Immobilized metal complexes in porous organic hosts: Development of a material for the selective and reversible binding of nitric oxide

Article abstract of DOI:10.1021/ja003282b

Delivery of NO to specific targets is important in fundamental studies and therapeutic applications. Various methods have been reported for delivery of NO in vivo and in vitro; however, there are few examples of systems that reversibly bind NO. Reported herein is the development of a new polymer (P-1[Co^II]) that reversibly binds NO. P-1[Co^II] has a significantly higher affinity for NO compared to 0₂, CO₂, and CO. The polymer is synthesized by template copolymerization methods and consists of a porous methacrylate network, containing immobilized four-coordinate Co^II sites. Binding of NO causes an immediate color change, indicating coordination of NO to the site-isolated Co^II centers. The formation of P-1[Co(NO)] has been confirmed by EPR, electronic absorbance, and X-ray absorption spectroscopies. Electronic and X-ray absorbance results for P-1[Co^II] and P-1[Co(NO)] show that the coordination geometry of the immobilized cobalt complexes are similar to those of their monomeric analogues and that NO binds directly to the cobalt centers. EPR spectra show that the binding of NO to P-1[Co^II] is reversible in the solid state; the axial EPR signal associated with the four-coordinate Co^II sites in P-1[Co^II] is quenched upon NO binding. At room temperature and atmospheric pressure, 40% conversion of P-1[Co(NO)] to P-1[Co^II] is achieved in 14 days; under vacuum at 120 °C this conversion is complete in ～1 h. The binding of NO to P-1[Co^II] is also observed when the polymer is suspended in liquids, including water.

Full text of DOI:10.1021/ja003282b

1072
J. Am. Chem. Soc. 2001, 123, 1072-1079
Immobilized Metal Complexes in Porous Organic Hosts:
Development of a Material for the Selective and Reversible Binding
of Nitric Oxide
Karen M. Padden,† John F. Krebs,† Cora E. MacBeth,† Robert C. Scarrow,‡ and
A. S. Borovik*,†
Contribution from the Departments of Chemistry, UniVersity of Kansas, Lawrence, Kansas 66045, and
HaVerford College, HaVerford, PennsylVania 19041-1392
ReceiVed September 5, 2000
Abstract: Delivery of NO to specific targets is important in fundamental studies and therapeutic applications.
Various methods have been reported for delivery of NO in vivo and in vitro; however, there are few examples
of systems that reversibly bind NO. Reported herein is the development of a new polymer (P-1[CoII]) that
reversibly binds NO. P-1[CoII] has a significantly higher affinity for NO compared to O2, CO2, and CO. The
polymer is synthesized by template copolymerization methods and consists of a porous methacrylate network,
containing immobilized four-coordinate CoII sites. Binding of NO causes an immediate color change, indicating
coordination of NO to the site-isolated CoII centers. The formation of P-1[Co(NO)] has been confirmed by
EPR, electronic absorbance, and X-ray absorption spectroscopies. Electronic and X-ray absorbance results for
P-1[CoII] and P-1[Co(NO)] show that the coordination geometry of the immobilized cobalt complexes are
similar to those of their monomeric analogues and that NO binds directly to the cobalt centers. EPR spectra
show that the binding of NO to P-1[CoII] is reversible in the solid state; the axial EPR signal associated with
the four-coordinate CoII sites in P-1[CoII] is quenched upon NO binding. At room temperature and atmospheric
pressure, 40% conversion of P-1[Co(NO)] to P-1[CoII] is achieved in 14 days; under vacuum at 120 °C this
conversion is complete in ∼1 h. The binding of NO to P-1[CoII] is also observed when the polymer is suspended
in liquids, including water.
Introduction
guanylyl cyclase. Nitric oxide also has several deleterious
effects, including inhibiting enzyme function, as well as inducing
DNA damage and lipid peroxidation.3,4
The design and synthesis of a new polymeric material that
reversibly binds nitric oxide (NO) is reported. The polymer
contains discrete metal sites with a high affinity for NO. It is
composed of isolated four-coordinate CoII complexes im-
mobilized in a highly cross-linked and porous methacrylate host.
An important consequence of the NO binding event is a rapid
and vivid color change of the material from orange to dark
brown-green. This dramatic color change leads to immediate
confirmation of the presence of nitric oxide in gas and solution
phases. Moreover, the immobilized CoII sites have a significantly
greater affinity for NO over other biologically important gaseous
compounds such as O2, CO2, and CO.
Interest in the chemistry of NO has intensified in recent years
because of its proposed role in numerous physiological func-
tions.1 The multifaceted biological importance of NO can be
separated into three categories including protective, regulatory,
and deleterious.2 Protective effects include nitric oxide as an
antioxidant agent and inhibitor to leukocyte adhesion. Among
the many regulatory effects of NO are its ability to control
vascular tone and permeability, cellular adhesion, and renal
function. Nitric oxide has also been established as a neurotrans-
mitter, bronchodilator, and inhibitor of platelet aggregation. In
addition, data strongly suggest that NO is the final common
effector molecule of nitrovasodilators, which activate soluble
It is evident from the varied biological effects listed above
that both the release and scavenging of nitric oxide are desirable
under physiological conditions. A variety of different systems
have been reported to dispense NO. The simplest cases use direct
inhalation of NO, which has been effective in reversing
pulmonary hypertension.5 Organic compounds have been de-
veloped that release NO via decomposition routes. For example,
anionic diazeniumdiolates6 dissociate in a first-order reaction
at constant pH to produce 2 equiv of NO. This class of
compounds has shown promise as prodrugs for protection of
cells from free radical mediated toxicity2 and for the relaxation
of vascular smooth muscle cells.6 Dispensing NO via photore-
lease mechanisms has also been explored; recent examples
include the photorelease of NO from nitrosothiol-derivatized
surfaces,7 Roussin’s salts,8 Fe-NO porphyrins,9 and CrIII-nitrite
complexes.10
(3) Culotta, E.; Koshland, D. E. Science 1992, 258, 1862-1865.
(4) Ignarrow, L. J. Pharm. Res. 1989, 6, 651-659.
(5) Rossaint, R.; Falke, K. J.; Lo´pez, F.; Slama, K.; Pison, U.; Zapol,
W. M. N. Engl. J. Med. 1993, 328, 399-405.
(6) For example see: Keefer, L. K.; Nims, R. W.; Davies, K. M.; Wink,
D. A. Methods Enzymol. 1996, 268, 281-293.
(7) Etchenique, R.; Furman, M.; Olabe, J. A. J. Am. Chem. Soc. 2000,
122, 3967-3968.
