Selective fluorescence sensing of copper(II) and water via competing imine hydrolysis and alcohol oxidation pathways sensitive to water content in aqueous acetonitrile mixtures

Article abstract of DOI:10.1021/ic402723c

Addition of hydrazines to a 1,8-disubstituted anthraquinone macrocycle containing a polyether ring produces site-selective imination, where hydrazone formation produces the more sterically hindered adduct. Reduction of the remaining carbonyl group to a secondary alcohol followed by addition of copper(II) ion causes intense yellow fluorescence to occur, which is selective for this metal cation and allows this system to be used as a fluorescence sensor. In the presence of water, a green-fluorescent intermediate appears, which slowly decomposes to produce the original starting anthraquinone. The addition of a large amount of water radically changes the reaction pathway. In this case, oxidation of the secondary alcohol is kinetically faster than hydrolysis of the hydrazone, although the same anthraquinone product is ultimately produced. Stern-Volmer data suggest that dioxygen quenches the green emission through both dynamic and static mechanisms; the static ground-state effect is most likely due to association of oxygen with the copper-bound fluorescent intermediate.

Full text of DOI:10.1021/ic402723c

Article
pubs.acs.org/IC
Selective Fluorescence Sensing of Copper(II) and Water via
Competing Imine Hydrolysis and Alcohol Oxidation Pathways
Sensitive to Water Content in Aqueous Acetonitrile Mixtures
Kadarkaraisamy Mariappan, Madhubabu Alaparthi, Gerald Caple, Vinothini Balasubramanian,
Mariah M. Hoffman, Mikaela Hudspeth, and Andrew G. Sykes*
The Department of Chemistry, University of South Dakota, Vermillion, South Dakota 57069, United States
S
* Supporting Information
ABSTRACT: Addition of hydrazines to a 1,8-disubstituted
anthraquinone macrocycle containing a polyether ring
produces site-selective imination, where hydrazone formation
produces the more sterically hindered adduct. Reduction of the
remaining carbonyl group to a secondary alcohol followed by
addition of copper(II) ion causes intense yellow fluorescence
to occur, which is selective for this metal cation and allows this
system to be used as a fluorescence sensor. In the presence of
water, a green-fluorescent intermediate appears, which slowly
decomposes to produce the original starting anthraquinone.
The addition of a large amount of water radically changes the
reaction pathway. In this case, oxidation of the secondary
alcohol is kinetically faster than hydrolysis of the hydrazone,
although the same anthraquinone product is ultimately produced. Stern−Volmer data suggest that dioxygen quenches the green
emission through both dynamic and static mechanisms; the static ground-state effect is most likely due to association of oxygen
with the copper-bound fluorescent intermediate.
alternative method for selective detection is the use of
fluorescence spectroscopy. Fluorescent sensors are advanta-
geous because they are often inexpensive, convenient, selective,
and sensitive.8,9 While numerous copper sensors have been
reported, many of these are fluorescence quenching sensors
(turn-off sensors) because of the paramagnetic nature of the
metal center.10In order to maximize spatial resolution,
fluorescence enhancement sensors (turn-on sensors) are
generally preferred over turn-off sensors.1 Cu(II) sensors with
turn-on fluorescence can be categorized primarily into
photoinduced electron transfer (PET)-type,11 rhodamine
spirolactam ring-opening-type,12 and internal charge transfer
(ICT)-type13 sensors. Other Cu(II) detection mechanisms
have included copper-catalyzed coupling reactions14 and the
use of nanoparticles.15
We recently reported a series of imines 2 based on site-
selective imination of the external carbonyl of quinone
macrocycle 1 (Scheme 1) using Ti(IV) as a catalyst.16
Reduction with NaBH4 yields alcohol 3, which undergoes
imine−enamine tautomerization, resulting in turn-on fluores-
cence, only in the presence of Zn(II).17 Very few analyte-
specific fluorescence sensor systems currently employ tautome-
rization as the operating photophysical mechanism. Surpris-
INTRODUCTION
■
The selective detection of metals is of particular importance to
the fields of environmental chemistry and biochemistry.
Detection of copper(II) is of particular interest as it is an
essential metal in all living organisms because of its catalytic
and oxidative properties.1 Copper is the third most abundant
transition metal in the human body, with only iron and zinc
being higher in concentration.2 Unregulated Cu(II) can cause
oxidative stress and has been shown to have connections to
aging and such neurodegenerative diseases as Menke’s,
Wilson’s, amytrophic lateral sclerosis, Alzheimer’s, prion, and
Parkinson’s diseases.1,3
The amount of biologically available copper has increased as
a result of anthropogenic activities such as smelting, mining,
fertilizing, and the use of fungicides and molluscicides.4
Increased copper levels have proven to be detrimental to the
growth of certain vegetation,5 stream insects,6 aquatic biota,
and microbes, among others.4Because of the toxicity of Cu(II),
the United States Environmental Protection Agency’s max-
imum contaminant level goal for Cu(II) in drinking water is 1.3
mg/L (∼1 ppm).7Detection of Cu(II) is critical for the
identification of polluted bodies of water and regulation of
human drinking water, and this ion has typically been detected
in the past by atomic absorbance spectroscopy, inductively
coupled plasma, voltammetry, and piezoelectric quartz crystals,
which are often expensive and time-consuming methods.8 An
Received: October 31, 2013
Published: March 6, 2014
© 2014 American Chemical Society
2953
dx.doi.org/10.1021/ic402723c | Inorg. Chem. 2014, 53, 2953−2962

Inorganic Chemistry
Article
Scheme 1. Site-Selective Syntheses of External Imines (216 and 317) and Internal Hydrazones (4 and 5, This Work)
Figure 1. (left) Thermal ellipsoid diagram (293 K, 30%) of 4b. Imine: C13−N1 = 1.301(7) Å; N1−N2 = 1.341(7) Å. Carbonyl: C34−O11 =
1.216(10) Å. Intramolecular hydrogen bond: N2−O7 = 2.637(7) Å, N2−H···O7 = 137(4)°. (right) Thermal ellipsoid diagram (100 K, 30%) of 5a·
CH2Cl2. Imine: C7−N1 = 1.285(4) Å; N1−N2 = 1.376(4) Å. Alcohol: C8−O7 = 1.414(4) Å. The dichloromethane solvate has been removed for
clarity.
ingly, we have found that the addition of hydrazines to 1
exclusively yields hydrazones 4 via condensation with the
opposing intraannular carbonyl instead. Reduction yields the
corresponding alcohol 5, which undergoes an intense
fluorescence enhancement in the presence of Cu(II) that is
dependent on the amount of water present in the solvent.
hydrazone formation at the internal carbonyl,19 and other prior
cyclic crown ethers had previously yielded external azophenolic
acerands.20 However, we have always found that the intra-
annular carbonyl is the most reactive carbonyl, undergoing a
kinetically facile reduction with NaBH4 in the case of 1 alone,
likely due to the electronic inductive effect of the nearby alkoxy
groups;21 thus, it is apparent that the “butterfly” motion of the
anthraquinone backbone is flexible enough to accommodate
“internal” hydrazone formation here.
