Alkylation of gold surface by treatment with C18H 37HgOTs and anodic Hg stripping

Article abstract of DOI:10.1021/ja3117125

Treatment of a gold surface with a solution of C₁₈H ₃₇HgOTs under ambient conditions results in the formation of a covalently adsorbed monolayer containing alkyl chains attached directly to gold, Hg(0) atoms, and no tosyl groups. It is stable against a variety of chemical agents. When the initial deposition is performed at a positive applied potential and is followed by oxidative electrochemical stripping, the mercury can be completely removed, leaving a gold surface covered only with alkyl chains. The details of the attachment structure are not known. The conclusions are based on infrared spectroscopy, X-ray and UV photoelectron spectroscopy, ellipsometry, contact angle goniometry, differential pulse polarography, and measurements of electrode blocking and electrochemical admittance.

Full text of DOI:10.1021/ja3117125

Subscriber access provided by LAURENTIAN UNIV
Article
Alkylation of Gold Surface by Treatment
with C18H37HgOTs and Anodic Hg Stripping
Malgorzata Mucha, Eva Kaletova, Anna Kohutova, Frank Scholz, Elizabeth S
Stensrud, Ivan Stibor, Lubomir Pospisil, Florian von Wrochem, and Josef Michl
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/ja3117125 • Publication Date (Web): 11 Feb 2013
Downloaded from http://pubs.acs.org on February 28, 2013
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 45
Journal of the American Chemical Society
1
2
3
Alkylation of Gold Surface by Treatment with C H HgOTs and Anodic Hg Stripping
18 37
4
5
6
7
8
a a a b a
Malgorzata Mucha, Eva Kaletová, Anna Kohutová, Frank Scholz, Elizabeth S. Stensrud, Ivan
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a a,c b
Stibor, Lubomír Pospíšil, Florian von Wrochem, and Josef Michl*
a,d
a
Institute of Organic Chemistry and Biochemistry, Academy of Sciences of the Czech Republic,
Flemingovo nám. 2, 16610 Prague 6, Czech Republic
b
Sony Deutschland GmbH, Materials Science Laboratory, Hedelfinger Strasse 61, 70327 Stuttgart,
Germany
c
J. Heyrovský Institute of Physical Chemistry, Academy of Sciences of the Czech Republic,
Dolejškova 3, 18223 Prague, Czech Republic
d
Department of Chemistry and Biochemistry, University of Colorado, 215 UCB, Boulder, Colorado
80309-0215, United States. E-mail: michl@eefus.colorado.edu
Abstract. Treatment of a gold surface with a solution of C H HgOTs under ambient conditions
18 37
results in the formation of a covalentlyadsorbed monolayer containingalkyl chains attached directly
to gold, Hg(0) atoms, and no tosyl groups. It is stable against a variety of chemical agents. When
the initial deposition is performed at a positive applied potential and is followed by oxidative
electrochemical stripping, the mercury can be completely removed, leaving a gold surface covered
only with alkyl chains. The details of the attachment structure are not known. The conclusions are
based on infrared spectroscopy, X-ray and UV photoelectron spectroscopy, ellipsometry, contact
angle goniometry, differential pulse polarography, and measurements of electrode blocking and
electrochemical admittance.
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 45
1
2
3
Introduction
4
5
6
1,2
It has been recently reported that disordered monolayers form on gold surface upon
7
8
treatment with solutions of trialkylstannyl tosylates, trifluoroacetates, or triflates under ambient
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
conditions. This discovery was preceded and motivated by accidental observations of attachment
3,4,5
of organomercury salts to the surface of gold
and even earlier investigations of adsorption of
6
organoplatinum complexes to platinum. The results indicated that such behavior may have some
generality and that other organometallics should be examined, but trialkylsilyl derivatives were
found to be unreactive. At the time, it was established that the leaving group did not remain in the
monolayer, but no direct information on the mode of attachment of the adsorbate to the gold surface
was available. We have since found that the process can be extended to additional structural types
7
of organometallics, carrying various main group metals. The organic groups are detached from the
metal atom during their transfer to the gold surface, and under ambient conditions the metal usually
ends up as an oxide.
Complementary evidence for transfer of alkyl groups from
alkyltrimethylstannanes to the atoms of a gold surface resulted from recent independent
8,9
measurements of single-molecule conductivity, and dovetails with our findings.
Presently, we take a closer look at the behavior of an organomercurial, C H HgOTs, and
18 37
report spectroscopic and electrochemical evidence that a treatment of a gold surface with a solution
of n-octadecylmercuric tosylate under ambient conditions covers the surface with a monolayer that
consists of n-octadecyl groups and mercury in its elemental form, and contains no tosylate. The
adsorbed atomic mercury can be removed by oxidative anodic stripping, leaving a coating of alkyl
10
groups on an otherwise clean gold surface. Elsewhere, we report analogous monolayer formation
with C H HgOTs and find that the removal of elemental mercury from the initially formed surface
4 9
2
ACS Paragon Plus Environment

Page 3 of 45
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
monolayers can be accomplished by thermal annealing of the monolayer. At this time, we do not
have direct spectroscopic evidence for the detailed structure of the direct attachment of the alkyl
groups to the surface. The simplest possibility is a single alkyl-Au bond, but we cannot exclude a
loss of one or more hydrogen atoms from the terminal or penultimate carbons of the alkyl chain and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
more complex bonding patterns such as alkylidene-Au , etc.
2
The use of main group organometallics for the direct transfer of organic residues to gold
11,12,13,14,15,16,17
surface promises to complement the established application of thiols
and analogous
18,19,20
compounds
that produces monolayers in which carbon is believed to be attached to the gold
surface through a mediating atom such as sulfur. Each choice appears to have certain disadvantages
21,22,23
(for thiols, sensitivity to aerial oxidation has been noted
), and a wider selection of options is
likely to be useful.
Results
Monolayer Formation and Ellipsometric Characterization. Figure 1 compares the
!5
gradual increase of the ellipsometric thickness of the surface layers formed from 10 M solutions
of C H HgOTs in THF and C H SH in ethanol. In both instances, the adsorption process is self-
18 37 18 37
limiting, and the growth of the adsorbed layer stops after about two hours at an ellipsometric
thickness of ~13 Å for C H HgOTs and ~21 Å for C H SH, regardless of whetherthe gold surface
18 37 18 37
was first cleaned with a hydrogen flame or a piranha solution. Although the growth limit is reached
after about 2 h in both cases, the ellipsometricallydetermined initial rate of growth of the monolayer
!3
is faster with C H SH than with C H HgOTs. At 10 M C H HgOTs concentration the same
18 37
18 37
18 37
limiting thickness is reached in ~1.5 h.
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 45
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Kinetics of monolayer formation followed by ellipsometric
thickness d for C H HgOTs (•,–) and C H SH (#,!) on gold
18 37
18 37
substrates cleaned by piranha (P) and hydrogen flame annealing (A).
Contact Angle Goniometry. Contact angles of water on monolayers from C H HgOTs
18 37
are 96° ± 4° and 101° ± 2° when the gold substrates are cleaned by hydrogen flame or piranha
solution, respectively. They are distinctly higher than the 17 ± 3° and 71 ± 4° angles measured on
the initial gold substrate cleaned by hydrogen flame or piranha, respectively, and are similar to the
1
values 92 – 97° found for layers generated from trialkylstannyl precursors. The angle is
significantly lower than the 113° we observe for a monolayer from C H SH adsorbed on gold
18 37
cleaned by piranha and close to the 100° we observe for a C H SH monolayer on flame-annealed
18 37
gold.
4
ACS Paragon Plus Environment