(8) Bourassa, J.; DeGraff, W.; Kudo, S.; Wink, D. A.; Mitchell, J. B.;
Ford, P. C. J. Am. Chem. Soc. 1997, 119, 2853-2860.
(9) Hoshino, M.; Ozawa, K.; Seki, H.; Ford, P. C. J. Am. Chem. Soc.
1993, 115, 9568-9575.
(10) De Leo, M.; Ford, P. C. J. Am. Chem. Soc. 1999, 121, 1980-
1981.
† University of Kansas.
‡ Haverford College.
(1) For examples see: Moncada, S.; Palmer, R. M. J.; Higgs, E. A.
Pharmacol. ReV. 1991, 43, 109-142.
(2) Wink, D. A.; Mitchell, J. B. Free Radical Biol. Med. 1998, 25, 434-
456.
10.1021/ja003282b CCC: $20.00 © 2001 American Chemical Society
Published on Web 01/23/2001

Immobilized Metal Complexes in Porous Organic Hosts
J. Am. Chem. Soc., Vol. 123, No. 6, 2001 1073
[Co1(dmap)2][PF6]. The following synthesis was conducted in the
drybox under an argon atmosphere. To a 100-mL single-neck flask
was added 0.991 g (1.86 mmol) of H21 that was partially dissolved in
30 mL of 1,2-dichloroethane to give a yellow suspension. The purple
methanolic (10 mL) Co(OAc)2 (0.329 g, 1.86 mmol) solution was added
to the suspension of the ligand with stirring. After addition was
completed, the reaction mixture became orange in color and an orange
precipitate began to form after approximately 5 min. 4-(Dimethylami-
no)pyridine, 0.455 g (3.72 mmol), in 10 mL of methanol, was then
added to the mixture and no appreciable color change was observed.
Ferrocenium hexafluorophosphate (0.616 g, 1.86 mmol) was partially
dissolved in 30 mL of methanol to give a deep blue solution, which
was then added to the reaction mixture. After approximately 1 h, the
reaction mixture turned from the orange color to a deep red-brown.
The reaction was then stirred for an additional 12 h after which volatile
compounds were removed under reduced pressure to yield a brown
solid. This solid was filtered and washed with diethyl ether until the
diethyl ether became clear. The red-brown solid was then washed three
times with a 3:1 diethyl ether:methanol solution and twice with diethyl
ether. The solid was dried under vacuum overnight. Yield of the
complex was 1.683 g (92%). Anal. Calcd for [Co1(dmap)2][PF6],
C48H50CoF6N6O4P: C, 58.90; H, 5.15; N, 8.59. Found: C, 58.93; H,
While there have been many advances in the technology of
NO delivery, there are few examples of systems that have the
capability to reVersibly bind NO.10,11 Useful applications exist
both in vivo and in vitro for materials that store and subsequently
release NO. Furthermore, a durable material that readily
scavenges NO would have important applications in cases where
high NO concentrations could lead to detrimental effects, such
as DNA damage. We have recently shown that immobilized
metal complexes in porous organic hosts can serve as sites for
the reversible binding of CO12 and O2.13 These materials are
formed by template copolymerization methods, a technique that
is used in making molecularly imprinted polymers.14 Materials
made in this way have a significant number of metal sites
(∼90%) that are isolated from each other by a porous host.15
Site isolation of the metal complexes is advantageous because
the formation of undesirable bimolecular metal-based adducts
is prevented and the function of the polymer is prolonged.
This copolymerization method has been used to develop
P-1[CoII], a material that binds NO to form P-1[Co(NO)].
Spectroscopic measurements confirm that NO binding occurs
at coordinatively unsaturated CoII sites, which are monodis-
persed within the porous host. The NO-loaded polymer, P-1[Co-
(NO)], is stable in the solid state at 25 °C for days. However,
slow loss of NO from P-1[Co(NO)] is observed at room
temperature with a 40% loss of NO occurring after 14 d. The
release rate of NO can be accelerated by increasing temperature
with a maximum loss observed within ∼1 h/50 mg of polymer,
after heating at 120 °C under vacuum. The recovered polymer
retains affinity for binding NO.
1
5.06; N, 8.25. H NMR (CDCl3): δ 7.88 (s, 2H, -NdCH(Ph)); 7.47
(d, 4H, J ) 7 Hz, dmapH); 7.42 (m, 8H, styryl phenyl); 7.07 (d, 2H,
J ) 9 Hz, salicyl phenyl); 6.82 (d, 2H, J ) 2 Hz, salicyl phenyl); 6.72
(dd, 2H, J ) 11, 18 Hz, H(Ph)CdCH2); 6.29 (d, 4H, J ) 7 Hz, dmapH);
6.24 (dd, 2H, J ) 9, 2 Hz, salicyl phenyl); 5.76 (d, 2H, J ) 18 Hz,
trans-H(Ph)CdCH2); 5.26 (d, 2H, J ) 11 Hz, cis H(Ph)CdCH2); 5.02
(s, 4H, PhO-CH2Ph); 4.01 (s, 4H, -CH2CH2-); 2.93 (s, 12H, dmap-
NCH3). IR (KBr): 3088 (w), 2924 (w), 2867 (w), 1625 (s), 1604 (s),
1527 (m), 1482 (w), 1430 (m), 1388 (m), 1306 (w), 1263 (w), 1226
(s), 1180 (m), 1143 (m), 1123 (m), 1063 (m), 1020 (m), 950 (w), 916
(w), 840 (bs) cm-1. UV-vis (CH2Cl2, λmax/nm (ꢀ, M-1 cm-1)): 234
(56700), 271 (109000), 380 (11600). Mp 221-222 °C dec.
Experimental Section
P-1[CoIII(dmap)2]. All polymers were synthesized with 5 mol %
of metal complex, 94 mol % of cross-linking agent, 1 mol % of initiator,
and a 3:1 v/v ratio of porogenic agent to cross-linking agent. All
reagents were added to a high-pressure glass reaction tube (Ace) and
sealed with a screw cap under an argon atmosphere. The complex,
[Co1(dmap)2][PF6] (0.660 g, 0.674 mmol), was added to the reaction
tube and ∼2/3 of the DMF (total DMF: 7.599 g, 104 mmol, 7.173
mL) was added. The remaining reagents were not added until the
complex was completely dissolved. The ethylene glycol dimethacrylate
(2.676 g, 13.5 mmol) and 2,2′-azobisisobutyronitrile (0.0235 g, 0.143
mmol) were then added using the remaining DMF. The tube was then
sealed and the reaction mixture placed in an oil bath at ∼60 °C for 24
h. After 24 h the DMF was removed under vacuum and the polymer
was ground with a mortar and pestle. The ground polymer was placed
in a Soxhlet apparatus and washed with methanol for 24 h. The resulting
polymer was then dried under vacuum, ground, and sieved to a particle
size of e75 µm. The yield of the red-brown polymer was 2.812 g.
