FerriNaphth: A fluorescent chemodosimeter for redox active metal ions

Article abstract of DOI:10.1039/c000248h

FerriNaphth, a fluorescent chemodosimeter for Fe^III, has been prepared and characterized. The probe consists of a catechol ligand linked to a naphthalimide fluorophore by an aniline nitrogen linker. Upon exposure to Fe^III, the aminocatechol of FerriNaphth is oxidized to the corresponding quinone, which in its imine-one tautomer, is hydrolyzed to liberate a fluorescent aminonaphthalimide derivative. The fluorescence behavior is consistent with oxidation being promoted by metal coordination.

Full text of DOI:10.1039/c000248h

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
FerriNaphth: A fluorescent chemodosimeter for redox active metal ions†
Randy K. Jackson, Yu Shi, Xudong Yao and Shawn C. Burdette*
Received 11th January 2010, Accepted 15th February 2010
First published as an Advance Article on the web 23rd March 2010
DOI: 10.1039/c000248h
FerriNaphth, a fluorescent chemodosimeter for FeIII, has been prepared and characterized. The probe
consists of a catechol ligand linked to a naphthalimide fluorophore by an aniline nitrogen linker. Upon
exposure to FeIII, the aminocatechol of FerriNaphth is oxidized to the corresponding quinone, which in
its imine-one tautomer, is hydrolyzed to liberate a fluorescent aminonaphthalimide derivative. The
fluorescence behavior is consistent with oxidation being promoted by metal coordination.
compound.12-14 These probes integrate a fluorescent reporting
group with an analyte recognition moiety linked covalently to the
Introduction
Fluorescent probes for metal ions like CaII and ZnII have become
spirolactam amide nitrogen atom. Metal coordination mediates
widespread in biological imaging because of their unique ability
the ring opening reaction that generates the emissive isomer of the
to quantify changes in the temporal and spatial concentrations
fluorophore.
of these analytes passively.1,2 The design of probes that exhibit
Sensors that emit at a different wavelength in the apo and
fluorescent enhancement upon exposure to FeII/III and CuI/II
bound forms possess advantages in biological imaging applica-
remains a significant challenge because redox active metal ions
tions. Ratiometric sensors often involve internal charge transfer
provide multiple non-radiative decay pathways for the excited
(ICT) to induce wavelength shifts. Several naphthalimide derived
sensors for CuII take advantage of ICT to shift absorption and
emission wavelengths,15-18 suggesting that dyads comprised of these
fluorophores and metal-binding ligands are good candidates for
fluorescent probes. We hypothesized that oxidizing a catechol
ligand conjugated to a naphthalimide fluorophore would result
in an ICT state that would shift the absorption and emission
wavelengths sufficiently to efficiently detect FeIII. Naphthalimide
fluorophores are prototypical donor–acceptor systems, where
conjugating an electron pair donor with one of the imide carbonyls
leads to the maximum fluorescence output.19 We reasoned that an
aniline nitrogen linker would be the best candidate to couple the
naphthalimide with the catechol because it would be conjugated
to both the fluorophore and the metal receptor (Fig. 1).
states of fluorophores. As a result, probes that show fluorescence
enhancement upon interaction with transition metal ions remain
scarce.3,4 Recently, we reported the design and synthesis of a
fluorescent probe for redox active metal ions to begin addressing
this problem.5 In FerriBRIGHT, FeIII and CuII oxidize a catechol
appended to a BODIPY fluorophore into a quinone. The oxidation
of the catechol interrupts photoinduced electron transfer (PeT)
quenching of the fluorophore, which enhances the intensity of
the probe’s emission. In addition to our system, a reversible
quinone/hydroquinone switch was used to construct a redox
probe, which further demonstrates the ability to use the redox
behaviour of a tethered substrate to elicit a fluorescence response.6
The design element consists of a 1,4-quinone coupled to a BOD-
IPY fluorophore via a phenylene spacer. The BODIPY-quinone
dyad fluoresces weakly owing to PeT from the quinone moiety to
the fluorophore. After exposure to a reducing agent, however,
fluorescence increases because formation of the corresponding
hydroquinone interrupts the PeT process. Additional fluorescent
probes for biological oxidants have been reported that use similar
BODIPY-quinone systems.7,8
Many probes for redox active species possess the characteristics
of chemodosimeters, since the fluorescence response is irreversible.
While several strategies for constructing chemodosimeters for FeIII
and CuII have been demonstrated,9-11 the majority utilize transition
metal-induced ring opening of a weakly fluorescent spirolactam
Department of Chemistry, University of Connecticut, Storrs, CT 06269-
3060, USA. E-mail: shawn.burdette@uconn.edu
† Electronic supplementary information (ESI) available: NMR spectra for
new compounds. Absorption spectra for CuII oxidation of FerriNaphth in
CH3CN and CH3OH. Absorption and emission spectra for FeIII oxidation
of FerriNaphth in CH3OH. Absorption spectrum for the titration of
FerriNaphth with GaIII. Absorption spectrum of oxidized FerriNaphth
at different concentrations and corresponding Beer’s law plot. Absorption
spectra in CH3CN of various FeIII and CuII salts used in these studies. See
DOI: 10.1039/c000248h
Fig. 1 Design and signalling mechanism of FerriNaphth. In the absence
of metal ions, the lone pair on the aniline nitrogen participates in a
donor–acceptor resonance interaction with the phthalimide carbonyl
group. After oxidation with FeIII, the aniline also can engage in a resonance
interaction with the quinone carbonyl, which would lead to different
absorption and emission characteristics by an ICT mechanism.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 4155–4161 | 4155
©