Page 5 of 45
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. ATR-FTIR spectra. A: C H SH (black and red) and
18 37
C H HgOTos (cyan and green) based monolayers adsorbed on
18 37
gold substrates cleaned by piranha (P) and hydrogen flame
annealing (A). B: Au plate treated with C H HgOTs before
18 37
(black) and after (red) electrochemical stripping.
Infrared Spectroscopy (IR). Single-reflectionattenuatedtotalreflectance(ATR)IR spectra
contain the typical peaks of alkyl chains (Figure 2A) and those of the tosylate residue are absent.
The spectra show only vibrations attributable to the CH and CH groups: stretching at 2853,
3 2
í (CH ), 2871, í (CH ), 2925, í (CH ), and 2958, í (CH ), and bending at 1378, 1418 and 1468 (all
s 2 s 3 3
as
2
as
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 45
1
2
3
-1 1 -1
in cm ). As in the monolayers from trialkylstannyl compounds, í (CH ) occurs 5 cm and í (CH )
2
s
2
as
4
5
6
-1
6 cm above their respective locations in the highly ordered monolayer of 1-octadecanethiol,
7
8
suggesting that the alkyl chain are not all-trans, but disordered. Upon heating, the monolayer
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
formed from octadecanethiol becomes disordered and the symmetric and antisymmetric CH
2
stretches move to higher frequencies. In the octadecylmercury derived monolayer these bands are
alreadyshifted at room temperature and no further shift occurs upon heating. As expected, the peaks
due to CH vibrations dominate considerably over those due to CH vibrations, more so than they
2
3
do for monolayers obtained from C H HgOTs (Supporting Information).
4 9
Photoelectron Spectroscopy: X-Ray (XPS) and UV (UPS). X-ray photoelectron spectra
show the presence of Hg and C and the absence of S at the gold surface in a fully grown
C H HgOTs derived monolayer, hence the tosylate residue is lost from C H HgOTs during the
18 37
18 37
1
self-assembly process, similarly as it is from organostannyl tosylates. In the XP spectra of the
pristine monolayers, the Hg atoms yield only a single set of peaks both in the 4f and in the 4d
24,25
regions. The core level energies (99.67 for Hg 4f ) fit the expectations for elemental Hg.
7/2
The stability of the Hg signals in time as a function of temperature was investigated
10
separately on very similar samples produced by treatment of a gold surface with C H HgOTs. In
4
9
high vacuum, the signals are stable indefinitely at ambient temperature and up to ~60 EC, but begin
to disappear rapidly at ~90 EC, and are inobservable above 120 EC. The disappearance of Hg from
the surface is presumablydue to diffusion into the bulk, but evaporation maycontribute as well. The
stability of the Hg signals at room temperature is important for the present purposes because it
permits us to use their intensities for a quantitation of the relative amounts of Hg and C present.
The results are collected in Table 1 and sample spectra are shown in Figure 3.
6
ACS Paragon Plus Environment

Page 7 of 45
Journal of the American Chemical Society
1
2
3
Table 1. XPS results for monolayers from C H HgOTs.
18 37
4
5
6
______________________________________________________________________________
7
8
a a
Sample R(Hg/Au) R(C/Au) A(Hg)/D
2 b
2 b
c
c
c
A(C )/D
18
E
/eV
E
/eV
E /eV
C 1s
Hg 4f7/2
Hg 4d5/2
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
______________________________________________________________________________
initial
0.073
0.64
13
36
99.67
358.16
284.63
stripped
-
0.68
-
34
-
-
284.50
______________________________________________________________________________
a
Elemental ratios from XPS peak areas after correction for instrumental sensitivity factors and for attenuation from the
b
Hg and S layers (attenuation factor is 0.69 for a dense Hg layer as in the initial SAM). Area/atom and area/molecule
26
obtained by comparison of Hg/Au and C/Au ratios with S/Au ratios of densely packed alkanethiol monolayers (S/Au
2
~ 0.045, area/molecule = 21.6 Å ) and by consideration of the photoemission attenuation due to the organic layer,
c
respectively. Chemical shifts derived from peak centers of Gaussian/Lorentzian fits to the XPS data.
The Hg 4d signals appear at binding energies of 358.16 eV (shoulder) and 378.06 eV for the
5/2 and the 3/2 components, respectively . The Hg 4d peaks are partially overlapped by strong Au
4d signals from the gold substrate at 334.98 and 353.09 eV (Au 4d and Au 4d , respectively). A
5/2
3/2
fit to the data, subject to peak area constraints imposed by angular momentum rules, permits an
evaluation of Hg abundance at the surface, assuming equal attenuation of Au and Hg photoelectrons
by the alkyl chain overlayer. With these assumptions, consistent results are obtained for Hg 4d and
Hg 4f lines. A comparison of the XPS intensity ratios of Hg and Au signals in C H Hg SAMs with
18 37
the intensity ratios of S and Au signals in densely packed dodecanethiol monolayers, which we use
2
as a reference, yields an area of ~13 Å per Hg atom for C H Hg SAMs. This corresponds to an
18 37
Hg surface density that is roughly twice as high as that of sulfur atoms in alkanethiol monolayers.
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 45
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. XPS in the Hg 4f (panel A) and Hg 4d (panel B) regions for monolayers formed by the
treatment of gold surface with C H HgOTs. Symbols represent data points, lines (red an blue)
18 37
are fitted curves. The black line is the background.
8
ACS Paragon Plus Environment

