Simple and modular design platform of bimodal turn-on chemodosimeters for oxophilic metal cations

Article abstract of DOI:10.1039/c8nj04198a

The selective chemodosimetric behaviour towards oxophilic metal cations of three compounds in one modular platform is explored. Each exhibits turn-on visible colour and emission, while two offer ratiometric emission profiles. The highest performer exhibits 3000-fold enhancement in visible-light emission per equivalent of metal cation.

Full text of DOI:10.1039/c8nj04198a

NJC
View Article Online
View Journal
LETTER
Simple and modular design platform of bimodal
turn-on chemodosimeters for oxophilic metal
cations†
Cite this:DOI: 10.1039/c8nj04198a
Received 16th August 2018,
Accepted 18th September 2018
David T. Hogan
and Todd C. Sutherland
*
DOI: 10.1039/c8nj04198a
rsc.li/njc
The selective chemodosimetric behaviour towards oxophilic metal
Following an unrelated investigation of radical stabilities,9
cations of three compounds in one modular platform is explored. we serendipitously uncovered synthetic intermediates that exhibited
Each exhibits turn-on visible colour and emission, while two offer emissive properties that suggested their utility as chemodosimeters.
ratiometric emission profiles. The highest performer exhibits 3000-fold As depicted in Scheme 1, N-methyl-9-phenylacridol (X = NCH3),
enhancement in visible-light emission per equivalent of metal cation.
9-phenylxanthenol (X = O), and 9-phenylthioxanthenol (X = S) were
converted into their respective cationic heteroaromatic form selec-
Selective detection of Al3+, In3+ and other oxophilic metal cations tively when exposed to either Al3+, In3+ or Zn2+ in CH3CN. This
is crucial from physiological and environmental perspectives caused a turn-on of visible colours, a red-shift in the emission bands
because over-exposure increases the risks of Alzheimer’s,1 and up to 3000-fold emission enhancement.
Parkinson’s,2 and kidney disorders,3 while seepage into soil
The chemodosimetric mechanism is dehydroxylation; coor-
disrupts normal root function and growth.4 Colorimetric and dination of the metal cation to the Lewis basic hydroxyl group
fluorometric detection of these species is preferred over methods is followed by C–O bond cleavage which leaves a resonance-
such as atomic absorption spectroscopy (AAS) or inductively- stabilized acridinium [X = NCH3]+, xanthenium [X = O]+, or
coupled plasma mass spectrometry (ICP-MS) due to the potential thioxanthenium [X = S]+ ion. To our knowledge, this is one of
for simple visual analysis of samples on-site. Organic chemo- very few reports10–12 of a chemodosimeter using this type of
dosimeters, which increase in either visible or emissive colour mechanism. The simple system in Scheme 1 incorporates three
upon exposure to a target analyte, are a promising solution. different heteroatoms, and we propose it could be used as a
Chemodosimeters are defined by near-irreversible reactions with modular platform to construct more elaborate, functionalized
the analyte, converting a masked organic chromo/fluorophore chemodosimeters. This platform boasts many of the above-
into its readily detectable form.5 This mode of reaction-based mentioned key design aspects for effective chemodosimeters:
detection has proven highly selective because the systems can be turn-on colorimetric and fluorometric response, potential for
designed in such a way that only desired analytes promote the
conversion reaction.6 An ideal chemodosimeter fulfils several
design criteria, such as high molar absorptivity, high fluores-
cence quantum yield, and turn-on visible/emissive colour
changes for increased instrumental sensitivity.7 Secondary, but
also significant criteria are a ratiometric optical response that
permits internal instrument calibration, regenerability of the
uncoloured form to allow re-usable assays, and simplistic design
for scale-able application.8
Department of Chemistry, University of Calgary, 2500 University Dr NW, T2N 1N4,
Calgary, Alberta, Canada. E-mail: todd.sutherland@ucalgary,ca
Scheme 1 Top: Reaction scheme of three chemodosimeters (X = NCH3,
O and S) in response to a variety of metals. Bottom: Key optical data of
absorption and emission lmax values (nm) and their associated colours in
solution for both off and on-states.
† Electronic supplementary information (ESI) available: UV-vis absorption and
emission of X = O, X = S and X = NCH3 and their triflate salts; UV-vis absorption
and emission metal cation titrations; UV-vis absorption regeneration titrations.