Reduction of 4a with NaBH4 produces compound 5a, in
which the external carbonyl has been reduced to the
corresponding alcohol (Scheme 1). IR data show the loss of
the carbonyl stretch and the introduction of a new OH stretch,
and the proton NMR spectrum shows the appearance of one
methine proton at 5.31 ppm (coupling constant J = 8 Hz)
coupled to the OH proton at 6.37 ppm (J = 8 Hz) (Figure S1
in the Supporting Information). In the structure of compound
5a (crystallized as the CH2Cl2 solvate; Figure 1 right), the C
N imine group reflects an unreduced −NN− double bond
(C7−N1 = 1.285 Å) while the carbonyl has been reduced to
the alcohol (C8−O1 = 1.414 Å). The N−H and O−H protons
in the structure form hydrogen bonds to crown ether oxygen
atoms of neighboring molecules. Unlike 4a, the dinitrophenyl
adduct 4b could not be reduced to the corresponding alcohol
using the same reaction conditions.
RESULTS
■
Synthesis. Commercially available aromatic hydrazines
condense with 1 in the presence of sulfuric acid to produce
bright-orange hydrazones 4, a familiar reaction typically used
for the qualitative identification of carbonyl groups in the
introductory organic chemistry laboratory (Scheme 1). Here,
however, the reaction is site-selective, condensing only with the
intraannular carbonyl group, even when the hydrazine is
present in excess. The resonances for the anthraquinone
protons remain relatively unchanged from those of the starting
compound 1, indicating that the external carbonyl remains
unreacted. X-ray analysis of crystals of 4b, the 2,4-dinitrophenyl
derivative, conclusively showed that the hydrazone formed
results from condensation only with the internal carbonyl
(Figure 1 left). The bond distances are typical for double bonds
for both the imine and carbonyl groups found within the
structure. Previously, this compound was incorrectly described
in the literature as the external adduct because of the possibility
that the crown ring, through steric interactions, would block
2954
dx.doi.org/10.1021/ic402723c | Inorg. Chem. 2014, 53, 2953−2962

Inorganic Chemistry
Article
Figure 2. Changes in the fluorescence of 5a (1.0 × 10−4 M) in acetonitrile upon the immediate addition of 2.0 equiv of metal cations, with excitation
at 360 nm. Inset: photograph showing solutions of 5a under irradiation before and immediately after addition of Cu(II).
Figure 3. Competition study: addition of two equivalents of selected metal cations to solutions of 5a (1.0 × 10−4 M). Fluorescence was monitored at
560 nm, and the excitation wavelength was 360 nm. Red and blue bars are before and after addition of Cu(II), respectively.
Spectroscopy. The addition of different metal cations to 5a
in acetonitrile was studied, and it was found that only Cu(II)
causes the immediate appearance of an intense yellow emission
(Figure 2). The addition of other cations prior to addition of
Cu(II) did show competition from the lower alkaline-earth
metals and paramagnetic transition-metal cations (Figure 3).
We previously demonstrated that oxophilic cations, such as
Ca(II) and Pb(II), have the highest affinity for the same crown
ether ring size and may block the association of Cu(II) within
the macrocycle.22,23
Final Product Identification. Under very dry conditions,
the yellow emission fades to form a nonemissive solution within
a few minutes (Figure 4 left). We identified the final
nonemissive product as compound 1, the original anthraqui-
none starting material as shown in Scheme 1, indicating that
both hydrolysis of the hydrazone and oxidation of the alcohol
back to the quinone occur upon the addition of Cu(II). We
isolated compound 1 as the final product by means of 1H NMR
spectroscopy (Figure S1) and X-ray crystallography (Figure
S2). Addition of CuCl2 to 5a in a 1:1 ratio yields yellow crystals
composed of 1·CuCl2 upon evaporation. The proton NMR
spectrum is identical to that of an authentic sample of 1, and
the X-ray structure shows a simple adduct of 1 cocrystallized
with CuCl2. Of interest is the fact that the copper metal center
remains in the +2 oxidation state with the presence of two
chloride anions bound in a linear arrangement around the metal
center.
However, when only a small amount of water (0.2−0.3% w/
v, ∼0.14 M) is present in the solution, the yellow emission is
rapidly replaced by an intense green emission at ∼490 nm
2955
dx.doi.org/10.1021/ic402723c | Inorg. Chem. 2014, 53, 2953−2962

Inorganic Chemistry
Article
Figure 4. (left) Fluorescence changes after addition of 2.0 equiv of Cu(II) to 1.0 × 10−4 M 5a in ∼1.4 × 10−3 M H2O in acetonitrile. (right)
Fluorescence changes after addition of 2.0 equiv of Cu(II) to 1.0 × 10−4 M of 5a in ∼0.14 M H2O in acetonitrile. The emission spectra were
collected every 15 s, and the excitation wavelength was 360 nm.
Scheme 2. Cu(II)-Catalyzed Oxidation and Hydrolysis of 4a
(Figure 4 right). Thus, compound 5a is not only a sensitive
fluorescent copper(II) sensor but also a fluorescent water
sensor. The green emission diminishes over a few minutes to
yield a nonemissive solution, in which the final product was
similarly identified as 1.
Reaction Mechanism. Scheme 2 represents two possible
pathways for the transformation of 5a to the final nonemissive
product 1. Potentially, either hydrolysis of the hydrazone (A →
B) or oxidation of the alcohol (A → C) could take place first.