Page 9 of 45
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
The C 1s and Hg 4f signal intensities indicate however that the surface concentration of carbon is
about 3 times lower than what is expected based on Hg density and on molecular composition
(C H Hg). Overall, the surface density of C units on Au is about 40% lower than in the densely
18 37
18
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
packed alkanethiols.
The location of the C 1s peak at a binding energy of 284.5 eV is in good agreement with its
position in alkanes. The relative abundance of surface-bound carbon (or carbons) is too low for
them to be discerned as a peak with a chemical shift different from the others.
The UV photoelectron spectrum (Supporting Information, Figure S9) of the monolayer
produced with C H HgOTs is verysimilar to that of a monolayer produced with dodecanethiol and
18 37
shows the characteristic alkyl valence band structure.
Chemical Stability. Chemical stability of the C H HgOTs derived monolayers was
18 37
-1
examined by monitoring the loss of their IR absorbance between 2800 and 3000 cm as a function
of time after exposure either to the laboratory atmosphere for a week or to various solvents and
reagents overnight. It is similar to the stability of monolayers produced from trialkylstannyl
1
precursors and generally slightly lower than that of monolayers obtained with C H SH (Figure 4).
18 37
Only the resistance to oxidants is significantly higher. Even after the monolayers are partially or
fully desorbed, no new absorption bands appear. There is not much difference between the two gold
cleaning procedures. In a sense, a stability comparison to monolayers obtained with C H SH is
18 37
unfair since the latter are denser and less permeable (cf. electrode blocking below). It is conceivable
that at similar density monolayers composed of directly bound alkyls would be considerably more
resistant than thiolate monolayers. This is not an issue that we can address presently and present
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 45
1
2
3
Figure 4 primarily as evidence that the alkyl groups in our monolayers are attached strongly and not
4
5
6
by mere physisorption.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Stability of adsorbed monolayers from C H SH
18 37
(black and red bars) and from C H HgOTos (cyan and green
18 37
bars) on gold surface after 20 - 23 h immersion at room
temperature, followed by rinsing and drying. R is the percent
monolayer remaining on the gold surface, calculated as the ratio
of the final to the initial integrated intensity of the 2800-3000
-1
cm bands. (A) Dry CH Cl , (B) wet CH Cl , (C) n-hexane, (D)
2
2
2
2
ethanol, (E) water, (F) 0.1 M H SO , (G) 0.1 M NaOH, (H) 1
2 4
mM KMnO , (I) 30% H O , (J) 10 mM NaBH , and (K) 7 days
4 2 2 4
in laboratory air. Gold substrates cleaned by piranha (P) and
hydrogen flame annealing (A).
Electrochemistry. The electrochemical stability range is similar for monolayers derived
from C H HgOTs (!1,38 to +1,40 V) and C H SH (!1,33 to +1,40 V). Cyclic voltammetry with
18 37 18 37
3- 4!
aqueous[Fe(CN) ] /[Fe(CN) ] asaredox probeshowsthatthelayerdepositedfromC H HgOTs,
6
6
18 37
1
like those obtained from trialkylstannnyl compounds, reduces the cyclic voltammetric peak height
to about two-thirds with respect to a clean Au sample (the electrode surface is ~35 % blocked,
3- 4!
Figure 5A). Since the electron transfer rate constant of [Fe(CN) ] /[Fe(CN) ] is high, cyclic
6
6
voltammetry may not be sensitive enough to the change of the electron transfer rate due to the
10
ACS Paragon Plus Environment

Page 11 of 45
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
blocking. A more sensitive method is impedance spectroscopy, which yields the charge transfer
resistance of fast electron transfer reactions, derived from the radius of a semicircle of the electrode
impedance plot. The impedance method can detect rate changes that are still within the diffusion
controlled limit on the time scale of voltammetry. Indeed, already a qualitative inspection indicates
a substantial increase of the semicircle relative to the one obtained on a clean Au surface (Figure
5B).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The blockingis incomplete even after an overnight immersion in a solution of C H HgOTs,
18 37
whereas a 2-h immersion in a solution of C H SH is sufficient to produce a monolayer that
18 37
completely suppresses the electrochemical response.
A more detailed electrochemical examination of the adsorbed monolayer took advantage of
the known properties of the Hg(0)/Hg(II) redox couple on gold electrode (Figure S1 in Supporting
Information). In aqueous media, its potential lies at about 0.1 V against the Ag|AgCl|3M LiCl
27
electrode. It was considered important to perform the electrochemical measurements on the very
same flat plates covered with a thin layer of Au that were used for other measurements, such as IR
2
and XPS. A special electrochemical cell that permits a measurement on a selected 4.02 mm area
of the plate was designed for the purpose (Supporting Information, Figures S2 and S3). We verified
that results obtained in this fashion were similar to those obtained on Au disc electrodes embedded
in a glass capillary. The measured complex electrochemical impedance is shown in the Supporting
Information (Figure S4).
28
Successive scans by differential pulse polarography (DPP) yielded mercury oxidation
signals of gradually decreasing size (Figure S5), suggesting that elemental mercury was being
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 45
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. Aqueous solution of 2 mM K [Fe(CN) ] and 0.1 M KNO . A: Cyclic voltammetry and B: local complex
3 6 3
2
impedance plot of a 4.02 mm spot on a Au plate (virgin, black; with an adsorbed layer of C H HgOTs, red; and with
18 37
an adsorbed layer of C H SH, magenta). C: Cyclic voltammetry and D: local complex impedance plot of an Au plate
18 37
(virgin, black; with an adsorbed layer of C H HgOTs, red; and after electrochemical Hg stripping, green).
18 37
removed. A determination averaged over four different spots on a plate yielded a peak potential
2
value of +135 ±5 mV and peak current of 11.9 ± 4.4 nA at an area of 4.02 mm .
When potential steps from -0.1 to +0.3 V were applied to an Au surface covered with a
monolayerproduced with C HgOTs andimmersedinto anaqueous0.1M KNO solution (immersed
18 3
2
area 0.32 cm ), transient currents i decayed in time (t) as 1/t (Figure S6), demonstrating that the
12
ACS Paragon Plus Environment