See DOI: 10.1039/c8nj04198a
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
New J. Chem.

View Article Online
Letter
NJC
ratiometric emission monitoring, regenerability of the masked thioxanthenol X = S was non-emissive. The excitation spectra
chromo/fluorophore by re-addition of hydroxide, and a simple of X = NCH3 and X = O mimicked their absorption spectra save
design based on optional functionalization. In this proof-of- for small changes in the relative intensities of certain bands.
principle communication, we compare the selectivity of three
chemodosimeters X = NCH3, X = O, and X = S for oxophilic dehydroxylation of the chemodosimeters, authentic samples of
metal cations in solution. N-methyl-9-phenylacridinium triflate [X = NCH3]OTf (ff = 0.05),
To support our hypothesis that the metal cations promoted
The synthetic methods have been described previously.9 9-phenylxanthenium triflate [X = O]OTf (ff = 0.63), and
Briefly, X = NCH3, X = O, and X = S were obtained by addition 9-phenylthioxanthenium triflate [X = S]OTf (ff = 0.05) were
of phenylmagnesium bromide to the corresponding diaryl synthesized.9 As depicted in Fig. S9–S12 (ESI†), in CH3CN
ketone. Assembling the molecules as outlined in Scheme 1 solution they appear identical to X = NCH3, X = O, and X = S
permits elaboration of either the incoming Grignard reagent with added Al3+ under both visible and UV lights. UV-vis absorp-
(Route A) or the diaryl ketone (Route B). In this way, site- tion and emission spectra, and quantum yields are consistent
specific functionality could be installed that would modify with literature13,14 which are available in Table S2 (ESI†).
the chemodosimetric or photophysical behaviour to fit a target
application.
Addition of Al3+ to a 20 mM CH3CN solution of X = O was
followed by UV-vis absorption spectroscopy, in Fig. 2a. Over the
After careful initial consideration (ESI†), NaCl, AlCl3, LiCl, course of the metal additions, bands at 447 nm, 373 nm, and
ZnCl2, and InCl3 were chosen for screening in CH3CN. As 259 nm grew, corresponding to those of an authentic cationic
represented by X = O in Fig. 1 top panel, the solution became [X = O]OTf, at the expense of the alcohol X = O bands.
visibly bright yellow in the presence of only Al3+ and In3+ and
The presence of isosbestic points at 295 nm, 274 nm,
remained clear and colourless with other metal cations. 245 nm, and 236 nm is an indication of clean conversion.
This turn-on visible colour was accompanied by bright green Signal saturation was observed after addition of 10 equivalents
emission when illuminated by a 365 nm UV lamp, as shown in of Al3+. Absorbance spectral changes associated with the
Fig. 1 bottom panel. The corresponding photographs for X = S remaining metal cations are in Fig. S19–S24 (ESI†). The only
and X = NCH3 are in the ESI.† Thioxanthenol X = S exhibited other metal that induced chemodosimetric behaviour was In3+,
the same selectivity, but solutions became rosy-orange with which required 19 equivalents to saturate the absorbance signal
pale-orange emission, while X = NCH3 responded to Al3+, In3+, for [X = O]+. UV-vis absorption changes of all metal cations with
and Zn2+ by gaining yellow colour and cyan-green emission. X = S and X = NCH3 can be found in Fig. S13–S30 with full
The screening is a visual representation of the selectivity of details in Table S3 (ESI†). Each responded as they had in the
these chemodosimeters: they respond to oxophilic metal cations visual metal screening and exhibited growth in three new
over the alkali metals and a negative control (Bu)4NCl.
The details of the spectroscopic studies on the chemodosi- expense of those from the masked tertiary alcohols.
