Selective reductive desorption of a SAM-coated gold electrode revealed using fluorescence microscopy

Article abstract of DOI:10.1021/ja0494095

The reductive desorption of a self-assembled monolayer (SAM) of a fluorescent thiol molecule (BodipyC10SH) from Au was characterized using electrochemistry and epi-fluorescence microscopy. Molecular luminescence is quenched near a metal surface, so fluorescence was only observed for molecules reductively desorbed and then separated from the electrode surface. Fluorescence imaging showed that reductive desorption was selective, with desorption occurring from different regions of the Au electrode depending on the extent of the negative potential excursion. When desorbed, the molecules were sufficiently mobile, diffusing away from the electrode surface, thereby preventing oxidative readsorption. At sufficiently negative desorption potentials, all of the thiol was desorbed from the electrode surface, resulting in fluorescence at the air/solution interface. The selective removal of the thiol monolayer from distinct regions was correlated to features on the electrode surface and was explained through potential-dependent interfacial energies. This in situ electrofluorescence microscopy technique may be useful in sensor development.

Full text of DOI:10.1021/ja0494095

Published on Web 06/09/2004
Selective Reductive Desorption of a SAM-Coated Gold
Electrode Revealed Using Fluorescence Microscopy
†
§
†
‡
Jeff L. Shepherd, Arnold Kell, Emily Chung, Chad W. Sinclar,
§
,†
Mark S. Workentin, and Dan Bizzotto*
Contribution from the Department of Chemistry, AdVanced Materials & Process Engineering
Laboratory (AMPEL), UniVersity of British Columbia, VancouVer, BC, Canada,
Department of Metals & Materials, AdVanced Materials & Process Engineering Laboratory
(
AMPEL), UniVersity of British Columbia, VancouVer, BC, Canada, and
Department of Chemistry, UniVersity of Western Ontario, London, ON, Canada
Received February 2, 2004; E-mail: bizzotto@chem.ubc.ca
Abstract: The reductive desorption of a self-assembled monolayer (SAM) of a fluorescent thiol molecule
(BodipyC10SH) from Au was characterized using electrochemistry and epi-fluorescence microscopy.
Molecular luminescence is quenched near a metal surface, so fluorescence was only observed for molecules
reductively desorbed and then separated from the electrode surface. Fluorescence imaging showed that
reductive desorption was selective, with desorption occurring from different regions of the Au electrode
depending on the extent of the negative potential excursion. When desorbed, the molecules were sufficiently
mobile, diffusing away from the electrode surface, thereby preventing oxidative readsorption. At sufficiently
negative desorption potentials, all of the thiol was desorbed from the electrode surface, resulting in
fluorescence at the air/solution interface. The selective removal of the thiol monolayer from distinct regions
was correlated to features on the electrode surface and was explained through potential-dependent interfacial
energies. This in situ electrofluorescence microscopy technique may be useful in sensor development.
1
. Introduction
The functionalization of metal surfaces by adsorbed organic
molecules is a rapidly expanding area of science because of its
Physically adsorbed organic molecules are also a viable, but
frequently overlooked candidate for the modification of surfaces
for device applications. In either case, the engineering of such
devices based on surface modification is hampered by our
limited understanding of nanoscale physical and chemical
processes such as adsorption and desorption of a surfactant.
Therefore, a better understanding of the universal properties of
adsorbed surfactants ranging from chemisorbed to physisorbed
will give us access to a wider range of exploitable properties.
Various probes have been employed to characterize the
quality of adsorbed surfactants including electrochemical tech-
niques, which through the control of interfacial energetics can
result in conformational changes and/or desorption of the organic
from the metal. The process of adsorption/desorption has been
observed to be very reproducible for physisorbed organic
molecules on Au(111) and Hg electrode surfaces in neutral
1
-8
potential application in sensors and in nanotechnology. The
most common organic-modified metal cited involves the forma-
tion of a robust self-assembled monolayer (SAM) of thiol
molecules on noble metals, most notably Au. With the ease of
forming such an interface, it has proven to be a popular subject
for the creation of an ideal environment for the study of
chemically tunable surfaces,1
of the monolayer has become the subject of much discussion.
9
0-12
and thus the reproducibility
13-17
†
Department of Chemistry, University of British Columbia.
Department of Metals & Materials, University of British Columbia.
University of Western Ontario.
‡
§
(
(
(
1) Ishida, T. Springer Ser. Chem. Phys. 2003, 70, 91-106.
2) Schwartz, D. K. Annu. ReV. Phys. Chem. 2001, 52, 107-137.
3) Brockman, J. M.; Nelson, B. P.; Corn, R. M. Annu. ReV. Phys. Chem. 2000,
18
electrolyte. In the case of the thiol SAM, very negative
5
1, 41-63.
(
4) Ulman, A.; Kang, J. F.; Shnidman, Y.; Liao, S.; Jordan, R.; Choi, G.-Y.;
Zaccaro, J.; Myerson, A. S.; Rafailovich, M.; Sokolov, J.; Fleischer, C.
ReV. Mol. Biotechnol. 2000, 74, 175-188.
potentials are required for reductive desorption, and therefore
to circumvent hydrogen evolution a strongly alkaline electrolyte
19-52
(
(
(
(
(
5) Badia, A.; Lennox, R. B.; Reven, L. Acc. Chem. Res. 2000, 33, 475-481.
is required.