[Co] ) 170-180 µmol of Co/g of polymer (range of values obtained).
Nitrogen analysis: 1.64% found; 1.40% calculated; 1.61% calculated
and corrected (the correction factor, 0.21%, is the average nitrogen
content of three different samples of “complex-free” poly(EGDMA)).
Fluoride analysis: 1.68% found; 1.89% calculated.
P-sal2. The polymer P-1[CoIII(dmap)2] (1.172 g) was treated with
50 mL of 0.1 M Na2EDTA in deionized water. The reaction mixture
was then refluxed for 24 h. The polymer was filtered and washed five
times with 5-mL aliquots of deionized water, three times with methanol,
and three times with diethyl ether. The polymer was then dried under
vacuum. The yield of the resulting light tan polymer was 0.977 g. [Co]
) 2-5 µmol of Co/g of polymer (range of values obtained). Nitrogen
analysis: 0.17% found; 0.04% calculated; 0.25% calculated and
corrected. Fluoride analysis: 0.05% found; 0.05% calculated.
Solvents and reagents were used as received from either Fisher or
Aldrich unless otherwise noted. Solvents used under an inert atmosphere
for synthesis or sample preparation were dried following standard
procedures.16 Ethylenediamine and 3,3′-diamino-N-methyldipropyl-
amine were distilled under nitrogen immediately before use. Co(OAc)2‚
4H2O was dehydrated by heating to 120 °C under vacuum for 48 h.
Synthesis of some complexes and preparation of polymerization reac-
tions were conducted in a Vacuum Atmospheres drybox under an argon
atmosphere. Standard Schlenk techniques were used during the workup
of some reactions and manipulations of polymer samples outside the
drybox. Elemental analyses were conducted by either Desert Analytics
(Tucson, AZ), the University of Kansas Department of Medicinal
Chemistry Microlab, or the analytical services laboratory of the Depart-
ment of Animal Sciences at Kansas State University. The compounds
2-hydroxy-4-(4-vinylbenzylmethoxy)benzaldehyde17 and bis[2-hydroxy-
4-(4-vinylbenzylmethoxy)benzaldehyde]ethylenediimine (H21)15,18 were
synthesized following literature methods.
(11) (a) Ribeiro, J. M. C.; Hazzard, J. M. H.; Nussenzveig, R. H.;
Champagne, D. E.; Walker, F. A. Science 1993, 260, 539-541. (b) Ding,
X. D.; Weichsel, A.; Andersen, J. F.; Shokhireva, T. K.; Balfour, C.; Pierik,
A. J.; Averill, B. A.; Montfort, W. R.; Walker, F. A. J. Am. Chem. Soc.
1999, 121, 128-138 and references therein.
(12) (a) Krebs, J. F.; Borovik, A. S. J. Am. Chem. Soc. 1995, 117, 10593-
10594. (b) Krebs, J. F.; Borovik, A. S. In Molecular and Ionic Recognition
with Imprinted Polymers; ACS Symp. Ser. No. 703; Bartsch, R. A., Maeda,
M., Eds.; American Chemical Society: Washington, DC, 1998; pp 159-170.
(13) Krebs, J. F.; Borovik, A. S. Chem. Commun. 1998, 553-554.
(14) (a) Wulff, G. Angew. Chem., Int. Ed. Engl. 1995, 34, 1812-1832
and references therein. (b) Mosbach, K.; Ramstrom, O. Biotechnology 1996,
14, 163-170. (c) Shea, K. J. Trends Polym. Sci. 1994, A32, 166-172. (d)
Molecular and Ionic Recognition with Imprinted Polymers; ACS Symp.
Ser. No. 703; Bartsch, R. A., Maeda, M., Eds.; American Chemical
Society: Washington, DC, 1998.
P-1. The P-sal2 polymer (1.127 g) was added to a 100-mL single-
neck flask fitted with a septum. The polymer was degassed (three times)
by successively applying a vacuum and refilling with nitrogen. To the
flask was added 10 mL of freshly distilled methanol via syringe. One
(15) Sharma, A. C.; Borovik, A. S. J. Am. Chem. Soc. 2000, 122, 8946-
8955.
(16) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals, 3rd ed.; Pergamon Press: New York, 1988.
(17) Daly, J.; Horner, L.; Witkop, B. J. Am. Chem. Soc. 1961, 83, 4787-
4792.
(18) Krebs, J. K. Ph.D. Thesis, 1998, Kansas State University

1074 J. Am. Chem. Soc., Vol. 123, No. 6, 2001
Padden et al.
equivalent of ethylenediamine (15.1 µL, 225 µmol, based on cobalt
analysis of P-1[Co(dmap)2][PF6]) was added and the mixture was stirred
for 6 h. The polymer was filtered and washed three times with diethyl
ether and dried under vacuum for 15 min to yield 0.9413 g of a yellow-
tan polymer. Nitrogen analysis: 0.76% found; 0.43% calculated; 0.64%
calculated and corrected. [Co] ) 3-6 µmol of Co/g of polymer (range
of values obtained).
P-1[CoII]. The polymer P-1 (0.901 g) was placed in a Schlenk tube
and degassed by applying a vacuum and refilling with nitrogen three
times. The polymer was then treated with 10 mL of a 0.1 M Co(OAc)2
solution under an argon atmosphere. The mixture was stirred for 6 h.
The polymer was then filtered into a fluted Whatman No. 41 (90 mm,
ashless) filter paper and placed in a glass thimble (with coarse frit).
The polymer was then washed in the Soxhlet apparatus with methanol
for 14 h. The resulting polymer was dried under vacuum to yield 0.294
g of an orange-red polymer. [Co] ) 170-230 µmol of Co/g of polymer
(range of values obtained). The pore diameter was 25 Å.
Sample Preparation. UV-vis spectra of the monomeric compounds
were obtained in solution with use of a 1.00-cm Suprasil quartz cell.
Polymer samples for electronic absorbance spectroscopy were prepared
by suspending the solid polymer in toluene under Ar in an anaerobic
1.00 cm quartz cell. Gaseous NO was added via syringe. Diffuse
reflectance samples were 25% w/w mixture of polymer in KBr. Nitric
oxide was added prior to mixing with KBr for these solid samples.
