Synthesis and properties of metalloporphyrin monolayers and stacked multilayers bound to an electrode via site specific axial ligation to a self- assembled monolayers

Article abstract of DOI:10.1021/ja973528l

A general method has been developed for the immobilization of metalloporphyrins at a gold electrode surface coated with a self-assembled monolayers (SAM). SAMs containing imidazole-terminated adsorbates are shown to bind a series of metalloporphyrins, including bis-acetonitrile octaethylporphyrinatoruthenium (II), Ru(OEP)(CH₃CN)₂; bis-acetonitrile octaethyltetraazaporphyrinato ruthenium(II), Ru(OETAP)(CH₃CN)₂, bis- acetonitrile tetra-(p-chlorophenyl)porphyrinatoruthrnium(II), Ru(Tp- CIPP)(CH₃CN)₂, bis-acetonitrile octaethylporphtrinatoosmium(II), Os(OEP)(CH₃CN)₂, and carbonyl meso-tetramesitylporphyrinatoruthenium(II), Ru(TMP)(CO). The SAM/metalloporphyrin films have been characterized by optical ellipsometry, contact angle goniometry, X-ray photoelectron spectroscopy, grazing angle FT-IR spectroscopy, transmission visible spectroscopy, and electrochemistry. The results indicated that the metalloporphyrins are chemisorbed via axial ligand substitution of the metal center with the porphyrin ring parallel to the surface and the second axial ligand position normal to the surface. Scanning tunneling microscopy images of Ru(TMP)(CO) bound to the SAM corroborate this model. Axial ligation of metalloporpyrins to SAMs serves as the basis for an iterative, defined approach to the preparation of stacked single component and mixed metalloporphyrin multilayers on SAms. In these materials, the bidentate ligand pyrazine serves as a bridge between successive metalloporphyrins in the stacks.

Full text of DOI:10.1021/ja973528l

4478
J. Am. Chem. Soc. 1998, 120, 4478-4487
Synthesis and Properties of Metalloporphyrin Monolayers and Stacked
Multilayers Bound to an Electrode via Site Specific Axial Ligation to
a Self-Assembled Monolayer
David A. Offord, Sandra B. Sachs, Matthew S. Ennis, Todd A. Eberspacher,
John H. Griffin,* Christopher E. D. Chidsey,* and James P. Collman*
Contribution from the Department of Chemistry, Stanford UniVersity, Stanford, California 94305
ReceiVed October 8, 1997
Abstract: A general method has been developed for the immobilization of metalloporphyrins at a gold electrode
surface coated with a self-assembled monolayer (SAM). SAMs containing imidazole-terminated adsorbates
are shown to bind a series of metalloporphyrins, including bis-acetonitrile octaethylporphyrinatoruthenium(II),
Ru(OEP)(CH3CN)2; bis-acetonitrile octaethyltetraazaporphyrinato ruthenium(II), Ru(OETAP)(CH3CN)2; bis-
acetonitrile tetra-(p-chlorophenyl)porphyrinatoruthenium(II), Ru(Tp-ClPP)(CH3CN)2; bis-acetonitrile octa-
ethylporphyrinatoosmium(II), Os(OEP)(CH3CN)2; and carbonyl meso-tetramesitylporphyrinatoruthenium(II),
Ru(TMP)(CO). The SAM/metalloporphyrin films have been characterized by optical ellipsometry, contact
angle goniometry, X-ray photoelectron spectroscopy, grazing angle FT-IR spectroscopy, transmission visible
spectroscopy, and electrochemistry. The results indicate that the metalloporphyrins are chemisorbed via axial
ligand substitution of the metal center with the porphyrin ring parallel to the surface and the second axial
ligand position normal to the surface. Scanning tunneling microscopy images of Ru(TMP)(CO) bound to the
SAM corroborate this model. Axial ligation of metalloporphyrins to SAMs serves as the basis for an iterative,
defined approach to the preparation of stacked single component and mixed metalloporphyrin multilayers on
SAMs. In these materials, the bidentate ligand pyrazine serves as a bridge between successive metalloporphyrins
in the stacks.
Introduction
Au(111)was used to anchor porphyrins directly to the gold
surface.6 Two clear advantages exist for the method presented
in this work over either EPGE or iodine-modified Au elec-
trodes: first, the role(s) played by surface functional groups in
metalloporphyrin adsorption and in modulation of redox proper-
ties may be better understood through the use of defined
electrodes which present a single type of axial ligand; second,
if the attachment site density can be changed, the final
concentration of the porphyrin on the surface will be controlled.
Electrodes coated with self-assembled monolayers (SAMs)
represent one such system, as the SAM end group functionality
is easily controlled.7 Due to their packing density and structural
rigidity, SAMs separate an electrode from a redox active species
via a thin hydrocarbon layer while still allowing for the passage
of current between the two. This has previously been demon-
strated by assembling alkyl mercaptans bearing covalently
attached electroactive end groups (e.g., ferrocene8 and ruthenium
pentaamine9) on gold electrodes. Metalloporphyrins have been
adsorbed to gold electrodes via alkyl mercaptan arms appended
to the porphyrin ring.10,11 Cobalt(II) porphyrins attached via
this method were found to catalyze the reduction of O2 to
H2O2.10b
The importance and versatility of the porphyrin macrocycle
and its metalated complexes in nature have inspired considerable
efforts to understand, mimic, and expand the role of metallo-
porphyrins through the use of model systems.1 The rich redox
chemistry of metalloporphyrins makes them ideal templates for
the design of new electron-transfer catalysts. Attaching redox
active metalloporphyrins to electrodes simplifies their electro-
chemical study and facilitates their use as catalysts in electro-
chemical cells.2 Most work in this area has involved metallo-
porphyrins adsorbed on edge-plane graphite electrodes (EPGE).3
Edge-plane graphite is a complex, heterogeneous surface
containing many functional groups (quinones, phenols, car-
boxylic acids, etc.),4 and it is believed that surface functionalities
influence the redox properties of EGPE-bound metallopor-
phyrins through axial ligation.5 Recently, iodine-modified
* To whom correspondence should be addressed.
(1) For a review, see: Collman, J. P.; Wagenknecht, P. S.; Hutchinson,
J. E. Angew. Chem., Int. Ed. Engl. 1994, 33, 1537-1554.
(2) Murray, R. W. Electroanalytical Chemistry: A Series of AdVances;
Bard, A. J., Ed.; Marcel Dekker: New York, 1984; Vol. 13.
(3) (a) Schick, G. A.; Schreiman, I. C.; Wagner, R. W.; Lindsey, J. S.;
Bocian, D. F. J. Am. Chem. Soc. 1989, 111, 1344-1350. (b) Wang, B.;
Cao, X. J. Electroanal. Chem. 1991, 309, 147-158. (c) Ramachandraiah,
G.; Bedioui, G.; Devynck, J. J. Electroanal. Chem. 1991, 319, 395-402.
(d) Ngameni, E.; Laoue´nan, A.; L′Her, M.; Hinnen, C.; Hendricks, N.;
Collman, J. P. J. Electroanal. Chem. 1991, 301, 207-226. (e) Collman, J.
P.; Chng, L.; Tyvoll, D. Inorg. Chem. 1995, 34, 1311-1324.
(4) Ni, C. L.; Anson, F. C. Inorg. Chem. 1985, 24, 4754-4756. (b) Shi,
C.; Anson, F. C. Inorg. Chem. 1990, 29, 4298-4305.
(6) Kunitake, M.; Akiba, V.; Batina, N.; Itaya, K. Langmuir 1997, 13,
1607-1615.
(7) For reviews of SAMs, see: (a) Whitesides, G. M.; Laibinis, P. E.
Langmuir 1990, 6, 87-96. (b) Ulman, A. An Introduction to Ultrathin
Organic Films from Langmuir Blodgett to Self-Assembly; Academic Press:
San Diego, 1991. (c) Dubois, L. H.; Nuzzo, R. G. Annu. ReV. Phys. Chem.
1992, 43, 437-463. (d) Ulman, A. Characterization of Organic Thin Films:
Materials Characterization Series-Surfaces, Interfaces, Thin Films;
Butterworth-Heinemann: Stoneham, MA, 1995.
(5) Bettelheim, A.; Chan, R. J. H.; Kuwana, T. J. Electroanal. Chem.
1979, 99, 391-397.