Under conditions with little water present (vide infra), we
propose that the hydrazone is first hydrolyzed by the Cu(II)
ion to form intermediate B, which then undergoes oxidation by
dissolved oxygen to produce the quinone (D).
99:1 acetonitrile/water solution (Figure 5) show the
appearance of an intermediate band at ∼330 nm, which we
attribute to the formation of the hydrolyzed intermediate B.
The intermediate appears immediately upon Cu(II) addition
and subsequently reacts to form product D with very clean
isosbestic points. Intermediate B happens to be the isomer of
compound 6 (Scheme 2), which we previously isolated and
found to be air-stable.21,24The absorption spectrum of
compound 6 (green trace) is also shown in Figure 5 and
exhibits an almost identical absorption band at 330 nm.
The final product spectral band centered at 370 nm that is
observed in Figure 5 is also identical to that of an authentic
sample of 1 (blue trace), providing additional evidence that 1 is
the final product. The corresponding NMR titration (Figure
S3) shows that the hydrazine NH proton disappears
immediately upon Cu(II) addition (fast hydrolysis of the
hydrazine), while two new resonances appear at 5.78 and 6.74
Evidence for this pathway comes from the UV/vis spectral
1
changes (Figure 5), an H NMR titration after addition of
Cu(II) (Figure S3), and mass spectrometry data (Figure S4).
The absorbance changes upon the addition of Cu(II) to 5a in
2956
dx.doi.org/10.1021/ic402723c | Inorg. Chem. 2014, 53, 2953−2962

Inorganic Chemistry
Article
Figure 5. UV/vis spectral changes (recorded every 2 min) upon the addition of 1 equiv of Cu(II) to 1.0 × 10−4 M 5a in a 99:1 acetonitrile/water
mixture.
Figure 6. UV/vis spectral changes (recorded every 2 min) upon the addition of 1 equiv of Cu(II) to 1.0 × 10−4 M 5a in a 90:10 acetonitrile/water
mixture.
ppm that correspond to methine and aromatic resonances for
intermediate B (Scheme 2), consistent with the A → B
pathway. Addition of Cu(II) to the final product 1 does not
produce any emission either, as we have previously reported.22
ESI-MS experiments (Figure S4) also showed that 1 is the final
product after addition of Cu(II) to 5a, with the appearance of
the sodiated peak at m/z 421 after several minutes. An
additional peak at m/z 208 is present, likely corresponding to
the hydrolyzed hydrazine associating with the copper cation,
[Cu(phenylhydrazine)(H2O)2]+ (m/z 208). MS−MS experi-
ments on peaks of larger mass did not produce a peak at m/z
208 either, indicating that this peak is not due to simple
fragmentation of larger species under these conditions.
Addition of Water. Of major significance is that upon
addition of even larger amounts of water to the reaction
medium, the spectroscopy completely changes when Cu(II) is
added. Figure 6 shows the changes in the UV/vis spectrum
when Cu(II) is added to a 90:10 acetonitrile/water mixture
(compared with a 99:1 acetonitrile/water mixture; Figure 5). In
this case, a broad peak at 460 nm slowly grows in with clean
isosbestic points, instead of the 330 nm peak observed earlier
under drier solvent conditions. The blue trace shows the
absorption spectrum of compound 4a, which has spectral
features almost identical to what has slowly grown in, indicating
that at higher water concentrations the reaction pathway
radically changes. We propose that in this case, oxidation of the
2957
dx.doi.org/10.1021/ic402723c | Inorg. Chem. 2014, 53, 2953−2962

Inorganic Chemistry
Article
Table 1. Hydrazine Hydrolysis and Alcohol Oxidation Rate Constants (min−1) of 5a and 4a after Addition of 1 and 2 equiv of
Cu(II) in Aqueous Acetonitrile Mixtures in Air
ACN/H2O volume
ratio
oxidation B → D
(326 nm↓)
oxidation A → C
(460 nm↑)
hydrolysis C → D
(460 nm↓)
hydrolysis C → D (4a→ 1)
(460 nm↓)
5a:Cu(II) hydrolysis A → B
100:1
1:1
1:2
1:1
1:2
1:1
1:2
1:1
1:2
1:1
1:2
1:1
1:2
fast
fast
fast
fast
fast
fast
0.061(4)
fast
fast
fast
99:1
0.056(3)
0.131(5)
0.024(3)
0.022(3)
fast
fast
98:2
fast
fast
95:5
0.63(10)
0.71(1)
0.19(1)
0.24(4)
0.012(2)
0.013(2)
0.022(3)
0.023(3)
0.096(10)
0.099(10)
0.039(5)
0.032(4)
0.017(2)
0.013(1)
90:10
50:50
alcohol takes place first (A → C; Scheme 2) to yield hydrazone
4a. Under these wet conditions, no fluorescence is observed;
however, the same quinone product (D, compound 1) is
produced after the solution is allowed to sit overnight. Figure
S5 shows the effect of adding metal cations to the unreduced
hydrazone macrocycle 4a. Almost instantaneous conversion of
4a to compound 1 (black trace) occurs in the presence of
Cu(II), indicating that Cu(II) can hydrolyze the unreduced
hydrazone as well. The addition of Hg(II) and Fe(III) also
produce hydrolysis, but the kinetics is slower. Similarly, no
luminescence changes were observed for the reaction of 4a with
Cu(II). Addition of Cu(II) to 5a in 90:10 acetonitrile/water
also revealed radically different peaks in the mass spectrum
(Figure S6) compared with what was previously observed in
dry acetonitrile (Figure S4). Compound 1 is still produced, but
there is no peak at m/z 208, indicating that the hydrazine is not
hydrolyzed when significant water is present. Also, there is also
a small peak at m/z 511, which is two mass units less than the
starting sodiated 5a peak (m/z 513). This suggests the presence
of the oxidized intermediate C (Scheme 2). In MS−MS
experiments, the peak at m/z 513 was shown not to produce
the m/z 511 fragment.
concentration of dioxygen and was analyzed by separate
Stern−Volmer investigations as discussed below.
Figure S10 shows a series of UV/vis spectra of 5a with 2
equiv of added Cu(II) collected at different volume percentages
of water in acetonitrile. Plotting the natural logarithm of the
absorbance change versus time using the absorbance at 326 nm
at low water concentration (Figure S11, following the decrease
in spectral changes from Figure S10B, representing the B → D
oxidation) or the absorbance at 460 nm at higher water
concentration (Figure S12, following the increase in spectral
changes from Figure S10E, representing the A → C oxidation)
yields straight lines representing first-order kinetics, and the
respective rate constants for these conversions are listed in
Table 1.