EPR spectroscopy samples were prepared by placing approximately
20 mg of solid sample into a quartz EPR tube under argon atmosphere.
Samples that required nitric oxide were packed under argon atmosphere
and sealed, and gaseous NO was added via syringe.
Aluminum sample holders for X-ray absorption measurements were
approximately 1 in. square by 1 mm thick and contained a 2 × 20 mm
slit. The solid samples were packed dry under either an argon
atmosphere (P-1[CoII]) or an atmosphere of NO gas (P-1[Co(NO)]).
Both samples were sealed with translucent electrical tape (3M#1205,
Minneapolis, MN).
Both transmission and fluorescence data were analyzed via general
procedures described elsewhere using Igor Pro (Wavemetrics, Lake
Oswego, OR) for data averaging and graphing.20 Definitions of ø
(EXAFS), k (the photoelectron wavenumber), and r′ (the apparent
Co-X distance in the Fourier transform) are in accord with general
usage as established by Scott.21
The data for P-1[CoII] and P-1[Co(NO)] from 7573 to 8465 eV
(excluding the 7720-7741 eV region at the top of the edge rise) were
fit to a cubic spline baseline with spline points at 7873 and 8173 eV;
a small (<1%) K-edge from contaminating nickel was also removed
as part of the baseline correction. After the baseline was fit, various
analyses of the EXAFS region were performed. The best fit to the data
was obtained by using a two-shell model (one N/O shell at ca. 2 Å
and one C shell at ca. 2.8 Å). The program FEFF, version 7.0, was
utilized during EXAFS analyses to obtain f and R as the average of
functions calculated for Co-O and Co-N paths;22 E0 was chosen as
7723 eV for the calcuation of k.23 The ionization potential for use with
the f and R functions was refined in the fits (77.20.5 eV for P-1[CoII]
2
and 7721.5 eV for P-1[Co(NO)]). An amplitude reduction factor (So )
of 0.85 (identical with that suggested for use in FEFF version 5)24 was
employed based on refinements of data from crystalline samples of
Co(acac)3, [Co(en)3]Cl3‚H2O, and [Co(NH3)6]Cl3. In the reported fits
the baselinge, edge, and EXAFS parameters were simultaneously refined
to minimize the residual function gof ) [sum{(XAS-XAScalc)/esddata}2/
(number of data - number of refined parameters)]1/2. The esddata was
taken to be 0.0005 below the edge and 0.05/k2 above the edge (this is
equivalent to refining k2ø). The reported gof has been normalized to
the refined edge height. The reported standard deviations were obtained
from the Igor Pro least-squares analysis.
Determination of gas binding was accomplished by using a gas
uptake apparatus, which is modeled after a system described by Taylor
and co-workers.25 The instrument consists of a series of chambers
separated by Teflon-wetted solenoids (NR Research, West Caldwell,
NJ). Five solenoids are used (Model 648T011, 12 V, two-way, normally
closed, 3.9 mm orifice diameter, 2 atm pressure maximum) to create
an upper (volume, 33.3 mL) and lower sample chamber (volume, 19.8
mL) and two gas-holding reservoirs. The apparatus utilizes a MKS
Instruments absolute pressure transducer (Model 122BA-0500BB)
interfaced to a MKS Type PDR-C-2C two-channel power supply/digital
output device. All pressures are measured in units of Torr. The
transducer is connected to a 316 stainless steel vacuum line (except
where Teflon adapters have been used to convert the 1/8 in. straight
threads of the solenoids to 1/8 in. NPT). The lower sample chamber
assembly consists of a removable miniature sample cylinder (Nupro,
10 mL, 316 stainless steel). A removable 7-µm filter has been fitted
directly above the sample cylinder to ensure that the samples do not
enter the upper chamber. The two gas reservoirs have been designed
to allow small quantities of gas into the system.
Samples were prepared as solids and placed in the sample chamber
with a sample mass of 0.153 g. The sample chamber was then evacuated
for a minimum of 30 min. Analyte gas (17 Torr of NO, CO2, CO, or
O2) was introduced into the upper sample chamber while the lower
sample chamber assembly remains under vacuum. When the lower
sample chamber is opened and exposed to the analyte gas (system
pressure now was at 10.3 Torr), pressure is recorded every 20 s until
equilibrium is reached (typically 5 min). Data corresponding to changes
in pressure were then converted to micromoles of analyte gas bound
after accounting for the change in pressure that results from the increase
in volume upon opening the sample chamber. Gas binding was
measured for P-1[CoII] ([CoII] ) 230 µmol of Co/g of polymer), P-1
([CoII] ) 3.4 µmol of Co/g of polymer), the polymer without CoII ions
bonded to the immobilized sites, and poly(EGDMA), a porous polymer
Physical Methods. All proton nuclear magnetic resonance (1H NMR)
spectra were collected on a 400-MHz Bruker spectrometer. Infrared
spectra were collected on a Mattson Genesis Series FT-IR spectrometer.
Solution and suspension UV-vis spectra were collected on a Cary 50
spectrophotometer. Diffuse reflectance spectra of the polymers were
obtained on a SLM-Aminco 3000 diode array spectrophotometer
equipped with a custom-made Harrick Scientific diffuse reflectance
accessory (DRA-4). This system only allowed for data collection in
the visible region of the spectrum (400-800 nm). Pore diameters and
surface areas were determined with a Gemini surface area microana-
lyzer.
EPR spectra were collected with use of a Bruker EMX spectrometer
equipped with an ER4102ST cavity. The instrument was previously
calibrated with DPPH. Spectra for the CoII samples were collected with
use of the following spectrometer settings: attenuation ) 25 dB,
microwave power ) 0.64 mW, frequency ) 9.31 GHz, sweep width
) 5000 G, modulation amplitude ) 5.02 G, gain ) 8.93 × 10-3
,
conversion time ) 81.92 ms, time constant ) 1.28 ms, and resolution
) 1024 points.
X-ray absorption data were collected at the National Synchrotron
Light Source at Brookhaven National Laboratory. Data for P-1[CoII]
was collected on beam line ×9B using a Si 220 monochromator low-
angle nickel mirror for harmonic rejection and a closed-cycle helium
cryostat to maintain sample temperature at ca. 25 K. P-1[Co(NO)] data
were collected at room temperature on beam line ×18B using a Si 111
monochromator. Transmission spectra (including the cobalt foil calibra-
tion spectrum) were obtained with use of gas ionization chambers. The
fluorescence spectra were obtained with 13-element Ge solid-state
detectors (Canberra); a deadtime correction of 3 µs was used, but the
effects of this were minimal since the total count rate was under 20000
s-1 for each detector. Beam energy was calibrated by assigning the
first inflection on the absorption edge of a cobalt foil to an energy of
7709.5 eV.19 The fluorescence mode XAS data were corrected for self-
absorption and incident detector efficiency; transmission data were used
in calculating the self-absorption correction.