View Article Online
Results and discussion
The desired fluorescent probe was prepared using a Pd-catalyzed
aryl amination between 4-bromo-N-butyl-1,8-naphthalimide (2)20
and 5-aminospiro(1,2-benzodioxole-2,1¢cyclohexane) (3)21 that
provided the precursor ligand in 58% yield after purification by col-
umn chromatography (Scheme 1). Removal of the cyclohexylidene
ketal with concentrated HCl provided FerriNaphth. The name
FerriNaphth is derived from the fluorophore (naphthalimide) and
the target analyte (ferric iron). FerriNaphth is a red crystalline
solid that is soluble in organic solvents, but only sparingly soluble
in aqueous solution. When dissolved in anhydrous acetonitrile,
FerriNaphth absorbs with a lmax at 441 nm (e = 16200 cm-1 M-1)
and displays weak emission with a lmax at 520 nm (U = 0.001).
Upon exposure of FerriNaphth to Fe(NO3)3 the absorption peak
at 441 nm erodes over a period of 3 min, with the concomitant
formation of a new peak at 368 nm (Fig. 2, top). An increase
in the emission intensity centered at 520 nm is observed after
the addition of H2O when excited at 400 nm (Fig. 2, bottom);
however, no significant changes in emission behavior are observed
under anhydrous conditions. Similar changes are observed in
the absorption spectra when CH3OH was used as a solvent;
however, the changes in emission are less pronounced. Using
the nitrate salt eliminates bands from FeCl3 and Fe(ClO4)3
that overlap the absorption peaks of FerriNaphth, but has
no significant impact on the fluorescence response. Absorption
bands blue-shift in typical ICT-based naphthalimide sensors,
but emission wavelengths do not red-shift.22 Since the changes
in the absorption spectra contradicted predicted ICT behavior,
the product of iron oxidation was interrogated. Analysis of
solutions containing the probe and FeIII by TLC revealed the
presence of a fluorescent compound and a bright red species that
decomposed upon solvent removal. The fluorescent compound
was isolated and identified as N-nbutyl-4-aminonaphthalimide
(7).19,23 The presence of 7 suggests that instead of an ICT system,
the fluorescence signal transduction mechanism of FerriNaphth
involves metal-promoted oxidation followed by a second chemical
transformation.
Fig. 2 Oxidation of 10 mM FerriNaphth in CH3CN with incremental
additions of Fe(NO3)3. Absorbance (top) measurements were recorded
after no additional changes were observed after each addition. Emission
spectra (bottom) were recorded over a period of 48 min after the addition
of 2 equivalents of Fe(NO3)3 and 1% H2O by volume. Excitation was
provided at 400 nm with an excitation slit width of 5.0 nm and an emission
slit width of 10 nm.
can exist in both a quinone (5) and an imine-one tautomer (6),
subsequently undergoes imine hydrolysis to provide the fluorescent
naphthalimide 7 (Scheme 2). In addition to this mechanistic
evidence, a charge transfer band at 562 nm corresponding to the
[Fe(ferrozine)3]4- complex forms when FerriNaphth is oxidized
by FeIII in the presence of ferrozine, a common FeII indicator
(Fig. 3). The catechol to quinone transformation requires a
2 electron oxidation, and the amount of ferrous iron present
after the oxidation corresponds to approximately 2 equivalents
of ferric iron reacting with every equivalent of FerriNaphth in
solution.
Aminoxyquinones like 5 are tautomeric red compounds,24 and
investigations revealed that the addition of water and 10 mol%
pyridyl p-toluenesulfonic acid accelerate the disappearance of
the red species and increase the intensity of emission from 7.
These observations suggest that oxidized FerriNaphth, which
Relatively few chemodosimeters have been designed where a
turn-on fluorescence signal results from hydrolysis of the weakly
fluorescent parent molecule.25-29 Typically these probes exploit
Scheme 1 Synthesis of FerriNaphth.
This journal is
4156 | Dalton Trans., 2010, 39, 4155–4161
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 2 Fluorescence transduction mechanism of FerriNaphth.
FeIII, the hydrolysis remains very slow even with the addition
of catalytic water and acid. Evaluation of the emission intensity
of FerriNaphth over time with FeIII in the presence of water,