S0002-7863(97)03528-2 CCC: $15.00 © 1998 American Chemical Society
Published on Web 04/28/1998

Site Specific Axial Ligation to a Self-Assembled Monolayer
J. Am. Chem. Soc., Vol. 120, No. 18, 1998 4479
electrochemistry were distilled under argon from the following drying
agents immediately prior to use: methylene chloride (phosphorus
pentoxide), acetonitrile (calcium hydride), benzene and THF (sodium/
benzophenone ketyl), methanol (magnesium turnings).
In this paper, we describe general methods for the controlled
preparation of defined metalloporphyrin monolayers and stacked
multilayers on SAM-modified gold electrodes. The core feature
of these methods is axial ligation of metalloporphyrins to
imidazole groups present at the surface of a SAM. This provides
a well-defined environment for analytical investigations, pas-
sivates the electrode surface, and allows for incremental
variation in metalloporphyrin surface coverage. Relative to
previous methods for attaching metalloporphyrins to gold
electrodes, our approach is attractive in that it eliminates the
need for chemical modification of the metalloporphyrin. We
present the results of detailed surface analytical investigations
of the monofacial metalloporphyrin acetonitrile octaethylpor-
phyrinatoruthenium(II), (Ru(OEP)(CH3CN)), ligated to imid-
azole-terminated, gold-supported SAMs. We demonstrate the
generality of this approach to metalloporphyrin surface im-
mobilization with a series of additional ruthenium and osmium
porphyrin complexes. Finally, we show that axially ligated
SAM/metalloporphyrin monolayers can serve as templates for
the efficient, stepwise elaboration of pyrazine bridged, single
component and mixed stacks of metalloporphyrins, which are
of interest for their structural and physical properties.12 This
iterative approach to building multilayers could be implemented
as a solid-phase synthesis of novel metalloporphyrin materials.12g
Adsorbates. n-Nonanethiol (Aldrich, 95%) and n-decanethiol
(Aldrich, 96%) were purified by bulb-to-bulb distillation at ∼0.075
Torr prior to use. 1-(10-mercaptodecyl)imidazole was prepared ac-
cording to a literature procedure.13 Final purification of this compound
was carried out on a Perkin-Elmer LC200 HPLC equipped with a Rainin
Dynamax 300A C18 preparatory column (25.4 cm × 21.4 mm i.d.,
12-µm particle size) and a gradient of acetonitrile in water (with 0.1%
trifluoroacetic acid, TFA). After an initial 5 min flow of 100% water
at 5 mL/min, an acetonitrile gradient was introduced at a rate of 0.25%/
min. The eluate was monitored at 207 nm, and the product eluted at
27% acetonitrile. The acetonitrile was removed under reduced pressure,
and the remaining solution was extracted with three 100 mL portions
of argon-sparged ethyl ether. The extracts were combined and washed
with 100 mL of argon-sparged saturated aqueous sodium bicarbonate
solution and three 100 mL portions of distilled water. To suppress
disulfide formation, special care was taken to exclude oxygen during
1
the isolation of the product. The H NMR, 13C NMR, and IR (KBr)
spectra for the compound matched those reported.13 Liquid secondary
ion mass spectrum (LSIMS): MH+ calcd m/z 241.2; found m/z 241.2.
Bis-acetonitrile Octaethylporphyrinatoruthenium(II), Ru(OEP)-
(CH3CN)2. Ru(OEP)(CH3CN)2 was synthesized by photolysis (mercury
arc lamp) of Ru(OEP)(CO)(MeOH) (60 mg, 0.086 mmol) in acetonitrile
(250 mL) under an argon purge following standard literature proce-
dures.14 The product was isolated in 65% yield (41 mg). UV-vis
(toluene) λmax 406 (Soret), 502, 528 nm. IR (KBr) 2259 cm-1 (νCN).
1H NMR (C6D6) Hmeso 9.98 (s, 4H); CH2 4.02 (q, 7.6 Hz, 16H); CH3
2.04 (t, 7.6 Hz, 24 H); CH3CN -2.60 (s, 6H) ppm. LSIMS: M+ (102Ru)
calcd m/z 716.3; found m/z 716.3.
Experimental Section
General. Manipulations of oxygen- and water-sensitive compounds
were performed in a nitrogen-filled Vacuum Atmospheres Co. drybox
maintained at or below 2 ppm O2. Oxygen levels were monitored with
an AO 316-C trace oxygen analyzer.
Bis-acetonitrile Octaethyltetraazaporphyrinatoruthenium(II),
Ru(OETAP)(CH3CN)2. Following a modified literature procedure,14
a 20 mL glass photolysis well equipped with a Teflon vacuum valve
was charged with Ru(OETAP)(CO)(CH3OH) (15 mg, 0.022 mmol) and
acetonitrile (12 mL). After two freeze/pump/thaw degassing cycles,
the sample was photolyzed for 20 min and subjected to another freeze/
pump/thaw cycle. The photolysis/freeze/pump/thaw procedure was
repeated three times. The photolysis well was then taken into an inert
atmosphere (N2) glovebox where the solvent was removed in vacuo.
The residue was purified by column chromatography on neutral alumina.
Using a 1:1 mixture of acetonitrile and tetrahydrofuran, the product
eluted as the major purple band. The solvent was removed in vacuo,
and the resulting solid was dissolved in a minimal amount of
acetonitrile. Upon drying, 14 mg (87% yield) of product was recovered.
UV-vis (CH2Cl2) λmax 560, 516, 540. IR (KBr) 2274 cm-1 (νCN). 1H
NMR (C6D6) CH2 3.99 (q, 7.6 Hz, 16H); CH3 2.10 (t, 7.6 Hz, 24 H);
Solvents. Ethanol (Gold Shield, 200 proof) was sparged with argon
before use. Deionized/distilled water was used for all washings. Water
purified by passage through a Milli-Q system was used for contact
angle measurements. Methylene chloride used for routine rinsing of
SAMs was obtained from Fisher and used as received. Solvents used
for metalloporphyrin purification, metalloporphyrin deposition, and
(8) (a) Chidsey, C. E. D.; Bertozzi, C. R.; Putvinski, T. M.; Mujsce, A.
M. J. Am. Chem. Soc. 1990, 112, 4301-4306. (b) Creager, S. E.; Rowe,
G. K. Anal. Chim. Acta 1991, 246, 233-239. (c) Chidsey, C. E. D. Science
1991, 251, 919-922. (d) Collard, D. M.; Fox, M. A. Langmuir 1991, 7,
1192-1197. (e) Rowe, G. K.; Creager, S. E. Langmuir 1991, 7, 2307-
2312. (f) De Long, H. C.; Donohue, J. J.; Buttry, D. A. Langmuir, 1991, 7,
2196-2202. (g) Groat, K. A.; Creager, S. E. Langmuir 1993, 9, 3668-
3675. (h) Curtin, L. S.; Peck, S. R.; Tender, L. M.; Murray, R. W.; Rowe,
G. K.; Creager, S. E. Anal. Chem. 1993, 65, 386-392. (i) Caldwell, W. B.;
Chen, K.; Herr, B. R.; Mirkin, C. A.; Hulteen, J. C.; Van Duyne, R. P.
Langmuir 1994, 10, 4109-4115. (j) Herr, B. R.; Mirkin, C. A. J. Am. Chem.
Soc. 1994, 116, 1157-1158.
(9) Finklea, H. O.; Snider, D. A.; Fedyk, J. Langmuir 1990, 6, 371-
378. (b) Finklea, H. O.; Hanshew, D. D. J. Am. Chem. Soc. 1992, 114,
3173-3181. (c) Finklea, H. O.; Hanshew, D. D. J. Electroanal. Chem. 1993,
347, 327-340. (d) Finklea, H. O.; Ravenscroft, M. S.; Snider, D. A.
Langmuir 1993, 9, 223-227. (e) Redepenning, J.; Tunison, H. M.; Finklea,
H. O. Langmuir 1993, 9, 1404-1407.
CH3CN -2.53 (s, 6H) ppm. LSIMS: M+
found m/z 720.2.
(
102Ru) calcd m/z 720.3;
Bis-acetonitrile octaethylporphyrinatoosmium(II), Os(OEP)-
(CH3CN)2 was prepared from Os(OEP)(CO)(MeOH) using a procedure
similar to that outlined for Ru(OETAP)(CH3CN)2. From 18 mg, 0.023
mmol, of starting material, 13 mg (70%) of the product was obtained.
