Deprotection, tethering, and activation of a one-legged metalloporphyrin on a chemically active metal surface: NEXAFS, synchrotron XPS, and STM study of [SAc]P-Mn(III)Cl on Ag(100)

Article abstract of DOI:10.1021/ja904664e

The structural and reactive properties of the acetyl-protected "one-legged" manganese porphyrin [SAc]P-Mn(III)Cl on Ag(100) have been studied by NEXAFS, synchrotron XPS and STM. Spontaneous surface-mediated deprotection occurs at 300 K accompanied by spre

Full text of DOI:10.1021/ja904664e

Published on Web 09/28/2009
Deprotection, Tethering, and Activation of a One-Legged
Metalloporphyrin on a Chemically Active Metal Surface:
NEXAFS, Synchrotron XPS, and STM Study of
[SAc]P-Mn(III)Cl on Ag(100)
Mark Turner, Owain P. H. Vaughan, Georgios Kyriakou, David J. Watson,
Lukas J. Scherer, Anthoula C. Papageorgiou, Jeremy K. M. Sanders, and
Richard M. Lambert*
Department of Chemistry, UniVersity of Cambridge, Cambridge CB2 1EW, United Kingdom
Received June 8, 2009; E-mail: rml1@cam.ac.uk
Abstract: The structural and reactive properties of the acetyl-protected “one-legged” manganese porphyrin
[SAc]P-Mn(III)Cl on Ag(100) have been studied by NEXAFS, synchrotron XPS and STM. Spontaneous
surface-mediated deprotection occurs at 300 K accompanied by spreading of the resulting thio-tethered
porphyrin across the metal surface. Loss of the axial chlorine ligand occurs at 498 K, without any
demetalation of the macrocycle, leaving the Mn center in a low co-ordination state. At low coverages the
macrocycle is markedly tilted toward the silver surface, as is the phenyl group that forms part of the tethering
“leg”. In the monolayer region a striking transition occurs whereby the molecule rolls over, preserving the
tilt angle of the phenyl group, strongly increasing that of the macrocycle, decreasing the apparent height
of the molecule and decreasing its footprint, thus enabling closer packing. These findings are in marked
contrast with those previously reported for the corresponding more rigidly bound four-legged porphyrin
[Turner, M.; Vaughan, O. P. H.; Kyriakou, G.; Watson, D. J.; Scherer, L. J.; Davidson, G. J. E.; Sanders,
J. K. M.; Lambert, R. M. J. Am. Chem. Soc. 2009, 131, 1910] suggesting that the physicochemical properties
and potential applications of these versatile systems should be strongly dependent on the mode of tethering
to the surface.
Introduction
Porphyrins have been attached to oxide surfaces for use as
sensors by means of organophosphonate linkers.12 Our interest,
however, lies in attaching metalloporphyrins to chemically active
metal surfaces, and we have reported a viable route for
deprotecting and covalently attaching Mn porphyrin molecules
to silver surfaces by means of four thio tethers.13 The orientation,
surface mobility, and degrees of freedom of such covalently
tethered porphyrins are expected to be important in determining
access and escape of molecules to/from the porphyrin metal
center and hence their effectiveness in a variety of applications.
One such application is catalysis, where Hulsken et al. have
elegantly demonstrated that a Mn metalloporhyrin adsorbed on
Au(111) catalyzed the aerobic epoxidation of stilbene in
solution.14 As extended Au surfaces cannot dissociatively adsorb
oxygen,15 a necessary precondition for epoxidation to occur,16
Porphyrins tethered to solid surfaces have potential uses in a
wide range of applications1-3 including light-harvesting arrays,4-6
optical switches, and photonic wires.2,3,7 Supramolecular arrays
are of interest with respect to the study of energy transfer and
photosynthetic mechanisms,7,8 while two-dimensional (2D)
porphyrin assemblies have attracted attention for possible
application as chemical sensors in molecular electronics or as
chemically switchable 2D rotors.3,9,10 Related to this, we recently
showed that a porphyrin-functionalized silver surface exhibits
ligand-binding and -unbinding reactions characteristic of the free
metalloporphyrin11 and showed that a well-chosen ligand can
strongly alter the dynamics of adsorbed porphyrins.10
(1) Drain, C. M.; Batteas, J. D.; Smeureanu, G.; Patel, S. Dekker
Encyclopedia of Nanoscience and Nanotechnology; Taylor & Francis:
New York, 2004.