CH3COOH, H3PO4 and HNO3 shows varying rates in the emission
increase, although hydrolysis requires a period of several hours and
remains incomplete even after several days. The slow fluorescence
response of FerriNaphth and the incompatibility of naphthalimide
fluorescence properties to aqueous conditions limit the potential
biological application of this probe, but represent a unique
signal transduction mechanism that could be adapted for such
purposes.
In order to confirm the fluorescence signaling mechanism
of FerriNaphth, we sought to confirm the identity of the
quinone/imine-one tautomeric intermediate. In thoroughly dried
CH3CN, the peak at 368 nm corresponding to oxidized Ferri-
Naphth remains stable for extended periods of time; however,
only 7 can be isolated when the solvent is removed. The
tautomeric species shows no appreciable fluorescence, which
suggests the increase in emission intensity in the fluorescence
studies can be attributed solely to 7. Recently, a naphthalimide-
based Zn2+ sensor has been reported that fluoresces at different
wavelengths in its two tautomeric forms.30 Since the interme-
diate remains stable in solution, tandem mass spectrometric
(MS/MS) analysis was performed (Fig. 4). In addition to de-
tecting a parent ion corresponding to a molecular weight of
2 mass units less than FerriNaphth, a fragmentation pattern
consistent with the step-wise loss of two CO units was also
observed. The combined experimental results suggest that catechol
oxidation followed by a hydrolysis of a tautomeric quinone/imine-
one correctly describes the mechanism of fluorescence
enhancement.
Fig. 3 Absorbance spectra of 10 mM FerriNaphth (CH3CN) in the
presence of 10 equivalents of FeIII or CuII and with either 10 or 20
equivalents of ferrozine or neocuproine respectively.
the Lewis acidity of a metal ion to promote the liberation of a
fluorophore that is quenched by the reactive bonding interactions.
In a probe that selectively responds to FeIII, Schiff bases from a
chelating ligand hydrolyze to release a highly fluorescent coumarin
fluorophore.26 While other fluorescent chemodosimeters for FeIII
based on oxidation chemistry and fluorophore isomerization have
been reported,9-11 there has been only one report of a system
utilizing both oxidation and hydrolysis to achieve an emission
enhancement in the presence of a redox active transition metal.10
In the squaramide hydroxamate system however, multiple products
from the redox process were observed and only a few that were
responsible for the emission enhancement were characterized
successfully.
To determine the extent of FerriNaphth’s ability to detect metal
oxidants, the probe was exposed to a variety of different oxidants at
several different concentrations. Cu(NO3)2 oxidizes FerriNaphth,
and the changes observed in the absorption spectra are similar
to those measured using FeIII. Oxidation of FerriNaphth using
stoichiometric amounts of CuII proceeds at rates similar to FeIII
although CuII has a standard reduction potential that is ~0.6 V
lower than FeIII. Since the oxidation is relatively rapid, it is
difficult to compare the rates of oxidation with the two metals
with standard UV-vis instrumentation. Stopped flow techniques
could elucidate more detailed kinetic information and may provide
a methodology to discriminate between the two metal ions
The emission intensity of 7 is polarity dependent, and has a
quantum yield of 0.74 in CH3CN.19 Repeating the fluorescence
assay in nonpolar solvents like THF and CH2Cl2 with 1% H2O
provides a more intense fluorescence signal. Experiments using
protic solvents like H2O and CH3OH only result in weak emis-
sion responses. Completely drying alcoholic solvents is difficult,
which explains why emission changes are observed in CH3OH
without the addition of exogenous water and why the absorption
spectra of the oxidized FerriNaphth erodes slowly over time.
Additional attempts to enhance the rate of the fluorescence
response from FerriNaphth were unsuccessful. While the initial
oxidation step is complete within minutes with stoichiometric
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 4155–4161 | 4157
©

