Effect of Large Electrolyte Anions on the Sequential Oxidations of Bis(fulvalene)diiron Attached to Glassy Carbon by an Ethynyl Linkage

Article abstract of DOI:10.1021/acs.langmuir.7b03649

Two ethynyl-derivatized isomers of bis(fulvalene)diiron (BFD, 1,1′-biferrocenylene) were prepared and covalently attached to glassy carbon electrodes through their ethynyl group by three different electrode modification methods. Cyclic voltammetry and square wave (SW) voltammetry were used to characterize surface coverages of 1.4-5.5 × 10^-10 mol cm^-2, the higher of these corresponding to roughly a monolayer, based on computation of an idealized close-packing structure for ethynylbis(fulvalene)diiron (E-BFD) on a solid surface. In a dichloromethane solution containing a smaller electrolyte anion such as [PF₆]^- or [ClO₄]^-, the E-BFD-modified electrodes exhibited two quasi-Nernstian one-electron oxidations. In contrast, the current for the second oxidation process, [E-BFD]^+/2+, was diminished in electrolytes containing one of the large fluoroaryl borate anions, [B(C₆F₅)₄]^- or [B(C₆H₃(CF₃)₂)₄]^-. The effect was enhanced for electrodes having higher surface coverages being probed at shorter voltammetric time scales. SW voltammetry showed that the diminished currents for [E-BFD]^+/2+ in large-anion electrolytes are not caused by slow electron transfer. Rather, they are attributed to mixed diffusivity of the counter-anions at the electrode/solution interface, as [E-BFD]⁺ and the anion form the optimum (lowest-energy) configuration of a 1:1 ion pair. The interior transport of the anion required to reach this configuration may be sterically encumbered, accounting for the diminished charge transfer observed with electrolytes containing large anions.

Full text of DOI:10.1021/acs.langmuir.7b03649

Subscriber access provided by READING UNIV
Article
The Effect of Large Electrolyte Anions on the Sequential Oxidations of
Bis(fulvalene)diiron Attached to Glassy Carbon by an Ethynyl Linkage
Matthew V. Sheridan, Paul Gamm, Severin T. Schneebeli,
Rochus Breuer, Michael Schmittel, and William E. Geiger
Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.7b03649 • Publication Date (Web): 08 Jan 2018
Downloaded from http://pubs.acs.org on January 8, 2018
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.
Langmuir 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 42
Langmuir
1
2
3
4
5
6
BFD_revised[1].docx
7
8
9
The Effect of Large Electrolyte Anions on the Sequential Oxidations of
Bis(fulvalene)diiron Attached to Glassy Carbon by an Ethynyl Linkage
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
Matthew V. Sheridan,a Paul Gamm,a Severin T. Schneebeli,a
Rochus Breuer,b Michael Schmittel,b and William E. Geigera *
a Department of Chemistry, University of Vermont, Burlington, Vermont 05405 USA
b Center of Microꢀ and Nanochemistry and Engineering, Organische Chemie 1, Universität
Siegen, AdolfꢀReichweinꢀStrasse 2, Dꢀ57068, Siegen, Germany.
* email: william.geiger@uvm.edu
ABSTRACT
Two ethynylꢀderivatized isomers of bis(fulvalene)diiron (BFD, 1,1’ꢀbiferrocenylene) were
prepared and covalently attached to glassy carbon electrodes through their ethynyl group by
three different electrode modification methods. Cyclic voltammetry and square wave (SW)
voltammetry were used to characterize surface coverages of 1.4ꢀ5.5 x 10ꢀ10 mol cmꢀ2, the higher
of these corresponding to roughly a monolayer, based on computation of an idealized closeꢀ
packing structure for ethynylbis(fulvalene)diiron (EꢀBFD) on a solid surface. In a
dichloromethane solution containing a smaller electrolyte anion such as [PF6]ꢀ or [ClO4]ꢀ, the Eꢀ
BFDꢀmodified electrodes exhibited two quasiꢀNernstian oneꢀelectron oxidations. In contrast, the
current for the second oxidation process, [EꢀBFD]+/2+, was diminished in electrolytes containing
one of the large fluoroaryl borate anions, [B(C6F5)4]ꢀ or [B(C6H3(CF3)2)4]ꢀ. The effect was
enhanced for electrodes having higher surface coverages being probed at shorter voltammetric
time scales. SW voltammetry showed that the diminished currents for [EꢀBFD]+/2+ in largeꢀanion
electrolytes are not caused by slow electronꢀtransfer. Rather, they are attributed to mixed
diffusivity of the counterꢀanions at the electrode/solution interface, as [EꢀBFD]+ and the anion
form the optimum (lowest energy) configuration of a 1:1 ion pair. The interior transport of the
anion required to reach this configuration may be sterically encumbered, accounting for the
diminished charge transfer observed with electrolytes containing large anions.
1
ACS Paragon Plus Environment