Page 13 of 45
Journal of the American Chemical Society
1
2
3
current transient is not controlled by diffusion from the bulk of the solution and that the decay
4
5
6
corresponds to the oxidation of surface confined Hg atoms. A similar hyperbolic i-t dependence for
7
8
29
an adsorbed species has been described before. Numerical integration yielded a charge of 48 mC,
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
which contains faradaic (Q ) and double layer (Q ) contributions. Repetition of the experiment with
F C
a clean Au surface of the same area yielded Q = 12.9 mC. The charging current decayed
C
exponentially in time (Figure S6). From Q = 35.1 mC, the surface concentration of Hg atoms is
F
-9 2 2
1.14×10 mol/cm , or 1 Hg atom per 15 Å .
Electrochemical Stripping of Mercury. A suitable stripping potential of 0.3 V against the
Ag|AgCl|3M LiCl electrode was estimated from the DPP results. When it was applied to the whole
electrode for 1 h, the DPP peak of mercury disappeared, but XPS indicated that 65 % of the
elemental Hg remained on the surface. Repeated scans of E from -1.0 V to +1.4 V lead to a gradual
decrease of the double layer capacitance C between -1 V and +1 V to a very low value, suggestive
of 100% coverage by a fairly compact layer (Figure 6). The decrease is especially apparent at
positive potentials, whereas at negative potentials the initial capacitance is already smaller. The
capacitance values show that outside this range C H HgOTs is still adsorbed to some extent, in a
18 37
form of a less compact adsorbate.
The C - E plots contain a sharp AC peak at 0.5 V against the Ag|AgCl|3M LiCl electrode,
absent at early and late times. Complex impedance at the potential of the peak maximum as a
function of the applied frequency (Figure S4) is a semicircle connected to a mass transfer line,
implying that the removal of mercury is a kinetically controlled process (for numerical fitting, see
Supporting Information).
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 45
1
2
3
4
5
6
7
8
9
The results obtained on gold electrodes are quite unlike those obtained on a glassy carbon
electrode, which we expect to be inert (Figure S7). On carbon the admittance data show no
maximum and no substantial time evolution of the adsorption zones upon repeated voltage scans.
Admittance at positive potentials never reaches very low and potential-independent values.
The results shown in Figure 6 suggested an improved protocol for possible electrochemical
removal of Hg from the surface. In a series of experiments Au plates were immersed for 2 h in a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
C H HgOTs solution. The plate was then transferred to an electrochemical cell filled with 0.1 M
18 37
TBAPF and a potential of 1.0 V was applied for 1 h. XPS analysis indicated that the content of Hg
6
on the surface dropped to about 20% of the original value.
Figure 6. Imaginary component of Au electrode admittance in 0.2 mM
C H HgOTs in acetonitrile and 0.1 M n-Bu PF . AC frequency: 64 Hz;
18 37
4
6
amplitude: 5 mV. Voltage scan start: -1.0 V.
An even higher degree of Hg removal was achieved by initially immersing the Au plate in
-5
the electrochemical cell containing 10 M C H HgOTs and in 0.1 M TBAPF in THF and allowing
18 37
6
14
ACS Paragon Plus Environment

Page 15 of 45
Journal of the American Chemical Society
1
2
3
the monolayer formation to take place at 1.0 V for a period of 4 h. The plate was then rinsed with
4
5
6
THF and kept at a potential of 1.0 V in a 0.1 M solution of TBAPF6 in acetonitrile or in a 0.1 M
7
8
9
solution of KF in water for approximately 1 h. This procedure yielded an adsorbed layer containing
no detectable Hg (at most 5% of the original content) but still containing the C(1s) peak,
demonstrating the presence of the monolayer (Figure 7).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Properties of Hg-Free Monolayers. The electrochemical removal of Hg on gold originally
cleaned by a hydrogen flame decreased the ellipsometric thickness of monolayer from ~13 Å to 9
Å and reduced the contact angle to ~53°, suggesting the presence of large islands of pure gold. The
C1s XPS data (Figure 7) confirmed that the same amount of carbon was still present after
electrochemical stripping.
IR spectra taken after the electrochemical Hg stripping resemble those taken before the
stripping. Stretching vibrations are observed at 2853, í (CH ); 2871, í (CH ); 2925, í (CH ); and
s 2 s 3 2
as
-1 -1
2958, í (CH ); all in cm , with í (CH ) and í (CH ) 3 cm above their respective locations before
2
as
3
s
2
as
-1
the stripping (Figure 2B). Bending vibrations are seen at 1378, 1418 and 1468 cm .
Figure 7. XPS in the Hg 4f (panel A) and C 1s (panel B) regions of a Au substrate treated
with C H HgOTs before (red) and after (black) electrochemical stripping of Hg.
18 37
15
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 45
1
2
3
4
5
6
Discussion
7
8
Pristine Monolayers. The formation of a monolayer on a gold surface upon treatment with
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a solution of C H HgOTs under ambient conditions proceeds well but more slowly than the
18 37
formation of a monolayer from C H SH (Figure 1). The ellipsometric thickness is smaller, only
18 37
about ~13 instead of ~21 Å, and the monolayers are slightly less resistant to most chemical reagents
tested and more resistant to oxidants. The ~13 Å thickness of the alkylmercury-based monolayer
derived from ellipsometry is probably fairly reliable since the refractive index of the monolayer is
not likely to be affected much by the presence of Hg on the gold surface (but might be affected by
a reduced packing density of the chains).
The ~13 Å thickness contrasts with the ellipsometric thickness of the various trialkylstannyl-
1
derived monolayers, which was only 6 - 7 Å, and suggests a considerably denser coverage for
alkylmercury-derived monolayers. However, this thickness is clearly smaller than that of
alkanethiol-based monolayers. This would be expected if the alkyl chains are attached directly to
the surface gold atoms and if some of the surface area is blocked by Hg atoms, reducing the
coverage by alkyl chains. From contact angle goniometry, the alkylmercury-derived monolayer
surface is found to be hydrophobic, consistent with methyl termination.
From both IR and XPS data, the tosylate leaving group is absent and alkyl chains are present.
Their presence is also apparent in UPS and IR. The observed IR frequencies are not compatible with
an all-anti alkyl conformation and suggest the presence of disordered chains. Disorder is also
suggested by the poor electrode blocking properties of the monolayer.
16
ACS Paragon Plus Environment