(20) Scarrow, R. C.; Trimitsis, M. G.; Buck, C. P.; Grove, G. N.;
Cowling, R. A.; Nelson, M. J. Biochemistry 1994, 33, 15023-15035.
(21) Scott, R. A. Methods Enzymol. 1985, 117, 414-459.
(22) Ankudinov, A. L.; Rehr, J. J. Phys. ReV. 1997, B56, R1712-R1715.
(23) Brennan, B. A.; Alms, G.; Nelson, M. J.; Durney, L. T.; Scarrow,
R. C. J. Am. Chem. Soc. 1996, 118, 9194-9195.
(24) O’Day, P. A.; Rehr, J. J.; Zabinsky, S. I.; Brown, G. E., Jr. J. Am.
Chem. Soc. 1994, 116, 2938-2949.
(25) Taylor, R. J.; Drago, R. S.; George, J. E. J. Am. Chem. Soc. 1989,
111, 6610-6615.
(19) Bearden, J. A.; Burr, A. F. ReV. Mod. Phys. 1967, 39, 78-124.

Immobilized Metal Complexes in Porous Organic Hosts
J. Am. Chem. Soc., Vol. 123, No. 6, 2001 1075
Scheme 1a
a Conditions: a, EGDMA, AIBN, DMF, N2, 60 °C; b, EDTA, H2O, ∆; c, C2H8N2, MeOH; d, Co(OAc)2, MeOH, N2.
without immobilized sites. Measurements were performed in triplicate;
samples were evacuated for at least 10 min between trials where only
nonselective physical adsorption is observed. In cases where chemical
adsorption is observed, each measurement was made on new samples,
which were thoroughly degassed prior to gas uptake. Gas uptake values
for P-1[CoII] have been corrected for nonselective physical adsorption
by the polymer host. This was accomplished for each gas by subtracting
the uptake data for P-1 from gas-binding data of P-1[CoII].
a porous polymeric host should allow for site isolation of the
metal centers, thus preventing the bimolecular pathway leading
to cobalt(III)-nitrite formation. With this pathway eliminated,
the immobilized [Co(salen)(NO)] is stable within the polymeric
host under ambient conditions. As a result, the composite
material P-1[CoII] can bind NO to form stable P-1[Co(NO)],
which then functions as a storage material for nitric oxide. An
advantage of this design is that subsequent removal of NO from
P-1[Co(NO)] regenerates P-1[CoII], which contains [CoII(salen)]
sites capable of rebinding additional NO molecules.
Results and Discussion
Scheme 1 represents the synthesis of P-1[CoII] utilizing the
general template copolymerization procedure for immobilization
of metal complexes.12,13,15 This procedure utilizes a template
complex (5 mol %), cross-linking agent (∼95 mol %), and
solvent, which serves as the porogen during the radical
polymerization process. In the design of P-1[CoII], the kinetically
inert CoIII complex, [CoIII1(dmap)2]+, is the template complex
used to form the immobilized sites. This complex contains the
Design and Synthesis of Polymers. The irreversible binding
of NO to four-coordinate CoII complexes is well docu-
mented.26,27 We have chosen one such complex, [CoII(salen)]
(salen, bis(2-hydroxy-benzaldehyde)ethylenediimine), to serve
as the sites for NO binding in our polymeric material. In
solution, monomeric [CoII(salen)] forms a stable nitrosyl
complex, [Co(salen)(NO)], under anaerobic conditions. In the
presence of dioxygen, however, [Co(salen)(NO)] is unstable and
is converted to the CoIII-nitrite complex, [CoIII(salen)(NO2)].
It has been proposed that the formation of [CoIII(salen)(NO2)]
proceeds through the dimeric intermediate [(salen)CoIII(ON-
O-O-NO)CoIII(salen)], which can readily form in solution at
room temperature.28 Immobilization of [Co(salen)(NO)] within
styrene-modified salen ligand [1]2- 13,29
which is covalently
,
attached to the porous methacrylate host by the procedure
shown. To ensure that the metal complex is retained within the
polymer host, two-point binding is utilized. P-1[CoIII(dmap)2]
typically has a cobalt concentration of between 170 and 180
µmol of Co/g of polymer with an average pore diameter of 25
Å. Note that immobilization of [CoIII1(dmap)2]+ in a porous
organic host with evidence of site isolation has been previously
reported.12
(26) (a) Enemark, J. H.; Feltham, R. D. Coord. Chem. ReV. 1974, 13,
339-406. (b) Feltham, R. D.; Enemark, J. H. Top. Stereochem. 1981, 12,
155-215. (c) Richter-Addo, G. B.; Legzdins, P. Metal Nitrosyls; Oxford
University Press: New York, 1992.
(27) Hoshino, M.; Konishi, R.; Tezuka, N.; Ueno, I.; Seki, H. J. Phys.
Chem. 1996, 100, 13569-13574.
(28) Clarkson, S. G.; Basolo, F. Inorg. Chem. 1973, 12, 1528-1534.
(29) Fujii, Y.; Matsutani, K.; Kikuchi, K. J. Chem. Soc., Chem. Commun.
1985, 415-417.

1076 J. Am. Chem. Soc., Vol. 123, No. 6, 2001
Padden et al.
Figure 2. Observed colors of polymers: P-1[CoII] (top) and P-1[Co-
(NO)] (bottom).
The differences in the absorbance properties of P-1[CoII] and
P-1[Co(NO)] lead to striking color changes in these materials.
This color conversion is shown in Figure 2. In the presence of
NO, the orange P-1[CoII] changes to the brown-green P-1[Co-
(NO)] polymer. The conversion back to the orange P-1[CoII] is
accomplished in ∼1 h when P-1[Co(NO)] is heated to 120 °C
under vacuum (0.05 Torr). Note that porous poly(EGDMA)
without immobilized CoII(salen) sites remains colorless in the
presence of NO.
Figure 1. Electronic absorbance spectra for P-1[CoII] (s) and P-1[Co-
(NO)] (--- ) (A) suspended in toluene and (B) solid-state diffuse
reflectance spectra.