At low water concentrations, hydrolysis of the hydrazine (A
→ B) is fast, and only oxidation of the alcohol (B → D) is
observed, as previously explained. There is an apparent small
decrease in the B → D oxidation rate as the water
concentration increases, and it appears that the rates remain
relatively constant at higher concentrations of Cu(II). At
roughly 5% added water, oxidation of the alcohol (A → C)
begins to compete with the hydrazine hydrolysis (A → B), as
the peak at 460 nm is seen to grow in. For 1:2 5a/Cu(II) in the
95:5 acetonitrile/water mixture (Figure S10D), both the
oxidation of A to C (growth of the peak at 460 nm) and the
hydrolysis of C to D (subsequent decrease in the peak at 460
nm) are observed. At even higher concentrations of water, the
conversion of C to D is not observable on this time scale,
although conversion to D does take place over numerous hours.
Oxidation of A to C appears to be more affected by the water
concentration than the oxidation of B to D. Since intermediate
C in Scheme 2 is identical to compound 4a, the rate data for
hydrolysis of the hydrazine by Cu(II) for 4a were also
determined (Figures S13 and S14). The rate decreases as
before with the addition of more water, and the rates are
approximately the same for the last two columns in the table
and are both lower than the rate measured for the oxidation of
A to C. We also tested the reaction of compound 6 with
Cu(II); however, there was no change in the reaction, even
after long periods of time, indicating that although intermediate
B and compound 6 are isomers, they have very different
oxidation potentials, as was previously reported.24We observed
that the rates for both hydrolysis and oxidation decrease with
added water, as shown in Table 1. Because of the poor
solubility of 4a in pure water, we synthesized compound 4c
Kinetics. We have shown qualitatively that altering the
amount of water has a profound effect not only on the initial
yellow- and green-emissive species when there is a relatively
small amount of water in the solvent (Figure 4) but also on the
relative rates of hydrolysis versus oxidation in this system when
larger amounts of water are added (Figures 5 and 6). We have
determined that the initial yellow fluorescence in very dry
acetonitrile decays by first-order kinetics and is independent of
the oxygen concentration. For N2, O2, and air-saturated
solutions of 5a + Cu(II), the initial yellow emission intensities
are identical, and the decay rate constants are also
approximately equal (1.20
0.12 min−1; Figure 4 left and
Figure S7). The UV/vis changes under very dry conditions
remain very similar to Figure 5 (Figure S8), indicating that the
reaction proceeds through the same basic steps, and plotting
the natural logarithm of the absorbance change versus time
gives a straight line and a rate constant of 0.53 min−1 (Figure
S9). In the presence of 0.25% water (∼0.14 M), the
disappearance of the green emission at band ∼490 nm also
follows first-order kinetics with a rate constant of ∼1.5 min−1 in
air (Figure 4 right and Figure S7). However, the initial emission
intensity of the green fluorescence does depend on the
2958
dx.doi.org/10.1021/ic402723c | Inorg. Chem. 2014, 53, 2953−2962

Inorganic Chemistry
Article
containing a carboxylic acid group that can form a carboxylate
salt and is soluble in water. As expected, 4c in the presence of
Cu(II) in a 90:10 H2O/acetonitrile mixture only very slowly
converts to 1 over the course of several hours (Figure S15).
Cyclic Voltammetry. Cyclic voltammetry data for com-
pounds 1, 4a, 5a, and 6 were recorded using tetrabutylammo-
nium perchlorate (TBAP) as the electrolyte and CH2Cl2 as the
solvent (Figure S16). Whereas compound 1 shows two one-
electron reductions at −1.07 and −1.51 V, which are
characteristic of the individual carbonyl reductions of the
unsymmetric 1,8-disubstituted anthraquinone,27compound 4a
shows only a single reversible one-electron reduction at −1.39
V and two irreversible oxidations at 0.95 and 1.63 V. We
attribute the reversible peak to the reduction of the lone
external carbonyl group in 4a and the irreversible peaks to the
hydrazine amine/imine oxidations. The reduction peak is
missing for the reduced compound 5a as no carbonyl is present,
whereas the irreversible amine oxidation potentials were
observed at similar potentials (0.93 and 1.40 V). Compound
6 shows one irreversible one-electron reduction peak for the
external carbonyl group, and neither 5a nor 6 have observable
oxidation peaks for the alcohol groups.
Stoichiometry. The reaction is not catalytic [addition of
5% Cu(II) ion produces only a small change in the total
concentration of 5a after several hours], and increasing the
copper concentration from 1 to 2 equiv does not appear to have
a significant effect on the rate (Table 1). We do know that
when the copper(II) salt with perchlorate (a noncoordinating
anion) is used, crystalline [Cu(CH3CN)4](ClO4) is isolated
from the reaction mixture.28To the best of our knowledge, the
reaction of compound 5a with Cu(II) (with noncoordinating
anions) involves hydrolysis and oxidation to form 1 followed by
slow oxidation of the hydrazine itself. Preliminary evidence
from preparatory-scale reactions of Cu(II) with free phenyl-
hydrazine or 2,4-dinitrophenylhydrazine in acetonitrile con-
firms that these hydrazines are oxidized, precipitating [Cu-
(CH3CN)4]+ and forming other phenolic byproducts as
determined by GC/MS (eq 1). This will be the subject of a
separate paper.
Figure 7. Stern−Volmer plot of 1 × 10−5 M 5a with 2.0 equiv of
added Cu(II) at 488 nm under pure N2, air, and pure O2 conditions.
Triangles: steady-state F0/F data. Squares: Lifetime data (τ0/τ).
dynamic quenching are involved. Lifetime data were also
collected for the same reaction (Figure S19 and Table S2), and
phase plane methods30 and curve simulations29for a single
exponential decay gave similar results. Plotting the lifetime data
according to the expression τ0/τ = 1 + KD[Q] produced the
dynamic component alone, and the slope gave the value KD ≈
250 M−1 (Figure 7). Substitution of KD allowed the calculation
of the static component as KS ≈ 300 M−1, which is of roughly
equal magnitude to the dynamic quenching portion. Also, the
bimolecular quenching constant (kq) is given by the expression
KD = kqτ0, from which the value kq = 1.2 × 1010 M−1 s−1 is
obtained, in excellent agreement with the diffusion-limited
bimolecular quenching constant for dioxygen. The lifetime of
the yellow emission at 560 nm was also determined to be short
(2.6 ns) and independent of the water concentration (Figure
S20 and Table S3).