The readsorption of the thiol can occur through
6) Ulman, A. Chem. ReV. 1996, 96, 1533-1554.
7) Honeychurch, M. J. Recent Res. DeV. Electroanal. Chem. 2000, 2, 1-16.
8) Everhart, D. S. Handb. Appl. Surf. Colloid Chem. 2002, 2, 99-116.
9) Bard, A.; Abruna, H. D.; Chidsey, C.; Faulkner, L. R.; Feldberg, S. W.;
Itaya, K.; Majda, M.; Melroy, O.; Murray, R. W.; Porter, M. D.; Soriaga,
M. P.; White, H. S. J. Phys. Chem. 1993, 97, 7147.
(14) Boubour, E.; Lennox, R. B. J. Phys. Chem. B 2000, 104, 9004-9010.
(15) Boubour, E.; Lennox, R. B. Langmuir 2000, 16, 7464-7470.
(16) Ma, F.; Lennox, R. B. Langmuir 2000, 16, 6188-6190.
(17) Boubour, E.; Lennox, R. B. Langmuir 2000, 16, 4222-4228.
(18) Bizzotto, D.; Yang, Y.; Shepherd, J. L.; Stoodley, R.; Agak, J. O.; Stauffer,
V.; Lathuilliere, M.; Ahktar, A.; Chung, E. J. Electroanal. Chem. 2004.
(19) Walczak, M. M.; Popenoe, D. D.; Deinhammer, R. S.; Lamp, B. D.; Chung,
C.; Porter, M. D. Langmuir 1991, 7, 2687-2693.
(20) Widrig, C. A.; Chung, C.; Porter, M. D. J. Electroanal. Chem. 1991, 310,
335-359.
(21) Weisshaar, D. E.; Lamp, B. D.; Porter, M. D. J. Am. Chem. Soc. 1992,
114, 5860-5862.
(10) Penn, S. G.; He, L.; Natan, M. J. Curr. Opin. Chem. Biol. 2003, 7, 609-
6
15.
(
11) Battaglini, N.; Klein, H.; Dumas, P.; Moustrou, C.; Samat, A. Appl. Surf.
Sci. 2003, 212-213, 481-484.
(
(
12) Sinner, E.-K.; Knoll, W. Curr. Opin. Chem. Biol. 2001, 5, 705-711.
13) Badia, A.; Back, R.; Lennox, R. B. Angew. Chem., Int. Ed. Engl. 1994,
3
3, 2332.
10.1021/ja0494095 CCC: $27.50 © 2004 American Chemical Society
J. AM. CHEM. SOC. 2004, 126, 8329-8335
9
8329

A R T I C L E S
Shepherd et al.
an oxidative process; however, this is strongly dependent on
the length of the alkyl chain comprising the thiol. Completely
irreversible desorption has been observed for short alkyl chain
from solution.39,59,60 The search continues for more flexible types
of chemical patterning, and the selective removal of surfactant
from certain areas of the surface offers one possibility. This
method of removing the bound thiol has been used to create a
multiple component thiol layer by reductively desorbing a
portion of the original SAM, and then allowing another thiol
thiols (n ) 2-8), while a reversible process was noted for thiols
of longer alkyl chains (n > 16).3
0,32,33,53
Furthermore, experi-
ments using single-crystal electrodes have shown that this
2
9,36,39,46,47,57,61
reductive desorption process is dependent on crystallographic
molecule to assemble into these void spaces;
orientation and on the length of the alkyl chain.2
5,28,30
This
however, achieving this with a desired pattern is difficult.
Reductive desorption of a thiol SAM from metals of low index
surfaces has been observed to occur at reasonably separated
dependence on surface crystallography has also been observed
for the physically adsorbed surfactants.54 Morin has completed
a series of studies illuminating this process based upon
2
8,62
cathodic potentials.
Thus, an ideally shaped nanocrystal in
2
6,31,34,41
electrochemical and in situ FTIR studies.
Reductive
a truncated octahedron geometry containing both the (111) and
the (100) surfaces may be exploited in creating a patterned
surface by the selective removal of thiol from one face. The
ability to selectively desorb certain regions of the thiol will be
examined in this report using a multifaceted substrate.
desorption has also been imaged with in situ STM.3
9,45,52,55-58
In all of these studies, the fate of the desorbed molecules is not
clearly understood and has only been indirectly experimentally
observed.
Sometimes a multifunctional or patterned surface with more
than one type of chemical functionality is required and has led
to the investigation of thiol exchange reactions where more
weakly bound thiols are replaced by a stronger interacting thiol
To fully characterize this process, we will combine electro-
chemical techniques with a recently developed in situ probe.
As mentioned earlier, electrochemical methods allow us to
control the properties of a modified electrode. Coupling
electrochemistry with in situ techniques can enable observation
of the direct response to potential perturbations. While fluo-
rescence is widely used in the microscale characterization of
biological systems, it has been underutilized at this scale for
the study of physical and chemical phenomena. Recent spatially
(
(
(
22) Weisshaar, D. E.; Walczak, M. M.; Porter, M. D. Langmuir 1993, 9, 323-
29.
23) Schneider, T. W.; Buttry, D. A. J. Am. Chem. Soc. 1993, 115, 12391-
2397.
24) Abbott, N. L.; Gorman, C. B.; Whitesides, G. M. Langmuir 1995, 11, 16-
8.