UV-vis (CH2Cl2) λmax 388 (Soret), 486, 514. IR (KBr) 2237 cm-1
(νCN). 1H NMR (C6D6) Hmeso 9.30 (s, 4H); CH2 3.91 (q, 7.5 Hz, 16H);
CH3 1.98 (t, 7.5 Hz, 24 H); CH3CN -2.80 (s, 6H) ppm. LSIMS: M+
(10) (a) Zak, J.; Yuan, H.; Ho, K.; Woo, L. K.; Porter, M. D. Langmuir
1993, 9, 2772-2774. (b) Hutchison, J. E.; Postlethwaite, T. A.; Murray, R.
W. Langmuir 1993, 9, 3277-3283. (c) Tyvoll, D. A., Ph.D. Thesis, Stanford
University, Stanford, CA, 1995.
(
192Os) calcd m/z 806.4; found m/z 806.3.
(11) Metalloporphyrins contained within enzymes have also been ad-
sorbed onto gold-thiolate SAMs. Attachment has been achieved by
electrostatic forces (Tarlov, M. J.; Bowden, E. F. J. Am. Chem. Soc. 1991,
113, 1847-1849) and by carbodiimide linkages through amino acid side
chains (Collinson, M.; Bowden, E. F. Langmuir 1992, 8, 1247-1250).
(12) Hopfield, J. J.; Onuchic, J. N. Beratan, D. N. Science 1988, 241,
817-820. (b) Chen, J.; Seeman, N. C. Nature 1991, 350, 631-633. (c)
Mathias, J. P.; Stoddart, J. F. Chem. Soc. ReV. 1992, 215-225. (d)
Kaszynski, P.; Friedli, A. C.; Michl, J. J. Am. Chem. Soc. 1992, 114, 601-
620. (e) Wasielewski, M. R. Chem ReV. 1992, 435-461 and references
therein. (f) Segawa, H.; Kunimoto, K.; Susumu, K.; Taniguchi, M.;
Shimidzu, T. J. Am. Chem. Soc. 1994, 116, 11193-11194. (g) Xiang, X.-
D.; Sun, X.; Briceqo, G.; Lou, Y.; Wang, K.-A.; Chang, H.; Wallace-
Freedman, W. G.; Chen, S.-W.; Schultz, P. G. Science 1995, 268, 1738-
1740.
meso-Tetrakis(pentafluorophenyl)porphyrinatoruthenium(II), Ru-
(TPFPP). The following procedure was adapted from the literature.15,16
Ru(TPFPP)(CH3CN)2 (18 mg, 0.016 mM) was lyophilized from benzene
(13) Lee, T. R.; Carey, R. I.; Biebuyck, H. A.; Whitesides, G. M.
Langmuir 1994, 10, 741-749.
(14) Anipas, A.; Buchler, J. W.; Gouterman, M.; Smith, P. D. J. Am.
Chem. Soc. 1978, 100, 3015-3024.
(15) Camenzind, M. J.; James, B. R.; Dolphin, D., J. Chem. Soc., Chem.
Commun. 1986, 1137-1139.
(16) (a) Lindsey, J. S.; Wagner, R. W.; J. Org. Chem. 1989, 54, 828-
836. (b) Chang, C. K.; Ebina, F. J. Chem. Soc., Chem. Commun. 1981,
778-779.

4480 J. Am. Chem. Soc., Vol. 120, No. 18, 1998
Offord et al.
(5 mL) and then pyrolyzed at 2 × 10-5 Torr and 230 °C for 5 h to
yield Ru(TPFPP) (8 mg, 0.0075 mmol, 47%). 1H NMR (C6D6) Hmeso
8.01 (s, 8 H) ppm. UV-vis (toluene) λmax 410 (Soret), 510, 530 nm.
Bis-dinitrogen meso-tetramesitylporphyrinatoruthenium(II),
Ru(TMP)(N2)2 was prepared according to the literature procedure.15
UV-vis (C6H6) λmax 408 (Soret), 509, 525 nm. 1H NMR (C6D6) Hâ
8.82 (s, 8 H); mesityl Hm 7.20 (s, 8 H); mesityl CH3 p-CH3 2.49 (s, 12
H), o-CH3 2.10 (s, 24 H) ppm.
chloride, distilled water, and ethanol or methanol. Substrates were then
blown dry under a stream of dinitrogen and characterized.
Porphyrin Deposition Procedure. In an inert atmosphere box,
metalloporphyrins were deposited onto SAMs from a stirred 1 mM
solution of metalloporphyrin in methylene chloride or benzene. The
gold surfaces were removed from the porphyrin solution after 4 h, rinsed
thoroughly, and allowed to dry in the box atmosphere prior to
characterization.
Carbonyl meso-Tetramesitylporphyrinatoruthenium(II), Ru-
(TMP)(CO).15 A sealed NMR tube containing Ru(TMP) (6 mg, 0.0068
mmol) and deuterated benzene (2.5 mL) was opened under a carbon
monoxide atmosphere after a freeze/pump/thaw cycle. The dark
brownish-red solution immediately turned a light orange-red. The
excess carbon monoxide was removed by two freeze/pump/thaw cycles.
The sample was taken into an inert (N2) box where the solvent was
removed in vacuo (6 mg, quantitative yield). UV-vis (CH2Cl2) λmax
412 (Soret), 528. IR (CH2Cl2) 1942 cm-1 (νCO). 1H NMR (C6D6) Hâ
8.77 (s, 8H); mesityl Hm 7.19 (s, 4H); mesityl Hm′ 7.05 (s, 4H); mesityl
CH3 p-CH3 2.43 (s, 12H), o-CH3 2.15 (s, 12H), o′-CH3 1.78 (s, 12H)
Multilayer Deposition Procedure. This procedure is exemplified
by the preparation of SAM/Ru(TMP)/pyrazine/Ru(TMP)(CO). In an
inert (N2) atmosphere box, Ru(TMP)(N2) was deposited onto a mixed
monolayer from a stirred 1 mM solution of Ru(TMP)(N2)2 in benzene.
The gold surface was removed from the porphyrin solution after 15
min and rinsed thoroughly with benzene and allowed to dry in the box
atmosphere. This surface was then placed in a stirred 1 mM solution
of pyrazine in methylene chloride for 15 min and rinsed thoroughly
with methylene chloride and benzene. This SAM/Ru(TMP)/pyrazine
surface was once again placed in a stirred 1 mM solution of Ru(TMP)-
(N2)2 in benzene for 15 min and rinsed thoroughly with benzene. The
resulting assembly was capped by placing the surface under an
atmosphere of pure CO (15 psig) for 1 h.
ppm. LSIMS: M+ 102Ru) calcd m/z 910.3; found m/z 910.2.
(
Carbonyl Octaethylporphyrinatoosmium(II), Os(OEP)(CO). [Os-
(OEP)]2 (7 mg, 0.0047 mmol) was quantitatively converted to Os(OEP)-
(CO) following a procedure described for the synthesis of Ru(TMP)-
(CO).15 UV-vis (CH2Cl2) λmax 390 (Soret), 508, 538 nm. IR (CH2Cl2)
1888 cm-1 (νCO). 1H NMR (C6D6) Hmeso 9.97 (s, 4 H); -CH2- 3.90
Contact Angle Goniometry and Optical Ellipsometry. Measure-
ment of contact angles and ellipsometric thicknesses are described
elsewhere21 and in the Supporting Information. A refractive index of
1.50 for the SAMs and SAM/metalloporphyrin films was assumed.22
(m, 16 H); CH3 1.88 (t, 24 H) ppm. LSIMS: M+ 192Os) calcd m/z
(
X-ray Photoelectron Spectroscopy (XPS). X-ray photoelectron
spectra were obtained on a Surface Science model 150 XPS spectrom-
eter equipped with an Al KR X-ray source, quartz monochromator,
concentric hemispherical analyzer operating in constant analyzer mode,
and a multichannel detector. The pressure in the analytical chamber
during analysis was approximately 2 × 10-8 Torr. A takeoff angle of
35° from the surface was employed. Spectra of the C(1s) and Ru(3d)
region (274-294 eV binding energy, 20 scans), the N(1s) region (388-
408 eV binding energy, 70 scans), the Cl(2p1/2,2p3/2) region (191-
211 eV binding energy, 20 scans), and the Au(4f) region (76-96 eV
binding energy, 10 scans) were recorded with a 50 eV pass energy and
a 150 × 800 µm spot size. The areas under the unsmoothed C(1s),
Ru(3d), N(1s), Cl(2p1/2, 2p3/2), and Au(4f) peaks were measured (after
a Shirley background subtraction) and corrected for the number of scans.
752.3; found m/z 752.3.
