil-Cyanide Reaction
Scheme 2. Synthetic Scheme
9
12164 J. AM. CHEM. SOC. VOL. 130, NO. 36, 2008

The Benzil-Cyanide Reaction
A R T I C L E S
Figure 1. (a) UV-vis spectra of indicator 7 (2.30 × 10-5 in MeOH/water (80:20, v/v)) recorded before and 10 min after the addition of sodium cyanide
(NaCN, 1 equiv and 10 µL of 1 N NaOH in water) to the cuvette containing 2 mL of the solution of 7. (b) Color changes observed 1 h after the addition
of varying quantities of NaCN (as 6.95 × 10-4 M stock solution in water) and 1 N NaOH (10 µL in water) to solutions of indicator 7 (7.20 × 10-6 M in
MeOH/water (70:30, v/v); initial volume of 7 ) 2 mL). From left to right: 0, 1.7, 3.4, 6.8, 13.6, and 0 µM of added [CN-], respectively.
ducted under the same conditions (i.e., 7, 2.30 × 10-5 in MeOH/
water (70:20, v/v); initial volume ) 2 mL) in the presence of
other anions, including OH- (10 µL of 1 N aqueous NaOH),
-
F- (as the trihydrate), N3-, BzO-, Cl-, HS2O4-, and H2PO4
as TBA salts (1 equiv of each anion) with or without sodium
cyanide (1 equiv). A discernible color change was only seen in
the presence of cyanide. Furthermore, this color change was
observed without any significant change in the reaction rate. It
is particularly noteworthy that the addition of aqueous NaOH
did not change the color of the solution even after 12 h, thus
Figure 2. Color changes observed upon the addition of various anions to
solutions of indicator 7 (2.30 × 10-5 in MeOH/water (80:20, v/v)) within
ruling out simple deprotonation effects, which are the basis of
a number of promising noncyanide anion indicators.15
10 min. From left to right: 7 only, 7 + NaCN (1 equiv) + NaOH (10 µL
of 1 N aqueous NaOH), 7 + NaOH (10 µL of 1 N aqueous NaOH), 7 +
The determinants of the cyanide-mediated cleavage reaction
TBA·F- (10 equiv), 7 + TBA·N3- (10 equiv), 7 + TBA·BzO- (10 equiv),
7 + TBA·H2PO4- (10 equiv). The initial solution volume of 7 was 2 mL.
were studied in detail using liquid chromatography-mass
spectrometry (LC-MS). The addition of cyanide and an additive
(10 µL of aqueous 1 N NaOH) to a 2 mL sample of sensor 7
gives rise to fragments in the LC-MS that are consistent with
the formation of benzaldehyde 7a and benzoate 7b (Figure 3).
The identity of these peaks was independently confirmed from
a small-scale benzil-cyanide reaction carried out in MeOH/
CH2Cl2 (2:1, v/v). These solvent conditions were chosen to
ensure a homogeneous reaction. The two major compounds were
isolated by preparative thin layer chromatography in 97% and
attained via simple naked-eye analysis in MeOH/water within
1 h (Figure 1b).
To evaluate the selectivity of indicator 7, various potentially
competing anions, including OH- (10 µL of 1 N aqueous
NaOH), F- (as the trihydrate), N3-, BzO-, and H2PO4- as TBA
salts (10 equiv of each anion was added from a stock solutions
of appropriate concentrations made up in MeOH) were added
to solutions of 7 (2.30 × 10-5 in MeOH/water (70:20, v/v);
initial volume ) 2 mL). In no case was a change in color
observed. In contrast, cyanide caused a complete loss in color.
Such absolute yes/no selectivity is generally not seen in
hydrogen-bonding-based receptors and is often precluded in
reaction-based indicator systems due to competing reactions
(e.g., nucleophilic addition of anions other than cyanide).14 More
importantly, direct anion interference experiments were con-
1
71% yield, respectively (Scheme 3). Standard H NMR spec-
troscopic and mass spectrometric analyses allowed these species
to be assigned as 7a and 7b, respectively.16
With authentic samples of 7a and 7b in hand, co-injection
experiments, involving of samples obtained under sensing
conditions, were carried out. As can be seen from an inspection
of Figure 4, LC-MS analysis revealed that the intensity of the
left-most peak (only) is increased upon the addition of authentic
7a, whereas the intensity of the right peak only increases upon
(10) (a) Kwart, H.; Baewky, M. J. Am. Chem. Soc. 1958, 80, 580–588. (b)
Trisler, J. C.; Frye, J. L. J. Org. Chem. 1965, 30, 306–307. (c)
Kuebrich, A. W.; Schowen, R. L. J. Am. Chem. Soc. 1971, 93, 1220–
1223. (d) Bwgstahler, A. W.; Walker, D. E., Jr.; Kuebrich, J. P.;
Schowen, R. L. J. Org. Chem. 1972, 37, 1272–1273. (e) Clerici, A.;
Porta, O. J. Org. Chem. 1994, 59, 1591–1592. (f) Okimoto, M.; Itoh,
T.; Chiba, T. J. Org. Chem. 1996, 61, 4835–4837.
(15) (a) Gunnlaugsson, T.; Kruger, P. E.; Jensen, P.; Pfeffer, F. M.; Hussey,
G. M. Tetrahedron Lett. 2003, 44, 8909–8913. (b) Bonizzoni, M.;
Fabbrizzi, L.; Taglietti, A.; Tiengo, F. Eur. J. Org. Chem. 2006, 16,
3567–3574. (c) Evans, L. S.; Gale, P. A.; Light, M. K.; Quesada, R.
Chem. Commun. 2006, 965–967. (d) He, X.; Hu, S.; Liu, K.; Guo,
Y.; Xu, J.; Shao, S. Org. Lett. 2006, 8, 333–336.
(11) Weil, T.; Wiesler, U. M.; Herrmann, A.; Bauer, R.; Hofkens, J.; De
Schryver, F. C.; Mu¨llen, K. J. Am. Chem. Soc. 2001, 123, 8101–8108.
(12) Wong, K. M. C.; Tang, W. S.; Lu, X. X.; Zhu, N.; Yam, V. W. W.
Inorg. Chem. 2005, 44, 1492–1498.
(16) The fact that compounds 7a and 7b were isolated in different relative
ratios is ascribed to difficulties in separation, rather than any intrinsic
differences in yield (the Rf values of the two products 7a and 7b are
too close to allow facile separation, even on a preparative silica gel
plate).
(17) The extent of the intensity increase was not the same in these two
experiments; this was not unexpected because these experiments were
only carried out in a qualitative manner without the concentrations of
the species involved being rigorously controlled.
(13) (a) Miyaji, H.; Sato, W.; Sessler, J. L. Angew. Chem., Int. Ed. 2000,
39, 1777–1780. (b) Anzenbacher, P., Jr.; Jursikova, K.; Sessler, J. L.
J. Am. Chem. Soc. 2000, 122, 9350–9351. (c) Sessler, J. L.; Cho, D.-
G.; Lynch, V. J. Am. Chem. Soc. 2006, 128, 16518–16519.
(14) Tomasulo, M.; Sortino, S.; White, A. J. P.; Raymo, F. M. J. Org.
Chem. 2005, 70, 8180–8189.
9
J. AM. CHEM. SOC. VOL. 130, NO. 36, 2008 12165