DISCUSSION
■
Corey and Knapp first reported Cu(II)-assisted hydrolysis of
imines in the 1970s,31 and Rezende and co-workers reported
that chelated structures greatly enhance the rate of hydrolysis of
imines by stabilizing the metal-associated transition state.32 For
5a, the polyether macrocycle that is positioned below the
hydrazone is in an ideal position to chelate/coordinate the
Cu(II) cation, including association with the imine nitrogen,
which would increase the positive charge on the imine carbon
though an inductive effect and promote attack by water on the
imine carbon to yield the hydrolysis products. Only recently has
hydrolysis of imines been employed in fluorescent chemo-
dosimeters to monitor Cu(II) concentration, including our own
previously reported work.24,33
5a + Cu(II) + O2 → 1 + Cu(I) + phenolics
(1)
This is not the case when the counteranion is chloride. Then
the product recovered after reaction of 5a with Cu(II) is 1·
CuCl2, which can be isolated in crystalline form. Here the
coordinating halogens reduce the oxidation potential of the
copper metal center but still allow hydrolysis.
Role of Oxygen. We investigated the quenching behavior
of oxygen for the green emission at 490 nm and determined
that oxygen has both dynamic and static quenching
components. When the solution is degassed with nitrogen
prior to the addition of Cu(II), the green emission intensity
increases over 10-fold compared with the intensity in air
(Figure S17). Under pure oxygen conditions, the opposite
occurs: the green emission intensity is significantly quenched
and quickly decreases over time (Figure S18). Figure 7 shows
upward curvature of a three-point Stern−Volmer plot of the
steady-state green emission intensity (488 nm) versus dissolved
oxygen concentration in acetonitrile under inert, air, and pure
oxygen conditions. This curvature follows a quadratic relation-
ship: F0/F = (1 + KS[Q])(1 + KD[Q]), where F0 is the
fluorescence intensity in the absence of quencher and KS and
KD are the static and dynamic quenching constants,
respectively.29 This indicates that both static quenching and
Normally an increase in the amount of water would promote
hydrolysis of imines; however the hydrolysis rate is also highly
dependent on the hydration shell around the Cu(II) ion. We
have observed that as very large amounts of water are added to
the acetonitrile solvent, the rate-determining step switches from
oxidation (B → D; i.e., hydrolysis of A to B is fast) to
hydrolysis (A → C; hydrolysis is slowed). We expect that as
more water is added, the equilibrium is shifted to a fully
hydrated Cu(II) coordination sphere, preventing direct
coordination with the hydrazone imine nitrogen and slowing
the hydrolysis. From Table 1, the oxidation of the alcohol (B →
D) is also kinetically slower than oxidation of the alcohol (A →
C). Under dry conditions, this may be due to initial binding of
the copper cation by macrocycle 5a, promoting very rapid
hydrolysis to produce intermediate B, which, because of the
more electronegative carbonyl group, has a higher oxidation
potential than A. We see this also from compound 6, which is
2959
dx.doi.org/10.1021/ic402723c | Inorg. Chem. 2014, 53, 2953−2962

Inorganic Chemistry
Article
spectrophotometer. Absorbance data were collected using an HP
8452A diode array spectrophotometer or a Varian Cary 50
spectrometer. Emission studies were conducted using a SPEX
Fluoromax fluorimeter. Mass spectrometry was conducted using a
Varian 500-MS IT electrospray ionization (ESI) mass spectrometer.
Cyclic voltammograms were recorded using a CH Instruments 660
electrochemical workstation. Melting points were determined using
open capillaries and are uncorrected. The synthesis of 1 has previously
been reported.27Copper(II) perchlorate trihydrate was purchased from
Aldrich and used without further purification.
Synthesis of 1,8-Oxybis(ethyleneethoxyethyleneethoxy)-9-
(phenylazo)-10-anthraquinone (4a). Phenylhydrazine (130 mg,
1.2 mmol) was suspended in 6 mL of methanol, and 0.4 mL of
concentrated H2SO4 was added. To the resulting clear solution was
added 1,8-oxybis(ethyleneethoxyethyleneethoxy)-9,10-anthraquinone
(1) (100 mg, 0.25 mmol) dissolved in 5−10 mL of methanol. The
resulting solution was stirred for 0.5 h and mixed with 100 mL of
distilled water, affording an orange solid, which was filtered off, washed
several times with distilled water, and air-dried. The compound was
purified on a silica gel column using a 98:2 methylene chloride/
methanol mixture. Yield: 100 mg (80%). Mp: 194−197 °C. Anal.
Calcd for C28H28N2O6: C, 68.84; H, 5.78; N, 5.73%. Found: C, 68.86;
H, 5.74; N, 5.68%. 1H NMR (CDCl3 at 25 °C): δ 3.06−4.40 (m, 16H,
CH2−O in polyether chain); 6.85−7.94 (m, 13H, ArH from
anthraquinone + phenyl group); 9.02 (bs, 1H, NH). 13C NMR
(CDCl3 at 25 °C): δ 69.28; 69.37; 69.72; 69.87; 69.98; 70.36; 70.53;
70.69; 114.02; 117.22; 117.96; 119.06; 120.92; 121.06; 127.80; 128.86;
129.89; 133.58; 134.75; 144.72; 153.58; 155.72; 184.78. IR: 1644 cm−1
(carbonyl); 3270 cm−1 (−NH−).
Synthesis of 1,8-Oxybis(ethyleneethoxyethyleneethoxy)-9-
(2,4-dinitrophenylazo)-10-anthraquinone (4b). 2,4-Dinitrophe-
nylhydrazine (238 mg, 1.2 mmol) was suspended in 6 mL of methanol
and mixed with 0.4 mL of concentrated H2SO4. To the resulting clear
solution was added 1 (100 mg, 0.25 mmol) dissolved in 5−10 mL of
methanol. A red-orange solid precipitated immediately, and the
mixture was cooled overnight and filtered. The solid was further
purified on a silica gel column using a 98:2 methylene chloride/
methanol mixture. Yield: 125 mg (85%). Mp: 218−220 °C. Anal.