(2) Kwok, K. S.; Ellenbogen, J. C. Mater. Today 2002, 5, 28.
(3) Carroll, R. L.; Gorman, C. B. Angew. Chem., Int. Ed. 2002, 41, 4379.
(4) Endo, M.; Fujitsuka, M.; Majima, T. Chem.sEur. J. 2007, 13, 8660.
(5) Nakano, A.; Osuka, A.; Yamazaki, I.; Yamazaki, T.; Nishimura, Y.
Angew. Chem., Int. Ed. 1998, 37, 3023.
(11) Williams, F. J.; Vaughan, O. P. H.; Knox, K. J.; Bampos, N.; Lambert,
R. M. Chem. Commun. 2004, 1688.
(12) Dubey, M.; Bernasek, S. L.; Schwartz, J. J. Am. Chem. Soc. 2007,
129, 6980.
(13) Turner, M.; Vaughan, O. P. H.; Kyriakou, G.; Watson, D. J.; Scherer,
L. J.; Davidson, G. J. E.; Sanders, J. K. M.; Lambert, R. M. J. Am.
Chem. Soc. 2009, 131, 1910.
(6) Prathapan, S.; Johnson, T. E.; Lindsey, J. S. J. Am. Chem. Soc. 1993,
115, 7519.
(7) Holten, D.; Bocian, D. F.; Lindsey, J. S. Acc. Chem. Res. 2002, 35,
57.
(14) Hulsken, B.; Van Hameren, R.; Gerritsen, J. W.; Khoury, T.;
Thordarson, P.; Crossley, M. J.; Rowan, A. E.; Nolte, R. J. M.;
Elemans, J.; Speller, S. Nat. Nanotechnol. 2007, 2, 285.
(15) Pireaux, J. J.; Chta¨ıb, M.; Delrue, J. P.; Thiry, P. A.; Liehr, M.;
Caudano, R. Surf. Sci. 1984, 141, 211.
(8) Kay, A.; Gra¨tzel, M. J. Phys. Chem. 1993, 97, 6272.
(9) Purrello, R.; Gurrieri, S.; Lauceri, R. Coord. Chem. ReV. 1999, 190-
192, 683.
(10) Vaughan, O. P. H.; Williams, F. J.; Bampos, N.; Lambert, R. M.
Angew. Chem., Int. Ed. 2006, 45, 3779.
(16) Grant, R. B.; Lambert, R. M. J. Chem. Soc., Chem. Commun. 1983,
662.
9
10.1021/ja904664e CCC: $40.75  2009 American Chemical Society
J. AM. CHEM. SOC. 2009, 131, 14913–14919 14913

A R T I C L E S
Turner et al.
the authors suggested that O2 dissociation was a cooperative
event involving two adjacent porphyrin molecules. Alternatively,
by using a silver surface to dissociatively chemisorb dioxygen
and deliver oxygen adatoms and a π-adsorbed alkene to the
active metal center of the porphyrin, one may hope to create a
hybrid, low-temperature, selective oxidation catalyst. Accord-
ingly, having investigated a rigidly tethered “four-legged” Mn
porphyrin on Ag(100),13 we report the properties of the
corresponding acetyl-protected “one-legged” Mn porphyrin on
the same surface. Substantial differences in mobility, spatial
distribution, acetyl deprotection, dechlorination, flexibility, and
orientational behavior as a function of coverage are found,
indicating that the physicochemical properties of porphyrin-
functionalized surfaces should be markedly dependent on the
mode of tethering. We also demonstrate that NEXAFS can be
used to provide very detailed information about porphyrin
adsorption geometry and changes in geometry with coverage,
key properties in most applications.