Langmuir
Page 2 of 42
1
2
3
4
I. INTRODUCTION
5
6
7
8
The interfacial microstructure and microenvironment of a surfaceꢀattached redoxꢀactive
molecule may have a significant effect on the kinetics and thermodynamics of its electronꢀ
transfer reactions.1ꢀ4 If at least one member of the redox couple is charged, the nature of the
supporting electrolyte salt can influence the electrochemical responses owing to the
thermodynamics and kinetics of ionꢀpairing interactions. A number of studies have appeared
addressing the effects of changes in electrolyte anion on the voltammetric response of ferroceneꢀ
terminated selfꢀassembled monolayers.5ꢀ18 We are interested in how variations in the electrolyte
might affect the redox properties of surfaceꢀbound species which undergo multiple stepwise oneꢀ
electron oxidations. Here, we focus on the effects of the size of the electrolyte anion on the
voltammetry of two isomers of ethynylꢀderivatized bis(fulvalene)diiron compounds, αΕꢀBFD
and βΕꢀBFD, that have been covalently attached to glassy carbon electrodes.
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
Fe
Fe
Fe
Fe
Fe
Fe
BFD
EꢀBFD
EꢀBFD
Reported originally in 1969,19,20 the parent complex bis(fulvalene)diiron (BFD)21 attracted
inquiries into the molecular and electronic structures of its oxidized forms.22ꢀ24 It was shown that
removal of one or two electrons from BFD results in wellꢀdefined changes in molecular
structure, including a modest reduction of the metalꢀmetal distance and a slight bending of the
cyclopentadienyl rings towards the metals in a plane parallel to the CꢀC linkage between the
fiveꢀmembered rings. 22ꢀ24 Relevant to the present work is the fact that that the formally mixedꢀ
valent monocation, BFD+, has been shown by both experiment25,26 and calculations22 to be
electronically fully delocalized. Thus, the positive charge in the BFD monocation is equally
distributed over the two ferrocenyl units, the SOMO of which is predominantly situated in the
d
x2ꢀy2 orbitals of the two metals.22 The monocations of monoꢀsubstituted BFD complexes retain
the property of full electronic delocalization.26
2
ACS Paragon Plus Environment

Page 3 of 42
Langmuir
1
2
3
4
5
6
The homogeneous electrochemistry of BFD and some of its monoꢀsubstituted derivatives
have also been thoroughly described.26ꢀ35 This family of compounds undergoes a sequence of
two quasiꢀNernstian oneꢀelectron oxidations (Eq 1) in both nonaqueous27ꢀ34 and aqueous35 media.
7
8
9
ꢀeꢀ
ꢀeꢀ
BFD
BFD+
BFD2+
(1)
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
E1/2(1)
E1/2(2)
The fact that the electronic properties of BFD are so well defined, and that it proceeds through
its threeꢀmember electronꢀtransfer series with minimal changes in molecular structure, make it a
promising model system for the study of multiꢀelectronꢀtransfer processes of surfaceꢀbound
species. Improved synthetic methods now provide access to monoꢀsubstituted BFD derivatives
having potential moleculeꢀtoꢀsurface linkages.23 To this end, selfꢀassembled monolayers (SAMs)
of BFD bonded to gold through an alkanethiol linkage have recently been reported.35 These
BFDꢀterminated systems exhibit the expected BFD0/+/2+ electrochemistry in CH2Cl2/[NBu4][A]
(A = [PF6]ꢀ, [BF4]ꢀ) and also show impressive resiliency of the BFD0/+ couple in aqueous media
over a full range of pH values.35
The present paper provides the first report of a BFDꢀmodified electrode in which the
molecule is covalently attached to the surface by a short linker and describes how changes in the
electrolyte medium affects its electrochemical properties. An advantage of this system is that
there is already available an extensive study of the influence of medium effects on the
homogeneous electrochemistry of bis(fulvalene)dinickel, Fv2Ni2,36 the cations of which have the
same basic molecular structures and identical chargeꢀtoꢀsize ratios as the BFD cations.
Owing to the popularity of ferroceneꢀterminated structures and the utility of the
ferrocene/ferrocenium couple in basic investigations of chemicallyꢀmodified electrodes (CMEs),
the ionꢀpairing literature in this area is dominated by studies of variations in the electrolyte
anion.5,6,8,10,11,14ꢀ18,37ꢀ40 Electrolyte changes have been shown to cause alterations not only in the
potential of the ferrocene/ferrocenium couple, but also in the shapes and peak currents of their
cyclic voltammetry (CV) waves. Previous studies have employed electrolytes containing halides
or small complex anions (e.g., [ClO4]ꢀ, [BF4]ꢀ, [PF6]ꢀ), the effects of which generally correlate
with the hydrophilicity of the anions, not surprising in that most of the tests were conducted in
aqueous media. Comparatively less attention has been paid to anion effects in nonaqueous
solutions, and to the best of our knowledge, no studies have been reported involving the types of
large, weaklyꢀcoordinating, anions (WCAs) (e.g., [B(C6F5)4]ꢀ and [B(C6H3(CF3)2)4]ꢀ) which are
3
ACS Paragon Plus Environment

Langmuir
Page 4 of 42
1
2
3
4
5
6
7
8
9
known to ionꢀpair less strongly than their more traditional smaller counterparts.36, 41ꢀ43 Spaceꢀ
filling models of four representative anions are shown in Figure 1, along with their computed
anionic volumes. The present study focuses primarily on a comparison of [PF6]ꢀ vs [B(C6F5)4]ꢀ ꢀ
based electrolytes in dichloromethane using [NBu4]+ as the electrolyte counterion.
As previously described,44ꢀ46 the ethynylꢀoxidation approach to electrode modification may
follow one of the three tactics outlined in Scheme 1, each of which is designed to provide a
reactive terminal ethynyl group, either as a radical (tactics IꢀLi and IꢀH) or as a formal biradical
(tactic IIꢀH). Reaction of the radical with a surface atom produces an ethynylꢀlinked molecular
moiety R that is strongly (most likely covalently)ꢀbound to the surface. In tactic IꢀLi, the Hꢀ
terminated ethynyl group is first activated by lithiation, after which anodic oxidation is employed
to produce the desired radical owing to the rapid loss of Li+ from the radical cation [RC≡CLi]+.
Tactics IꢀH and IIꢀH circumvent the ethynyl lithiation step by employing direct anodic oxidation
of the molecule to induce deprotonation at its Hꢀterminated ethynyl group. If the ethynyl group is
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
Scheme 1.
oxidized more readily than the R group (tactic IꢀH), the desired radical may be produced by loss
of proton from the highly acidified ethynyl group in the radical cation [RC≡CH]+. If the nonꢀ
ethynyl ‘core’ of the molecule, R, is oxidized more readily, loss of the ethynyl proton may
require removal of two electrons from RC≡CH (tactic IIꢀH). We use the phrase “direct oxidation
methods” for both IꢀH and IIꢀH in order to separate them conceptually from the chemical
activation method IꢀLi. A more detailed comparison of the three methods is available.46
The approaches used for anodic electrografting of the ethynylꢀBFD system onto glassy carbon
are shown below, where BFD* stands for the Fe2(C10H8)(C10H7) “core” of the molecule. Eq 1
describes deposition method IꢀLi for this system. Eq 2 is a multiꢀelectron variation of process
4
ACS Paragon Plus Environment