3
1
1
(
(
(
(
(
25) Yang, D. F.; Wilde, C. P.; Morin, M. Langmuir 1996, 12, 6570-6577.
26) Yang, D. F.; Morin, M. J. Electroanal. Chem. 1997, 429, 1-5.
27) Finklea, H. O. Electroanal. Chem. 1996, 19, 109-335.
resolved observations of electrochemically induced changes in
63,64
an adsorbed monolayer
using epi-fluorescence microscopy
28) Zhong, C.-J.; Zak, J.; Porter, M. D. J. Electroanal. Chem. 1997, 421, 9-13.
29) Imabayashi, S.; Hobara, D.; Kakiuchi, T.; Knoll, W. Langmuir 1997, 13,
demonstrate its versatility in the characterization of interfaces.
This fluorescence technique is useful because molecular lumi-
nescence near metal surfaces is strongly related to the distance
separating the fluorophore from the metal. At close proximity
(<50 nm), fluorescence is efficiently quenched by nonradiative
energy transfer into surface plasmons and electron hole pairs
4
502-4504.
(
30) Zhong, C.-J.; Porter, M. D. J. Electroanal. Chem. 1997, 425, 147-153.
(
31) Yang, D. F.; Al-Maznai, H.; Morin, M. J. Phys. Chem. B 1997, 101, 1158-
1
166.
(
(
(
32) Yang, D. F.; Wilde, C. P.; Morin, M. Langmuir 1997, 13, 243-249.
33) Yang, D. F.; Morin, M. J. Electroanal. Chem. 1998, 441, 173-181.
34) Byloos, M.; Al-Maznai, H.; Morin, M. J. Phys. Chem. B 1999, 103, 6554-
6
561.
6
5,66
(
(
(
35) Badia, A.; Arnold, S.; Scheumann, V.; Zizlsperger, M.; Mack, J.; Jung,
in the metal.
Therefore, fluorescence will only be observed
G.; Knoll, W. Sens. Actuators, B 1999, B54, 145-165.
if the desorbed molecule is far from the metal surface. Thus,
the reductive desorption of a fluorescent thiol monitored by this
technique will result in an understanding of the desorption
process and fate of desorbed thiolates. By employing a
multifaceted metal substrate, the selective reductive desorption
will be illuminated at different potentials due to the different
energetics of each face. We describe such a process by
monitoring the potential-induced desorption of a fluorescent-
labeled thiol from various facets of a polycrystalline Au bead.
Selective desorption from specific regions of the electrode
surface, controlled by potential, will be demonstrated, and the
fate of these desorbed molecules will be monitored. Finally,
we will link these results on the thiol SAM with the previously
published work on the electrochemical and spectroelectrochemi-
cal characterization of physically adsorbed monolayers to reveal
some similarities between the two. The ability to selectively
desorb the thiol from certain facets of the metal surface is
36) Kuwabata, S.; Kanemoto, H.; Oyamatsu, D.; Yoneyama, H. Electrochem-
istry (Tokyo) 1999, 67, 1254-1257.
37) Kawaguchi, T.; Yasuda, H.; Shimazu, K.; Porter, M. D. Langmuir 2000,
1
6, 9830-9840.
(
(
38) Wong, S.-S.; Porter, M. D. J. Electroanal. Chem. 2000, 485, 135-143.
39) Kakiuchi, T.; Sato, K.; Iida, M.; Hobara, D.; Imabayashi, S.; Niki, K.
Langmuir 2000, 16, 7238-7244.
(
(
(
40) Calvente, J. J.; Andreu, R.; Gil, M.-L. A.; Gonzalez, L.; Alcudia, A.;
Dominguez, M. J. Electroanal. Chem. 2000, 482, 18-31.
41) Byloos, M.; Al-Maznai, H.; Morin, M. J. Phys. Chem. B 2001, 105, 5900-
5
905.
42) Mulder, W. H.; Calvente, J. J.; Andreu, R. Langmuir 2001, 17, 3273-
3
280.
(
43) Imabayashi, S.; Hobara, D.; Kakiuchi, T. Langmuir 2001, 17, 2560-2563.
(
44) Esplandiu, M. J.; Hagenstroem, H.; Kolb, D. M. Langmuir 2001, 17, 828-
8
38.
(
(
(
45) Wano, H.; Uosaki, K. Langmuir 2001, 17, 8224-8228.
46) Satjapipat, M.; Sanedrin, R.; Zhou, F. Langmuir 2001, 17, 7637-7644.
47) Oyamatsu, D.; Kanemoto, H.; Kuwabata, S.; Yoneyama, H. J. Electroanal.
Chem. 2001, 497, 97-105.
(
48) Byloos, M.; Rifai, S.; Al-Maznai, H.; Laferriere, M.; Morin, M. Langmuir
2
001, 17, 2478-2484.
(
49) Kakiuchi, T.; Usui, H.; Hobara, D.; Yamamoto, M. Langmuir 2002, 18,
5
231-5238.
(
50) Plummer, S. T.; Bohn, P. W. Langmuir 2002, 18, 4142-4149.
51) Brask, J.; Wackerbarth, H.; Jensen, K. J.; Zhang, J.; Chorkendorff, I.;
Ulstrup, J. J. Am. Chem. Soc. 2003, 125, 94-104.
(
(59) Felgenhauer, T.; Rong, H.-T.; Buck, M. J. Electroanal. Chem. 2003, 550-
551, 309-319.