In earlier studies, self-assembled porphyrin monolayers have
been deposited on gold surfaces from solution.14,18,19 Although
these results are certainly interesting, the approach is not suited
to our purposesa SAM blankets the metal surface and excludes
it from adsorbing reactants or participating in their further
conversion to products. On the other hand deposition by vacuum
evaporation allows close control of porphyrin coverage from
submonolayers into the multilayer regime, making it the method
of choice for our purposes. Often, porphyrin tilt angles have
been inferred by considering factors such as packing density
and surface periodicity without resort to direct spectroscopic
measurements. Although infrared spectroscopy20 does provide
a reliable technique for determining porphyrin tilt angles,
sensitivity is low, again precluding characterization of sub-
monolayer coverages, which are the focus of our interest. So
we have used near-edge X-ray absorption fine structure (NEX-
AFS) spectroscopy, high-resolution synchrotron XPS, and STM
in conjunction with porphyrin deposition carried out under
conditions of ultra high vacuum. NEXAFS provides extremely
high sensitivity, thus allowing investigation of submonolayer
coverages, yielding orientational information about both the
macrocycle and the attached phenyl group that forms part of
the tethering leg; the ability to probe both N and C K-edge
transitions is an added bonus.
Figure 1. Molecular structure and 3D space-filling model for the “one-
legged” porphyrin, [SAc]P-Mn(III)Cl.
acid/ acetic anhydride ) 4:1 at 110 °C using MnCl2 ·4H2O as the
manganese source. After completion of the reaction the remaining
inorganic salts were filtered off, leaving the pure [SAc]P-Mn(III)Cl
porphyrin (Figure 1).
High-resolution XPS and NEXAFS measurements were carried
out on the SuperESCA beamline at the ELETTRA synchrotron
radiation source in Trieste, Italy. Spectra were collected using a
single-pass 32-channel concentric hemispherical electron analyzer.
The excitation energies used for acquisition of the C 1s, Ag 3d, Cl
2p, S 2p, Mn 2p, and N 1s spectra were 350, 470, 302.5, 251, 720,
and 500 eV, respectively; the dwell time for signal averaging was
0.1 s. The angle between the analyzer entrance lens and the
incoming photon beam was 70° in the horizontal plane. The
Ag(100) crystal was attached to a motorized manipulator Via a
tantalum backplate fitted with a T1T2 thermocouple and could be
heated resistively to 900 K or cooled to 77 K. STM experiments
were carried out in Cambridge, UK, with an Omicron variable-
temperature ultra-high vacuum STM operated in constant current
mode using etched tungsten tips. The Ag(100) sample was cleaned
by repeated cycles of Ar+ sputtering (99.999% Messer) followed
by annealing at 600 K until a clean, atomically flat surface was
obtained, as monitored by XPS and LEED (Trieste) or LEED,
Auger electron spectroscopy, and STM (Cambridge).
[SAc]P-Mn(III)Cl was deposited onto the Ag surface by means
of a resistively heated collimated evaporation source fitted with a
T1T2 thermocouple. The (very small) amounts of porphyrin used
were injected into the sublimation source as solutions in dichlo-
romethane which were then evaporated to dryness before mounting
in the vacuum chamber. Calibration of the surface coverage was
achieved by following the uptake of the porphyrin on Ag(100) using
XPS13 (Trieste) or estimated from STM images (Cambridge).
Typically, porphyrin was deposited at room temperature followed
by annealing at 473 K for 15 min to disperse initially formed islands
and desorb any multilayer material. This method provided a
convenient and reliable way of preparing any given coverage in
the submonolayer to monolayer regime.13
Experimental Methods
Results and Discussion
The free-base acetyl-protected porphyrin [SAc]P was synthesized
following the procedure described by Ryppa et al.21 Briefly, the
acid-catalyzed condensation of dipyrromethane, pyrrole-2-carbal-
dehyde, and S-acetylthiobenzaldehyde in a “[2 + 1 + 1]” approach
was used to produce two different meso substituted porphyrins: (i)
[SAc]P porphyrin which has one substituent at position 5 (9.7%
yield) and (ii) [SAc]2P with two substituents at positions 5, 10
(11.4% yield). The two porphyrins were then separated by column
chromatography. After crystallization from dichloromethane/
methanol, the compounds showed no detectable impurities. Meta-
lation of the free-base porphyrin was achieved in a mixture of acetic
BehavioratLowCoverage.Acetyl-protected[SAc]P-Mn(III)Cl
was deposited on clean Ag(100) over 20 min, yielding an
estimated submonolayer coverage of ∼0.5 ML. Figure 2 shows
N 1s XP spectra acquired (i) immediately after dosing and (ii)