View Article Online
of the probe into several unidentifiable fluorescent products
suggesting FerriNaphth is incompatible with especially strong
oxidants.
Since the rapid oxidation of FerriNaphth precludes examination
of coordination chemistry, the redox inert structural analog of
FeIII, GaIII, was employed for additional studies. In addition to
coordination chemistry, GaIII allows the examination of ICT states
produced by metal ion binding when FerriNaphth is not oxidized.
Titration of FerriNaphth with up to 20 equivalents of Ga(NO3)3
results in a slight red shifting (~7 nm) and a decrease in intensity
of the absorption band centered at 441 nm and an increase in the
absorption band at 360 nm. GaIII titrations were carried out with
the nitrate salt in MeOH because of limited solubility in CH3CN;
in addition, there are no readily available GaIII salts that are com-
patible with the procedures used to assess FerriNaphth oxidation.
When the FerriNaphth-GaIII species is excited at 450 nm a modest
2-fold increase in the emission intensity can be measured (Fig. 6).
The spectroscopic changes suggest GaIII binds to the catechol of
FerriNaphth generating an ICT state; however, the interaction
does not appear to be very strong. At higher concentrations of
Ga(NO3)3, evidence of FeriNaphth oxidation exists. Under acidic
conditions, nitrate is a good oxidizing agent. GaIII coordination
to FerriNaphth lowers the pKa of the coordinated phenols, which
generates the conditions necessary for nitrate to act as an electron
acceptor. Excess KNO3 under acidic conditions also slowly
oxidizes FerriNaphth. By adding additional acid to a solution
containing FerriNaphth and 30 equivalents of Ga(NO3)3, the slow
oxidation accelerates slightly. Fluorescence studies with GaIII were
carried out with 3 equivalents of Hunig’s base to differentiate
metal binding-based ICT states from ones resulting from oxida-
tion. While some evidence of complexation between GaIII and
FerriNaphth was acquired, the persistence of oxidation reactions
makes characterizing these weakly binding complexes difficult and
ambiguous.
Fig. 4 MS/MS analysis of oxidized FerriNaphth. The parent quinone
anion as well as peaks corresponding to the loss of CO units verify the
identity of the oxidation product.
by rate analysis. In a manner analogous to the FeIII/ferrozine
assay, oxidation of FerriNaphth with CuII in the presence of
neocuproine, a common indicator for CuI, results in a strong
band at 480 nm that is indicative of [Cu(neocuproine)]+ indicating
approximately 2 equivalents of CuII are consumed in the oxidation
(Fig. 3).
FerriNaphth does not oxidize in the presence of AgNO3, which
has a slightly more positive standard reduction potential than FeIII,
even at high concentrations of the metal ion. AgI does not coor-
dinate strongly with catechol ligands like those in FerriNaphth,
suggesting that coordination of the metal ion to ligand facilitates
the oxidation. This conclusion is supported further by the results
3-
of assays using Fe(CN)6 and [Co(NH3)5Cl]Cl2 as oxidants. Ex-
posure of FerriNaphth to 10 or more equivalents of coordinatively
saturated Fe(CN)63- shows no evidence of oxidation. FerriNaphth
slowly oxidizes in the presence of [Co(NH3)5Cl]Cl2 but the reaction
remains incomplete after 48 h (Fig. 5). Even though CoIII is a strong
oxidant, slow ligand exchange31 with [Co(NH3)5Cl]2+ results in
decreased rates of oxidation. Unlike the other common oxidants
screened, ceric ammonium nitrate causes the decomposition
Fig. 6 Titration of 10 mM FerriNaphth in CH3OH with Ga(NO3)3.