Bis-pyrazine octaethylporphyrinatoruthenium(II), Ru(OEP)-
(pyrazine)2, was prepared according to the literature procedure.17 UV-
vis (toluene/pyrazine) λmax 402 (Soret), 494, 520 nm. 1H NMR (C6D6)
Hmeso 9.82 (s, 4 H); CH2CH3 3.91 (m, 16 H); CH2CH3 3.88 (t, 24 H);
H
(pyrazine) 5.20 (d, 4 H); H(pyrazine) 1.79 (d, 4 H) ppm. LSIMS: M+ (102Ru)
calcd m/z 794.3; found m/z 794.4.
Substrates. Polished silicon(100) wafers (2 in. diameter, Pure-Sil
or 4 in. diameter, SiliconValley Microelectronics, Inc.) were cleaned
by immersion in “piranha” solution (1:1 vol/vol 30% aqueous H2O2:
H2SO4) at 20 °C for 10 min. (WARNING: Piranha solution reacts
violently, even explosively, with organic materials.18 It should not be
stored or combined with significant quantities of organic material.)
Clean substrates were rinsed thoroughly with deionized/distilled water,
dried in a 2-propanol vapor bath, blown dry under a stream of argon,
and transferred to an electron-beam vapor deposition chamber (5 ×
10-7 Torr). Gold(111) substrates19 were prepared by evaporating 10
nm of titanium followed by 100 nm of gold. Gold(111) substrates for
STM study were prepared by electron-beam evaporation of 150 nm of
gold (0.3 nm/s) onto freshly cleaved mica. The gold-coated mica was
removed from the deposition chamber and annealed20 in a quartz tube
furnace at 480 °C for 24 h under a continuous flow of argon (scrubbed
by hot titanium turnings).
Monolayer Formation. Mixed SAMs were formed in a nitrogen-
filled inert atmosphere drybox by immersing freshly prepared gold
substrates in solutions containing mixtures of mercaptans in oxygen-
free ethanol or methanol (35 mL volume, 1 mM total mercaptan
concentration). Each deposition solution was capped and stirred for
15 min before a freshly prepared gold substrate was introduced. The
container was capped and stored at room temperature for 48 h, at which
time the samples were removed from the drybox and rinsed successively
with approximately 10 mL each of ethanol or methanol, methylene
Grazing Angle FT-IR. Grazing angle (85° off surface normal)
infrared spectra were obtained using p-polarized light on a Mattson
RS 10000 FT-IR spectrometer. The spectra were taken in a dry air-
filled sample compartment (water and carbon dioxide was removed
by a Balston 74-5041 pure air generator). Five thousand scans were
recorded at a resolution of 2 cm-1
. A narrow-band MCT detector,
cooled with liquid nitrogen, was used to detect the reflected light. The
moving mirror speed was 2.5 cm/s (40 kHz) in the forward (sampling)
direction. The SAM spectra were ratioed to a background spectrum
of a perdeuterated hexadecyl mercaptan SAM on gold.
Electrochemical Measurements. Surface electrochemical measure-
ments were made in a cell formed by pressing a bored-out cone of
poly(tetrafluoroethylene) against the monolayer. This defined the
electrode area (0.43 cm2).8a The electrolyte (0.2 M tetrabutylammonium
hexafluorophosphate in methylene chloride), the counter electrode
(platinum wire), and the reference electrode (Ag/Ag+) were then placed
in the bore. Solution electrochemistry was performed in a 3 mL conical
vial with the same electrolyte, counter electrode, and reference electrode
used in the surface electrochemistry, but employing a gold disk as the
working electrode. Reference electrodes in both experiments were
calibrated by measuring the ferrocene (Fc) redox potential which is
known to be (+0.66V vs NHE)23 in methylene chloride. All electro-
chemical measurements were made in an inert (N2) atmosphere box
(17) Collman, J. P.; McDevitt, J. T.; Yee, G. T.; Leidner, C. R.;
McCullough, L. G.; Little, W. A.; Torrance, J. B. Proc. Natl. Acad. Sci.
U.S.A. 1986, 83, 4581-4585.
(18) (a) Dobbs, D. A.; Bergman, R. G.; Theopold, K. H. Chem. Eng.
News 1990, 17, 2. (b) Wnuk, T. Chem. Eng. News 1990, 26, 2. (c) Matlow,
S. L. Chem. Eng. News 1990, 30, 2.
(19) Polycrystalline gold substrates prepared by cold evaporative tech-
niques have been shown to have extremely strong (>80%) Au(111)
texture: (a) Chidsey, C. E. D.; Loiacono, D. N.; Sleator, T.; Nakahara, S.
Surf. Sci. 1988, 200, 45-66. (b) Chidsey, C. E. D.; Loiacono, D. N.
Langmuir 1990, 6, 682-691. (c) Nuzzo, R. G.; Fusco, F. A.; Allara, D. L.
J. Am. Chem. Soc. 1987, 109, 2358-2368.
(21) Offord, D. A.; John, C. M.; Griffin, J. H. Langmuir 1994, 10, 761-
766.
(22) Swalen, J. D.; Santo, R.; Take, M.; Fischer, J. IBM J. Res. DeV.
1977, 21, 169-175. The refractive index of metalloporphyrins on the SAM
is unknown. Therefore, the observed thickness increases are qualitative rather
than quantitative measurements.
(20) DeRose, J. A.; Thundat, T.; Nagahara, L. A.; Lindsay, S. M. Surf.
Sci. 1991, 256, 102-108.
(23) Astruc, Didier. Electron Transfer and Radical Processes in Transi-
tion-Metal Chemistry; VCH Publishers: New York, 1995; p 144.

Site Specific Axial Ligation to a Self-Assembled Monolayer
J. Am. Chem. Soc., Vol. 120, No. 18, 1998 4481
Figure 1. Deposition procedure and proposed mode of attachment of
M(porphyrin)(L2) to a mixed SAM of C9-SH and imid-SH on a planar
gold substrate. L1 is the more labile ligand.
Figure 2. Comparison of the XPS N1s between several monolayers
made on Au(111). (a) A C9-SH monolayer after immersing in 1 mM
Ru(OEP)(CH3CN)2 for 4 h. (b) A mixed monolayer of C9-SH and imid-
SH formed where øsoln ) 0.5. (c) The same monolayer described in (b)
after immersing in 1 mM Ru(OEP)(CH3CN)2 for 4 h.
with a PAR 273 potentiostat interfaced to a personal computer. The
PAR 273 was modified to provide an input potential that was a smooth
ramp in time.
Scanning Tunneling Electron Microscopy. Images were obtained
in air on a Nanoscope-III instrument (Digital Instruments) equipped
with a Pt/Ir wire tip sharpened by mechanical cutting. All images were
obtained in height mode (constant current) and low-pass filtered. The
x and y distances of the piezo scanner were calibrated with a pyrolytic
graphite surface. The z distance was calibrated to the atomic steps on
annealed gold substrates. Voltages and currents for individual images
are given in the figure heading. To avoid destructive imaging, very
large tunneling impedances were employed. Generally, sample-to-tip
resistances were in the range of 0.6 and 33 GΩ.
SH (øsoln ) [Imid-SH]/([Imid-SH] + [C9-SH])) was varied from
0 to 1 in order to produce monolayers with a range of surface
compositions (øSAM ) [Imid-S-Au]/([Imid-S-Au] + [C9-S-Au])).
The resulting series of mixed SAMs was then immersed in a 1
mM solution of Ru(OEP)(CH3CN)2 in methylene chloride under
a dinitrogen atmosphere for 4 h24 at room temperature to yield
the SAM/Ru(OEP)(CH3CN) series. Monolayers were charac-
terized by X-ray photoelectron spectroscopy (XPS), optical
ellipsometry, contact angle goniometry, grazing angle FT-IR
spectroscopy, electrochemistry, transmission visible spectros-
copy,25 and scanning tunneling microscopy (STM).
Results
I. Axially Ligated Metalloporphyrin Monolayers. Ap-
proach. In the approach to metalloporphyrin surface im-
mobilization presented here, a specific end group of a previously
formed SAM serves as the axial ligand for the metal complex
(Figure 1). An imidazole was chosen as the end group for these
studies due to its high affinity for the metalloporphyrins of
interest and its lack of competitive chemistry. An imidazole-
terminated SAM is formed by immersing a gold surface in a
solution containing a long-chain adsorbate bearing mercaptan
and imidazole terminal groups. The SAM is then immersed in
a solution of a metalloporphyrin bearing weakly bound or no
axial ligands. The metalloporphyrin is immobilized by axial
ligation to the SAM. The hydrocarbon backbone of the SAM
holds the complex at a defined distance from the electrode and
passivates the electrode surface.