Once immobilized, the cobalt(III) ions and dmap ligands in
P-1[CoIII(dmap)2] are removed under acidic conditions to afford
P-sal2.30 These conditions hydrolyze 1, as is evident by the lack
of nitrogen in the polymer and the characteristic salicylaldehyde
ν(CdO) at 1521 cm-1. P-sal2 contains immobilized sites with
two salicylaldehyde moieties that are predisposed to bind
ethylenediamine to reform the immobilized tetradentate salen
ligand. The resulting polymer, P-1, readily binds CoII ions to
form P-1[CoII], which contains immobilized four-coordinate CoII
centers. The cobalt concentration in P-1[CoII] ranges from 180
to 230 µmol of Co/g of polymer with an average pore diameter
of 25 Å.
The reversible binding of NO to P-1[CoII] can also be
observed by EPR spectroscopy. The immobilized low-spin CoII-
(salen) complexes in P-1[CoII] are EPR active with signals
originating from the (1/2 doublet ground state. Figure 3A shows
the X-band EPR spectrum of solid P-1[CoII] measured at 77 K.
An axial spectrum is obtained with g-values at 3.83 and 1.98.
This spectrum is consistent with a square-planar coordination
geometry about the CoII ions. Addition of NO to form P-1[Co-
(NO)] quenches the EPR signal. The lack of signal for P-1[Co-
(NO)] is consistent with the immobilized metal centers being
diamagnetic, which could result from antiferromagnetic coupling
of the unpaired electrons on NO and the Co center.32 Removal
of NO from P-1[Co(NO)] by heating under vacuum (vide supra)
affords the axial EPR signal of P-1[CoII]. The immobilized
cobalt sites in this regenerated form of P-1[CoII] are still capable
of binding NO as shown by the quenching of the axial EPR
signal upon the addition of NO. Under conditions of room
temperature and pressure, 40% conversion of P-1[Co(NO)] to
P-1[CoII] is observed within 14 d, as determined by comparison
of the relative areas of the collected EPR signals (Figure 3B).
Similar EPR properties are observed for suspensions of
P-1[CoII] and P-1[Co(NO)] in noncoordinating solvents. For
example, suspensions of P-1[CoII] and P-1[Co(NO)] in toluene
give identical spectra to those measured in the solid state. More
complicated EPR spectra are obtained for suspensions of
P-1[CoII] in coordinating solvents, such as water and methanol.
Binding of NO to P-1[CoII]. A. Electronic and EPR
Spectral Properties. The absorbance spectra for suspensions
of P-1[CoII] and P-1[Co(NO)] in toluene are shown in Figure
1A.31 The absorbance spectrum of the P-1[CoII] suspension has
two bands at 350 and 410 nm with a shoulder centered at 495
nm. For P-1[Co(NO)], which was formed by treating the
P-1[CoII] suspension with gaseous NO, two prominent bands
at 310 and 400 nm are observed. The spectrum of P-1[Co(NO)]
also contains a weaker absorbance at ∼600 nm. Note that the
visble absorbance peaks of the monomeric complex, Co(salen)
in ethanol, are at 353 and 415 nm and those of Co(salen)(NO)
are at 410 and ∼550 nm. The solid-state absorbance spectra
for P-1[CoII] and P-1[Co(NO)] are shown in Figure 1B and are
qualitatively similar to those obtained as suspensions. The
diffuse reflectance visible spectrum of P-1[CoII] in KBr has a
shoulder centered at 500 nm. The tail of the stronger absorbance
band, originating from the UV region, is also observed. The
diffuse reflectance visible absorbance spectrum of P-1[Co(NO)]
has a feature at ∼650 nm; the 500-nm shoulder found in
P-1[CoII] is absent in the spectrum of P-1[Co(NO)].
(32) (a) Earnshaw, A.; Hewlett, P. C.; Larkworthy, L. F. J. Chem. Soc.
1965, 4718-4723. (b) Earnshaw, A.; Hewlett, P. C.; King, E. A.;
Larkworthy, L. F. J. Chem. Soc. 1968, 241-246. (c) Duffin, P. A.;
Larkworthy, L. F.; Mason, J.; Stephen, A. N.; Thompson, R. M. Inorg.
Chem. 1987, 26, 2034-2040.
(30) Attempts to reduce the cobalt centers with ascorbic acid or dithionite
are not as efficient in removing the cobalt and dmap as that described.
(31) P-1[CoII] and P-1[Co(NO)] suspensions in toluene produce optically
transparent mixtures.

Immobilized Metal Complexes in Porous Organic Hosts
J. Am. Chem. Soc., Vol. 123, No. 6, 2001 1077
addition, the Co-NO sites in P-1[Co(NO)] are stable in the
presence of O2, whereas in solution Co(salen)(NO) complexes
are unstable and form (salen)CoIII-NO2 complexes.28 The
difference in Co-NO stability can be explained by the site
isolation properties provided by the organic host in P-1[CoII].
As a result, the formation of dimeric cobalt species, essential
for the formation of the CoIII-NO2 complexes, is prevented
and removal of NO from the polymer is possible.
B. X-ray Absorption Spectroscopy. The cobalt K-edge
X-ray absorption spectra (XAS) of P-1[CoII] and P-1[Co(NO)]
were measured to further investigate the geometric properties
of the immobilized cobalt sites. X-ray absorption near edge
structure (XANES) spectra and extended X-ray absorption fine
structure (EXAFS) were obtained for frozen samples of solid
P-1[CoII] and P-1[Co(NO)]. Figure 4a shows the XANES
spectra for P-1[CoII] and P-1[Co(NO)], which have edge
positions of 7719.48(2) and 7720.86(2) eV, respectively. The
lowest energy preedge peak in each spectrum is assigned to
the 1s f 3d transition. The area of this peak is sensitive to the
symmetry of the cobalt complexes, and thus can be used to probe
the coordination number and geometry of the cobalt centers.
For P-1[CoII], the 1s f 3d preedge peak is found at 7709.9(1)
eV with a refined area/edge height of 0.06(1) eV, while that
for P-1[Co(NO)] occurs at 7710.47(1) eV with a refined area/
edge height of 0.220(3) eV (Figure 4b). The relatively small
value for the area of the 1s f 3d peak in P-1[CoII] indicates
that its cobalt centers are centrosymmetric. A square-planar
coordination geometry for the immobilized cobalt centers is
consistent with this symmetry requirement. Support for this
assignment is provided by the large shoulder (area/edge height
of 0.32 eV) observed at 7715.30(2) eV, which is attributed to
the 1s f 4pz transition.33 In contrast, the significantly larger
area observed for the 1s f 3d preedge peak in P-1[Co(NO)] is
consistent with a loss of centrosymmetry caused by the
formation of a five-coordinate complex in the presence of NO.