meters in the absence of metal cations are contained in Tables
The same X = O solution during Al3+ addition was followed
bands corresponding to [X = S]OTf and [X = NCH3]OTf at the
S1 and S2 and Fig. S4–S8 of the ESI.† All three exhibited only UV by fluorescence spectroscopy, as shown in Fig. 2b. Here,
absorption bands. In acetonitrile, acridol X = NCH3 was emis- two independent emission peaks can be monitored. The UV
sive in the near-UV and tailing into the visible region (ff = 0.05), emission of X = O at 302 nm (lex = 280 nm), which decreased with
xanthenol X = O only emitted in the UV (ff = 0.11) while added metal, concomitant with the growth of the emission band at
513 nm (lex = 373 nm), which matches the xanthenium derivative,
[X = O]OTf. The emission peaks at 513 nm (cation, increase) and
302 nm (alcohol, decrease) was saturated after 10 equivalents of
Al3+. The same saturation behaviour was observed when 19
equivalents of In3+ were added. Fluorometric titration data of
all metal cations into X = S and X = NCH3 are available in
Fig. S32–S60 (ESI†), which show that emission bands matching
[X = S]OTf and [X = NCH3]OTf appeared during the titrations
with the same metals as in the UV-vis titrations. As demonstrated
in Fig. 2b, this platform offers the possibility of monitoring two
wavelengths for ratiometric detection.
Of note for all three chemodosimeters is the sigmoidal
response of both absorbance and emission to added titrant,
represented by Fig. 2c. Typical chemodosimeters exhibit bounded
exponential growth as the amount of titrant approaches
saturation.15,16 However, for X = NCH3, X = O, and X = S the first
additions do not cause the same magnitude of change as those
Fig. 1 Visual responses of 0.1 mM X = O in the presence of different metal
that come mid-way through the metal salt additions. We offer two
possible explanations for this induction period. Literature has
shown the sigmoidal response curve has been attributed to
salts in CH3CN. The first vials in each series represents the ‘on-state’ [X = O]+
via its triflate salt. Top: Colour change under ambient light. Bottom:
Fluorescence under 365 nm UV light.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
NJC
Letter
Fig. 3 Responsiveness of the three chemodosimeters to metal salts
monitored by both absorption and emission. (a) Changes in absorption
spectra at 424, 447 and 493 nm for X = NCH3, X = O and X = S,
respectively, at signal saturation per equivalent of metal added. (b)
Changes in emission spectra at 500, 513 and 563 nm for X = NCH3,
X = O and X = S, respectively, at signal saturation per equivalent of metal
added.
added to each chemodosimeter and act as a qualitative respon-
siveness figure of merit. Table 1 tabulates the numerical
responsiveness of the three chemodosimeters to the various
metal ions. Using either UV-vis absorption or emission to
follow the metal additions, the three chemodosimeters exhibit
similar responsiveness trends which are Al3+ 4 In3+ 4 Zn2+ 4
Na+ B Li+ B (Bu)4N+. However, there are subtle difference
between both the optical methods (Abs. vs. Em.) and chemo-
dosimeter structure.
Fig. 2 Optical responses of 20 mM CH3CN solutions of X = O with up to
10 equivalents of AlCl3. (a) UV-vis absorption changes upon addition of
AlCl3 in dry CH3CN. Arrows over peaks indicate the direction of peak
evolution. (b) Fluorescence spectral changes showing both decreasing
X = O content (lex = 280 nm) concomitant with [X = O]+ growth (lex = 373 nm).
(c) Normalized peak intensities of the optical data as a function of metal
equivalents added. Left: Absorption response. Right: Emission response.
Using absorption spectroscopy to follow peak maxima in the
visible region (4400 nm), the X = NCH3 derivative, shown in
Fig. 3a, is 6Â more responsive to Al3+ than X = O and 500Â more
chemodosimeters that have a two-stage function,17,18 although than X = S. The improved responsiveness is because fewer
this situation is unlikely here given the presence of many iso- equivalents of metal are required to convert X = NCH3 into
sbestic points. Alternatively, we suggest hydration of the metal [X = NCH3]+. Even Zn2+ induces a 7Â optical increase in the
cations from adventitious water in the solvent; some metal salts X = NCH3 derivative compared to either X = O or X = S. Sodium
are coordinated by water and rendered less active towards the cation and Li+ exhibit a null-response when compared to the
chemodosimeter. After the water is removed, then fresh titrant negative control, (Bu)4N+. Within the series of metal ions that
reacts with the chemodosimeter. All solvents used were triply turn-on, there is a consistent trend in absorption responsive-
dried over flamed molecular sieves and handled using Schlenk ness that follows X = NCH3 4 X = O 4 X = S attributed to two
techniques, as detailed in the ESI.†
factors: stability of the cationic ‘‘on’’ state and equivalents of
Fig. 3 displays the responsivity of X = NCH3, X = O, and X = S metal ions to reach saturation. However, from the practical
to Na+, Al3+, Li+, Zn2+ and In3+ by dividing the magnitude of perspective for visual indication, X = O is the best performer
enhancement in optical intensity (I/I0) by the equivalents of with highest ratio between ‘‘on’’ and ‘‘off’’ states at 650-fold
metal cation that causes signal saturation. This approach was with In3+, albeit with less responsivity to the metal ion concen-
taken to account for the differing amounts of metal cations tration than X = NCH3. Enhancement ratios (I/I0) for turn-on
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
New J. Chem.