Page 5 of 42
Langmuir
1
2
3
4
5
6
IIꢀH, in which anodic oxidation of the ethynyl group is preceded by the two oneꢀelectron
oxidations of the BFD molecular core. Both deposition methods were used successfully in the
electroꢀgrafting of these systems.
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
The surfaceꢀdeposited ethynylꢀBFD systems were characterized by both CV and by two
pulsed methods less frequently applied to CMEs: differential pulse voltammetry (DPV) and
squareꢀwave voltammetry (SWV). The limitations of CV in characterizing nonꢀideally shaped
surface waves, especially those observed in the presence of appreciable nonꢀfaradaic charging
currents, have been noted, and both doubleꢀpotential step chronoamperometry and staircase
cyclic voltammetry have been evaluated as alternatives to CV analysis.17, 47, 48 However, when
two closelyꢀspaced waves are involved, and especially when the electronꢀtransfer kinetics of the
surface species become a point of inquiry, the DPV and SWV methods have advantages, owing
to their pseudoꢀGaussian wave shapes, their discrimination against charging currents, and, in the
case of SWV, its sensitivity to chargeꢀtransfer kinetics.49,50 The literature offers papers on both
the theory and application of SWV for surfaceꢀbound species.51ꢀ59
II. EXPERIMENTAL
General Experiments were carried out under nitrogen, either in a drybox or under Schlenk
conditions. Solvents were purified by passing them through aluminaꢀbased solvent systems under
argon. Glassware was dried overnight at 110 °C. Tetrabutylammonium ions of [PF6]ꢀ, [ClO4]ꢀ,
and [CF3SO3]ꢀ were used as received. [NBu4][B(C6F5)4] was prepared as previously described41
and recrystallized from dichloromethane/diethyl ether. [NBu4][B(C6H3(CF3)2)4] was prepared by
metathesis of [NBu4]Br and Na[B(C6H3(CF3)2)4] (Boulder Scientific Corporation) in methanol
and purified by passing it through a small column of neutral activated alumina, using
dichloromethane as the eluent. Voltammetric measurements were carried out using a PARC
Model 273A potentiostat, interfaced to a personal computer. The experimental reference
electrode was a silver chlorideꢀcoated silver wire in the electrolyte solution, separated from the
working compartment by a fineꢀporosity glass frit. However, all potentials in this paper are
5
ACS Paragon Plus Environment

Langmuir
Page 6 of 42
1
2
3
4
5
6
7
8
9
referenced to the ferrocene/ferrocenium couple. The required potential conversions were carried
out by first measuring the potential of the analyte against the in-situ decamethylferrocene0/+
potential, and then converting that value to the ferrocene0/+ reference value by subtracting 0.56 V
for [PF6]ꢀ or [ClO4]ꢀ solutions, or 0.61 V for [B(C6F5)4]ꢀ or [B(C6H3(CF3)2)4]ꢀ solutions in
dichloromethane. Decamethylferrocene was used as the in-situ standard owing to the fact that the
ferrocene oxidation occurs at a potential between the first and second oxidations of BFD and its
derivatives. Glassy carbon (gc) electrodes (Bioanalytical Systems) were polished before use with
Bühler diamond paste, then sonicated and finally dried under antechamber vacuum. Additional
details about the electrochemical procedures are available in earlier papers.46,60
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
Electrografting procedures Electrodeꢀmodification based on oxidation of lithioꢀactivated EꢀBFD
compounds (method IꢀLi) followed the ex-situ method described elsewhere in detail.46 The
essential aspects are that a 4 mM solution of the EꢀBFD compound in dry THF was allowed to
react for about one minute at 273 K with a slight excess of nꢀBuLi in hexanes. A sample was
then transferred via syringe to a solution of THF containing either 0.1 M [NBu4][PF6] or 0.05 M
[NBu4][B(C6H3(CF3)2)4]. The resulting electrolyte solution contained 3ꢀ4 mM LiEꢀBFD, which
was then subjected to anodic cyclic voltammetry scans which resulted in grafting of the EꢀBFD
onto the electrode surface. An alternative approach to electrografting employing a direct
oxidation procedure (method IIꢀH vide ante) which utilized 3ꢀ5 mM solutions of EꢀBFD in
dichloromethane containing either 0.1 M [NBu4][PF6] or 0.05 M [NBu4] [B(C6H3(CF3)2)4] and
involved CV scans past the third anodic oxidation of the compound (vide infra). After
modification, the working electrode was removed from solution, washed with dichloromethane,
sonicated in acetone for 2 min, washed with dichloromethane, and vacuum dried. We note also
that the soꢀcalled “spontaneous” deposition method, in which the lithiated ethynyl compound in
THF is simply allowed to react with an unconnected electrode for 10 min,61 also produced a Eꢀ
BFDꢀmodified carbon electrode, albeit with a somewhat diminished surface coverage.
Preparation of ethynylꢀBFD compounds The synthesis of αꢀ and βꢀethynylbis(fulvalene)diiron
(αEꢀBFD and βEꢀBFD) was achieved according to a modified literature procedure.62 The
structural identification of the two constitutional isomers was derived from characteristic
multiplets of their fulvalene ring protons in the 1H NMR spectra as described in an earlier
6
ACS Paragon Plus Environment