The loss of centrosymmetry allows mixing between the 3d and
4p orbitals, which results in area increases of the 1s f 3d
preedge peak.
Figure 3. (A) X-band EPR spectra measured at 77 K for two complete
cycles of NO addition to solid P-1[CoII] and subsequent removal at
120 °C under vacuum (total of five spectra reported): initially recorded
spectrum of P-1[CoII] (s), first addition and removal cycle (‚‚‚), and
second NO addition and removal cycle for P-1[Co(NO)] (---). Note
that for each cycle there are two spectra, one showing the quenching
of the signal after NO addition and the second the axial signal after
NO removal. (B) X-band EPR spectra measured at 77 K for solid
P-1[CoII] (s) and P-1[Co(NO)] after exposure to room temperature
and pressure for 14 d (---). (C) X-band EPR spectra measured at 77 K
for a water suspension of P-1[CoII] (s) and after treating the polymer
with NO (---).
The results of EXAFS spectral analysis for P-1[CoII] and
P-1[Co(NO)] (Figure 5) support the coordination changes
proposed between the two polymers and the direct binding of
NO to the immobilized cobalt centers. The best fit of the EXAFS
data for P-1[CoII] gives 3.7(2) nitrogen and/or oxygen ligands
bonded to the CoII centers at a distance of 1.859(4) Å. These
metrical values are in agreement with those reported from X-ray
diffraction studies of molecular square planar CoII complexes
with similar tetradentate ligands. For example, the average Co-
N/O bond distance in CoIIsalen‚CHCl3 is 1.849(6) Å.34 The
EXAFS analysis for P-1[Co(NO)] indicates that the cobalt
centers have increased their coordination number by 20%; the
refined number being 4.6(1). The average Co-N/O bond
distance in P-1[Co(NO)] is 1.886(2) Å, which is similar to the
1.862(2) Å average bond distance found for [Co(salen)(NO)]
by X-ray diffraction methods.35
Figure 3C shows the EPR spectrum of P-1[CoII] suspended in
water. The spectrum suggests that water molecule(s) have
coordinated to some of the immobilized CoII centers. This EPR
signal is completely quenched when NO is added to the
suspension, indicating that the immobilized sites are capable
of binding NO in an aqueous medium.
C. Gas Uptake Measurements. The immobilized sites in
P-1[CoII] selectively bind NO relative to the biologically
important gases O2, CO2, and CO. This selectivity is demon-
strated by gas uptake measurements, results of which are shown
The spectral properties of P-1[CoII] and P-1[Co(NO)] are
nearly identical with those reported for monomeric CoII(salen)
and Co(salen)(NO) complexes in solution.27 This indicates that
the coordination geometries of the immobilized cobalt com-
plexes in the polymeric hosts are similar to those of their
molecular analogues. However, immobilization of CoII(salen)
complexes in a porous host has a dramatic effect on the binding
of NO to the cobalt centers. The reversible binding of NO
observed for P-1[CoII] contrasts the irreversible binding reported
for monomeric CoII(salen) in solution and the solid state.27 In
(33) This assignment is made based on analogy to peaks found in square-
planar NiII complexes: Colpas, G. J.; Maroney, M. J.; Bagyinka, C.; Kumar,
M.; Willis, W. S.; Suib, S. L.; Baidya, N.; Mascharak, P. K. Inorg. Chem.
1991, 30, 920-928.
(34) Schaefer, W. P.; March, R. E. Acta Crystallogr. 1969, B25, 1675-
1682.
(35) Haller, K. J.; Enemark, J. H. Acta Crystallogr. 1978, B34, 102-
109.

1078 J. Am. Chem. Soc., Vol. 123, No. 6, 2001
Padden et al.
Figure 4. (Left) Comparison of XANES spectra for solid P-1[Co(NO)] (s) and P-1[CoII] (---). (Right) Comparison of preedges of solid P-1[Co-
(NO)] (top: offset by +0.25) and solid P-1[CoII] (bottom: no offset). Five plots are presented for each polymer: the measured preedge data with
- -
baseline subtracted and normalized to the edge height (s), the fits to the edge by the integral of a 75% Gaussian/25% Lorenzian peak (- -),
the difference when the edge fit is subtracted from the data (‚‚‚), and the deconvolution of the preedge peaks into two Gaussian peaks of the same
width (---).
Figure 5. (Left) EXAFS (k2ø) of solid P-1[Co(NO)] (top: offset by +2) and solid P-1[CoII] (no offset). EXAFS data are shown by circles, solid
lines represent the computer generated fits, and the difference is shown with dots. (Right) Comparison of Fourier filtered k3ø (data from k ) 1-14
Å-1 are filtered with a 5% window). Solid lines are Fourier transforms (FT) of data, dashed lines are FT of the fits, and dotted lines are FT of the
residual from the fits. The top shows P-1[CoII(NO)], and the bottom shows P-1[CoII].
in Figure 6. Solid samples of P-1[CoII] were exposed to 10.3
Torr (30.5 µmol) of gas and uptake was monitored as a change
in pressure with respect to time. The uptake data presented in
Figure 6A indicate that the cobalt sites in P-1[CoII] bound 21
µmol of NO within 2 min. This accounts for a 70% uptake of
the available NO, which corresponds to 60% of the immobilized
cobalt sites binding NO. In contrast, the immobilized CoII sites
in P-1[CoII] display no detectable affinity for O2, CO2, and CO
when exposed to 30.5 µmol of each gas (Figure 6A).36 The
importance of the CoII ions in the NO binding process was
examined by gas uptake to the apo polymer, P-1, a polymer
without CoII ions coordinated to the immobilized salen sites
(Scheme 1). P-1 takes up only ∼3 µmol of NO; similar results
were obtained for the uptake of O2, CO2, and CO to P-1 (Figure
6B). The gas uptake values for P-1 are nearly identical with
those obtained for solid poly(EGDMA),15 a porous polymer
without immobilized sites (Figure S1). These results establish
that for P-1, only nonselective physical adsorption of gas is
occurring, and that intact CoII(salen) sites are necessary for
selective binding of NO in these polymers.
These findings are consistent with the binding properties
reported for similar square-planar CoII complexes. Square-planar
CoIIsalen complexes are very weak binders of O2; a change in
the cobalt coordination geometry to square pyramidal is needed
to allow O2 to bind at an appreciable level.13,15,37 The low
binding affinity of P-1[CoII] for CO2 and CO is also consistent
with the binding properties reported for other monomeric
analogues,38 where square-planar CoII complexes weakly bind
these two gases.