after annealing to 473 K for 15 min. In both cases only a single
N 1s peak appearedscharacteristic of a metalated porphyrin in
which all N atoms in the macrocycle are equivalent. This is
important because it confirms that no demetalation occurred
upon adsorption or heating: the free base porphyrin would
exhibit two distinct N 1s signals corresponding to pyrrolic and
iminic species.22,23 The N 1s binding energies of these species
are ∼400 and ∼397 eV, respectively, so that they would have
been readily resolved in our experiment. The weak Cl 2p
emission and the undetectability of Mn 2p emission are
(17) Yasseri, A. A.; Syomin, D.; Malinovskii, V. L.; Loewe, R. S.; Lindsey,
J. S.; Zaera, F.; Bocian, D. F. J. Am. Chem. Soc. 2004, 126, 11944.
(18) Shimazu, K.; Takechi, M.; Fujii, H.; Suzuki, M.; Saiki, H.; Yoshimura,
T.; Uosaki, K. Thin Solid Films 1996, 273, 250.
(19) Zak, J.; Yuan, H.; Ho, M.; Woo, L. K.; Porter, M. D. Langmuir 1993,
9, 2772.
(20) Wei, L.; Syomin, D.; Loewe, R. S.; Lindsey, J. S.; Zaera, F.; Bocian,
D. F. J. Phys. Chem. B 2005, 109, 6323.
(22) Flechtner, K.; Kretschmann, A.; Bradshaw, L. R.; Walz, M.-M.;
Steinru¨ck, H.-P.; Gottfried, J. M. J. Phys. Chem. C 2007, 111, 5821.
(23) Gottfried, J. M.; Flechtner, K.; Kretschmann, A.; Lukasczyk, T.;
Steinru¨ck, H.-P. J. Am. Chem. Soc. 2006, 128, 5644.
(21) Ryppa, C.; Senge, M. O.; Hatscher, S. S.; Kleinpeter, E; Wacker, P.;
Schilde, U.; Wiehe, A. Chem.sEur. J. 2005, 11, 3427.
9
14914 J. AM. CHEM. SOC. VOL. 131, NO. 41, 2009

One-Legged Metalloporphyrin on a Metal Surface
A R T I C L E S
Figure 2. N 1s XP spectra of a low (submonolayer) dose of one-legged
porphyrin acquired before (left) and after annealing to 473 K (right) for 15
min.
Figure 3. Cl 2p region of the XP spectrum of the one-legged porphyrin
after initial dosing (left), after annealing to 473 K (center), and after
annealing to 498 K (right).
consistent with the low-number density of Cl and Mn atoms
on the surface at saturation coverage, even though both were
studied at the optimum photon energies that maximized the
respective photoionization cross sections: Cl 2p ) 3 Mb at 251
eV, Mn 2p ) 1 Mb at 720 eV.13 Both spectra in Figure 2 may
be fitted with a single Gaussian centered at 397.2 eV (fwhm )
0.9 eV), assigned to Mn porphyrin molecules at submonolayer
coverage. Annealing to 473 K caused no loss of N 1s intensity,
confirming the absence of porphyrin multilayers. This behavior
contrasts strongly with that of the corresponding four-legged
Mn porphyrins which at 300 K form three-dimensional ag-
gregates at low coverage, as observed by STM24 and confirmed
by XPS.13 The implication is that in the present case reduced
tethering results in increased mobility so that porphyrin spread-
ing over the surface occurs even at room temperature.
The temperature dependence of the Cl 2p emission for the
same porphyrin coverage is shown in Figure 3; the second and
third spectra were obtained after annealing at 473 and 498 K,
respectively, for 15 min. It is clear that much of the axial
chlorine ligand was lost during the first stage of annealing with
no demetalation having occurred (N 1s XPS unchanged) and
by 498 K chlorine loss was complete as a result of desorption,13
perhaps accompanied by some dissolution25 into the Ag, leaving
the Mn center in a low-coordination state.13 This behavior is
somewhat different from that of the corresponding four-legged
porphyrin ([SAc]4P-Mn(III)Cl)13 which was readily and com-
pletely dechlorinated on Ag(100) at 473 K.
Figure 4. (a) N K-edge NEXAFS spectra acquired at five angles of photon
incidence (θ) for a ∼0.5 ML coverage of [SAc]P-Mn(III)Cl on Ag(100).