Self-Assembled Mixed Monolayers (SAMs) and Self-
Assembled Mixed Monolayers with Bound Ru(OEP)-
(CH3CN)-SAM/Ru(OEP)(CH3CN). A series of mixed SAMs
formed from the combination of n-nonanethiol (C9-SH) and
1-(10-mercaptodecyl)imidazole (imid-SH) were prepared and
characterized. Monolayers were deposited by immersing freshly
prepared gold(111)-coated silicon wafers in ethanol or methanol
solutions of the adsorbates (1 mM total concentration) for 2
days at room temperature. The solution mole fraction of Imid-
Due to the large van der Waals area of the porphyrin
macrocycle (200 Å2)26 relative to that of alkyl mercaptan
adsorbates (21.4 Å2),27 it is expected that only a fraction of the
imidazoles in a øsoln ) 1.0 SAM could ligate a metalloporphyrin
as indicated in Figure 1. If the imidazole-terminated adsorbates
were perfectly distributed, maximum metalloporphyrin coverage
would be reached at øSAM ) 0.11.
XPS. The nonanethiol (C9-SH) and 1-(10-mercaptodecyl)-
imidazole (imid-SH) SAMs and the imid-SH SAM/Ru(OEP)-
(CH3CN) and SAM/Ru(Tp-ClPP)(CH3CN) series were charac-
terized by XPS. We used this technique to determine the surface
atomic compositions which show the presence or absence of
certain species in the monolayer under various deposition
conditions. Figure 2 compares the N(1s) regions of the XPS
spectra of three different monolayers: C9-SH SAM submersed
in Ru(OEP)(CH3CN) for 4 h, C9-SH/imid-SH mixed monolayer,
and C9-SH/imid-SH monolayer submersed in Ru(OEP)(CH3CN)
for 4 h. Each of the three monolayers were treated with the
(24) Longer deposition times (up to 24 h) resulted in no noticeable
increase in the amount of adsorbed metalloporphyrin.
(25) These data are presented in the Supporting Information.
(26) Collman, J. P.; Eberspacher, T. A., unpublished crystallographic
data.
(27) Strong, L.; Whitesides, G. M. Langmuir 1988, 4, 546-558.

4482 J. Am. Chem. Soc., Vol. 120, No. 18, 1998
Offord et al.
Figure 3. Comparison of the XPS C1s region between several
monolayers made on Au(111). (a) A C9-SH monolayer after immersing
in 1 mM Ru(OEP)(CH3CN)2 for 4 h. (b) A mixed monolayer of C9-
SH and imid-SH formed where øsoln ) 0.5. (c) The same monolayer
described in (b) after immersing in 1 mM Ru(OEP)(CH3CN)2 for 4 h.
Figure 4. Ellipsometric thickness and contact angle goniometry data
as a function of øsoln for mixed monolayers of C9-SH and imid-SH
before (circles) and after (squares) immersing in 1 mM Ru(OEP)-
(CH3CN)2 for 4 h: (a) ellipsometric thickness; (b) advancing water
contact angles (θa(H2O)). Curves are provided as guides to the eye.
same rinse procedure. In Figure 2a, one notes the distinct
absence of nitrogen signal for the C9 monolayer, as expected,
since there is no ligand anchor present to dock the porphyrin,
the only nitrogen-containing species present in this experiment.
Figure 2b shows that the imidazole nitrogens in the monolayer
matrix yield a signal indicative of the presence of only one
species of nitrogen. In Figure 2c, the nitrogen signal is
partitioned into two discernible peaks: the sharper peak at lower
binding energy is assigned as the porphyrin nitrogens while the
broader peak at higher binding energy corresponds to the
imidazole nitrogens as seen in panel b. One can conclude
qualitatiVely that there is less than one porphyrin for every
imidazole present in the monolayer because the integrals of the
peaks in Figure 2c are roughly the same. If there were a
metalloporphyrin for every imidazole, the number of nitrogens
represented in the integral would be four to two. However,
that analysis does not take into account attenuation factors for
the imidazole nitrogens which are buried under the porphyrin.
Any correction of this type would only strengthen the qualitative
argument. We conclude that the porphyrin coverage is not
limited by a shortage of ligand sites.
and on an imidazole-containing monolayer, can be found in the
Supporting Information.
Ellipsometry. A series of SAMs and the resulting SAM/
Ru(OEP)(CH3CN) films were characterized by optical ellip-
sometry to determine film thickness (Figure 4a). Thicknesses
for the SAMs increased from 13 to 19 Å as øsoln was increased
from 0 to 1. The thickness of the SAM/Ru(OEP)(CH3CN) films
increased from 14 to 32 Å over the same range of øsoln. The
ellipsometric thickness values at øsoln ) 0 for the SAM, and
the SAM/Ru(OEP)(CH3CN) films were the same within ex-
perimental precision ((2 Å). Thicknesses of the SAM series
that had been exposed to Ru(OEP)(CH3CN)2 increased at a
greater rate than the SAM series until a maximum difference
(∆d ) 13 Å) was reached with the øsoln ) 1 surface.
Contact Angle Goniometry. The SAM series and the
resulting SAM/Ru(OEP)(CH3CN) series were characterized by
water contact angle goniometry to determine wettability. The
observed contact angles are plotted as a function of øsoln in
Figure 4b. SAM advancing water contact angles decreased with
øsoln from 115° (øsoln ) 0) to 59° (øsoln ) 1). Following
immersion of the SAMs in the Ru(OEP)(CH3CN)2 solution, the
contact angles initially decreased with øsoln but approached a
limiting value of 85° near øsoln ) 0.50. Thus, SAM/
Ru(OEP)(CH3CN) surfaces reach limiting wetting properties at
moderate surface concentrations of imidazole groups. This
behavior indicates that the surface is covered with metallopor-
phyrins near øsoln ) 0.50; therefore øsoln ) 0.5 was used for
monolayer formation in all XPS and electrochemical studies.
Electrochemistry. Figure 5a presents the cyclic voltammo-
gram (CV) of Ru(OEP)(CH3CN)2 in solution as well as the CV
of the blank electrolyte. Figure 5b shows Ru(OEP)(CH3CN)
after attachment to a C9-SH/imid-SH monolayer along with two
control experiments: a C9-SH/imid-SH monolayer without
Figure 3 shows the carbon region for the same set of
monolayers described above. In Figure 3a, the carbon signal
for a C9-SH monolayer is shown. The second plot, Figure 3b,
shows a relative increase of the carbon signal to that shown in
Figure 3a when a mixed monolayer of C9-SH/imid-SH is
presentsmore carbons are in the monolayer; therefore an
increase is expected. An imid-SH/Ru(OEP)(CH3CN) monolayer
is shown in Figure 3c. The signature feature in this panel is
the presence of a distinct peak at lower binding energy than
the carbon. This peak is assigned to Ru(3d) whose only origin
can be from the porphyrin. The nitrogen and carbon XPS data
show that the porphyrin only attaches to the monolayer surface
when there is a site present to which it can bind. XPS data for
another porphyrin, Ru(Tp-ClPP)(CO)(CH3CN) in a KBr matrix

Site Specific Axial Ligation to a Self-Assembled Monolayer
J. Am. Chem. Soc., Vol. 120, No. 18, 1998 4483
Table 1. Advancing Water Contact Angles and Ellipsometric
Thicknesses of Ru(OEP)(CH3CN), Ru(OETAP)(CH3CN),
Os(OEP)(CH3CN), and Ru(TMP)(CO) Metalloporphyrins Bound to
a Monolayer (øsoln ) 0.75) Compared to the Corresponding SAM
without Metalloporphyrin
surface
θa(H2O)
thickness (Å)
SAM
79
88
89
90
93
15
25
28
24
25
SAM/Ru(OEP)(CH3CN)
SAM/Ru(OETAP)(CH3CN)
SAM/Os(OEP)(CH3CN)
SAM/Ru(TMP)(CO)
Integration of the first set of redox peaks at 0.103 V and
normalization by the scan rate and electrode area gives an
average charge density of 5.3 × 10-6 C/cm2. Using the same
procedure on the second set of peaks32 at 1.263 V yields an
average charge density of 5.7 × 10-6 C/cm2. Dividing the
average of these values by the charge on an electron, an effective
area of 291 Å2 per Ru(OEP)(CH3CN) is calculated. As
described in the Supporting Information, one physical interpre-
tation of the electrochemical data comes from a proposed
x
x
packing model involving a
13x 13R13.9° superlattice
which allows commensurate packing of porphyrins 200 Å2 in
x
x
size on the 13x 13R30° lattice. The model gives an
effective area of 281 Å2 per porphyrin, in good agreement with
our experimental result, suggesting that our electrochemical
coverage is indeed reasonable.