Page 17 of 45
Journal of the American Chemical Society
1
2
3
From both XPS and DPP data, the surface contains a large amount of elemental mercury
4
5
6
before the electrochemical stripping. The position of the narrow XPS peaks in the spectrum exactly
7
8
24,25
matches the binding energy of elemental mercury
and gives no indication, within the 0.1 eV
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
resolution for peak centers, that Hg is present in more than one kind of environment. The
polarographic peak corresponds to Hg(0)/Hg(II) oxidation. The areas per mercury atom deduced
2
from the two methods are similar, ~13 (XPS) and 15 (DPP) Å /Hg atom. The Q = 35.1 mC value
F
probably underestimates the amount of Hg present, since some of the surface area is blocked by the
2
adsorbed layer, and the electrochemical value of 15 Å /Hg atom is an upper limit, whereas the XPS
2
value of 13 Å /Hg atom is more reliable. To derive the latter value, equal attenuation of Hg 4f and
Au 4f photoelectrons by the alkyl overlayer is assumed and the attenuation from the Hg layer is
explicitly included. An Hg adlayer on Au has precedent from a previous study of underpotential
30
deposition.
2
The surface area per alkyl chain deduced from the C 1s XPS signal is ~36 Å /chain. This
is about 66% more than the area per sulfur atom and thus per alkyl chain in a close-packed
2
31
dodecanethiolate monolayer, which is 21.6 Å /S atom, and roughly 3 times the surface area per Hg
atom. Since small differences in molecular coverage can strongly affect the chain tilt angle, the
reduced thickness observed by ellipsometry is easily explained by the lower packing density, under
the assumption that the C H chains are attached directly to Au.
18 37
2
2
The factor of ~3 observed between the surface areas of ~13 Å /Hg atom and 36 Å /alkyl
chain precludes the possibilitythat each single mercuryatom carries an alkyl chain. Although about
1/3 of the Hg atoms could be attached to an alkyl chain based on the evidence discussed so far, it
would be unlikely for XPS signals from Hg atoms in C H Hg to be identical with the signals from
18 37
17
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 45
1
2
3
4
5
6
7
8
9
elemental Hg within the 0.1 eV resolution. It appears most probable that during the self-assembly
some of the alkyl chains shift from Hg to Au whereas the remaining alkyls are desorbed. We have
no spectroscopic evidence for the mode of attachment of the alkyl group to the surface gold atoms,
and the simplest such attachment would be through a single C-Au bond. An analogous alkyl shift
from an Sn atom to a Au surface, inducing C-Au bond formation, has been proposed recently to
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
8,9
explain observations of enhanced single-molecule conductivity in alkyl chains.
The compactness of an adsorbed monolayer is usually tested by its inhibiting properties.
3-/4-
Simple electron transfer reactions, such as the one-electron exchange in [Fe(CN) ] , often show
6
a complete elimination of the faradaic current. This is not true in the present case and the monolayer
produced by deposition of C H HgOTs only blocks ~35% of the surface. The anodic and cathodic
18 37
peak potentials are practically unchanged. This observation is in agreement with standard
32,33
assumptions concerning the influence of uncharged surfactants on the electron transfer rate.
According to a simplified model the effective electron transfer rate constant k is a linear
eff
function of the degree of coverage è
k = k (1 – è) + k è
eff 0 1
where k is the rate constant on the free surface and k is the rate constant on a fully covered surface.
0 1
Our observation suggests that prior to the electrochemical stripping treatment the Hg atoms form
islands of amalgamated Au surface on which k is as high as on bare Au and the voltammetric peak
0
separation is unchanged. Alkyl chains block the electron transfer and reduce the active electrode
area, causing a decrease of voltammetric peak heights (Figure 5). The rate constant k is very likely
1
18
ACS Paragon Plus Environment

Page 19 of 45
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
small, since otherwise a larger separation of voltammetric peaks would be detected. The removal
of Hg from the adsorbed layer will not have much effect on the total active area but may lead to a
more evenlydistributed adsorbate and to a higher porosityof the surface layer. The voltammograms
show an almost negligible inhibition, but simulations show that for all heterogeneous rate constants
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
-1
higherthan0.01 cm.s identical voltammograms areobtained. The application of a fastertechnique,
impedance spectroscopy, revealed that the electron transfer is inhibited even after the removal of
Hg from the adsorbed layer (Figure 5, C and D).
Similarly as the detailed nature of the attachment of the alkyl chains to the surface gold
atoms, the mechanism by which they are transferred from the Hg atom to the Au atom or atoms is
not known. There seems to be little if any precedent for the process. A distant analogy might be
drawn to gas-phase S 2 reactions in which a nuclophile such as NH displaces a metal atom M (M
N
3
+ + 34
= Zn, Cd, or Hg) from a carbon in the ion CH M to yield CH NH . In our case, the Au surface
3
3
3
would act as a nucleophile and the displacement would occur on the á carbon of C H HgOTs,
18 37
+ ! + !
which would act as C H Hg TsO to yield Au -C H + Hg + TsO . Instances in which a gold
18 37
18 37
n
35
atom or a small gold cluster
36,37
located on a surface acts as an electron donor are known.
Monolayers after Oxidative Mercury Stripping. The imaginary component of electrode
admittance is proportional to the double layer capacitanceC and sensitivelyresponds to the structure
of the interface. Comparison of repeated scans from -1.0 V to +1.4 V measured without and with
C H HgOTs in the bulk of the solution shows a remarkable gradual decrease of the double layer
18 37
capacitance C in the presence of the adsorbate (Figure 6) as layer adsorbed to the surface becomes
progressively more compact, until at the end the capacitance is very low and almost potential
19
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 45
1
2
3
38,39
independent. This is characteristic of a compact layer.
The slow initial growth of the peak at
4
5
6
7
8
0.5 V in Figure 7 is attributed to a relatively slow formation of the initial monolayer.
The attribution of the admittance maximum at 0.5 V to electrochemical stripping of Hg is
supported by the shape of the complex impedance plot (Figure S4), which implies a kinetically
controlled process, most likely electron transfer from Hg hindered by the adsorbed layer and shifted
to a higher potential relative to DPP on a clean surface. The disappearance of the maximum at the
end of the stripping experiment suggests that the removal of Hg ceases.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
IR and XPS show that after stripping, alkyl chains are still adsorbed to the surface, although
the mercury is desorbed. At this point, most or all alkyls must be directly connected to the gold
surface atoms, even those previouslyattached to the Au surface through Hg atoms, if there were any.
The ellipsometric thickness is reduced upon stripping, which might be in part due to the loss of an
important constituent of the original monolayer, elemental Hg, but in part also due to the loss of
some fraction of the alkyls. From the present data, it is however not at all obvious that the alkyls
are attached to the surface through a single C-Au bond.
Conclusions
In conclusion, we have demonstrated that the treatment of a gold surface with a solution of
C H HgOTs produces sturdy monolayers containing elemental mercury and disordered alkyl
18 37
chains, attached directly to gold surface atoms in an unknown mode. Oxidative stripping removes
the mercury from the monolayer, and when it is started concurrently with the deposition, it leaves
no detectable mercury on the surface. The result is a layer of n-octadecyl chains attached to an
otherwise clean gold surface.
20
ACS Paragon Plus Environment