(
(
(
52) Loglio, F.; Schweizer, M.; Kolb, D. M. Langmuir 2003, 19, 830-834.
53) Aoki, K.; Kakiuchi, T. J. Electroanal. Chem. 1998, 452, 187-192.
54) Bizzotto, D.; Noeel, J. J.; Lipkowski, J. Thin Solid Films 1994, 248, 69-
(60) Chung, C.; Lee, M. J. Electroanal. Chem. 1999, 468, 91-97.
(61) Shimazu, K.; Kawaguchi, T.; Isomura, T. J. Am. Chem. Soc. 2002, 124,
652-661.
7
7.
(
(
(
(
55) Hobara, D.; Miyake, K.; Imabayashi, S.; Niki, K.; Kakiuchi, T. Langmuir
(62) Walczak, M. M.; Alves, C. A.; Lamp, B. D.; Porter, M. D. J. Electroanal.
Chem. 1995, 396, 103-114.
1
998, 14, 3590-3596.
56) Schneeweiss, M. A.; Hagenstrom, H.; Esplandiu, M. J.; Kolb, D. M. Appl.
Phys. A 1999, 69, 537-551.
57) Hobara, D.; Sasaki, T.; Imabayashi, S.; Kakiuchi, T. Langmuir 1999, 15,
(63) Shepherd, J.; Yang, Y.; Bizzotto, D. J. Electroanal. Chem. 2002, 524-
525, 54-61.
(64) Shepherd, J. L.; Bizzotto, D. J. Phys. Chem. B 2003, 107, 8524-8531.
(65) Pineda, A. C.; Ronis, D. J. Chem. Phys. 1985, 83, 5330-5337.
(66) Chance, R. R.; Prock, A.; Silbey, R. In AdVances in Chemical Physics;
Prigogine, I., Ed.; Interscience Publishers: New York, 1978; Vol. 37, pp
1-65.
5
073-5078.
58) Dunbar, T. D.; Cygan, M. T.; Bumm, L. A.; McCarty, G. S.; Burgin, T.
P.; Reinerth, W. A.; Jones, L.; Jackiw, J. J.; Tour, J. M.; Weiss, P. S.;
Allara, D. L. J. Phys. Chem. B 2000, 104, 4880-4893.
8330 J. AM. CHEM. SOC.
9
VOL. 126, NO. 26, 2004

Reductive Desorption of a SAM-Coated Gold Electrode
A R T I C L E S
Figure 1. Schematic representation of the epi-fluorescence setup used for imaging during electrochemical measurements. (a) SEM of the bead electrode;
b) brightfield image of the bead electrode taken just before fluorescence measurements; the facet located on the bottom right is outlined.
(
Scheme 1. The Synthesis of BiodipyC10SH
attributed to the electric variable having control over the tension
of each face (electrocapillarity).
and positioned just above the microscope objective port (see Figure 1)
in an arrangement suitable for in situ investigation.
67
2.3. Electron Backscattered Diffraction (EBSD). EBSD was used
2
. Experimental Section
.1. Synthesis of BodipyC10SH. The fluorophore thiol precursor
utilized as the probe in this study, 4,4-difluoro-1,3,5,7-tetramethyl-8-
(10-mercapto)]-4-bora-3a,4a-diaza-s-indacene (BodipyC10SH), was
to determine the crystallographic surface normal for various regions
of the Au electrode. Indexing of the backscattered Kikuchi pattern was
performed via an automated software algorithm (HKL Channel 5
software). Measurement of the crystallographic surface normal neces-
sitates that the facet be precisely tilted within the microscope such that
each facet faces the detector and is at 70° to the horizontal plane. While
this technique has previously been used for the study of faceted
surfaces,68 the difficulty in precisely aligning each of the facets decreases
the accuracy of the technique. The magnitude of this error was judged
by measuring the known surface normal for a series of Si single crystals
mounted such that they were oriented at random angles to the horizontal.
It was found that the error in manually positioning the facets to 70°
within the microscope introduced an error of no more than (10° in
the measured crystallographic surface normal. While this error is large
relative to the normal systematic error involved in EBSD measurements
2
[
synthesized using the procedure outlined in Scheme 1. Full experimental
and characterization details are provided in the Supporting Information.
2
.2. Electrochemistry. Electrochemical investigations were per-
formed on an electropolished Au bead working electrode (SEM image
shown in Figure 1a), a Au coil counter electrode, and a saturated
calomel (SCE) reference electrode connected to the working solution
through a salt bridge. The 0.1 M NaOH supporting electrolyte (Fluka,
suprapur used as received) was prepared in Millipore water (>18 MΩ
cm) and purged with Ar (Praxair, cleaned with a charcoal filter,
Supelco). A constant blanket of Ar was maintained above the electrolyte
throughout the experiment. Solution cleanliness was confirmed by cyclic
voltammetry (CV) and differential capacitance measurements of the
Au bead in NaOH electrolyte prior to the surface modification. The
SAM was formed by introducing the clean, dry bead into a solution of
dissolved thiol in chloroform (3 mg/mL) for roughly 10 min. The bead
was then rinsed with water, followed by sonication in a methanol
solution to remove the physically adsorbed thiol. After evaporation of
methanol, the SAM-coated electrode was introduced to the electrolyte
(
∼1-2°), it was felt to be sufficient for the current study.
2
2
.4. Spectroelectrochemistry.