View Article Online
Letter
NJC
Table 1 Responsivity data for chemodosimeters X = NCH3, X = O and
X = S
trials also supported our proposed mechanism because addi-
tion of hydroxide reformed the original species.
The goal of this communication was to evaluate the ability
of the modular platform to produce oxophilic metal cation-
selective chemodosimeters. We have shown that X = NCH3,
X = O, and X = S relied on a revertible dehydroxylation
mechanism to form [X = NCH3]+, [X = O]+, and [X = S]+ in
CH3CN. Due to planarization and extended conjugation, these
new forms became both coloured and emissive in the visible
region. Selective, absorptive colouration in the presence of Al3+
or In3+ was best followed using X = NCH3 while the most
responsive, ratiometric emissive chemodosimeter was X = O.
Signal enhancement ranged from a moderate 40-fold increase
to an unprecedented 3000-fold for X = O. This platform meets
many key design criteria for effective chemodosimeters.
Exploiting the built-in modularity of the platform, our lab
continues to improve upon recognized shortcomings, such as
Eq. to
I/I0 in I/I0 in Abs. I/I0 in I/I0 in Em.
M+ a saturate Abs.b per M+ Eq. Em.b
per M+ Eq.
X = NCH3 Al3+
0.63
21.0
4.25
251
155
164
398
7
39
48
53
42
77
3
10
Zn2+
In3+
X = O
Al3+
Zn2+
In3+
10.0
618
1
657
62
2970
69
3230
297
1
162
50.0b
20.0
o1
33
X = S
Al3+
20.0
151
1
372
8
61
1
73
3
Zn2+
50.0b
o1
o1
In3+ 220
2
o1
Na+, Li+ and (Bu)4N+ are excluded for clarity as all responses are E1
(ESI). Data available in graphical form in Fig. S31 (ESI).
a
b
chemodosimeters are used frequently in the literature to water-intolerance, by appending mesomeric electron-donating
compare different systems even though concentrations range groups. The findings will be presented in due course.
from 1 mM to 50 mM, which report enhancements from 32 to
582-fold.11,16,19 Given these comparator enhancement ratios,
our chemodosimeters relate favourably with literature.
Conflicts of interest
When measuring emission in Fig. 3b, X = O is the most
There are no conflicts to declare.
responsive. It is over 4Â more responsive to Al3+ and In3+ than
X = NCH3 and over 100Â more than X = S because the ff is
greater (0.63 vs. 0.05). Xanthenol derivative, X = O, is also over
Notes and references
300Â more responsive to Al3+ than Na+, Li+ and (Bu)4N+
although it has a slight turn-on for Zn2+ (Table 1). Thioxanthenol
derivative X = S is the least responsive emission chemodosimeter
because 20 equivalents of Al3+ and 220 equivalents of In3+ are
needed to reach saturation. The enhancement ratio in Table 1 of
the X = S and X = NCH3 emission is poorer than that for
absorption, perhaps due to low ff. However, for X = O the
enhancement ratio is much greater at about 3000-fold, which
is superior to literature reports in our concentration range.11,16,19
All three chemodosimeters are more selective for oxophilic metal
cations Al3+ and In3+, and X = NCH3 is more responsive in
absorption measurements while X = O is more responsive for
emission.
1 T. P. Flaten, Brain Res. Bull., 2001, 55, 187.
2 E. Hirsch, J. P. Brandel, P. Galle, F. Javoy-Agid and Y. Agid,
J. Neurochem., 1991, 56, 446–451.
3 F. P. Castronovo and H. N. Wagner, Br. J. Exp. Pathol., 1971,
52, 543–559.