Page 7 of 42
Langmuir
1
2
3
4
5
publication.23 The starting compounds, 2ꢀ and 3ꢀacetylꢀ1,1´ꢀbiferrocenylene (αꢀacetylꢀBFD and
βꢀacetylꢀBFD), were prepared according to a previously published procedure.23
6
7
8
9
αꢀEthynylbis(fulvalene)diiron (αEꢀBFD). Lithium diisopropylamide (LDA) (2.20 mmol, 3.01
eq.) in THF (1.10 mL) was added dropwise to a solution of 2ꢀacetylꢀ1,1´ꢀbiferrocenylene (300
mg, 0.73 mmol) in THF (60 mL) at –78 °C under nitrogen. Stirring was continued for 1 hour at –
78 °C, subsequently diethylchlorophosphate (177 mg, 1.03 mmol, 1.40 eq.) was added. The
reaction temperature was maintained at –78 °C for one hour. Within the following 180 min the
reaction mixture was allowed to warm up to room temperature. After cooling down to –78 °C a
second batch of LDA (5.00 mmol, 6.83 eq.) in THF (2.5 mL) was added dropwise. Immediately
after completion of the addition the reaction mixture was warmed up to room temperature within
150 min and finally quenched with water. The separated aqueous layer was extracted with 100
mL of CH2Cl2. The combined organic layers were washed with brine (3 x 150 mL) and finally
dried over Na2SO4. After evaporation of the solvent under reduced pressure, the solid was
purified by flash column chromatography (silica: dichloromethane/nꢀhexane, 50/50 v/v; Rf =
0.70) furnishing additionally 50% of the starting material (2ꢀacetylꢀ1,1´ꢀbiferrocenylene: Rf =
0.29 in dichloromethane). Yield: 90.0 mg (31%). IR (KBr, cm−1): 3287 (≡C–H, s), 3095 (C–HAr,
m), 2103 (C≡C, m), 1407 (w), 1380 (m), 1259 (m), 1028 (s), 847 (m), 808 (s), 659 (m), 584 (m),
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
493 (s). 1H NMR (600 MHz, C2D2Cl4): = 6.12 (dt, 3J = 2 Hz, 4J = 1 Hz, 1H, 10ꢀH), 5.64 (dt, 3J
δ
= 2 Hz, 4J = 1 Hz, 1H, 10´ꢀH) 5.44 (dt, 3J = 2 Hz, 4J = 1 Hz, 1H, 7ꢀH), 5.33 (dd, 3J = 3 Hz, 4J = 1
Hz, 1H, 5ꢀH), 5.32 (dt, 3J = 2 Hz, 4J = 1 Hz, 2H, 7´ꢀH, 5´ꢀH), 5.14 (dt, 3J = 2 Hz, 4J = 1 Hz, 1H,
2´ꢀH), 4.17 (dd, 3J = 3 Hz, 4J = 1 Hz, 1H, 3ꢀH), 4.05 (td, 3J = 2 Hz, 4J = 1 Hz, 1H, 3`ꢀH), 3.93
(td, 3J = 2 Hz, 4J = 1 Hz, 2H, 8ꢀH, 9´ꢀH), 3.85 (dd(t), 3J = 3 Hz, 1H, 4ꢀH), 3.83 (td, 3J = 2 Hz, 4J
= 1 Hz, 1H, 9ꢀH), 3.82 (td, 3J = 2 Hz, 4J = 1 Hz, 1H, 4´ꢀH), 3.80 (td, 3J = 2 Hz, 4J = 1 Hz, 1H, 8´ꢀ
H), 3.35 (s, 1H, 2´´ꢀH). 13C NMR (100 MHz, C2D2Cl4):
δ = 97.5, 96.8, 94.4, 94.0, 84.7, 77.2,
76.4, 71.4, 70.4, 70.2, 70.1, 69.8, 69.7, 69.0, 67.7, 67.1 (2C), 66.9, 66.8, 66.5, 66.4, 64.7. Anal.
Calcd for C22H16Fe2: C, 67.40; H, 4.11. Found: C, 67.48; H, 4.12.
βꢀEthynylbis(fulvalene)diiron (βEꢀBFD). Under nitrogen, LDA (4.00 mmol, 2.05 eq.) in THF
(2.00 mL) was added dropwise to a solution of 3ꢀacetylꢀ1,1´ꢀbiferrocenylene (800 mg, 1.95
mmol) in THF (160 mL) at –78 °C. Stirring was continued for 1 hour at –78 °C and subsequently
diethylchlorophosphate (468 mg, 2.71 mmol, 1.39 eq.) was added. The reaction temperature was
7
ACS Paragon Plus Environment