(b) Curve-fitting analysis of the photon angle dependence of π* resonances
A and C to estimate the corresponding macrocycle tilt angle, R.
Table 1. Peak Assignments for the N K-Edge NEXAFS of
[SAc]P-Mn(III)Cl on Ag(100)
peak
energy/eV
assignment
A
B
C
D
E
398.4
400.7
401.4
406.5
414.8
N 1s f π*
N 1s f π*
N 1s f π*
N 1s f σ*
N 1s f σ*
N K-edge NEXAFS spectra acquired at five angles of photon
incidence for the same low coverage (∼0.5 ML) are shown in
Figure 4a. Clear resonances due to N 1s f π* and N 1s f σ*
transitions are observed which display pronounced dependence
on the photon incidence angle; peak assignments and associated
transitions are presented in Table 1.26-28
(26) Narioka, S.; Ishii, H.; Ouchi, Y.; Yokoyama, T.; Ohta, T.; Seki, K. J.
Phys. Chem. 1995, 99, 1332.
(24) Vaughan, O. P. H.; Turner, M.; Williams, F. J.; Hille, A.; Sanders,
J. K. M.; Lambert, R. M. J. Am. Chem. Soc. 2006, 128, 9578.
(25) Goddard, P. J.; Lambert, R. M. Surf. Sci. 1977, 67, 180.
(27) Okajima, T.; Yamamoto, Y.; Ouchi, Y.; Seki, K. J. Electron Spectrosc.
Relat. Phenom. 2001, 114-116, 849.
9
J. AM. CHEM. SOC. VOL. 131, NO. 41, 2009 14915

A R T I C L E S
Turner et al.
Figure 5. Two possible orientations of the one-legged porphyrin tethered
to Ag(100) corresponding to the same macrocycle tilt angle of ∼25°-30°:
(a) porphyrin tilted upward and (b) porphyrin tilted downward.
The principal resonances A and C occur before the step edge
and correspond to transitions from N 1s to each of the first two
available π* orbitals, respectively. Resonance B, also due to a
π* transition, appears as weak shoulder to C; two broad σ*
resonances D and E occur beyond the step edge. A more detailed
assignment of the final-state orbitals and their symmetry would
require comparison with simulated NEXAFS spectra29 and is
not necessary for our purposes. As expected, both π* and σ*
resonances display distinct dichroism in their angular depen-
dence. The absence of a strong resonance between C and D (at
∼403 eV, characteristic of pyrrolic N in the metal-free base)
confirms that the porphyrin is metalated and that the manganese
metal center is fully coordinated by the four central N atoms.
Since the N atoms reside within the center of the macrocycle
and are approximately coplanar, the angular dependences
observed in the N K-edge spectra provide a powerful means of
determining the orientation of the metalloporphyrin with respect
to the surface. Figure 4b presents such an analysis for the
principal π* resonances A and C, the observed normalized
intensities being overlaid with best-fit theoretical curves for the
molecular tilt angle (R) with respect to the surface.30 (In
principle, σ* resonances may also be subjected to a similar
analysis, but their width and superposition on the step back-
ground preclude accurate fitting.) A least-squares fit gives
macrocycle tilt angles of 29.4° for A and 23.5° for C; the values
are consistent within experimental error ((5°) and indicate that
the porphyrin is very appreciably inclined with respect to the
surface.
Figure 6. (a) C K-edge NEXAFS spectra acquired at five angles of photon
incidence (θ) for a ∼0.5 ML coverage of [SAc]P-Mn(III)Cl on Ag(100).
(b) Curve fitting analysis of the photon angle dependence of π* resonances
a and b to determine the corresponding macrocycle and phenyl tilt angles
(R) respectively.