Spectra were taken at 20 mM increments up to 360 mM of Ga3+. Above
360 mM of Ga(NO3)3, nitrate mediated oxidation of FerriNaphth starts
to occur. Each spectrum was corrected for dilution by multiplying the
measured absorption and integrated emission intensity by the inverse
of the dilution factor. The emission intensity was integrated from 470
to 650 nm. Excitation was provided at 450 nm with an excitation slit
width of 5.0 nm and an emission slit width of 10.0 nm. Inset: changes
in integrated fluorescence emission with increasing concentrations of
Fig.
5
Oxidation of 10 mM FerriNaphth with 6 equivalents of
[Co(NH3)5Cl]Cl2 in CH3CN. The absorption spectra were automatically
recorded every hour for 48 h. Inset: changes in the absolute value of the
absorbance at 442 nm as a function of time.
Ga3+
.
4158 | Dalton Trans., 2010, 39, 4155–4161
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
4 H), 4.20 (t, J = 6.92, 2 H), 1.99-1.96 (m, 4 H), 1.78-1.69 (m,
6 H), 1.49-1.41 (m, 2 H), 1.29-1.25 (m, 1 H), m 1.01-0.96 (m,
3 H). 13C NMR (100 MHz, CDCl3) d 164.8, 164.2, 184.8, 184.3,
145.9, 133.9, 132.8, 131.5, 130.1, 126.2, 125.4, 121.2, 117.10,
112.6, 108.8, 108.1, 106.1, 40.2, 35.4, 34.9, 35.4, 34.9, 30.5,
25.5, 24.7, 23.4, 20.6, 14.1. IR (thin film, cm-1), 3349.2, 2935.9,
1684.79, 1640.91, 1581.7, 1535.4, 1429.5, 1390.4, 1360.99, 1344.6,
1240.7. HRMS (+ESI): Calcd. for M Na+.479.1947; Found,
479.1993.
Conclusions
In summary we have prepared and characterized a new fluorescent
chemodosimeter for metal based oxidants based on a tandem
oxidation–hydrolysis reaction of a catechol linked napthalimide
dyad. FerriNaphth has a weak emission that increases significantly
upon conversion into N-nbutyl-4-aminonaphthalimide in the
presence of FeIII and CuII. Non-coordinating metal oxidants and
redox inactive metal ions have only a minimal effect on the
absorption and emission properties of FerriNaphth. The absorp-
tion changes are similar to some recently described CuII sensors
that possess similar aniline components and are deprotonated
by metal ion coordination,15 but additional studies are needed
to make meaningful comparisons. The signaling mechanism has
been confirmed by isolating the fluorescent species, detecting
reduced FeII in the fluorescence assay, and characterizing the
oxidized intermediate by mass spectroscopy. Future efforts will
involve designing a chemodosimeter with an ICT mechanism by
stabilizing the linkage between catechol and fluorophore to inhibit
hydrolysis.
2-Butyl-6-(3,4-dihydroxy-phenylamino)-benzo[de]isoquinoline-
1,3-dione (FerriNaphth, 1). Concentrated hydrochloric acid
(16 mL) was added to a stirred suspension of 5 (300 mg, 0.65 mmol)
in absolute ethanol (25 mL). The mixture was refluxed for 90 min
followed by the removal of half of the solvent by vacuum.
Water was added to precipitate a red solid. This was filtered
and washed (2 ¥ 5 mL) with water. The solid was collected and
chromatographed on silica using 100% dichloromethane to yield a
red crystalline solid (185 mg, 75%).TLC Rf 0.2, dichloromethane.
1H NMR (400 MHz, DMSO) d 9.18 (s, 2 H), 8.97 (s, 1 H), 8.82
(d, J = 8.7 Hz, 1 H), 8.48 (d, J = 7.1 Hz, 1 H), 8.22 (d, J =
8.3 Hz, 1 H), 7.76 (t, J = 8.0 Hz, 1 H), 6.96 (d, J = 8.5 Hz, 1
H), 6.84 (d, J = 8.2 Hz, 1 H), 6.78, (s, 1 H), 6.67 (d, J = 8.4 Hz,
1 H), 4.04 (t, J = 7.5 Hz, 2 H), 1.63 (sextet, J = 6.8 Hz, 2 H),
1.34 (septuplet, J = 7.4, 2 H), 0.94 (t, J = 7.2, 3 H). 13C NMR
(400 MHz, DMSO) d 164.4, 163.5, 150.4, 146.7, 143.9, 134.4,
131.7, 131.6, 130.3, 129.5, 125.4, 122.6, 121.3, 116.7, 116.4, 113.29,