Langmuir
Page 8 of 42
1
2
3
4
5
6
7
8
9
maintained at –78 °C for 1 additional hour. Within the following 180 minutes the reaction
mixture was allowed to warm up to room temperature. After cooling down to –78 °C a second
batch of LDA (10.0 mmol, 5.13 eq.) in THF (5.0 mL) was added dropwise. Immediately after
completion of the addition the reaction mixture was slowly warmed up to room temperature
(over 150 minutes) and finally quenched with water. The separated aqueous layer was extracted
with 150 mL of CH2Cl2. The combined organic layers were washed with brine (3 x 200 mL) and
finally dried over Na2SO4. After evaporation of the solvent under reduced pressure, the solid was
purified by flash column chromatography (silica: dichloromethane/nꢀhexane, 50/50 v/v; Rf =
0.65). Yield: 583 mg (76%). IR (KBr, cm−1): 3297 (≡C–H, s), 3100 (C–HAr, m), 2109 (C≡C, m),
1268 (m), 1171 (w), 1035 (s), 936 (w), 847 (s), 808 (s), 657 (m), 584 (m), 534 (m), 489 (s). 1H
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
NMR (600 MHz, C2D2Cl4): δ
= 5.71 (dd(t), 4J = 1 Hz, 1H, 2ꢀH), 5.65 (dt, 3J = 2 Hz, 4J = 1 Hz,
1H, 2´ꢀH), 5.45 (dt, 3J = 2 Hz, 4J = 1 Hz, 1H, 10´ꢀH), 5.37 (dt, 3J = 2 Hz, 4J = 1 Hz, 1H, 10ꢀH),
5.36 (m, 2H, 5ꢀH, 7ꢀH), 5.33 (dt, 3J = 2 Hz, 4J = 1 Hz, 1H, 5´ꢀH), 5.29 (dt, 3J = 2 Hz, 4J = 1 Hz,
1H, 7´ꢀH), 4.12 (td, 3J = 2 Hz, 4J = 1 Hz, 1H, 4´ꢀH), 4.10 (dd, 3J = 3 Hz, 4J = 1 Hz, 1H, 4ꢀH),
3.87 (td, 3J = 2 Hz, 4J = 1 Hz, 1H, 9´ꢀH), 3.86 (td, 3J = 2 Hz, 4J = 1 Hz, 1H, 8ꢀH), 3.80 (td, 3J = 2
Hz, 4J = 1 Hz, 1H, 9ꢀH), 3.78 (td, 3J = 2 Hz, 4J = 1 Hz, 1H, 8´ꢀH), 3.61 (td, 3J = 2 Hz, 4J = 1 Hz,
1H, 3`ꢀH), 2.70 (s, 1H, 2´´ꢀH). 13C NMR (100 MHz, C2D2Cl4):
δ = 96.5, 96.2, 94.2, 94.0, 82.6,
74.8, 72.8, 72.2, 70.6, 69.8, 69.7, 69.5, 69.4, 69.3, 69.1, 68.9, 68.1, 67.0 (2 C), 66.6, 66.5, 62.5.
Anal. Calcd for C22H16Fe2: C, 67.40; H, 4.11. Found: C, 67.60; H, 4.03.
Computations. To model the packing of βEꢀBFD on a flat carbon surface, we turned to quantum
mechanical calculations using semiempirical NDDO63 theory. First, a 3x3 array of βEꢀBFD
molecules was built as the input structure for optimization. This input array of molecules was
constructed with all acetylene substituents of the βEꢀBFD molecules aligned in the same plane,
vertical to the xꢀy plane. Next, the array of βEꢀBFD molecules was minimized with the PM7
Hamiltonian64 with the MOPAC201665 software package. In order to constrain the molecules to
the same surface, all coordinates, except the zꢀcoordinates of the terminal hydrogens of the
acetylene groups, were allowed to optimize.
8
ACS Paragon Plus Environment

Page 9 of 42
Langmuir
1
2
3
4
III. RESULTS AND DISCUSSION
5
6
III. A. Solution voltammetry of EꢀBFD and LiEꢀBFD
7
8
9
In homogeneous solution, CVs of the α and β isomers of ethynylbis(fulvalene)diiron (Eꢀ
BFD) are similar in CH2Cl2/0.1 M [NBu4][PF6] (Figure 2). In addition to the two quasiꢀNernstian
oneꢀelectron oxidations at E1/2(1) = ꢀ0.29 V and E1/2(2) = 0.38 V (Eqs 3 and 4) which are
inherent to BFD itself 22ꢀ35, a third anodic feature is seen at Epa(3) = 1.60 V (ν = 0.4 V sꢀ1), which
we ascribe to removal of an electron from the ethynyl moiety (Eq 5), the consequence of which
is loss of the ethynyl proton (Eq 6), as shown previously to occur for the oxidation of
ethynylferrocene45,46 [in Eq 6, EꢀBFD has been rewritten as BFD*ꢀC≡CꢀH in order to emphasize
the loss of the terminal hydrogen atom]. The sharp cathodic feature seen in the reverse scan in
Figure 2 at Epc ≈ 1.40 V is ascribed to the surfaceꢀconfined oxidation product. As discussed in
section III.B, scanning through the third anodic wave of either αEꢀBFD or βEꢀBFD produces a
glassy carbon surface to which ethynylꢀBFD species are strongly (most likely covalently)
bonded.
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
EꢀBFD
EꢀBFD+ + eꢀ
EꢀBFD2+ + eꢀ
EꢀBFD3+ + eꢀ
E1/2(1)
E1/2(2)
Epa(3)
(3)
(4)
(5)
(6)
EꢀBFD+
EꢀBFD2+
[BFD*ꢀC≡CꢀH]3+
[BFD*ꢀC≡C]2+ + H+
Surface deposition of the ethynylꢀBFD compound is also observed when its ethynyl group is
anodically oxidized in CH2Cl2/0.05 M [NBu4][B(C6F5)4]. However, in this case the third anodic
wave of the ethynylꢀBFD compound is not directly observed in the CV scan in homogeneous
solution (Figure S1). The visual absence of the ethynyl oxidation wave under these conditions is
explained by the fact that the [B(C6F5)4]ꢀ anion is both easier to oxidize and less strongly ionꢀ
pairing than the [PF6]ꢀ anion:36,42 the first property makes the anodic “window” of the electrolyte
less positive, and the latter causes the potential for oxidation of the EꢀBFD dication, Epa(3), to
move positive. We will return to these points when describing the preparation of the chemicallyꢀ
modified surfaces.
9
ACS Paragon Plus Environment