Table 2. Peak Assignments for the C K-Edge NEXAFS of
[SAc]P-Mn(III)Cl on Ag(100)
peak
energy/eV
assignment
a
b
c
d
e
283.6
284.8
287.3
293.4
300.6
C 1s f π* (porphyrin)
C 1s f π* (phenyl)
C 1s f π* (porphyrin)
C 1s f σ*
C 1s f σ*
NEXAFS cannot distinguish between molecular tilt angles
toward or away from the surface and Figure 5 shows the two
possibilities corresponding to the present case. Of these, the
second, b, is intuitively the more likely as it maximizes
molecule-surface interaction. Taking account of the dimensions
of the macrocycle and the tethering “leg”, simple trigonometry
confirms that a porphyrin downward tilt angle of 25°-30° is
possible (5.9 Å leg length, 11.0 Å macrocycle width, ∼100°
angle between leg and macrocycle). Moreover, STM data to be
presented below are consistent with the downward tilt model.
Figure 6 shows low coverage (∼0.5 ML) C K-edge NEXAFS
spectra acquired at five photon incidence angles. Five resonances
(28) Wende, H.; Bernien, M.; Luo, J.; Sorg, C.; Ponpandian, N.; Kurde,
J.; Miguel, J.; Piantek, M.; Xu, X.; Eckhold, Ph.; Kuch, W.;
Baberschke, K.; Panchmatia, P. M.; Sanyal, B.; Oppeneer, P. M.;
Eriksson, O. Nat. Mater. 2007, 6, 516.
(29) Polzonetti, G.; Carravetta, V.; Iucci, G.; Ferri, A.; Paolucci, G.;
Goldoni, A.; Parent, P.; Laffon, C.; Russo, M. V. Chem. Phys. 2004,
296, 87.
(30) Sto¨hr, J.; Outka, D. A. Phys. ReV. B 1987, 36, 7891.
9
14916 J. AM. CHEM. SOC. VOL. 131, NO. 41, 2009

One-Legged Metalloporphyrin on a Metal Surface
A R T I C L E S
Figure 7. STM images and line profiles of a submonolayer coverage of the one-legged porphyrin acquired after annealing to 473 K. (a) (360 × 360 Å2,
Vtip ) 1.00 V, I ) 1.02 nA). Two cleaved acetyl groups circled in green. (b) Small-scale image showing positions of line scans taken through chlorinated
and dechlorinated molecules (75 × 75 Å2, Vtip ) 1.00 V, I ) 1.01 nA). (c) Line scans taken at the positions indicated in (b).
are apparent, three C 1s f π* at lower photon energy and two
C 1s f σ* at higher photon energy, beyond the step edge. (Table
2 lists peak assignments.) The improved s/n ratio compared to
the N K-edge spectra reflects the ∼7:1 C:N ratio in the molecule.
Unlike the N K-edge spectrum, the C spectrum contains
contributions from the phenyl substituent which forms part of
the tethering leg. In free porphyrins with phenyl substituents in
the meso position, the latter are generally strongly tilted with
respect to the central ring, so the angular dependence of the π*
resonances is more complicated than in the N K-edge spectrum.
In the past, this has led many groups to dismiss C K-edge spectra
for the orientational analysis of phenyl-substituted porphyrins.26
However careful inspection and assignment of each π* reso-
Figure 8. N 1s XP spectra for a high coverage of porphyrin as deposited
(left) and after annealing to 473 K (right).
nance do yield orientational information for both ring systems.31,32
Peaks a and c display the same type of angular dependence
as the N spectrum π* resonances and are assigned to C 1s f
Figure 3, and in good accord with the line scans in Figure 7c
for chlorinated and dechlorinated molecules. Also visible are a
number of smaller features (green circles, Figures 7a) identified
as acetyl groups24 resulting from spontaneous surface-mediated
deprotection of the thiol group by cleavage of the acetyl group
initially present at the end of the leg. The dimensions of these
smaller features (∼4.5 Å lateral, apparent height 0.65 Å) are
commensurate with those of an acetyl group.24 Figure 7a is a
larger-scale image of porphyrin molecules in the contact layer
and the cleaved acetyl groups: examination of several images
with a total area of 2760 nm2 showed that the ratio of porphyrin
molecules to cleaved acetyl groups was ∼1:1.4 in reasonable
accord with the expected value of 1:1. Thiol-deprotection was
confirmed by the S 2p XP spectrumsa single S 2p3/2 component
at a binding energy of 162.3 eV appeared, corresponding to
deprotected molecules bound to the metal surface.13
π* transitions involving the porphyrin ring. Peak b clearly
exhibits a different angular dependence, the signal varying little
with θ; it is therefore assigned to C 1s f π* transitions of the
phenyl substituent. The results of a full analysis of the angular
dependence of resonances a and b are shown in Figure 6b.