Calcd for C28H26N4O10: C, 58.13; H, 4.53; N, 9.68%. Found: C, 58.21;
H, 4.49; N, 9.59%. 1H NMR (CDCl3 at 25 °C): δ 3.26−4.49 (m, 16H,
CH2−O in polyether chain); 7.19−7.91 (m, 8H, ArH from
anthraquinone + DNP); 9.08 (bs, 1H, NH); 11.55 (s, 1H, ArH
from DNP). 13C NMR (CDCl3 at 25 °C): δ 68.03; 68.74; 69.10;
69.35; 69.53; 69.65; 70.61; 70.68; 109.73; 115.99; 117.44; 117.65;
119.41; 120.45; 123.02; 129.26; 129.74; 130.48; 132.01; 134.06;
134.83; 138.32; 144.66; 153.75; 155.49. IR: 1674 cm−1 (carbonyl);
3310 cm−1 (−NH−).
Synthesis of 1,8-Oxybis(ethyleneethoxyethyleneethoxy)-9-
(4-carboxylphenylazo)-10-anthraquinone (4c). In a round-
bottom flask, 15 mL of water and 5 mL of concentrated H2SO4
were added, followed by 4-hydrazinobenzoic acid (0.478 g, 3.14
mmol). The solution was heated to 65 °C and stirred for 15−20 min
to give a clear solution. 1 (0.500 g, 1.26 mmol) was added at the same
temperature, and the solution was stirred for 10 min, after which the
temperature was increased to 90 °C. This temperature was maintained
for 12 h, followed by suction filtration of the hot solution and washing
of the solid with hot water (2 × 50 mL). The product was dried and
purified by recrystallization from hot acetonitrile. Yield: 0.350 g (52%).
Mp: 247−249 °C. Anal. Calcd for C29H27N2NaO8: C, 62.81; H, 4.91;
N, 5.05%. Found: C, 62.26; H, 5.39; N, 5.09%. ESI-MS: calcd for
resistant to oxidation by Cu(II) and dioxygen. Ultimately the
different hydrolysis and oxidation rates under wet and dry
conditions have value to synthetic chemists who wish to
selectively oxidize or hydrolyze products with the same types of
functional groups.
It is likely that the initial yellow emission observed when 5a
and Cu(II) are combined is due to the complexation of Cu(II)
with 5a, inducing an ICT photophysical change, as previously
observed for these types of macrocycles.22,23,25,26 The presence
of slightly more water, which results in the sudden appearance
of the green emission, may be due to simple differences in the
coordination shell of the copper ion involving substitution of
acetonitrile by water. Evidence to support this comes from the
respective rate constants for the decay of the yellow and green
emission intensities compared with the corresponding UV−vis
changes under the same conditions. Both the yellow and green
emission bands disappear at ∼20−30 times greater rates than
the corresponding UV/vis kinetic changes, indicating that the
fluorescence changes likely follow the rate of hydrazine
hydrolysis while the UV/vis changes follow the oxidation of
the alcohol (intermediate B) to the final product. From the
Stern−Volmer fluorescence data, the static quenching observed
is likely due to complexation of oxygen with a copper cation
ligated in some fashion to the green-fluorescent intermediate,
which is consistent with formation of a metal complex
promoting luminescence in the first place. Site-specific damage
of DNA by phenylhydrazine in the presence of Cu(II)
presumably involves oxygen-derived active species (copper−
oxygen complexes) implicated in the mechanism, similar to the
results presented here.34 Work to identify the phenylhydrazine
byproducts is ongoing and will be reported elsewhere.
CONCLUSION
■
We have demonstrated that selective imination of the internal
carbonyl of 1 occurs upon the addition of hydrazines, as
opposed to imination at the external carbonyl as previously
observed with ketamine formation involving transition-metal-
catalyzed imination. The resulting phenylhydrazone produces a
selective chemodosimeter (irreversible) sensitive only to
Cu(II), proceeding though two intense fluorescence inter-
mediates that are themselves water-dependent. Decomposition
of hydrazones via two distinct pathways in the presence of
Cu(II)/O2 is also dependent on larger water content, which
leads ultimately to the same product. This chemistry has
consequences for synthetic applications, depending on whether
hydrolysis or oxidation is the desired outcome. Under dry
conditions, the hydrazone is initially cleaved, followed by
oxidation of the alcohol. Under wet conditions, the opposite
occurs. Of particular interest is that the intermediate green
emission under dry conditions has both dynamic and static
quenching components with dioxygen. The static quenching
component suggests that the green-fluorescent intermediate
involves complexation of dioxygen with a copper metal center
coordinated to the fluorescent macrocycle involved in the slow
oxidation of the secondary alcohol.
protonated 4c 533.56, found 533.3; calcd for sodiated 4c 555.5, found
1
555.3. H NMR (CDCl at 25 °C): δ 3.14−3.22 (m, 1H, CH2−O,
3
EXPERIMENTAL SECTION
General Procedures. Elemental analyses were conducted using an
Exeter CE-440 elemental analyzer. Elemental analysis results are
reported for samples and copper perchlorate previously dried in an
polyether); 3.29−3.49 (m, 5H, CH2−O, polyether); 3.63−3.69 (m,
2H, CH2−O in polyether chain); 3.77−4.65 (m, 8H, CH2 −O in
polyether chain);7.37−7.52 (m, 5H, ArH from anthraquinone +
phenyl ring); 7.63−7.71 (m, 2H, ArH from anthraquinone); 7.78−
7.83 (m, 1H, ArH from phenyl ring); 7.94−7.97 (d, 2H, ArH from
anthraquinone); 9.61 (s, 1H, NH). IR: 1648 cm−1 (carbonyl); 3243
cm−1(−NH−).