109.9, 107.1 IR (thin film, cm-1), 3390.07, 3290.8, 2956.55, 2925.2,
1685.5, 1674.5 1638.3, 1612.1, 1580.5, 1565.4, 1542.1, 1530.1,
1520.6. HRMS (+ESI): Calcd. for M Na+.399.1321; Found,
399.1352.
Experimental
Synthesis
General procedures. All materials listed below were of research
grade or of spectrograde of highest purity available from TCI
America or Acros Organics. Dichloromethane (CH2Cl2), toluene
(C6H5CH3) and tetrahydrofuran (THF) were sparged with argon
and dried by passage through a Seca Solvent Purification Sys-
tem. Chromatography and TLC were performed on silica (230–
400 mesh) obtained from Silicycle. TLCs were developed with
mixtures of EtOAc/hexanes or CH2Cl2 unless otherwise stated.
4-Bromo-N-butyl-1,8-naphthalimide (2)20 and 5-aminospiro(1,2-
benzodioxole-2,1-cyclohexane) (3)21 were prepared according to
Spectroscopy
General methods. All solutions were prepared with spec-
trophotometric grade solvents. FerriNaphth was dissolved in
DMSO to make a 10 mM stock solution. A 3 mL aliquot of
FerriNaphth stock was placed in a quartz cuvette and diluted with
3 mL of CH3CN to provide a 10 mM solution for spectroscopy
unless otherwise noted. Stock solutions (10 mM) of each metal
ion were prepared in 3/2 EtOH–CH3CN unless otherwise noted.
Absorption spectra were recorded on a Cary 50 UV-visible
spectrophotometer operated by a PC equipped with Pentium-
IV processor. Spectra were taken at 25 ◦C in 1-cm path length
cuvettes. Fluorescence spectra were recorded on a Hitachi F-
4500 spectrophotometer operated by a PC equipped with a
Pentium-IV processor, running the FL solutions 2.0 software. A
150 W Xe lamp operating at 5 A provided excitation. Spectra
were acquired in a quartz cuvette with a 1-cm path length.
Slit widths are 5 nm for excitation and 10 nm for emission
with a photomultiplier tube voltage of 700 V, unless stated
otherwise.
literature procedures. H and 13C were recorded using a Bruker
1
400 MHz NMR instrument. Chemical shifts are reported in
ppm relative to tetramethylsilane. IR spectra were recorded
on a Nicolet 205 FT-IR Instrument and the samples were
prepared as KBr pellets. High resolution mass spectra were
recorded at the University of Connecticut mass spectrometry
facility using a micromass Q-Tof-2TM operating in positive
mode.
2-Butyl-6-spiro
(1,3-benzodioxole-2,1¢-cyclohexane-phenyl-
amino)-benzo[de]isoquinoline-1,3-dione (4). A Schlenk tube was
charged with 600 mg (1.18 mmol) of 2, 408 mg (2.0 mmol) of 3,
26 mg of BINAP (2.3 mol%), 18 mg Pd (dba)3, 480 mg (5.0 mmol)
of sodium tertiary butoxide and toluene (100 mL). The process
of freeze–pump–thaw was repeated and the tube was back-filled
with nitrogen, then sealed and heated to 90 ◦C while stirring
for 48 h. The mixture was filtered through Celite, followed by
repeated washing of the residue with dichloromethane. Solvent
was removed by reduced pressure to yield an orange-brown solid.
Flash chromatography on silica using ethyl acetate–hexanes
(4 : 1) yielded an orange solid. (775 mg, 58%). TLC Rf 0.25,
UV-Vis experiments with Cu(NO3)2 and Fe(NO3)3. The initial
spectrum of 10 mM FerriNaphth solution was recorded. Metal ion
stock was added incrementally up to 21.6 mM and spectra were
recorded after equilibrium was reached following each addition
(~ 5 min). Experiments in MeOH were performed using identical
procedures.
1
4/1 ethyl acetate–hexanes. H NMR (400 MHz, CDCl3) d 8.47
(d, J = 7.2, 1 H), 8.41 (d, J = 8.4, 1 H), 8.27 (d, J = 8.52, 1
H), 7.7 (t, J = 7.52, 1 H), 7.1 (d, J = 8.6, 1 H), 6.80-6.72 (m,
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 4155–4161 | 4159
©