Resonance a corresponds to a porphyrin ring tilt angle of 25.2°s
within experimental error fully consistent with the value derived
from analysis of the N K-edge data. Resonance b yields a tilt
angle of 50.3° for the phenyl ring; as expected, this indicates
significant tilt of the phenyl ring with respect to the plane of
the porphyrin ring. Taking account of the macrocycle tilt of
∼30°, we conclude that the phenyl group lies at ∼80° to the
porphyrin planesi.e. much as would be expected for the free
molecule and consistent with the illustration in Figure 5b.
STM imaging of 300 K as-deposited submonolayer coverages
showed that no morphological changes occurred upon annealing
to 473 K, in good accord with the corresponding N 1s XPS
results, and typical images are shown in Figure 7. The lateral
dimensions of individual features correspond to those expected
for porphyrin molecules. Their apparent height of ∼1.9 Å is
comparable to the apparent height of ∼2.3 Å recorded for the
analogous 4-legged porphyrin on Ag(100)24sconsistent with a
downwardly tilted macrocyle ring in the present case. An
upwardly tilted molecule would have an apparent height of about
twice this value. Most molecules are characterized by a bright
central spot which we associate with the Cl ligand attached to
the Mn center, consistent with the Cl XP spectrum shown in
Behavior at High Coverages. Behavior at higher coverages
was investigated by depositing more porphyrin on top of the
300 K ∼0.5 ML deposit, and N 1s XP spectra taken before and
after annealing to 473 K are shown in Figure 8. Three
components may be discerned in the 300 K as-deposited film:
one centered at 397.2 eV corresponding to the low-coverage N
1s spectrum (cf. Figure 2), a principal component centered at
397.7 eV that we assign to a dense monolayer phase in contact
with the Ag surface, and one at 398.5 assigned to molecules in
a multilayer configuration. Annealing to 473 K resulted in the
N 1s spectrum shown in the right panel of Figure 8. The
multilayer component is strongly attenuated, the submonolayer
contribution is absent, and molecules in the dense contact layer
are the predominant species; heating has resulted in spreading
of the initially inhomogeneous film. This observation is in good
agreement with the STM image (Figure 9) acquired in Cam-
¨
(31) de Jong, M. P.; Friedlein, R.; Sorensen, S. L.; Ohrwall, G.; Osikowicz,
W.; Tengsted, C.; Jo¨nsson, S. K. M.; Fahlman, M.; Salaneck, W. R.
Phys. ReV. B 2005, 72, 035448.
(32) Li, N. Y.; Su, Z.; Coppens, P.; Landrum, J. J. Am. Chem. Soc. 1990,
112, 7294.
9
J. AM. CHEM. SOC. VOL. 131, NO. 41, 2009 14917

A R T I C L E S
Turner et al.
Figure 9. STM image (300 × 300 Å2, Vtip ) 1.21 V, I ) 0.84 nA) of the
densely packed contact layer of the one-legged porphyrin acquired after
multilayer deposition and annealing to 473 K. White blobs are molecules
in the second layer, as judged by apparent height.
Figure 11. (a) C K-edge NEXAFS spectra acquired at five angles of photon
incidence θ for a ∼1 ML coverage of the one-legged porphyrin
[SAc]P-Mn(III)Cl on Ag(100). (b) Curve fitting analysis of the angle
dependence of π* resonances a and b to determine the corresponding
π-system tilt angles R.
bridge for a nominal porphyrin coverage of ∼1.1 ML (estimated
by STM). Line scans through molecules in the dense contact
layer indicate an apparent molecular height of 1.6 Å, which is
smaller than that found for molecules in the low-coverage
regime (1.9 Å); as we shall see, this finding assists interpretation
of the corresponding NEXAFS results.
N and C K-edge NEXAFS spectra acquired from the
as-deposited high-coverage film showed no energy shifts or
changes in peak shape compared to the submonolayer spectra,
indicating minimal changes in electronic structure with coverage.
However, the high-coverage spectra exhibit little dependence
on photon incidence angle, indicating significant disorder in the
inhomogeneous as-deposited porphyrin film (see Supporting
Information).