.4.1. Epi-fluorescence Imaging. Fluorescence imaging was per-
formed on an Olympus IX70 inverted microscope equipped with a 10×
(
67) Randle, V.; Engler, O. Introduction to Texture Analysis; Gordon and
Breach: Amsterdam, 2000.
(68) Davies, P. A.; Randle, V. J. Microsc. 2001, 204, 29-38.
J. AM. CHEM. SOC.
9
VOL. 126, NO. 26, 2004 8331

A R T I C L E S
Shepherd et al.
objective (NA ) 0.3, wd ) 10 mm) and a SPOT RT digital camera.
The inverted arrangement enabled looking up onto the surface of the
working electrode as shown in Figure 1. SEM and brightfield images
of the electrode are shown. All images of fluorescence reported were
collected using a Xe arc lamp and a U-MWIBA filter cube (Olympus:
BP 460, DM 505, BA 515lF), resulting in images of roughly 1 × 1.5
mm with a resolution of 1.93 µm/pixel (2 × 2 binning, 8 bit). Each
image was acquired with an exposure time of 50 ms and a gain of 1.
One potential step-imaging cycle consisted of a series of potential
steps. Starting at 0 mV/SCE, the potential was incrementally stepped
(∆E ) -50 mV) to a predetermined negative potential, and then the
potential was stepped back to 0 mV/SCE (∆E ) 50 mV). After every
step, interfacial capacitance and fluorescence images were recorded.
The potential steps were separated by 2 s, resulting in a scan rate of
roughly 25 mV/s. These potential-imaging cycles were applied to the
modified Au bead electrode with each successive cycle employing a
more negative turn-around potential (-0.950, -1.250, -1.300, -1.350,
and -1.400 V/SCE). In addition, once reaching the negative potential
of each cycle, the potential was held for 6 s to allow for acquisition of
three images and capacitance. Before each potential step-imaging
cycling procedure, one CV cycle using the same negative potential limit
was recorded (20 mV/s).
2
.4.2. Image Analysis. All images were analyzed using dip_image69
for MATLAB and Image Pro Plus 4.5. The average gray scale, which
was assumed to be proportional to the fluorescence intensity, was
calculated using a MATLAB procedure after subtracting the 0 mV/
SCE fluorescence image (to correct for the small background due to
scattered light).
3. Results and Discussion
3
.1. Electrochemical and Epi-fluorescence Investigations.
Figure 2. Capacitance measured during fluorescence imaging for the same
modified electrode surface with a sequentially changing negative potential
limit starting from (a) -0.950, (b) -1.250, (c) -1.300, (d) -1.350, and
The BodipyC10SH-modified gold bead electrode was introduced
to the electrochemical cell at E ) 0 V/SCE. The stability of
the SAM was first probed in a limited potential range using
cyclic voltammetry (CV) prior to fluorescence imaging and
capacitance measurement. To ensure that the fluorescent head-
group remained stable, the measurements were conducted in
less basic pH conditions than those typically used for reductive
desorption studies of alkanethiols. Therefore, the reductive
desorption peak reported in these studies would have been
obscured by hydrogen evolution at this lower pH. Capacitive
measurements will be presented because we are primarily
interested in these changes that are more sensitive and reliable
than the currents measured in CV. Figures 2 and 3 show results
for concurrent measurements of capacitance and fluorescence
during each potential-imaging cycle. A selection of fluorescence
images are shown in Figure 4, which correspond to the letter
legend in Figure 3. It is important to reinforce that the same
SAM-coated surface was probed using a series of capacitance
and fluorescence experiments with a progressively more negative
potential limit.
(e) -1.400 V/SCE. The dotted line represents the capacitance scan measured
for the bare Au electrode in contact with 0.1 M NaOH. Positive going
(closed symbols) and negative going scans (open symbols) for the modified
electrode are also shown. Capacitance is calculated assuming a series RC
circuit, 25 Hz, 5 mV rms.
capacitance was measured as the potential was stepped in a
negative direction in 50 mV intervals from the initial potential
to the first negative potential limit of -0.950 V/SCE. At this
negative potential, three fluorescence images were acquired. The
potential was then stepped positively in 50 mV increments back
to the initial potential. A small increase in the capacitance at
the positive limit was observed after one potential cycle,
suggesting some disruption or disorder of the SAM. The
corresponding fluorescence measurements show a small increase
in the fluorescence at the negative potential limit. This potential-
dependent fluorescence was used in the study of physically
6
3,70,71
adsorbed surfactants on gold electrodes
as well as by
Gaigalas in the study of the dynamics of immobilized fluoro-
7
2,73
phores
and fluorescein-labeled single-strand oligonucleotides
A freshly modified electrode surface immersed at E ) 0
7
4
immobilized onto gold. The changes in fluorescence are due
to a slight increase in the distance between the fluorophore and
the electrode surface. The increase in interfacial capacitance
suggests a partial desorption of some thiol, with the desorbed
molecules moving away from the interface.
-2
V/SCE had a capacitance which was much lower (1.2 µF cm )
than the uncoated electrode surface, demonstrating the formation
of a well-ordered layer SAM despite the bulky headgroup and
short assembly time (10 min). Longer assembly times (>20 h)
were not used because the minimum capacitance did not
significantly change and presumably a less organized layer
would be more responsive to potential, and less influenced by
other effects (e.g., strong hydrophobic intermolecular interac-
tions). Fluorescence from the electrode was recorded, and the
(
70) Bizzotto, D.; Lipkowski, J. J. Electroanal. Chem. 1996, 409, 33-43.