Langmuir
Page 10 of 42
1
2
3
4
5
6
7
8
9
Owing to the fact that the two reversible BFDꢀcentered oxidations provide a critical basis for
voltammetric characterization of EꢀBFDꢀmodified electrodes, it was important to examine how
changes in the electrolyte medium influence the potential separation, ꢀE1/2 (= E1/2(2) ꢀ E1/2(1)), of
the two oxidations. ꢀE1/2 values were determined for both [PF6]ꢀ and ][B(C6F5)4]ꢀ ꢀbased
electrolytes in three different solvents, namely the two relatively low polarity solvents
dichloromethane ( strong acceptor) and THF (strong donor), along with the more polar solvent
nitromethane. The measured values of E1/2(1) and E1/2(2) are collected in Table 1 (for
dichloromethane) and Table 2 (for THF and nitromethane), and selected ꢀE1/2 values are given in
Table 3. In agreement with the more comprehensive study of medium effects for the oxidation of
the nickel analogue Fv2Ni2 36, ꢀE1/2 values for ethynylꢀBFD oxidations are quite large under all
the conditions that were employed (≥ 560 mV). Furthermore, they increase when TFAB is used
in place of [PF6]ꢀ by the amounts [ ꢀꢀE1/2 (Table 3)] of ~ 330 mV in dichloromethane and 160
mV in THF. The value of ꢀꢀE1/2 falls to 20 mV in nitromethane, in which ionꢀpairing effects are
diminished. These trends indicate that the electronic effects accompanying the oxidation of
ethynylꢀBFD are very similar to those of both BFD itself and bis(fulvalene)dinickel. We
conclude that EꢀBFD+, like the parent BFD+, has its positive charge equally distributed on the
two “ferrocenyl” halves of the molecule, a model that is consistent with spectroscopic and
calculational results on other monosubstituted BFD+ systems.66
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
When one equivalent of nꢀBuLi is added to a THF solution of either EꢀBFD isomer, a
chemically irreversible anodic feature appears at a less positive potential than E1/2(1), owing to
oxidation of the lithioethynyl complex, LiEꢀBFD, formed in solution (Eq 7).67 As the potential
scan is continued positive, the reversible EꢀBFD 0/+ couple is observed (Figure 3). As explained
BFD*ꢀC≡CꢀH + nꢀBuLi
BFD*ꢀC≡CꢀLi + nꢀbutane
(7)
previously for lithioethynylferrocene,44,46 the initial oneꢀelectron oxidation of the lithiated
complex involves rapid loss of lithium ion (Eq 8), giving the neutral radical, in this case BFD*ꢀ
C≡C•, which either bonds to the electrode surface or abstracts a hydrogen atom from solvent.
BFD*ꢀC≡CꢀLi
BFD*ꢀC≡C• ꢀ eꢀ ꢀ Li+
(8)
The latter reaction reproduces the original terminal ethynyl complex, EꢀBFD, (Eq 9) in the
electrode reaction layer, and is responsible for the reversible wave at E1/2(1) in Figure 3. The Epa
10
ACS Paragon Plus Environment

Page 11 of 42
Langmuir
1
2
3
4
BFD*ꢀC≡C• + solvent
BFD*ꢀC≡CH + side product
(9)
5
6
7
8
value of ꢀ0.36 V for the oxidation of the α isomer of LiEꢀBFD is 170 mV negative of the E1/2(1)
value of αΕ−ΒFD, similar to the shift reported for the oxidation of lithioethynylferrocene
compared to ethynylferrocene.44,46 The return scan in Figure 3 thus contains two overlapped
cathodic features, one for reduction of homogeneous [αEꢀBFD]+ and the other for reduction of
any surfaceꢀbound cation.
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
III.B. Anodicallyꢀinduced electrografting of EꢀBFD onto electrode surfaces
III.B.1. Oxidation after Lithiation (Scheme 1, method IꢀLi).
CV scans to 0.1 V at a gc electrode in a 4 mM solution of LiEꢀBFD (Figure 3) resulted in an
EꢀBFDꢀmodified electrode. When the scanned electrode was removed, washed, sonicated, and
transferred to a clean electrolyte solution, two reversible BFD oxidation waves were observed
(Figure 4) having peak currents which increased in proportion to scan rate.68 The surface waves
were generally wellꢀshaped, but their full widths at half maximum (FWHM) varied markedly
with changes in the electrolyte anion, as will be discussed. The surface coverage, Γ, (determined
by integration of the current/time curve for the first anodic wave) increased progressively with
the number of scans of the LiEꢀBFD solution, going from about Γ = 2 x 10ꢀ10 mol cmꢀ2 for a
single scan to a maximum of Γ = 5.5 x 10ꢀ10 mol cmꢀ2 after 10 scans.
The extent of surface coverage and chemical layering in modified electrodes is often
discussed in terms of the apparent number of monolayers deposited on the surface, assuming an
idealized packing geometry. A simple drawing (Scheme 2) of EꢀBFD molecules linked to a
surface through their ethynyl groups suggests that an EꢀBFD monolayer might occupy a surface
area not unlike that of thioalkylꢀlinked ferrocenes on gold, generally accepted to be about 4.5 x
10ꢀ10 mol cmꢀ2.69
11
ACS Paragon Plus Environment

Langmuir
Page 12 of 42
1
2
3
4
5
6
7
8
Scheme 2. Comparative sketches of monolayers of EꢀBFD on carbon and thioalkylferrocene on
gold.
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
Fe
Fe
Fe
Fe
Fe
Fe
S
S
E-BFD on carbon
thioalkylferrocene on Au
This model was verified by calculating the optimized structure of an array of nine EꢀBFD
molecules fixed to a single plane through ethynylꢀtoꢀsurface bonds. A side view of the computed
structure is shown in Figure 5 (a), and a top view using a spaceꢀfilling representation is shown in
Figure 5 (b). Based on the minimum separations computed for this molecular array, the
maximum concentration of a closelyꢀpacked monolayer of EꢀBFD is approximately 5 x 10ꢀ10 mol
cmꢀ2. Although one must be cognizant of the wellꢀknown irregularities of glassy carbon surfaces,
this value gives us a starting point for considering the ways in which the surfaceꢀbound species
may interact with each other and offers an idea about the amount of open space available for
solvent and counterions. The latter becomes important when accounting for experimental
variations in the current passed for second oxidation of the surfaceꢀbound species. Based on the
estimated value of 5 x 10ꢀ10 mol cmꢀ2 for a monolayer of EꢀBFD, the deposition methods
described in the present paper do not appear to give rise to significant multiꢀlayering effects,
such as that reported for other radicalꢀbased electrografting processes,70 including that of
lithioethynylferrocene on glassy carbon.46
III.B.2. Direct oxidation method (scheme 1, method IIꢀH). Both EꢀBFD isomers were also
grafted onto glassy carbon electrodes by the direct anodic oxidation approach in
dichloromethane. In [PF6]ꢀ electrolytes, this method relies on scanning into the third (ethynylꢀ
12
ACS Paragon Plus Environment