■
1
Abderhalden flask at 100 °C to minimize the moisture content. H
NMR spectra (200 MHz) were obtained at room temperature in
CDCl3. Infrared spectra were obtained using a Bruker Alpha
2960
dx.doi.org/10.1021/ic402723c | Inorg. Chem. 2014, 53, 2953−2962

Inorganic Chemistry
Article
Synthesis of 1,8-Oxybis(ethyleneethoxyethyleneethoxy)-9-
(phenylazo)-10-hydroxyanthrone (5a). Method A. 4a (200 mg,
0.41 mmol) was suspended in 20 mL of 95% ethanol and mixed with
NaBH4 (0.016 g, 0.41 mmol). The solution was stirred for 2 h, mixed
with 100 mL of distilled water, and extracted with 50 mL of methylene
chloride. The methylene chloride solution was dried with anhydrous
sodium sulfate and filtered, and most of the solvents were evaporated
under reduced pressure. Diethyl ether was diffused into the methylene
chloride solution, yielding a cream-colored solid.
ACKNOWLEDGMENTS
■
The authors thank NSF-EPSCOR (EPS-0554609) and the
South Dakota Governor’s 2010 Initiative for financial support
and the purchase of a Bruker SMART APEX II CCD
diffractometer. We also thank Aravind Baride and Dr. P.
Stanley May at the University of South Dakota for nanosecond
lifetime measurements.
Method B. 4a (200 mg, 0.41 mmol) was suspended in 20 mL of
THF under a N2 atmoshphere and mixed with LiAlH4 (0.016 g, 0.41
mmol). The solution was stirred for 3 h, slowly mixed with 100 mL of
saturated ammonium chloride solution, and extracted with 50 mL of
methylene chloride. The methylene chloride solution was dried with
anhydrous sodium sulfate and filtered, and most of the solvents were
evaporated under reduced pressure. Diethyl ether was diffused into the
methylene chloride solution, yielding a cream-colored solid.
REFERENCES
■
(1) Domaille, D. W.; Que, E. L.; Chang, C. J. Nat. Chem. Biol. 2008,
4, 168.
(2) Barceloux, D. G.; Barceloux, D. Clin. Toxicol. 1999, 37, 217.
(3) (a) Que, E. L.; Domaille, D. W.; Chang, C. J. Chem. Rev. 2008,
108, 1517. (b) Gaggelli, E.; Kozlowski, H.; Valensin, D.; Valensin, G.
Chem. Rev. 2006, 106, 1995.
Melting point and spectroscopic characterization revealed that both
methods yielded 5a. Yield: 150 mg (70%). Mp: 145−148 °C. Anal.
Calcd for C28H30N2O6: C, 68.59; H, 6.11; N, 5.71%. Found: C, 68.35;
H, 6.15; N, 5.48%. ESI-MS: calcd for protonated 5a 491.28, found
491.30. 1H NMR ([CD3]2SO at 25 °C): δ 3.08−4.35 (m, 16H, CH2−
O in polyether chain); 5.29−5.33 (d, 1H, anthrone); 5.75 (s, 2H,
solvated CH2Cl2); 6.35−6.39 (d, 1H, OH); 6.72−7.41 (m, 11H, ArH
from anthraquinone + phenyl group); 9.01 (bs, 1H, NH). IR: 3313
cm−1 (−NH−), 3372 cm−1 (−OH). No carbonyl stretching was
observed.
Crystallography. X-ray-quality crystals were obtained by diffusing
diethyl ether into methylene chloride/methanol solutions. Crystallo-
graphic data were collected at 293 K (4b) or 100 K (5a) using Mo Kα
radiation. Cell constants were determined after integration from
typically more than 9000 reflections.35 Structures were solved by direct
methods using SIR9736 and refined using SHELXL-97.37 Data
reduction and refinement were completed using the WinGX suite of
crystallographic software.38 All non-hydrogen atoms were refined with
anisotropic displacement parameters, and all hydrogen atoms were
placed in ideal positions and refined as riding atoms with relative
isotropic displacement parameters, with the exception of hydrogen-
bonded protons, which were found from the difference map. Table S1
in the Supporting Information lists additional crystallographic and
refinement information.
The space group P21/c did provide a solution for 4b; however, the
disorder of the anthraquinone macrocycle could not be satisfactorily
modeled to produce a reasonable R value. The space group Pc resulted
in two different macrocycles within the asymmetric unit with different
ring conformations, which were modestly restrained using five C−O
and C−C DFIX instructions, yielding a satisfactory result. Low-
temperature data for 4b improved the thermal parameters but left
higher R values for both the P21/c and Pc solutions. An absolute
configuration for this chiral molecule could not be determined. Three
neighboring anthraquinone carbon atoms in the structure of 5a were
essentially two-dimensional and were refined as isotropic only.
(4) Flemming, C.; Trevors, J. Water, Air, Soil Pollut. 1989, 44, 143.
(5) Chu, L.; Liu, D.; Wang, Y.; Li, Y.; Liu, H. Chin. J. Appl. Ecol. 2004,
15, 119.
(6) Schultheis, A. S.; Sanchez, M.; Hendricks, A. C. Hydrobiologia
1997, 346, 85.
(7) EPA National Primary Drinking Water Regulations. http://water.
epa.gov/drink/contaminants/index.cfm#1 (accessed Dec 4, 2012).
(8) Quang, D. T.; Kim, J. S. Chem. Rev. 2010, 110, 6280.
(9) Valeur, B. Molecular Fluorescence: Principles and Applications;
Wiley-VCH: Weinheim, Germany, 2001.
(10) (a) Chen, C.; Wang, R.; Guo, L.; Fu, N.; Dong, H.; Yuan, Y.
Org. Lett. 2011, 13, 1162. (b) Wang, H. H.; Xue, L.; Fang, Z. J.; Li, G.
P.; Jiang, H. New J. Chem. 2010, 34, 1239. (c) Zhao, Y.; Zhang, X. B.;
Han, Z. X.; Qiao, L.; Li, C. Y.; Jian, L. X.; Shen, G. L.; Yu, R. Q. Anal.
Chem. 2009, 81, 7022. (d) Ajayakumar, G.; Sreenath, K.; Gopidas, K.
́
R. Dalton Trans. 2009, 1180. (e) Zheng, Y.; Gattas-Asfura, K. M.;
Konka, V.; Leblanc, R. M. Chem. Commun. 2002, 2350. (f) De Santis,
G.; Fabbrizzi, L.; Licchelli, M.; Mangano, C.; Sacchi, D.; Sardone, N.
Inorg. Chim. Acta 1997, 257, 69. (g) Corradini, R.; Dossena, A.;
Galaverna, G.; Marchelli, R.; Panagia, A.; Sartor, G. J. Org. Chem. 1997,
62, 6283.