Figure 5. CVs showing redox couples of (a) solid line, Ru(OEP)-
(CH3CN)2 in 0.2 M tetrabutylammonium hexafluorophosphate/meth-
ylene chloride solution; dotted line, 0.2 M tetrabutylammonium
hexafluorophosphate in methylene chloride; (b) solid line, a øsoln
)
Chemisorption versus Physisorption of Metalloporphyrins
at the SAM Surface. Experiments were carried out to
determine whether specific ligation to the imidazole terminus
of the SAM was critical for metalloporphyrin binding. First,
attempts to deposit Ru(OEP)(CH3CN)2 from methylene chloride
solution in the presence of 1 mM trifluoroacetic acid were
unsuccessful. The metalloporphyrin was determined to be
0.5 SAM/Ru(OEP)(CH3CN); dotted line, C9-SH soaked in Ru(OEP)-
(CH3CN)2 for 4 h and rinsed as per the procedure described; dot-
dashed line, SAM (øsoln ) 0.5) without porphyrin soak. The scan rate
in all cases was 100 mV/s.
porphyrin and a C9-SH monolayer soaked in Ru(OEP)(CH3-
CN)2. The solution electrochemistry (Figure 5a) shows three
reversible redox couples of Ru(OEP)(CH3CN)2 with peaks at
0.304, 1.174, and 1.650 V vs NHE as well as an irreversible
set straddling 0.9 V vs NHE.28 The first oxidation peak at 0.304
V is known to be the RuIII/II couple,29 while the second and
third peaks are most likely the RuIV/III and porphyrin-centered
oxidations;29-31 however, the electrochemical assignment of
peaks to particular chemical species is unresolved.30,31 When
Ru(OEP)(CH3CN) is attached to the SAM surface, we assign
the RuIII/II peak to be at 0.103 V vs NHE (Figure 5b) and the
second peak at 1.263 V vs NHE to be either the RuIV/III or a
porphyrin-centered oxidation. The surface-bound porphyrin CV
has symmetric peaks reflecting the fact that the redox species
no longer have to diffuse to the electrode; it also shows that
the electron transfer is kinetically fast on the time scale of the
experiment, otherwise the oxidation and reduction waves would
occur at different potentials. Ten CV scans caused no noticeable
loss of peak area over the potential range in Figure 5b, indicating
reasonable stability of the monolayer. The C9-SH/imid-SH
monolayer CV has a larger background current than the
unfunctionalized C9-SH SAM presumably because the imid-
azoles in the mixed monolayer do not pack as well as a pure
C9-SH SAM. This allows electrolyte ions to approach closer
to the electrode and produce a larger charging current.19b
From the surface-bound Ru(OEP)(CH3CN) CV in Figure 5b,
we can quantify the porphyrin coverage on the monolayer.
1
unaffected by the dilute acid through H NMR and UV-vis
analysis. Characterization of the protonated SAM revealed that
the dilute acid did not disrupt the monolayer but did reduce the
advancing water contact angle as expected if the imidazole
termini were protonated. Second, attempts to deposit Ru(OEP)-
(1,5-dicyclohexylimidazole)2, in which the axial ligands are
much less labile than those of Ru(OEP)(CH3CN)2, were
unsuccessful. Finally, the XPS results and electrochemistry data
for C9-SH monolayers treated with porphyrin that were
discussed previously bolster the argument that physisorption
onto methyl-terminated monolayers is not taking place, implying
that it is chemisorption that occurs when adsorbed porphyrin is
observed.
Other Metalloporphyrins. Following the procedure de-
scribed for preparation of SAM/Ru(OEP)(CH3CN), bis-aceto-
nitrile octaethyltetraazaporphyrinatoruthenium(II) Ru(OETAP)-
(CH3CN)2, bis-acetonitrile octaethylporphyrinatoosmium(II)
Os(OEP)(CH3CN)2, and carbonyl meso-tetramesitylporphyrin-
atoruthenium(II) Ru(TMP)(CO) were adsorbed on øsoln ) 0.75
SAMs. The advancing water contact angles and ellipsometric
thicknesses for these surfaces and SAM/Ru(OEP)(CH3CN) are
summarized in Table 1. The advancing water contact angles
and thicknesses for the various SAM/metalloporphyrins are
similar, indicating that the metalloporphyrins form similar
structures on the SAM.
SAM/Ru(OETAP)(CH3CN) (0.58 V, 1.21 V vs NHE) and
SAM/Os(OEP)(CH3CN) (0.59 V, 1.25 V vs NHE) each
(28) These extraneous peaks are attributable to a small amount of impurity
in the porphyrin and, in our hands, are always observed.
(29) Brown, G. M.; Hopf, F. R.; Ferguson, J. A.; Meyer, T. J.; Whitten,
D. G. J. Am. Chem. Soc. 1973, 95, 5939-5942.
(30) Kadish, K. M. Progress in Inorganic Chemistry; Lippard, S. J., Ed.;
John Wiley & Sons: New York, 1986; pp 437-460, 549-559.
(31) Bond, A. M.; Khalifa, M. Aust. J. Chem. 1988, 41, 1389-1406.
(32) Not knowing the specific identity of this species does not hinder
the analysis here. Since we have a 1-electron oxidation of either RuIV/III or
the porphyrin which are present in a 1:1 stoichiometry, the CV area still
leads to a porphyrin coverage.

4484 J. Am. Chem. Soc., Vol. 120, No. 18, 1998
Offord et al.
Figure 6. Grazing angle FT-IR spectra of a øsoln ) 0.75 SAM/Ru-
(TMP)(CO). The stretching frequency of SAM/Ru(TMP)(CO), 1967
cm-1 (νCO), was shifted from the Ru(TMP)(CO) methylene chloride
solution value of 1942 cm-1. The negative peaks around 2150 cm-1
are due to the deuterated hexadecyl mercaptan SAM used as a
background spectrum.
displayed two reversible redox waves. SAM/Ru(TMP)(CO) did
not display reversible redox chemistry, which is also true of
Ru(TMP)(CO)(L).
Grazing Angle FT-IR. The C-N stretch (νCN) of the
acetonitrile axial ligands remaining after binding of
Ru(OEP)(CH3CN)2, Ru(OETAP)(CH3CN)2, and Os(OEP)-
(CH3CN)2 to imidazole-terminated SAMs were too weak to be
seen by p-polarized, grazing angle FT-IR. However, the C-O
stretch (νCO) of the SAM/Ru(TMP)(CO) was clearly visible as
a single, sharp peak at 1967 cm-1 (Figure 6), significantly shifted
from that observed for this complex in methylene chloride
solution (1942 cm-1).
Scanning Tunneling Microscopy (STM). Ru(TMP)(CO)
was adsorbed onto a SAM with low surface imidazole coverage
(øsoln ) 0.25) and imaged by constant current STM. Images
were obtained in air for both the metalated and nonmetalated
surfaces (Figure 7). High tunneling impedances of 0.6-33 GΩ
were employed to minimize interactions between the tip and
the surface.33 Repeated scans of the same area reproducibly
created the same image and showed no evidence of tip-induced
damage to the substrate.
A 100 × 100 nm2 image of the SAM with no adsorbed
metalloporphyrin (Figure 7a) displayed characteristic “etch pits”
formed during the chemisorption process as well as multiple
steps in the underlying gold substrate.33 A 100 × 100 nm2
image of the SAM that had been treated with a Ru(TMP)(CO)
solution, SAM/Ru(TMP)(CO), shows etch pits and gold steps
similar to those of the nonmetalated SAM but also many disklike
features (Figure 7b) consistent with metalloporphyrins. The
disks also appear to be randomly spaced, suggesting that the
imid-SH and C9-SH mercaptan components of this SAM are
not highly phase separated.34 A high-resolution image of a
cluster of the disks, each approximately 2 nm in diameter, is
shown on a 10 × 10 nm2 scale in Figure 7c. Figure 8 shows
a cross sectional plot through the middle of the central disk of
Figure 7c taken along the horizontal scan direction. This cross
Figure 7. Scanning tunneling microscopy image of (a) a 100 × 100
nm2 region (z scale ) 3 nm, 30.52 pA, 1 V) of a øsoln ) 0.25 SAM; (b)
a 100 × 100 nm2 image (z scale ) 1.5 nm, 500 pA, 400 mV) of a øsoln
) 0.25 SAM/Ru(TMP)(CO); and (c) a 10 × 10 nm2 image (z scale )
0.6 nm, 500 pA, 300 mV) of a øsoln ) 0.25 SAM/Ru(TMP)(CO). All
images were obtained in height mode (constant current) and displayed
in height-mapped gray scale with white corresponding to the highest
and black the lowest features.