Page 13 of 42
Langmuir
1
2
3
4
5
6
7
8
9
based) oxidation wave of EꢀBFD (Figure 2). One or two scans into the rising portion of this wave
(reversing at 1.5 V), followed by the usual workup procedure, afforded electrodes with wellꢀ
formed EꢀBFD surface waves very similar to those described above for the oxidation method
B.1. Interestingly, similar surface coverages were obtained from EꢀBFD solutions in
dichloromethane containing [NBu4][B(C6H5)4] as the electrolyte, even though the CV grafting
scans had to be extended into the anodic background to reach a potential at which the ethynyl
group would be oxidized. Thus, eight scans to 2.3 V for a 1 mM solution of αEꢀBFD in this
medium reproducibly gave a modified electrode having a surface coverage of about 3.8 x 10ꢀ10
mol cmꢀ2.
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
As will be shown in section III.D, pulsed methods were of value in probing the voltammetric
responses of the two EꢀBFD oxidation waves relative to each other. Their SW voltammograms at
frequencies of 10 Hz to120 Hz were very similar when the modified electrode was scanned in an
electrolyte containing a traditional anion (either [PF6]ꢀ or [ClO4]ꢀ , Figure 6). The FWHM of the
second wave was usually somewhat larger than that of the first (200 mV vs 135 mV in Figure 6),
but the peak currents were generally the same, within about a 10% variation. As will be
discussed below, this result changed dramatically if the SWV scans were recorded when the
supporting electrolyte was [NBu4][B(C6H5)4] or [NBu4][B(C6H3(CF3)2)4].
DPV scans of the chemicallyꢀmodified electrodes showed two anodic waves of similar peak
height in all the electrolyte solutions (Figure S2) and were used to determine the E1/2 values of
the second oxidation process in the largerꢀanion electrolytes.
III.C. Chemical (“Spontaneous”) grafting of EꢀBFD onto glassy carbon
A nonꢀelectrochemical (i.e. “spontaneous”) approach to grafting, described earlier for
lithioethynylferrocene,61 proved successful for the α isomer of EꢀBFD, as well. In this process,
an unconnected gc electrode was immersed in a 4 mM solution of LiEꢀBFD in THF for 10 min,
then removed, sonicated, washed, and finally transferred to a fresh electrolyte solution. CVs of
the modified electrode (Figure 7) at different scan rates confirmed the presence of surfaceꢀbound
αEꢀBFD and indicated a below ꢀmonolayer surface coverage of 1.4 x 10ꢀ10 mol cmꢀ2.
III.D. Effects of electrolyte anion on voltammetry of surfaceꢀbound species.
13
ACS Paragon Plus Environment

Langmuir
Page 14 of 42
1
2
3
4
5
6
7
8
9
III.D.1 Redox Potentials. For electrolytes containing the smaller anions ([PF6]ꢀ, [BF4]ꢀ,
[ClO4]ꢀ, and [CF3SO3]ꢀ), the formal potentials of surfaceꢀbound EꢀBFD were virtually unchanged
from the values measured in homogeneous solution. The ꢀE1/2 values (= E1/2(2) ꢀ E1/2(1)) of Eꢀ
BFD fell in the range 570ꢀ630 mV in dichloromethane/[NBu4][PF6], with only a 20ꢀ30 mV
difference between the surfaceꢀbound and homogeneous samples (see second column in Table
4). Owing to the lower ionꢀpairing strengths of the large anions ([B(C6F5)4]ꢀ and
[B(C6H3(CF3)2)4]ꢀ)42,71 ꢀE1/2 values recorded in their solutions were greater (820ꢀ950 mV, third
column in Table 4) than those in [PF6]ꢀ. There was a somewhat greater difference in the ꢀE1/2
values when going from homogeneous to surfaceꢀbound species. The values of ꢀꢀE1/2 given in
the last column of Table 4 show, for example, that the difference in ꢀE1/2 for αEꢀBFD in the
surfaceꢀbound species is only 220 mV, compared to a difference of 340 mV for this compound in
solution. The diminished ꢀꢀE1/2 values arise largely from a positive shift of the surfaceꢀbound
E1/2(1) values, suggesting that the ionꢀpairing constant of 1:1 [EꢀBFD]+:[B(C6F5)4]ꢀ may be less
for the surfaceꢀbound system than for the homogeneous system.
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
D.2. Charges Passed. Drastic changes were seen in the shapes of waves and the amounts of
current passed for the surfaceꢀbound species when the electrolyte anion was changed from small
to large. When recorded in smallꢀanion electrolyte solutions, the CVs of EꢀBFDꢀmodified
electrodes are well representative of two highly reversible oxidations (solid, [PF6]ꢀ, and dashed,
[ClO4]ꢀ, black lines of Figure 8). Although the FWHM value of the second oxidation peak was
somewhat larger than that of the first (typically 200 mV vs 140 mV), integration of the charge
passed for the second oxidation (Q2) was always within 10% of that (Q1) passed for the first,
independent of the grafting method or the depth of surface coverage. However, there were
striking changes in the CV waves when the electrolyte anion was either [B(C6F5)4]ꢀ or
[B(C6H3(CF3)2)4]ꢀ. In those cases, the surface waves became somewhat broader and the charge
passed for the second oxidation was drastically lower than that passed for the first oxidation, the
latter being an effect that became exaggerated at higher surface coverages. Referring to the solid
gray line of Figure 8, although the first peak was broader in a [B(C6F5)4]ꢀ electrolyte (FWHM =
290 mV) than in [PF6]ꢀ or [ClO4]ꢀ (170 mV), it retained a symmetric shape and passed charge
that was reasonably close to the value of Q1 determined from scans in [ClO4]ꢀ (Γ = 3.8 x 10ꢀ10
mol cmꢀ2). The more dramatic effect of the [B(C6F5)4]ꢀ electrolyte was on the charged passed for
14
ACS Paragon Plus Environment