Page 21 of 45
Journal of the American Chemical Society
1
2
3
Experimental Methods
4
5
6
7
8
WARNING. Organomercury compounds are highly toxic and require the use of protective
garments and extreme care in handling. Contact with skin must be carefully avoided. Waste needs
to be treated separately from ordinary laboratory waste and disposed of properly.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Materials. Octadecylmercury(II) bromide (1). Magnesium (100 mg, 4 mmol) was heated
to 340 °C under reduced pressure for several minutes. After cooling to room temperature by flow
of argon a small piece of iodine was added. The flask was heated again until it turned purple (250
°C). Octadecyl bromide (1.0 g, 3 mmol) and tetrahydrofuran (5 mL) were added. The solution was
refluxed for 2 h, cooled, filtred through glass wool and slowly added to a solution of HgBr (1.1 g,
2
3 mmol) in dry diethyl ether. After stirring overnight at room temperature under argon atmosphere
chloroform (50 mL) was added and the solution was filtered over cellite. The solid was washed with
chloroform and combined filtrates were evaporated to give crude product (1.6 g), which was then
40
recrystallized twice from hexanes. Yield 0.9 g (56%). M.p. 112.0 - 114.0 °C (lit. 110 - 111 °C).
1
Elemental analysis: calcd.: 40.49 % C, 6.98 % H; found: 40.16 % C, 6.82 % H. H NMR (CDCl )
3
d 0.88 (t, J = 7 Hz, CH , 3H), 1.21 - 1.39 (m, s CH , 28H), 1.57 (br s, CH , 2H), 1.78 (quintet, J =
3 2 2
7 Hz, CH , 2H), 2.14 (d,t J
2
= 190 Hz, J
= 7 Hz, CH Hg, 2H).
2
HCHg
HCCH
Octadecylmercury (II) p-toluenesulfonate (2). 1 (680 mg, 1.28 mmol) was dissolved in
ethanol (99.5%, 100 mL) and silver p-toluene sulfonate (356 mg, 1.28 mmol) was added in one
portion. The reaction mixture was stirred at room temperature overnight and filtered over cellite.
The solid was washed with ethanol (99.5%, 30 mL). The solvent was evaporated, and the product
was recrystallized from petroleum ether. Yield 480 mg (60 %). M.p. 74 °C. Elemental analysis:
calcd.: 48.02 % C, 7.09 % H, 5.13 % S, 32.08 % Hg; found 47.90 % C, 6.98 % H, 5.10 % S, 32.42
21
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 45
1
2
3
1
% Hg. H NMR (CDCl ) d 0.88 (t, J = 7 Hz, CH , 3H), 1.20 - 1.42 (m, s CH , 30H), 1.70 (quintet,
2
3
3
4
5
6
J = 8 Hz, CH , 2H), 2.23 (d,t J
2
= 213 Hz, J
= 8 Hz, CH Hg, 2H), 2.29 (s, CH –Ar, 3H), 7.13,
2 3
HCHg
HCCH
7
8
7.15, 7.48, 750, AB system, Ar).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Adsorbed Monolayer Formation. Glass substrates coated with 2000 Å of gold were
purchased from Platypus Technologies and were cleaned in one of two ways. In the first procedure,
the gold substrates were cleaned by immersion in piranha solution (3:1 sulfuric acid / hydrogen
peroxide) at 90 °C for 30 s, rinsed with copious amounts of 18.2 MÙ H O and absolute ethanol, and
2
dried under a stream of nitrogen. In the second procedure, substrates were flame annealed prior to
use with a hydrogen torch for ca 30 s. Samples for XPS and STM measurements were prepared on
gold on mica. Atomically flat Au(111) surfaces were obtained by flame annealing of the substrates.
-5
Monolayers of 1-octadecanethiol were prepared byimmersion of a gold substrate in a 1 × 10
M solution in absolute ethanol for ~2 h. Octadecylmercury tosylate monolayers were formed by
-3
-5
immersing the gold substrates in a 1 × 10 M or 1 × 10 M solution of a octadecylmercury(II)
tosylate in dryTHF for ~2 h. After removal from solution, the gold substrates were rinsed three times
with THF and dried under a stream of nitrogen prior to analysis.
Electrochemistry. Electrochemical measurements were performed using an Autolab
PGSTAT30 potentiostat/galvanostat equippedwith a frequencyresponse module (Metrohm Autolab,
The Netherlands). A four-electrode electrochemical cell for measurements of local properties of Au
plates was constructed (Supporting Information, Figures S2 and S3). The design was aimed at
measurements of voltammetryand electrochemical impedance spectroscopyin a small area of a plate
coated with a gold layer. The cell can be moved to different locations on the Au plate for improved
2
statistics. The working electrode was Au(111) deposited on a glass substrate, with a 0.64 cm total
22
ACS Paragon Plus Environment