(33) (a) Widrig, C. A.; Alves, C. A.; Porter, M. D. J. Am. Chem. Soc.
1991, 113, 2805-2810. (b) Haussling, L.; Michel, B.; Ringsdorf, H.; Rohrer,
H. Angew. Chem., Int. Ed. Engl. 1991, 30, 569-572. (c) Schonenberger,
C.; Sondag-Huethorst, J. A. M.; Jorritsma, J.; Fokkink, L. G. J. Langmuir
1994, 10, 611-614. (d) Poirier, G. E.; Tarlov, M. J.; Rushmeier, H. E.
Langmuir 1994, 10, 3383-3386.
section represents the height of the tip above the surface. A
similar analysis of each disk in Figure 7c afforded an average
height of 0.15 ( 0.04 nm and a width of 2.28 ( 0.05 nm. Using
the X-ray structure of Ru(TMP)(THF)(N2) for comparison,35
(34) Folkers, J. P.; Laibinis, P. E.; Whitesides, G. M.; Deutch, J. J. Phys.
Chem. 1994, 98, 563-571.

Site Specific Axial Ligation to a Self-Assembled Monolayer
J. Am. Chem. Soc., Vol. 120, No. 18, 1998 4485
Figure 8. Cross sectional plot representing the tip height across the
center of one of the features assigned to the metalloporphyrin of Figure
7c.
these dimensions are consistent with a metalloporphyrin bound
by one axial site with its ring parallel to the SAM surface and
the remaining axial site exposed and normal to the surface. A
closer look at Figure 8 reveals that two cross sectional features
were displayed by each disk: a set of inner peaks spaced 0.49
( 0.07 nm apart and a set of outer peaks spaced 1.69 ( 0.04
nm apart. These peaks represent regions of high electron
conductivity. The distance between the two inner peaks
corresponds to the diameter of the inner porphyrin ring. The
distance between the two outer peaks corresponds closely to
the distance between opposite mesityl rings.
II. Axially Ligated, Bridged Metalloporphyrin Multilay-
ers. Approach. The efficiency and generality of axial ligation
of metalloporphyrins to imidazole-terminated SAMs suggest that
these systems might serve as foundations for the construction
of defined metalloporphyrin multilayer structures. The specific
approach is outlined using bis-dinitrogen meso-tetramesitylpor-
phyrinatoruthenium(II), Ru(TMP)(N2)2, and the bidentate ligand
pyrazine (pyz) in Figure 9. Here, ligation of Ru(TMP)(N2)2 to
the SAM (A) affords SAM/Ru(TMP)(N2), which still bears a
second labile axial ligand. Replacement of this ligand with
pyrazine (B) results in SAM/Ru(TMP)/pyz, which presents a
new axial ligand for a metalloporphyrin at the surface. Reaction
of this assembly with Ru(TMP)(N2)2 (C) generates the bridged
bilayer SAM/Ru(TMP)/pyz/Ru(TMP)(N2), which can be elabo-
rated to higher order multilayers through iteration of steps B
and C, or capped with strongly binding axial ligands such as
carbon monoxide (D). Use of the same metalloporphyrin in
each cycle leads to the formation of single component stacked
multilayers, while use of different metalloporphyrins provides
mixed multilayer stacks. We have prepared examples of both
single component and mixed multilayer metalloporphyrin stacks
by this method. These materials have been characterized by
transmission visible spectroscopy,25 optical ellipsometry, XPS,
advancing water contact angle goniometry, and grazing angle
FT-IR.
Figure 9. Deposition procedure and proposed mode of attachment of
the Ru(TMP) double layer, capped with carbon monoxide, to a mixed
SAM (øsoln ) 0.50).
Table 2. Number of Porphyrin Layers, Ellipsometric Thicknesses
for the SAM/Ru(TMP)[(pyz)Ru(TMP)]n(CO) Series (n ) 0, 1, 2, 3,
4) Compared with a øsoln ) 0.50 SAM
porphyrin thickness
surface
layers
Å
øsoln ) 0.5 SAM
0
1
2
3
4
5
14
24
34
43
48
55
SAM/Ru(TMP)(CO)
SAM/[Ru(TMP)(pyz)]1Ru(TMP)(CO)
SAM/[Ru(TMP)(pyz)]2Ru(TMP)(CO)
SAM/[Ru(TMP)(pyz)]3Ru(TMP)(CO)
SAM/[Ru(TMP)(pyz)]4Ru(TMP)(CO)
Ru(TMP)(N2) were added by immersing pyrazine-terminated
assemblies in a benzene solution of 1 mM Ru(TMP)(N2)2.
Capping was carried out by placing the resulting surface under
a CO atmosphere. After each step, the surfaces were thoroughly
rinsed to remove excess ligand or metalloporphyrin.
Table 2 presents thicknesses as determined by ellipsometry
data for the series of pyrazine-bridged Ru(TMP) multilayers.
Increases in apparent film thickness of 5-10 Å were observed
after each cycle, consistent with the incremental addition of
pyrazine/Ru(TMP) units. XPS provided corroborating evidence
for multilayer formation. The ratios of C(1s) and Ru(3d) peak
intensities to Au(4f) peak intensities increased in step with the
number of added metalloporphyrin layers.
Mixed Metalloporphyrin Bilayers. To explore the versatil-
ity of the general multilayer synthetic scheme, two bilayers
containing Ru(TMP) and a second, different metalloporphyrin
were prepared and compared to SAM/Ru(TMP)/pyz/Ru(TMP)-
(CO). SAM/Ru(TMP)/pyz/Ru(TPFPP)(CO) was formed by
addition of (meso-tetrakis(pentafluorophenyl)porphyrinato)-
Ru(TMP) Multilayers. Samples of SAM/Ru(TMP)(N2)
were prepared by immersing a øsoln ) 0.50 mixed monolayer
in a 1 mM solution of Ru(TMP)(N2)2 in benzene under an
atmosphere of dinitrogen. The following capped stacks were
assembled according to the scheme outlined in Figure 9: SAM/
Ru(TMP)(CO), SAM/[Ru(TMP)/pyz/]nRu(TMP)(CO), n ) 1,
2, 3, and 4. Replacement of dinitrogen axial ligands with
pyrazine was effected by placing the surfaces in a methylene
chloride solution of 1 mM pyrazine. Additional layers of
(35) Camenzind, M. J.; James, B. R.; Dolphin, D.; Sparapany, J. W.;
Ibers, J. A. Inorg. Chem. 1988, 27, 3054-3057.

4486 J. Am. Chem. Soc., Vol. 120, No. 18, 1998
Offord et al.
Table 3. Ellipsometric Thicknesses, νCO Stretching Frequencies,
and Advancing Water Contact Angles for the Metalloporphyrin
Double Layers and the Corresponding CO-Capped Metalloporphyrin
Building Blocks on øsoln ) 0.50) SAM
angles. Goniometry indicates that a uniform surface was
achieved by metalloporphyrin adsorption to a øsoln ) 0.50
surface, suggesting that, at higher values of øSAM, some of the
bound metalloporphyrins were physisorbed. It may be that
specifically bound metalloporphyrins serve as nucleation sites
for the formation of physisorbed multilayer structures. The
value of øSAM calculated to give complete coverage of the SAM
by Ru(OEP)(CH3CN) (øSAM ) 0.11 if the imid-SH adsorbates
are regularly distributed in the SAM) is expected to increase if
the imid-SH adsorbates are randomly distributed in the SAM.
Our results suggest that SAMs with øsoln < 0.50 will afford
metalloporphyrins bound in the most uniform manner.
thickness
(Å)
νCO
surface
(cm-1
)
θa(H2O)
SAM
SAM/Ru(TMP)(CO)
SAM/Ru(TMP)(pyz)Ru(TMP)(CO)
SAM/Os(OEP)(CO)
SAM/Ru(TMP)(pyz)Os(OEP)(CO)
SAM/Ru(TPFPP)(CO)
14
24
35
25
36
27
30
88.5
88
88
88
81
97
83
1967
1974
1917
1978
1998
2002
SAM/Ru(TMP)(pyz)Ru(TPFPP)(CO)
Axial Substitution. Four results provide evidence that
metalloporphyrins are axially ligated to the imidazole groups
presented at the surface of a SAM. First, the location, strength,
and sharpness of the νCO stretch in the grazing angle FT-IR
spectrum of SAM/Ru(TMP)(CO) provide important information
about the orientation and environment of the CO at the surface.