71) Sagara, T.; Zamlynny, V.; Bizzotto, D.; McAlees, A.; McCrindle, R.;
Lipkowski, J. Isr. J. Chem. 1997, 37, 197-211.
(
(
72) Gaigalas, A. K.; Li, L.; Ruzgas, T. Curr. Top. Colloid Interface Sci. 1999,
3, 83-101.
(
73) Li, L.; Meuse, C.; Silin, V.; Gaigalas, A. K.; Zhang, Y. Langmuir 2000,
16, 4672-4677.
(
69) van Vliet, L. J. DIPLib, the Delft Image Processing Library; http://
(74) Wang, L.; Silin, V.; Gaigalas, A. K.; Xia, J.; Gebeyehu, G. J. Colloid
Interface Sci. 2002, 248, 404-412.
www.ph.tn.tudelft.nl/DIPLib/DIPLib.html.
8332 J. AM. CHEM. SOC.
9
VOL. 126, NO. 26, 2004

Reductive Desorption of a SAM-Coated Gold Electrode
A R T I C L E S
desorption at this potential. Some surfactant was desorbed from
the metal surface, and, for fluorescence to be observed, this
surfactant must be separated from the metal by some as yet
undetermined distance.
These results for the desorption of a SAM exactly parallel
the results previously published for the desorption of physically
1
8
adsorbed lipid-like surfactants. These weakly adsorbed sur-
factants are displaced from the electrode surface by less negative
potentials and are driven by changes in potential-dependent
interfacial tension (electrocapillarity). The desorption from Au
(111) occurred over a very small potential region, and the
readsorption of these insoluble surfactants was found to be
complete. A mechanism was proposed to explain these potential-
dependent changes based upon electrochemical and spectro-
electrochemical studies. These fluorescence results have shown
that the reductive desorption of the SAM is very similar and
the existence of desorbed molecules near the electrode surface
can be explained using the same mechanism developed for the
physically adsorbed surfactants. The oxidative readsorption
observed for long-chain alkylthiols can be conveniently ex-
plained through the existence of these desorbed molecules near
3
1
the electrode surface.
The fluorescence image of the desorbed fluorescing surfactant
at -1.250 V/SCE, shown in Figure 4a, displays significant heter-
ogeneity. A striking correlation can be made between the bottom
right of these fluorescence images and a facet observed in a
brightfield image of the electrode surface taken before potential
cycling (outlined region in Figure 1b). Desorption of the mole-
cules seems to be occurring from particular regions of the Au
bead electrode. The images taken while holding at the negative
potential limit illustrates the mobility of the desorbed surfactant.
Features in the image become less sharp due to a slow diffusion
away from the region near the electrode, seen most easily by
comparing images B1, B2, and B3. In our previous investiga-
Figure 3. Potential-dependent average fluorescence intensity measured from
the same electrode surface with sequentially changing negative potential
limit starting at (a) -0.950, (b) -1.250, (c) -1.300, (d) -1.350, and (e)
-
1.400 V/SCE. Positive (open symbols) and negative potential scans (closed
symbols) are also shown.
6
3,64
tions,
a decrease in fluorescence intensity for the positive
potential scan direction was attributed to readsorption of the
surfactant. However, diffusion was not evident in those experi-
ments, and the minimum capacitance did not increase after the
negative potential excursion. The decrease in fluorescence inten-
sity observed in this case was mainly attributed to loss of the
surfactant from the interface, because the increase in the mini-
mum capacitance indicates incomplete readsorption (via oxida-
tive process) of the surfactant. Also, readsorbed molecules or
those adsorbed molecules able to migrate to the desorbed regions
would be expected to desorb on the next cycle, and, as we shall
demonstrate, regions where fluorescence was observed during
this cycle did not fluoresce on subsequent cycles. The hysteresis
in the return potential scan can also be explained by the
combination of incomplete readsorption and loss of surfactant.
Figure 4. Fluorescence images from the same electrode for various
potentials (letters correspond to potentials and intensities shown in Figure
3
) obtained during potential scans to differing negative potentials: (a)
-
1.250, (b) -1.300, (c) -1.350V/SCE.
The negative limit for the subsequent potential scan was made
more negative (-1.250 V/SCE), resulting in a significant
increase in capacitance demonstrating the instability of the
monolayer toward large negative potentials. The minimum
capacitance of the SAM-coated interface has increased to 3 µF
cm (measured at the positive potential limit). At the negative
potential limit, the capacitance increased strongly, indicating
the penetration of the SAM by the electrolyte through potential
created defects in the layer caused by the initial stages of
reductive desorption. The change in capacitance at this potential
was accompanied by a marked increase in the average fluores-
cence intensity, also consistent with significant reductive
The third potential-imaging cycle was performed with a
negative potential limit of -1.300 V/SCE. At this negative
potential, the capacitance approached the value of the unmodi-
fied electrode surface. After potential cycling, the SAM was
significantly perturbed, resulting in a minimum capacitance of
6 µF cm and a capacitance scan that begins to resemble the
uncoated Au electrode. The fluorescence intensity closely
follows the cathodic scan of the previous run, but remains
consistently below until reaching the negative potential limit
where the maximum in fluorescence intensity became larger.