4 J. F. Ma, P. R. Ryan and E. Delhaize, Trends Plant Sci., 2001,
6, 273–278.
5 D.-G. Cho and J. L. Sessler, Chem. Soc. Rev., 2009, 38, 1647.
6 D. T. Quang and J. S. Kim, Chem. Rev., 2010, 110, 6280–6301.
7 Y. Yang, Q. Zhao, W. Feng and F. Li, Chem. Rev., 2013, 113,
192–270.
8 J. Chan, S. C. Dodani and C. J. Chang, Nat. Chem., 2012, 4,
973–984.
9 D. T. Hogan and T. C. Sutherland, J. Phys. Chem. Lett., 2018,
9, 2825–2829.
The bulk of literature on chemodosimeters use complex
reactions that irrevocably change the system.20–22 Our platform
relies simply on removal of a masking hydroxyl group – a
´
´˜
transformation that could be reverted in the proper conditions 10 M. Lo Presti, S. El Sayed, R. Martınez-Manez, A. M. Costero,
´
to regenerate the original chemodosimeter. Regeneration holds
promise for re-usable assays and has therefore been built into
S. Gil, M. Parra and F. Sancenon, New J. Chem., 2016, 40,
9042–9045.
chemodosimeters for other analytes,17,23 but only rarely for 11 M. Alaparthi, K. Mariappan, E. Dufek, M. Hoffman and
metal cations.15,24 Regeneration trials were conducted by satur-
ating X = NCH3, X = O, and X = S with ample Al3+ and then 12 E. E. Nekongo, P. Bagchi, C. J. Fahrni and V. V. Popik,
adding (CH3)4NOH and following spectral evolution by UV-vis Org. Biomol. Chem., 2012, 10, 9214.
absorption spectroscopy. The hydroxide competed for both 13 S. A. Jonker, F. Ariese and J. W. Verhoeven, Recl. Trav. Chim.
metal and organic cations in solution, so excess hydroxide Pays-Bas, 2010, 108, 109–115.
was required to regenerate the original tertiary alcohols. 14 J. M. Bedlek, M. R. Valentino and M. K. Boyd, J. Photochem.
Fig. S61–S63 (ESI†) show the absorption bands for [X = NCH3]+,
Photobiol., A, 1996, 94, 7–13.
[X = O]+, and [X = S]+ decrease as the bands for X = NCH3, X = O, 15 Y. Yang, F. Huo, C. Yin, A. Zheng, J. Chao, Y. Li, Z. Nie,
A. G. Sykes, RSC Adv., 2016, 6, 11295–11302.
´
´˜
and X = S increase to saturation. Besides proving that this
platform can be used for regenerable chemodosimeters, the
R. Martınez-Manez and D. Liu, Biosens. Bioelectron., 2013,
47, 300–306.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
NJC
Letter
16 J. Wang, Y. Li, K. Li, X. Meng and H. Hou, Chem. – Eur. J., 21 Y.-Y. Lv, W. Gu, J.-B. Wang, W.-B. Huang, W.-X. Shen and
2017, 23, 5081–5089. X.-Y. Sun, Sens. Actuators, B, 2017, 246, 1017–1024.
17 A. K. Mahapatra, K. Maiti, R. Maji, S. K. Manna, S. Mondal, 22 P. Das, T. Chaudhuri, A. Karmakar, S. Saha, S. Sengupta
S. S. Ali and S. Manna, RSC Adv., 2015, 5, 24274–24280.
18 C. Wang and K. M.-C. Wong, Inorg. Chem., 2013, 52,
13432–13441.
Bandyopadhyay and C. Mukhopadhyay, Asian J. Org. Chem.,
2016, 5, 1492–1498.
23 W. Zhang, T. Liu, F. Huo, P. Ning, X. Meng and C. Yin, Anal.
Chem., 2017, 89, 8079–8083.
19 P. Du and S. J. Lippard, Inorg. Chem., 2010, 49, 10753–10755.
´
´˜
20 S. Yao, Y. Qian, Z. Qi, C. Lu and Y. Cui, New J. Chem., 2017, 24 J. V. Ros-Lis, M. D. Marcos, R. Martinez-Manez, K. Rurack
41, 13495–13503. and J. Soto, Angew. Chem., Int. Ed., 2005, 44, 4405–4407.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
New J. Chem.