The change in the carbonyl IR stretching frequency from the
free Ru(TMP)(CO) to the surface-bound Ru(TMP)(CO) is 25
cm-1. This indicates a distinctly different ruthenium carbonyl
bonding environment for the two species. Furthermore, the
direction of the change implies that the ruthenium carbonyl bond
is stronger in the free Ru(TMP)(CO) species than in the surface-
bound species as expected from the trans influence of a
ruthenium imidazole bond.37 p-Polarized light interacts with
dipoles that have a component perpendicular to the surface.38
The strength of the CO signal compared to the methylene
stretches indicates that the CO dipole has a large component
normal to the surface. In addition, the surface CO stretch was
very sharp, suggesting a single CO containing species. These
IR data are consistent with a SAM-bound metalloporphyrin
having the porphyrin ring oriented parallel and the CO ligand
normal to the surface, as depicted in Figure 1. Second,
protonation of the imidazole of the SAM completely inhibited
its reaction with Ru(OEP)(CH3CN)2. Third, attaching strongly
bound axial ligands to the metalloporphyrin also prevented
adsorption to the SAM. Fourth, STM analysis of Ru(TMP)-
(CO) adsorbed on SAMs gave images consistent with a model
in which metalloporphyrins lie flat on the SAM. At the low
surface imidazole concentration employed in these experiments,
the metalloporphyrins appear as randomly spaced disks on the
SAM surface. At high resolution, the sizes of the disks are in
agreement with crystal structure analysis of Ru(TMP)(CO).
Generality and Advantages. Several metalloporphyrins
were found to bind to mixed imidazole/methyl-terminated SAMs
under the same conditions employed in the SAM/Ru(OEP)-
(CH3CN) study. Variation of the porphyrin ring (from OEP to
TMP, OETAP, and TPFPP) and the metal (from Ru to Os) did
not significantly hinder the formation of surface-bound com-
plexes. Our results indicate that this method can be used for
the general binding of any metalloporphyrin that displays high
affinity for the ligand presented by a self-assembled monolayer.
Indeed, this method could be generalized to many other
situations where metal ions or complexes exhibit significant
affinity for ligands which can be presented at the surface of a
SAM. Due to the site specific binding of the metalloporphyrin
and the sensitivity of the various surface analytical techniques
available, this method requires only small amounts of material
and no synthetic modification of the metalloporphyrin.
ruthenium(II) (Ru(TPFPP)) to SAM/Ru(TMP)/pyz, followed by
capping with CO. SAM/Ru(TMP)/pyz/Os(OEP)(CO) was
formed by addition of five-coordinate Os(OEP)(CO) to SAM/
Ru(TMP)/pyz.
Table 3 presents ellipsometric thicknesses, CO stretching
frequencies, and advancing water contact angles for both the
single component and mixed metalloporphyrin bilayers and
monolayers bearing the corresponding metalloporphyrin building
blocks. SAM/Ru(TPFPP)(CO) was prepared by axial ligation
of Ru(TPFPP) to a øsoln ) 0.50 SAM and CO capping. The
differences in observed thicknesses for SAM/Ru(TMP)/pyz/Os-
(OEP)(CO) versus SAM/Os(OEP)(CO) and SAM/Ru(TMP)/
pyz/Ru(TMP)(CO) versus SAM/Ru(TMP)(CO) were identical
(11 Å). The presence of Ru(TPFPP)(CO) on the surface was
clearly indicated by the νCO observed at 2002 cm-1 for this
system relative to 1998 cm-1 for SAM/Ru(TPFPP)(CO). In
addition, a large F(1s) photoelectron peak was observed in the
XPS of this surface along with splitting of the C(1s) peak due
to the electron-withdrawing effect of the fluorine substituents
attached to carbon. In each of the bilayer structures, CO
stretching frequencies were found at higher frequencies than
those in the corresponding monolayers. This shift was quite
dramatic with SAM/Ru(TMP)/pyz/Os(OEP)(CO) versus SAM/
Os(OEP)(CO), where νCO values were 1978 and 1917 cm-1
,
respectively. The shifts to higher frequency are consistent with
CO being bound opposite to pyrazine, which is a stronger π-acid
than imidazole.
We have also prepared the mixed bilayer SAM/Ru(TPFPP)/
pyz/Ru(TMP)(CO) and the alternating trilayer SAM/Ru(TMP)/
pyz/Ru(OEP)/pyz/Ru(TMP)(CO).36 In synthesizing the trilayer,
the bis(pyrazine) complex Ru(OEP)(pyz)2 was adsorbed directly
onto SAM/Ru(TMP)(N2), and Ru(TMP)(CO) was added to cap
the SAM/Ru(TMP)/pyz/Ru(OEP)/pyz assembly. Thus, we were
able to bypass several steps of the general multilayer synthetic
scheme by using Ru(OEP)(pyz)2 as a bidentate ligand with a
metalloporphyrin core.
Discussion
Site Specific Binding. An important conclusion that can be
drawn from the XPS, electrochemistry, ellipsometry, and
goniometry data is that Ru(OEP)(CH3CN) binds specifically to
the imidazole terminus of the 1-(10-mercaptodecyl)imidazole
and not to the methyl terminus of the unsubstituted alkanethiol.
No evidence for Ru(T-p-ClPP)(CH3CN) deposition on the full
C9-SH SAM (øSAM ) 0) was found. The amount of bound
Ru(OEP)(CH3CN) was directly related to the quantity of 1-(10-
mercaptodecyl)imidazole in the SAM, reaching a maximum with
the full 1-(10-mercaptodecyl)imidazole SAM (øSAM ) 1) surface
as shown by ellipsometry but not by advancing water contact
(37) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G.; Principles
and Applications of Organotransition Metal Chemistry; University Science
Books: Mill Valley, CA 1987.
(38) (a) Porter, M. D. Anal. Chem. 1988, 60, 1143A-1155A. (b) Pearce,
H. A.; Sheppard, N. Surf. Sci. 1976, 59, 205-217.
(36) Ennis, M. S., Ph.D. Thesis, Stanford University, Stanford, CA, 1995.

Site Specific Axial Ligation to a Self-Assembled Monolayer
J. Am. Chem. Soc., Vol. 120, No. 18, 1998 4487
Single Component and Mixed Multilayer Metallopor-
phyrin Stacks. By taking advantage of the high affinity of
metalloporphyrins for the bidentate ligand pyrazine, we have
developed a simple and flexible route to electrode-bound
metalloporphyrin multilayers in which a gold-supported SAM/
metalloporphyrin serves as the base. The pyrazine-bridged
multilayers prepared are immobilized analogues of the highly
conductive, soluble metalloporphyrin polymers reported by
Collman et al.39 The controlled, iterative, and general nature
of the method defines it as a form of solid-phase synthesis and
suggests that it could be exploited in a combinatorial fashion12g
for the preparation and screening of new metalloporphyrin-based
materials.12
possible through funds from the NSF-MRSEC program. We
acknowledge the Mass Spectrometry Facility of the University
of California at San Francisco, supported by NIH Division of
Research Resources Grant RR01614 and by NSF Grant
DIR8700766. We also thank Dr. H. T. Fish and Dr. J. Kendall
for providing the H2TMP and H2OETAP ligands, respectively.
Supporting Information Available: Grazing angle FTIR
and transmission visible spectra of SAM/metalloporphyrins
complexes discussed in this paper as well as the relevant
experimental detail and discussion. XPS spectra of Ru(Tp-
ClPP)(CO)(CH3CN) in a KBr matrix and on an imidazole
surface are presented. Additionally, a proposed hexagonal
packing arrangement of the porphyrins on an ideal SAM surface
is shown, along with geometric arguments for calculating the
“ideal” effective area of porphyrins arranged in this way. (12
pages, print/pdf). See any current masthead page for ordering
information and Web access instructions.
Acknowledgment. This work was supported by the National
Science Foundation and the Arnold and Mabel Beckman
Foundation. We also thank the Center for Materials Research
at Stanford University for use of the STM and the XPS made
(39) Collman, J. P.; McDevitt, J. T.; Leidner, C. R.; Yee, G. T.; Torrance,
J. B.; Little, W. A. J. Am. Chem. Soc. 1987, 109, 4606-4614.
JA973528L