The fluorescence images are similarly heterogeneous as the
-
2
-
2
J. AM. CHEM. SOC.
9
VOL. 126, NO. 26, 2004 8333

A R T I C L E S
Shepherd et al.
previous potential scan, but fluorescence was observed from
distinctly different regions of the electrode. The reductively
desorbed molecules were also observed to diffuse away from
the surface of the electrode shown by the three images at the
negative potential limit (E1, E2, and E3). The anodic scan shows
a strong decrease in fluorescence in addition to the decrease in
capacitance; as before, the incomplete readsorption of the
desorbed molecules partially explains the loss of fluorescence
intensity, but diffusion away from the electrode plays the main
role.
The fourth potential-imaging cycle used a negative potential
of -1.350 V/SCE. The capacitance resembles an unmodified
electrode interface, indicating a very strong disruption in the
adsorbed monolayer. The minimum capacitance in this case was
-
2
1
3 µF cm . The corresponding fluorescence intensity was
Figure 5. Fluorescence images (first column) obtained at various negative
potential limits: (a) -1.250, (b) -1.300, (c) -1.350 V/SCE. Fluorescent
regions were outlined (middle column), and then applied to the image
recorded at -1.300 V/SCE (last column) to illustrate the regional specificity
of the desorption process.
weaker, and the fluorescence image was again not uniform,
clearly showing fluorescence in a region that was dark in the
previous cycle. Diffusion of the fluorescing molecules was also
observed at this potential (H1, H2, and H3).
The potential scan out to -1.400 V/SCE (fifth potential-
imaging scan) revealed an electrode surface with a capacitance
practically identical to that of the uncoated electrode, and, as
expected, no fluorescence was observed. All of the molecules
composing the initially complete SAM have been reductively
desorbed from the metal surface. Interestingly, at the end of
the experiment, fluorescence was observed at the electrolyte
surface, strongly suggesting the segregation of the desorbed
molecules (now presumably thiolates) to the gas|solution (G|S)
interface. An examination of the time lapse fluorescence images
in Figure 4 shows that, when reductively desorbed, the
molecules stay near the electrode surface and only slowly diffuse
away on the time scale of the fluorescence measurements. It is
expected that the surfactant would oxidatively readsorb if it
remained in the vicinity of the metal surface as in the reductive
desorption experiments done on thiols with long alkyl chains.
Oxidative readsorption onto bare gold regions would result in
fluorescence during the subsequent desorption potential scan.
This phenomenon was not observed in these studies, but it could
be achieved by using a thiol with a longer alkyl chain, resulting
in less solubility in the electrolyte. As the current study
demonstrates, there exists a strong similarity between the
potential-driven desorption from the electrode surface of co-
valently attached thiols and of physisorbed lipid-like molecules.
Further experiments are planned using a more insoluble
fluorescent thiol to confirm that the readsorption process is, like
the desorption process, general for both chemisorbed and
physisorbed surfactants. One important observation in the current
study made clear by our in situ fluorescence imaging is the large-
scale heterogeneity in these fluorescence images that was not
present in the previous single-crystal studies. The region of the
electrode where fluorescence was observed, and therefore where
desorption occurred, was strongly influenced by potential and,
as we shall show below, electrode surface crystallography or
morphology.
Figure 5. Outlines of each image (shown in the second column
of each row) were created using a simple segmentation
procedure and were overlaid onto the -1.300 V/SCE image,
displayed in the third column.
Fluorescence from different regions of the surface is clearly
related to the value of the negative potential limit. The small
facet observed on the bottom right was desorbed at the least
negative potential (-1.250 V/SCE). The image taken at the
negative potential limit of the subsequent scan (-1.300 V/SCE)
shows no fluorescence from this region, indicating the irrevers-
ible loss of the surfactant due to desorption from the facet. The
control that potential offers is even clearer when comparing the
images taken at -1.350 and -1.300 V/SCE. The fluorescent
regions observed for -1.350 V/SCE were not observed to be
fluorescent at the other less negative desorption potentials.
Conversely, all regions would show fluorescence if the desorp-
tion potential limit is pushed to -1.350 V/SCE on the first cycle.
The sensitivity toward the value of the potential driving these
changes is clear and can be understood in terms of the energetics
of the electrode/electrolyte interface. Bead electrodes are
typically composed of regions of differing surface crystal-
7
5
lography, which can be visualized by etching in aqua regia.
The SEM and optical image (Figure 6a and b, respectively) of
the etched Au bead show the heterogeneity of our electrode
surface with some features outlined in Figure 6c. Because the
bead electrode was not a single crystal, the presence of grain
boundaries and defects can be used in comparing the fluores-
cence images with the electrode surface morphology. The SEM
and optical image clearly show the facet (labeled I), also
observed in the brightfield image shown in Figure 1b. In this
region, fluorescence was observed for the least negative
desorption potential. Because studies of reductive desorption
of thiol SAMs on Au single crystals have shown the dependence
of reductive desorption potential on surface crystallography, we
anticipate that the regions first desorbed were from surfaces with
a (111) orientation. This was confirmed for the facet on the
bottom right (I) by EBSD.
3
.2. Selective Reductive Desorption. To define in detail the
relationship between surface crystallography and desorption
potential, image analysis was performed on the first images
acquired at each negative potential limit for the second, third,
and fourth cycles (negative limits are -1.250, -1.300, and
Desorption from other regions of the electrode surface are
most likely also dependent on surface crystallography, but EBSD
(
75) Perdriel, C. L.; Arvia, A. J.; Ipohorski, M. J. Electroanal. Chem. Interfacial
-1.350 V/SCE, respectively), as shown in the first column in
Electrochem. 1986, 215, 317-329.