Figure 10. (a) N K-edge NEXAFS spectra acquired at five angles of photon
incidence (θ) for a ∼1 ML coverage of [SAc]P-Mn(III)Cl on Ag(100).
(b) Curve fitting analysis of the θ dependence of π* resonances A and C
to estimate the tilt angle (R) of the macrocycle.
After 473 K annealing both the N K-edge and C K-edge
NEXAFS exhibited significant angular variation (Figures 10 and
11) consistent with removal of multilayer aggregates and
9
14918 J. AM. CHEM. SOC. VOL. 131, NO. 41, 2009

One-Legged Metalloporphyrin on a Metal Surface
A R T I C L E S
the surface, resulting in decreased relaxation of the metal’s
valence electrons around the N core hole.
Conclusions
The one-legged acetyl-protected manganese porphyrin
[SAc]P-Mn(III)Cl may be deposited without decomposition or
demetalation on Ag(100) and subsequently deprotected, leading
to covalently bound thioporphyrin molecules. These in turn may
be dechlorinated, resulting in formation of stable, surface-
tethered, low-coordinate Mn centers. At high coverage an
unusual orientational transition occurs whereby the molecule
rolls over, preserving the tilt angle of the phenyl group toward
the surface, strongly increasing that of the macrocycle, decreas-
ing the apparent height of the molecule and decreasing its
footprint.
Compared to its more rigidly bound four-legged counterpart,
the one-legged porphyrin exhibits substantial differences in
mobility, spatial distribution, acetyl deprotection, dechlorination,
flexibility and orientational behavior as a function of coverage.
These properties are expected to be important in determining
access and escape of molecules to/from the porphyrin metal
center and hence their effectiveness in a variety of applications.
Thus, the physicochemical properties of porphyrin-functional-
ized surfaces should be markedly dependent on the mode of
tethering the macrocycle.
Figure 12. Proposed adsorption geometry of the one-legged porphyrin in
the dense contact layer tethered to Ag(100). Macrocycle and phenyl tilt
angles with respect to the surface are ∼50° and ∼53° respectively.
smoothing of the contact layer, in line with both the XPS and
STM results. XPS data taken before and after the acquisition
of the NEXAFS spectra indicated no significant beam damage
of the ∼1 ML porphyrin film. Detailed analysis of the N K-edge
resonances A and C, (Figure 10) yields estimated macrocycle
tilt angles of ∼48° and ∼51°, respectively, Very substantially
larger than that found for the low-coverage case (∼29.4° and
∼23.5° respectively). Analysis of the corresponding C K-edge
data for resonances a and b yields estimated tilt angles of ∼51°
and ∼53° for the macrocycle and the phenyl group, respectively,
the markedly increased tilt derived for the former being in good
agreement with the N K-edge results (Figure 10). In other words,
at high coverage a transition occurs in the contact layer:
compared to the low-coverage case, the tilt of the phenyl group
with respect to the surface remains unchanged, whereas that of
the macrocycle increases substantially. This change in adsorption
geometry may be accounted for by allowing the molecule to
roll over, as illustrated in Figure 12, preserving the tilt angle of
the phenyl group, increasing that of the macrocycle, decreasing
the apparent height of the molecule, and at the same time
decreasing its footprint, thus enabling closer packing. Note that
the increased N 1s XPS binding energy relative to that of the
low-coverage value (397.7 eV versus 397.2 eV) may also be
rationalized in terms of this change in adsorption geometry; at
high coverage the nitrogen atom ends up further away from
Acknowledgment. We are grateful to Silvano Lizzit, Sandra
Gardonio, and Michele Tranquilln for their assistance during the
synchrotron experiments. M.T. and O.P.H.V. acknowledge financial
support from Johnson Matthey plc and King’s College, Cambridge,
respectively. L.J.S. acknowledges financial support from the Swiss
National Foundation and the Novartis Foundation. D.J.W, G.K.,
and A.C.P. acknowledge financial support from the U.K. Engineer-
ing and Physical Sciences Research Council.
Supporting Information Available: N and C K-edge NEX-
AFS spectra acquired from the as-deposited high-coverage
porphyrin film. This material is available free of charge via the
Internet at http://pubs.acs.org.
JA904664E
9
J. AM. CHEM. SOC. VOL. 131, NO. 41, 2009 14919