View Article Online
UV-Vis experiments with K3Fe(CN)6, AgNO3 and
[Co(NH)5Cl]Cl2. For K3Fe(CN)6 and AgNO3, the initial
spectrum of a 10 mM solution of FerriNaphth was recorded
which was followed by the addition of a 2 equivalents of the
metal complex. The spectra were monitored for changes over a
period of 20 min. An additional 8 equivalents of metal complex
was added to the cuvette and spectra were monitored for an
additional 10 min. The experiment was repeated using a 1.71 mM
stock solution of [Co(NH)5Cl]Cl2 which was prepared in 2/1/1
MeOH–DMSO–H2O. The initial spectrum of a 10 mM solution
of FerriNaphth was recorded followed by the addition of a
6-fold excess of metal. Spectra were recorded every hour for
48 h.
mass spectrometer (QTOFmicro, Waters, Milford, MA), equipped
with an in-house modified electrospray (ESI) source. The com-
pound was dissolved in CH3CN, and diluted with 10 mM
ammonium bicarbonate in CH3CN–H2O (1 : 1, v/v). The sam-
ple solution was infused to a mass spectrometer with a flow
rate of 50 mL min-1. Key instrument parameters for the mass
spectrometer, run under negative ion mode, were as follows:
capillary voltage 2800 V, cone voltage (CV) 15 V, extraction voltage
2 V, collision energy (CE) offset 25 V. The MS/MS spectrum
was obtained by selecting the negatively charged molecular ion
([M - H]-, m/z 373.1) as precursor. The proposed fragmenta-
tion mechanism (loss of CO) was based on the accurate mass
measurement.
UV-Vis experiments with ferrozine and neocuproine. Commer-
cially available ferrozine indicator was dissolved in a 4/1 MeOH–
H2O to make 10 mM stock solutions. A 1 mM stock solution
of Fe(NO3)3 was prepared in 9/1 CH3CN–EtOH. Solutions
were prepared by mixing 10 mM of FerriNaphth in CH3CN
with a 10-fold excess of both indicator and metal ion. The
mixture was allowed to equilibrate for 0.5 h and the spectrum
was recorded. Commercially available Neocuproine indicator
was dissolved in CH3CN to make a 10 mM solution, and
analogous procedures to the ferrozine assay were utilized for the
neocuproine experiment except a 20-fold excess of indicator was
used.
Notes and references
1 P. Jiang and Z. Guo, Coord. Chem. Rev., 2004, 248, 205–
229.
2 R. Y. Tsien, Annu. Rev. Neurosci., 1989, 12, 227–253.
3 J. L. Bricks, A. Kovalchuk, C. Trieflinger, M. Nofz, M. Buschel, A. I.
Tolmachev, J. Daub and K. Rurack, J. Am. Chem. Soc., 2005, 127,
13522–13529.
4 E. L. Que, D. W. Domaille and C. J. Chang, Chem. Rev., 2008, 108,
1517–1549.
5 D. P. Kennedy, C. M. Kormos and S. C. Burdette, J. Am. Chem. Soc.,
2009, 131, 8578–8586.
6 A. C. Benniston, G. Copley, K. J. Elliott, R. W. Harrington and W.
Clegg, Eur. J. Org. Chem., 2008, 2705–2713.
7 Z. N. Sun, F. Q. Liu, Y. Chen, P. K. H. Tam and D. Yang, Org. Lett.,
2008, 10, 2171–2174.
8 Z. N. Sun, H. L. Wang, F. Q. Liu, Y. Chen, P. K. H. Tam and D. Yang,
Org. Lett., 2009, 11, 1887–1890.
9 R. Kikkeri, H. Traboulsi, N. Humbert, E. Gumienna-Kontecka, R.
Arad-Yellin, G. Melman, M. Elhabiri, A. M. Albrecht-Gary and A.
Shanzer, Inorg. Chem., 2007, 46, 2485–2497.
10 N. C. Lim, S. V. Pavlova and C. Bruckner, Inorg. Chem., 2009, 48,
1173–1182.
Emission studies with Fe(NO3)3. Spectra of 10 mM Ferri-
Naphth in CH3CN was recorded before and after oxidation with
Fe(NO3)3. After the addition of 30 mL of H2O, an emission
spectrum was recorded every 8 min for 48 min. The excitation
wavelength was lex = 400 nm. The excitation slit width was 5 nm
and emission 10 nm.
Emission studies with Ga(NO3)3. Stock solution of 10 mM
Ga(NO3)3 was prepared in a 9/1 MeOH–DMSO. The initial
emission spectrum of a 10 mM solution of FerriNaphth in
MeOH was recorded followed by the addition of metal in
20 mM increments. Spectra were recorded after each addition. The
excitation wavelength was lex = 450 nm. The excitation slit width
was 5 nm and emission 10 nm. The area under the emission spectra
was integrated between 470–650 nm. The absorption spectra
were recorded after the addition of each 20 mM increment of
metal.
11 Y. Xiang and A. J. Tong, Org. Lett., 2006, 8, 1549–1552.
12 V. Dujols, F. Ford and A. W. Czarnik, J. Am. Chem. Soc., 1997, 119,
7386–7387.
13 L. M. Hyman, C. J. Stephenson, M. G. Dickens, K. D. Shimizu and
K. J. Franz, Dalton Trans., 2010, 568–576.
14 H. N. Kim, M. H. Lee, H. J. Kim, J. S. Kim and J. Yoon, Chem. Soc.
Rev., 2008, 37, 1465–1472.
15 J. H. Huang, Y. F. Xu and X. H. Qian, Dalton Trans., 2009, 1761–
1766.
16 J. H. Huang, Y. F. Xu and X. H. Qian, Org. Biomol. Chem., 2009, 7,
1299–1303.
17 Z. C. Xu, Y. Xiao, X. H. Qian, J. N. Cui and D. W. Cui, Org. Lett.,
2005, 7, 889–892.
18 Z. C. Xu, X. H. Qian and J. N. Cui, Org. Lett., 2005, 7, 3029–
Quantum yield. Quantum yields were calculated by measuring
the integrated emission area of the corrected spectra and compar-
ing that value to the area measured for Quinine in 0.5 M H2SO4
when excited at 365 nm (Ufl = 0.54).32 The quantum yields for
FerriNaphth were calculated using eqn (1), where F represents
the area under the emission spectra for the standard samples, h
is the refractive index of the solvent and Abs is the absorbance at
the excitation wavelength selected for the standard and samples.
3032.
19 S. Saha and A. Samanta, J. Phys. Chem. A, 2002, 106, 4763–
4771.
20 B. Liu and H. Tian, Chem. Commun., 2005, 3156–3158.
21 R. L. Clark, A. A. Pessolano, T.-Y. Shen, D. P. Jacobus, H. Jones,
V. J. Lotti and L. M. Flataker, J. Med. Chem., 1978, 21, 965–
978.
22 D. Srikun, E. W. Miller, D. W. Dornaille and C. J. Chang, J. Am. Chem.
Soc., 2008, 130, 4596–4597.
23 M. S. Alexiou, V. Tychopoulos, S. Ghorbanian, J. H. P. Tyman, R. G.
Brown and P. I. Brittain, J. Chem. Soc., Perkin Trans. 2, 1990, 837–
842.
Emission spectra were integrated between 440–640 nm (lex
=
400 nm, 10-6 M/abs400 ~ 0.1).
24 D. J. Liberato, V. S. Byers, R. G. Dennick and N. C. Jr, J. Med. Chem.,
sample
sample
standard
sample