8334 J. AM. CHEM. SOC.
9
VOL. 126, NO. 26, 2004

Reductive Desorption of a SAM-Coated Gold Electrode
A R T I C L E S
The fluorescence observed on the right-hand side is constrained
between the facet and the grain boundary outlining region II.
Finally, overlaying of the mask created from the -1.350 V/SCE
image is shown in Figure 6f. The curved region outlined
correlates well with the lines that decorate the electrode surface
(labeled VI). The small circular region next to the triangular
feature is encompassed in the outlined area, and the grain
boundary that curves off to the left from the triangular region
(III) and (V) is also captured in the outlined region.
Overall, the correlation of the fluorescence features to the
electrode surface features is quite close, and, although an exact
correspondence of fluorescence to surface features was not
possible, a number of regions of fluorescence were “assigned”
to surface features. The potential-dependent fluorescence fea-
tures seem to match well with the electrode surface structure
from which desorption took place.
Figure 6. Correlation between surface morphology (revealed through
etching) and fluorescence features at increasingly negative potentials. (a)
SEM image of etched electrode; (b) brightfield image with labeled features;
and (c) brightfield image with features outlined. Outlined brightfield image
with masks created from (d) -1.250, (e) -1.300, and (f) -1.350 V/SCE
fluorescence images.
4. Conclusions
The reductive desorption of a SAM was monitored for the
first time with electrofluorescence microscopy, and the fate of
the desorbed molecules was determined. This union of electro-
chemical techniques and spatially resolved fluorescence has
proven to be a powerful and nondestructive method of monitor-
ing the potential-induced changes in adsorbed organic layers.
We have shown the reductive desorption of a SAM of a slightly
soluble fluorescent thiol from selective regions of a polycrys-
talline Au bead. The observation of intense fluorescence at
negative potentials displays the desorption of the surfactant from
the metal surface. The slight disruption of this desorbed layer
was observed through both capacitive measurements and
diffusion of the thiol. A fluorescence increase at desorption has
been observed in the case of physically adsorbed surfactants
and is strikingly similar to the desorption process for these
covalently bound molecules. This implies that the mechanism
of desorption is general and may hold for other systems as well.
The readsorption process of thiols may have the same generality
if a less soluble thiol is used; however, the slight solubility of
the surfactant in this experiment had the unique advantage of
allowing selective exposure of different regions of the electrode
through control of the potential. This selective desorption was
further defined using the techniques of EBSD with the conclu-
sive assignment of desorption from the (111) facet at -1.250
V/SCE. Furthermore, image-masking techniques led to the
assignment of desorption from areas of the electrode surface
where EBSD was unable to unambiguously assign a crystal face.
This selective desorption may have profound application in the
creation of a thiol-modified surface with a desired pattern and
multifunction, and in the electrofluorescence technique in the
detection strategy used for SAM array sensors.
was unable to unambiguously determine the surface normals
for curved portions of the surface or for regions that did not
exhibit clear facets. Smooth reflective features at the top of the
image resembling an elongated triangle (labeled II) with a small
circular region on its left are obvious. Starting from the
triangular region, a grain boundary (labeled III) curves off to
the left and then downward toward another half-moon feature
(labeled IV). This boundary continues on the bottom of the
image toward the facet on the right (labeled V). These key
features allow overlaying of the fluorescence images with the
optical image. This was accomplished by starting with the facet
(I) and overlaying the first desorption image using the oblong
shape of the facet in orienting the optical image. We are
confident in our analysis because SEM images of the complete
bead surface did not reveal any other large oblong facet. This
optical image of the electrode was created by overlaying the
in-focus regions from a number of pictures with differing focal
planes to facilitate comparison with the SEM image. This was
not done for the fluorescence images. Fluorescence will be
measured from regions not in the focal plane, so a direct
comparison is qualitative but still very useful.
Figure 6d shows the mask created from the fluorescence
image recorded at -1.250 V/SCE overlaid on the outlined
optical image of the electrode. The optical image was rotated
to optimally fit the shape of the facet. The fluorescent feature
protruding from the center top matches well with the extended
triangular feature on the optical image (II). Unfortunately, the
feature in the center of the image does not seem to correlate
well with any surface irregularities. Overlaying the mask created
from the -1.300 V/SCE is shown in Figure 6e. The facet is no
longer fluorescent, but the regions immediately around the facet
are fluorescent, allowing a clear fit with the optical image. The
grain boundary on the bottom of the image (V) correlates with
the observed fluorescence. The region outlined at the top left
also correlates well with the area outlined by the grain
boundaries that extend from the triangular region to the left.
Acknowledgment. This work was supported by the NSERC
(Canada). J.L.S. was supported by a NSERC PGS B graduate
student fellowship.
Supporting Information Available: Synthesis of 4,4-difluoro-
1
,3,5,7-tetramethyl-8-[(10-mercapto)]-4-bora-3a,4a-diaza-s-in-
dacene (BodipyC10SH). This material is available free of charge
via the Internet at http://pubs.acs.org.
JA0494095
J. AM. CHEM. SOC.
9
VOL. 126, NO. 26, 2004 8335