A R T I C L E S
Cho et al.
Figure 3. Peaks in the liquid chromatograms that undergo change upon the addition of cyanide anion (1 equiv) and 10 µL of 1 N aqueous NaOH to 2 mL
of a solution of indicator 7 (2.30 × 10-5 M in MeOH-water (80:20, v/v)) (left) and their corresponding mass spectra as determined by LC-mass spectrometry
(right).
Figure 4. (a) LC profiles obtained from the sample used to produce Figure 3. (b) The sample of experiment (a) after being co-injected with a sample of 7a
produced from an independent synthesis and purification protocol. (c) The sample of experiment (a) after being co-injected with a sample of 7b produced
from an independent synthesis and purification protocol.
Scheme 3. Synthetic Scheme
co-injection of authentic 7b.17 Taken in concert, these results
are fully consistent with the proposal that 7 undergoes a
benzil-cyanide reaction to produce 7a and 7b when exposed
to small quantities of cyanide anion in aqueous methanol. The
rupture of the π-framework that occurs as a result thus provides
a mechanistic rationale for the proposed signal-producing
process, that initial chromophore-containing solution (pale
yellow in color) is transformed into one that lacks any
appreciable visible absorption bands (and thus appears clear).
Conclusion
The solubilized benzil derivative 7 was found to act as a
highly selective reaction-based colorimetric indicator for cyanide
in aqueous methanolic media. The high selectivity for cyanide
9
12166 J. AM. CHEM. SOC. VOL. 130, NO. 36, 2008

The Benzil-Cyanide Reaction
A R T I C L E S
is thought to reflect the fundamental nature of the benzil-cyanide
reaction. Under the reaction conditions used to establish the
selectivity and limits of detection, the transformations produced
by adding cyanide to aqueous methanolic solutions of 7 do
indeed follow the pathway expected for a benzil-cyanide
reaction, as established by a combination of LC-mass spec-
trometric analysis and independent synthesis. Presumably as a
consequence, the modified benzil indicator 7 is a highly
promising cyanide anion indicator, in terms of ease-of-use,
selectivity, tolerance for aqueous environments, as well as visual
detection limits. It is expected that appropriate modifications
may allow this system to be optimized further, and work along
these lines is currently in progress.
Acknowledgment. This work was supported by the National
Institutes of Health (grant GM 58907) and the Robert A. Welch
Foundation (grant F-1018).
Supporting Information Available: Synthetic experimental
and characterization data for compounds 7, 7a, and 7b. This
material is available free of charge via the Internet at http://
pubs.acs.org.
JA8039025
9
J. AM. CHEM. SOC. VOL. 130, NO. 36, 2008 12167