Page 23 of 45
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
area. It was mechanically attached to a microscope table equipped with two manipulating screws
enabling the choice of a spot on Au(111) to be electrochemically tested. The cell containing the
solution was a glass tube mounted on a holder, which can be vertically moved with a fine thread
screw. The lower part of the cell tube was fitted to a piece of Teflon with an O-ring. Two reference
electrode wires were mounted through the Teflon body. The DC reference electrode was an Ag|AgCl
wire. The high-frequency reference electrode was a Pt wire coupled to the DC reference electrode
via a 0.1 ìF condenser. The distance between the two reference electrodes and the working electrode
was approximately 1 mm. The auxiliary electrode was a Pt net mounted on a Pt wire and immersed
in the cell tube through its upper end, which was closed by a septum. The area of the net was
sufficiently larger than the working electrode area to meet the requirement for obtaining a cell
impedance that corresponds to the impedance of the working electrode. Prior to measurements the
cell tube assembly was moved down, gently pressed against the Au plate, and filled with the test
solution. The entire setup showed no liquid leaks and a good reproducibility of the tested area. The
adsorption and the compactness of adsorbed layers were followed by an established method using
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4-/3-
inhibition of electron transfer of the couple [Fe(CN) ] in aqueous 0.1 M KCl or KNO . Oxygen
6
3
was removed from the solution bypassing a stream of argon. Cyclic voltammetrywas measured with
a scan rate of the applied DC potential of 0.007 V/s. Inhibition or desorption was followed by
4-/3-
changes of the kinetic parameters of the redox exchange of [Fe(CN) ] . Rate constants were
6
41
evaluated by digital simulation using finite difference elements. Electrochemical impedance
spectroscopy was measured in the range 100 kHz to 0.8 Hz using the amplitude of the AC signal 5
mV. The size of data collection was 120 points with a logarithmic distribution over the whole
frequency range. Faradaic charge transfer resistance, reflecting quantitatively inhibition efficiency,
23
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 45
1
2
3
4
5
6
7
8
9
was evaluated bya simulation program Nova 1.8 supplied bythe manufacturer of the electrochemical
instrument. Layers showing a high degree of inhibition were evaluated by voltammetry, whereas at
low inhibition the impedance method proved to be more sensitive. The time dependence of the
double layer capacity C of Au and glassy carbon (GCE) electrodes was measured using small
electrode discs (0.5 mm diameter) sealed in a glass capillary. Phase-sensitive AC polarography used
a sine wave of 64 Hz frequency and 10 mV amplitude. The change of C–E dependence during the
adsorption process was measured in 0.1 M tetrabutylammonium hexafluorophosphate in acetonitrile.
All solution components were dried.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Electrochemical stripping of Hg was performed by two different procedures with similar
results. In the first one, samples were prepared by immersion of an Au plate that had been treated with
C H HgOTsintoanelectrochemicalcellcontaining0.1Mtetrabutylammoniumhexafluorophosphate
18 37
in acetonitrile and a specified potential was applied. In the other procedure, monolayers on a gold
substrate were prepared by immersion of a clean Au plate into an electrochemical cell containing 0.1
M tetrabutylammonium hexafluorophosphate or aqueous 0.1 M KF and various amounts (0.05 to 0.2
mM) of C H HgOTs in acetonitrile. Again, the potential was either applied or repeatedly scanned
18 37
within selected limits.
Ellipsometry. All measurements were made using a variable-angle Stokes ellipsometer
(Gaertner Scientific) with a 633 nm HeNe laser with the incident angle adjusted to 70°. Optical
constants of the gold substrates were taken for all freshly cleaned substrates. An index of refraction
of 1.47 was assumed for the films. Ellipsometry measurements were taken at a minimum of five
different areas on each sample.
24
ACS Paragon Plus Environment

Page 25 of 45
Journal of the American Chemical Society
1
2
3
Contact Angle Measurement. A static contact angle for 18.2 MÙ H O was found with a
2
4
5
6
CAM101 instrument (KSV Instruments) using a 1 to 2 ìL drop of water. Measurements were taken
7
8
at a minimum of five different areas for each sample.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
-1
Infrared Spectroscopy. FTIR-ATR spectra (800 scans, 4 cm resolution) were recorded
using a Nicolet 6700 FT-IR spectrometer (Thermo Electron Corporation) with a liquid-N2- cooled
-1
MCT detector in the range of 650-4000 cm . The data were collected with p-polarized light at 45°
incidenceusing a Seagull variable-angle accessory(HarrickScientificInc.)and a Ge hemisphere(12.5
mm diameter). Prior to each measurement, the Ge crystal was cleaned with ethanol and a reference
spectrum of the crystal in contact with air was measured.
X-ray Photoelectron Spectroscopy. XPS was recorded with a Kratos Axis Ultra instrument
using a monochromated Al K emission source (1486.6 eV) operated at 15 kV and 180 W. The
á
photoelectrons were collected by the spectrometer in normal emission geometry. With an X-ray
monochromator and a pass energy of 40 eV for the analyzer, an instrumental energy resolution of ~0.5
eV was achieved. The energy scale is referenced to the Au 4f7/2 line at a binding energy (BE) of 84.0
eV. For all samples, a survey spectrum and high resolution spectra of the Hg 4d, Hg 4f, S 2p, C 1s,
O 1s, and Au 4f regions were acquired. The spectra were fitted using a linear background for all
elements except Au, where Shirleybackground was used. Voigtfunctions employing a 50:50 Lorentz-
Gaussian ratio, including a slight asymmetryfactor (instrumental) were used as fit functions. The line
shape parameters were determined by least squares fitting to carbon or sulfur core level lines from
known reference samples.
UV Photoelectron Spectroscopy. UPS was obtained using a helium UV lamp as a source.
The gas pressure in the lamp was adjusted in such a way that He I (hí = 21.2 eV) and He II (hí = 40.8
25
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 26 of 45
1
2
3
4
5
6
7
8
9
eV) light was emitted in a ratio of approximately 4:1. The light was incident at an angle of 55° from
the sample normal and the photoelectrons were collected by an energy dispersive hemispherical
analyzer at a takeoff angle of 90°. The analyzer was set to a pass energy of 5 eV, providing an
instrumental resolution of about 0.14 eV. Binding energies were referenced to the Fermi level of a
clean, argon ion-etched Au surface and defined as positive for occupied states below the Fermi level.
Monolayer Stability. The IR spectrum of each self-assembled monolayer was recorded. For
stability measurements, the gold slide was immersed for 20 - 24 h at room temperature in dry CH Cl ,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
2
wet CH Cl , n-hexane, ethanol, water, 0.1 M H SO , 0.1 M NaOH, 1 mM KMnO , 30% H O , or 10
2 2 2 4 2
4
2
mM NaBH , removed from solution, rinsed thoroughly with either an appropriate solvent (CH Cl ,
4 2 2
hexane, or ethanol) or with copious amounts of 18.2 MÙ H O and absolute ethanol in the case of water
2
solutions, and dried under a stream of nitrogen before an IR spectrum was recorded. The samples
were also exposed to the ambient laboratory atmosphere for 7 days and then rinsed with absolute
ethanol and dried before the IR spectrum was measured. The thermal desorption of the monolayers
was followed by keeping the monolayer under argon for 1 h each at 80, 140, and finally 200 °C. Prior
to the IR measurements, the gold slides were rinsed with absolute ethanol and dried under a stream
of nitrogen.
Acknowledgement. The research leading to these results has received funding from the
European Research Council under the European Community’s Seventh Framework Programme
(FP7/2007-2013 FUNMOL 213382 and ERC Grant Agreement 227756). Initial efforts were
supported by the Grant Agency of the Czech Republic (203/07/1619 and 203/09/0705) and the
Institute of Organic Chemistry and Biochemistry (RVO: 61388963). The authors thank Dr. P. Saidl
26
ACS Paragon Plus Environment