(37) (a) Cesarotti, E.; Gullotti, M.; Pasini, A.; Ugo, R. J. Chem. Soc.,
Dalton Trans. 1977, 757-763. (b) Meier, I. K.; Pearlstein, R. M.;
Ramprasad, D.; Pez, G. P. Inorg. Chem. 1997, 36, 1707-1714 and
references therein.
(36) These values are corrected for physical gas adsorption. See the
legend in Figure 6 and Experimental Section for details.

Immobilized Metal Complexes in Porous Organic Hosts
J. Am. Chem. Soc., Vol. 123, No. 6, 2001 1079
difference is attributed to the presence of the porous poly-
methacrylate host in these materials. The ability of the host to
isolate the metal sites from each other leads to different reactivity
than is observed for the corresponding monomers. Similar types
of results have been obtained for dioxygen binding polymers
made by template copolymerization methods.13,15 These results
highlight the importance of the immobilization process in tuning
metal ion reactivity.
The vivid color change resulting from the conversion of
P-1[CoII] to P-1[Co(NO)] suggests the potential of this system
for the detection of NO. Numerous methods are available for
detecting NO;40,41 however, most do not directly detect NO but
rather, sense adducts including NO2. In addition, few of the
methods reported to date are as durable as the highly cross-
linked polymer, P-1[CoII]. This polymer can be recycled,
retaining high binding affinity for NO, and functions equally
well in the solid state or as a suspension in a liquid. These
properties lend this system to use in a variety of detection
applications and efforts to incorporate materials of this kind
into sensor technology are ongoing.
Acknowledgment. We thank K. Trafford for assistance in
collecting the XAS data and Professor C. Loudon for helpful
discussions. Financial support of this work was provided by
ONR-DEPSCoR and NIH. The Bruker EMX-EPR spectrometer
was purchased with funds obtained from the ONR-DURIP
program.
Figure 6. Gas binding versus time for P-1[CoII] (A) and P-1 (B).
Legend of analyte gases: NO (O), O2 (4), CO2 (0), and CO (]). The
gas uptake plots in part A have been corrected for physical adsorption
by subtracting the values of P-1 binding from those of P-1[CoII] for
each gas, respectively.
Supporting Information Available: Table of XAS param-
eters (Table S1) and gas uptake to poly(EGDMA) (Figure S1)
(PDF). This material is available free of charge via the Internet
at http://pubs.acs.org.
JA003282B
Summary and Conclusions. Immobilization of metal com-
plexes in porous organic hosts from soluble molecular precursors
has proven to be an effective way of generating new functional
materials. The results presented in this paper demonstrate the
usefulness of this approach in developing a polymer for the
storage and release of NO. Spectroscopic measurements show
that the CoII(salen) complexes can be immobilized within a
porous polymethacrylate host and serve as sites for NO binding.
The CoII(salen) polymer, P-1[CoII], binds NO but is relatively
inert toward other biologically important gaseous compounds
such as O2, CO2, and CO. Nitric oxide is slowly released from
P-1[Co(NO)] under ambient conditions; ∼80% of NO is lost
from P-1[Co(NO)] after 30 days. The slow release of NO as
observed for P-1[Co(NO)] is advantageous for certain applica-
tions, such as the prevention of restenosis after angioplasty.39
The NO binding properties of P-1[CoII] and P-1[Co(NO)]
contrast those found for their molecular analogues. This
(39) Smith, D. J.; Chakravarthy, D.; Pulfer, S.; Simmons, M. L.; Hrabie,
J. A.; Citro, M. L.; Saavedra, J. E.; Davies, K. M.; Hutsell, T. C.; Mooradian,
D. L.; Hanson, S. R.; Keefer, L. K. J. Med. Chem. 1996, 39, 1148-1156.
(40) Selected references. Spectrophotometric: Green, L. C.; Wagner, D.
A.; Glogowski, J.; Skipper, P. L.; Wishnok, J. S.; Tannenbaum, S. R. Anal.
Biochem. 1982, 126, 131-138. Luminescence methods: Kikuchi, K.;
Nagano, T.; Hayakawa, H.; Hirata, Y.; Hirobe, M. Anal. Chem. 1993, 65,
1794-1799. Leone, A. M.; Furst, V. W.; Foxwell, N. A.; Cellek, S.;
Moncada, S. Biochem. Biophys. Res. Commun. 1996, 221, 37-41. Zhou,
X.; Arnold, M. A. Anal. Chem. 1996, 68, 1748-1754. Kojima, H.;
Nakatsubo, N.; Kikuchi, K.; Kawahara, S.; Kirino, Y.; Nagoshi, H.; Hirata,
Y.; Nagano, T. Anal. Chem. 1998, 70, 2446-2453. Barker, S. L. R.; Zhao,
Y.; Marletta, M. A.; Kopelman, R. Anal. Chem. 1999, 71, 2071-2075.
Electrochemical methods: Malinski, T.; Taha, Z. Nature 1992, 358, 676-
677. Pariente, F.; Alonso, J. L.; Abrun˜a, H. D. J. Electroanal. Chem. 1994,
379, 191-197. Friedemann, M. N.; Robinson, S. W.; Gerhardt, G. A. Anal.
Chem. 1996, 68, 2621-2628. Maskus, M.; Pariente, F.; Wu, Q.; Toffanin,
A.; Shapleigh, J. P.; Abrun˜a, H. D. Anal. Chem. 1996, 68, 3128-3134.
Mitchell, K. M.; Michaelis, E. K. Electroanalysis 1998, 10, 81-88. EPR
spectroscopy: Yoshimura, T.; Yokoyama, H.; Fujii, S.; Takayama, F.;
Oikawa, K.; Kamada H. Nature Biotech. 1996, 14, 992-994.
(38) Buckinham, D. A.; Clark, C. R. In ComprehensiVe Coordination
Chemistry; Wilkinson, G., Gillard, R. D., McCleverty, J. A., Eds.; Pergamon
Press: New York, 1987; Vol. 4, pp 635-900. Sweany, R. L. In
ComprehensiVe Organometallic Chemistry II; Abel, E. W., Stone, F. G.
A., Wilkinson, G., Eds.; Pergamon Press: New York, 1995; Vol. 8, p 8.
(41) For recent examples of molecular systems that induce spectral
changes upon NO binding see: (a) Franz, K. J.; Singh, N.; Lippard, S. J.
Angew. Chem., Int. Ed. Engl. 2000, 39, 2120-2122. (b) Rathore, R.;
Lindeman, S. V.; Rao, K. S. S. P.; Sun, D.; Kochi, J. K. Angew. Chem.,
Int. Ed. Engl. 2000, 39, 2123-2127.