(11) (a) Hennrich, G.; Walther, W.; Resch-Genger, U.;
Sonnenschein, H. Inorg. Chem. 2001, 40, 641. (b) Martínez, R.;
Zapata, F.; Caballero, A.; Espinosa, A.; Tar
Lett. 2006, 8, 3235.
́
raga, A.; Molina, P. Org.
(12) (a) He, G.; Zhang, X.; He, C.; Zhao, X.; Duan, C. Tetrahedron
2010, 66, 9762. (b) Zhou, Y.; Wang, F.; Kim, Y.; Kim, S. J.; Yoon, J.
Org. Lett. 2009, 11, 4442. (c) Kim, Y. R.; Kim, H. J.; Kim, J. S.; Kim, H.
Adv. Mater. 2008, 20, 4428. (d) Lee, M. H.; Kim, H. J.; Yoon, S.; Park,
N.; Kim, J. S. Org. Lett. 2008, 10, 213. (e) Xiang, Y.; Tong, A.; Jin, P.;
Ju, Y. Org. Lett. 2006, 8, 2863.
(13) (a) Zhang, Y. J.; He, X. P.; Hu, M.; Li, Z.; Shi, X. X.; Chen, G. R.
Dyes Pigm. 2011, 88, 391. (b) Wen, Z.-C.; Yang, R.; He, H.; Jiang, Y.-
B. Chem. Commun. 2006, 106. (c) Singh, A.; Yao, Q.; Tong, L.; Still,
W. C.; Sames, D. Tetrahedron Lett. 2000, 41, 9601. (d) Ambrosi, G.;
Ciattini, S.; Formica, M.; Fusi, V.; Giorgi, L.; Macedi, E.; Micheloni,
M.; Paoli, P.; Rossi, P.; Zappia, G. Chem. Commun. 2009, 7039.
(e) Amendola, V.; Bergamaschi, G.; Buttafava, A.; Fabbrizzi, L.;
Monzani, E. J. Am. Chem. Soc. 2010, 132, 147.
ASSOCIATED CONTENT
■
S
* Supporting Information
(14) Zhou, Z.; Fahrni, C. J. J. Am. Chem. Soc. 2004, 126, 8862.
(15) Lu, D.; Lei, J.; Tian, Z.; Wang, L.; Zhang, J. Dyes Pigm. 2012, 94,
239.
(16) Basa, P. N.; Bhowmick, A.; Schulz, M. M.; Sykes, A. G. J. Org.
Chem. 2011, 76, 7866.
NMR, MS, and crystallographic characterization data and
kinetic, cyclic voltammetric, and spectroscopic data. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(17) Basa, P. N.; Bhowmick, A.; Medicine Horn, L.; Sykes, A. G. Org.
Lett. 2012, 14, 2698.
AUTHOR INFORMATION
■
(18) (a) Xu, Z.; Baek, K. H.; Kim, H. N.; Cui, J.; Qian, X.; Spring, D.
R.; Shin, I.; Yoon, J. J. Am. Chem. Soc. 2010, 132, 601. (b) Wang, Q.;
Xie, Y.; Ding, Y.; Li, X.; Zhu, W. Chem. Commun. 2010, 46, 3669.
(c) Shao, J.; Yu, X.; Lin, H.; Lin, H. J. Mol. Recognit. 2008, 21, 425.
(19) Allen, J. R.; Cynkowski, T.; Desai, J.; Bachas, L. G.
Electroanalysis 1992, 4, 533.
Corresponding Author
*E-mail: asykes@usd.edu.
Notes
The authors declare no competing financial interest.
2961
dx.doi.org/10.1021/ic402723c | Inorg. Chem. 2014, 53, 2953−2962

Inorganic Chemistry
Article
(20) Kaneda, T.; Umeda, S.; Ishizaki, Y.; Kuo, H. S.; Misumi, S.; Kai,
Y.; Kanehia, N.; Kasai, N. J. Am. Chem. Soc. 1989, 111, 1881.
(21) Kadarkaraisamy, M.; Caple, G.; Gorden, A. R.; Squire, M. A.;
Sykes, A. G. Inorg. Chem. 2008, 47, 11644.
(22) Kadarkaraisamy, M.; Sykes, A. G. Inorg. Chem. 2006, 45, 779.
(23) Kadarkaraisamy, M.; Sykes, A. G. Polyhedron 2007, 26, 1323.
(24) Basa, P. N.; Sykes, A. G. J. Org. Chem. 2012, 77, 8428.
(25) Young, V. G., Jr.; Quiring, H. L.; Sykes, A. G. J. Am. Chem. Soc.
1997, 119, 12477.
(26) Chemosensors: Principles, Strategies, and Applications; Wang, B.,
Anslyn, E. V., Eds.; John Wiley & Sons: Hoboken, NJ, 2011.
(27) Delgado, M.; Gustowski, D. A.; Yoo, H. K.; Gatto, V. J.; Gokel,
G. W.; Echegoyen, L. J. Am. Chem. Soc. 1988, 110, 119.
(28) Csoregh, I.; Kierkegaard, P.; Norrestam, R. Acta Crystallogr.
1975, B31, 314.
(29) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 3rd ed.;
Springer Scientific: New York, 2006; p 277.
(30) Adamson, A. W.; Demas, J. N. J. Phys. Chem. 1971, 75, 2463.
(31) Corey, E. J.; Knapp, S. Tetrahedron Lett. 1976, 17, 3667.
(32) Machado, V. G.; Nascimento, M. G.; Rezende, M. C. J. Braz.
Chem. Soc. 1993, 4, 76.
(33) Kim, M. H.; Jang, H. H.; Yi, S.; Chang, S. K.; Han, M. S. Chem.
Commun. 2009, 4838.
(34) Yamamoto, K.; Kawanishi, S. Chem. Res. Toxicol. 1992, 5, 440.
(35) SAINT, version 6.1; Bruker Analytical X-ray Systems: Madison,
WI, 1999.
(36) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.;
Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.;
Spagna, R. J. Appl. Crystallogr. 1999, 32, 115.
(37) Sheldrick, G. M. SHELX97: Programs for Crystal Structure
Analysis, release 97-2; Universitat Gottingen: Gottingen, Germany,
̈
̈
̈
1998.
(38) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837.
2962
dx.doi.org/10.1021/ic402723c | Inorg. Chem. 2014, 53, 2953−2962