Page 27 of 45
Journal of the American Chemical Society
1
2
3
for performing some of the XPS measurements and to P. Poncar for constructing a special
4
5
6
electrochemical cell.
7
8
Supporting Information. Figures S1 - S9 and a description of the electrochemical cell and
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
equivalent circuit. This material is available free of charge via the Internet at http://pubs.acs.org.
References
1.
Khobragade, D.; Stensrud, E. S.; Mucha, M.; Smith, J.; Pohl, R.; Stibor, I.; Michl, J.
Langmuir 2010, 26, 8483.
2.
3.
Michl, J.; Stibor, I., CZ Patent No. 302441 B6, April 8, 2011.
Zheng, X.; Mulcahy, M. E.; Horinek, D.; Galeotti, F.; Magnera, T. F.; Michl, J. J. Am. Chem.
Soc. 2004, 126, 4540.
4.
Mulcahy, M. E.; Magnera, T. F.; Michl, J. J. Phys. Chem. C. 2009, 113, 20698.
5.
Mulcahy, M. E.; Bastl, Z.; Stensrud, K. F.; Magnera, T. F.; Michl, J. Phys. Chem. C. 2010,
114, 14050.
6.
7.
Lee, T. R.; Whitesides, G. M. Acc. Chem. Res. 1992, 25, 266.
Kaletová, E.; Kohutová, A.; Si, Z.; Kaleta, J.; Scholz, F.; Hajduch, J.; Pospisil, L.; von
Wrochem, F.; Bastl, Z.; Michl, J. Book of Abstracts, 4th EuCheMS Congress, Aug. 26-30,
2012, Prague, Czech Republic, P-0888.
8.
Cheng, Z. L.; Skouta, R.; Vázquez, H.; Widawsky, J. R.; Schneebeli, S.; Chen, W.;
Hybertsen, M. S.; Breslow, R.; Venkataraman, L. Nat. Nanotechnol. 2011, 6, 353.
9.
Chen, W.; Widawsky, J. R.; Vázquez, H.; Schneebeli, S. T.; Hybertsen, M. S.; Breslow, R.;
Venkataraman, L. J. Am. Chem. Soc. 2011, 133, 17160.
27
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 28 of 45
1
2
3
10.
Scholz, F.; Kaletová, E.; Stensrud, E. S.; Ford, W. E.; Mucha, M.; Stibor, I.; Michl, J.; von
4
5
6
Wrochem, F., submitted for publication.
7
8
11.
Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo R. G.; Whitesides, G. M. Chem. Rev. 2005,
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
105, 1103.
12.
13.
14.
Schreiber, F. Prog. Surf. Sci. 2000, 65, 151.
Ulman, A. Chem. Rev. 1996, 96, 1533.
Slowinski, K.; Chamberlain, R. V.; Bilewicz R.; Majda, M. J. Am. Chem. Soc. 1996, 118,
4709.
15.
16.
17.
Nuzzo, R. G.; Dubois L. H.; Allara, D. L. J. Am. Chem. Soc. 1990, 112, 558.
Chidsey, C. E. D.; Loiacono, D. N. Langmuir 1990, 6, 682.
Bain, C. D.; Troughton, E. B.; Tao, Y. T.; Evall, J.; Whitesides, G. M.; Nuzzo, R. G. J. Am.
Chem. Soc. 1989, 111, 321.
18.
19.
Protsailo, L. V.; Fawcett, W. R.; Russell, D.; Meyer, R. L. Langmuir 2002, 18, 9342.
von Wrochem, F.; Gao, D.; Scholz, F.; Nothofer, H. G.; Nelles, G.; Wessels, J. M. Nature
Nanotech. 2010, 5, 618.
20.
21.
22.
Colorado, R.; Villazana, R. J.; Lee, T. R. Langmuir 1998, 14, 6337.
Joseph, Y.; Guse, B.; Nelles, G.; Chem. Mater. 2009, 21, 1670.
Willey, T. M.; Vance, A. L.; van Buuren, T.; Bostedt, C.; Terminello, L. J.; Fadley, C. S.
Surf. Sci. 2005, 576, 188.
23.
24.
Huang, F. K.; Horton, R. C.; Myles, D. C. Jr.; Garrell, R. L. Langmuir 1998, 14, 4802.
Svenson, S.; Martensson, N.; Basilier, E.; Malquist, P. A.; Gelius, U.; Siegbahn, K. J. El.
Spectrosc. Rel. Phenomena 1976, 9, 51.
25.
Morris, T.; Szulczewski, G. Langmuir 2002, 18, 2260.
28
ACS Paragon Plus Environment

Page 29 of 45
Journal of the American Chemical Society
1
2
3
4
5
26.
27.
Chidsey, C. E. D.; Liu, G. Y.; Rowntree, P.; Scoles, G. J. Chem. Phys. 1989, 91, 4421.
Heyrovský, J.; Kùta, J. Principles of Polarography, Publishing House of the Czechoslovak
6
7
8
Academy of Sciences, Prague 1965, p. 538.
9
nd
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
28.
Bard A. J.; Faulkner L. R. Electrochemical Methods. Fundamentals and Applications, 2
edition, J. Wiley, New York 2001, p. 286-293.
29.
30.
31.
32.
33.
34.
35.
Koper, M. T. M. Z. Phys. Chem. 2003, 217, 547.
Li J.; Abruna, H. D. J. Phys. Chem. B 1997, 101, 2907.
Chidsey, C. E. D.; Liu, G. Y.; Rowntree, P.; Scoles, G. J. Chem. Phys. 1989, 91, 4421.
Schmidt, R. W.; Reilley, C. N. J. Am. Chem. Soc. 1958, 80, 2087.
Weber, J.; Koutecký, J.; Koryta, J. Z. Elektrochem. 1959, 63, 583.
Kretschmer, R.; Schlangen, M.; Schwarz, H. Angew. Chem. Int. Ed. 2011, 50, 5387
Risse, T.; Shaikhutdinov, S.; Nilius, N.; Sterrer, M.; Freund, H.-J. Acc. Chem. Res. 2008, 41,
949, and references therein.
36.
Brown, M. A.; Ringleb, F.; Fujimori, Y.; Sterrer, M.; Freund, H.-J.; Preda, G.; Pacchioni,
G. J. Phys. Chem. C 2011, 115, 10114.
37.
Brown, M. A.; Fujimori, Y.; Ringleb, F.; Shao, X.; Stavale, F.; Nilius, N.; Sterrer, M.;
Freund, H.-J. J. Amer. Chem. Soc. 2011, 133, 10668.
38.
39.
Sridharan, R.; de Levie, R. J. Phys. Chem. 1982, 86, 4489.
de Levie, R. Advances in Electrochemistry and Electrochemical Engineering, J. Wiley: New
York, 1986; p 1.
40.
41.
Meals, R. N. J. Org. Chem. 1944, 9, 211.
BASi DigiSim Simulation Software for Cyclic Voltammetry (Bioanalytical Systems, Inc.,
West Lafayette, IN, U.S.A.).
29
ACS Paragon Plus Environment