Abs


1981, 24, 28–33.
F
h
h
sample =Fstandard











F

standard  standard 
 
(1)
fl
fl
25 R. Ohshima, M. Kitamura, A. Morita, M. Shiro, Y. Yamada, M.
Ikekita, E. Kimura and S. Aoki, Inorg. Chem., 2010, 49, 888–
889.






Abs


F

Mass spectroscopy. Tandem mass spectrometric (MS/MS)
analysis was performed on a quadrupole-time-of-flight tandem
26 W. Y. Lin, L. Yuan, J. B. Feng and X. W. Cao, Eur. J. Org. Chem., 2008,
2689–2692.
4160 | Dalton Trans., 2010, 39, 4155–4161
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
27 S. Aoki, K. Sakurama, R. Ohshima, N. Matsuo, Y. Yamada, R.
Takasawa, S. I. Tanuma, K. Talkeda and E. Kimura, Inorg. Chem.,
2008, 47, 2747–2754.
28 X. Qi, E. J. Jun, L. Xu, S. J. Kim, J. S. J. Hong, Y. J. Yoon and J. Y.
Yoon, J. Org. Chem., 2006, 71, 2881–2884.
30 Z. Xu, K.-H. Baek, H. N. Kim, J. Cui, X. Qian, D. R.
Spring, I. Shin and J. Yoon, J. Am. Chem. Soc., 2010, 132,
ASAP.
31 H. Diebler, M. Eigen, G. Ilgenfritz, G. Maass and R. Winkler, Pure
Appl. Chem., 1969, 20, 93–115.
29 A. Mokhir and R. Kramer, Chem. Commun., 2005, 2244–
2246.
32 G. A. Crosby and J. N. Demas, J. Phys. Chem., 1971, 75, 991–
1024.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 4155–4161 | 4161
©