Page 15 of 42
Langmuir
1
2
3
4
5
6
7
8
9
the second oxidation process. Although it is difficult to calculate an accurate value Q2 from the
CV scan in [B(C6F5)4]ꢀ, Q2 is obviously much less than (perhaps only 20ꢀ25% of) Q1. At much
lower coverages, the relative current for the second oxidation increased, the charge Q2 becoming
roughly 60% of Q1 in [B(C6F5)4]ꢀ electrolyte when Γ = 1.4 x 10ꢀ10 mol cmꢀ2 (see Figure 7). In
more limited examinations of [B(C6H3(CF3)2)4]ꢀꢀbased electrolytes, the second oxidation seemed
to be even more strongly affected by the size of the anion. An example is given by the dashed
gray line in Figure 8, in which the second oxidation wave was barely visible.
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
Although squareꢀwave voltammograms do not offer a direct quantitation of the surface
coverage, they proved to be advantageous in evaluating the relative currents of the two oxidation
processes. As discussed in section III.B.2 and shown in both Figure 6 and in the solid black line
of Figure 9 (see also Figure S3), for both αEꢀBFD and βEꢀBFDꢀmodified electrodes, the net peak
current in a SW scan of the second oxidation wave in [PF6]ꢀ solutions, ꢀI2, remained essentially
the same as that of the first oxidation wave, ꢀI1, over the frequency range 10ꢀ60 Hz. In contrast,
the ratio of the peak currents, ꢀI2/ꢀI1, decreased markedly at higher SW frequencies for solutions
containing the large anions. For example, it was only 0.17 in [B(C6F5)4]ꢀ solutions at a SW
frequency of 60 Hz (see solid gray line in Figure 9). We will return to consideration of how the
relative peak heights change with SW frequency when discussing mechanistic aspects of the E.T.
reactions in section III.E.
To summarize the most important effects of counterꢀanions on the voltammetric responses for
both homogeneous solutions and chemically modified electrodes: 1) the ꢀE1/2 values for the two
EꢀBFD oxidations increase, as expected, in electrolytes containing larger, more weakly ionꢀ
pairing, anions. Slightly smaller increases in ꢀE1/2 are seen for surfaceꢀbound species; 2) the
surface charges passed in the first oxidation (EꢀBFD0/+) and the second oxidation (EꢀBFD+/2+) are
roughly equal when the electrolyte anion is [PF6]ꢀ or another small anion; 3) the surface charge
passed for the EꢀBFD+/2+ couple becomes significantly less, relative to the EꢀBFD0/+ couple, in a
largeꢀanion electrolyte. This effect becomes greater for electrodes having higher surface
coverages; 4) under largeꢀanion conditions, the ratio of the currents passed for the two redox
couples is sensitive to the inherent time scale of the voltammetric sampling method. Thus, when
a higher CV scan rate or SW frequency was employed, the current for the second (EꢀBFD+/2+
)
oxidation process was diminished compared to that of the first (EꢀBFD0/+) process. Further to
15
ACS Paragon Plus Environment

Langmuir
Page 16 of 42
1
2
3
4
5
6
7
8
9
this point, the peak height ratios were near unity in both [PF6]ꢀ and [B(C6F5)4]ꢀ solutions when
the scans were carried out using differential pulse voltammetry, which has an inherently longer
time scale than CV or SW voltammetry [Note that DPV was convenient for measuring accurate
E1/2(2) potentials in largeꢀanion media. For an example, see Figure S5]; 5) electrodes with
significantly lower surface coverages (e.g., ≤ 1.4 x 10ꢀ10 mol cm2, see section III.C.) had second
oxidation currents which were less sensitive to the size of the electrolyte anion (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
III.E. Model for the Largeꢀanion Effect
The most striking effect of the large electrolyte anion is in the diminution of the transient
current for the second oxidation of surfaceꢀbound EꢀBFD. A number of cases have been
previously reported in which changes in an electrolyte counterion resulted in diminished charge
transfer (sometimes termed “partial” oxidation or reduction) for the redoxꢀactive moiety on a
CME. For example, only oneꢀhalf of the available ferrocenes are electrochemically active for a
ferrocenylꢀterminated SAM when the counterion is the [ꢀOSO3]ꢀ group of a sodium nꢀalkyl
sulfate surfactant, CH3(CH2)nOSO3Na. The surfactant anions are thought to adsorb onto the
surface of the SAM by interdigitated ionꢀpairing interactions that have the effect of blocking
electronꢀtransfer to half of the surfaceꢀbound ferrocenes.16 Conceptually similar arguments have
been made to account for the lower currents for ferrocenylꢀterminated “click”ꢀattached
monolayers in roomꢀtemperature ionic liquids.18 A related investigation of binary SAMS having
a mixture of ferrocenylꢀ and alkylꢀend groups reported much lower anodic currents when the
[BF4]ꢀ anion in the liquid was replaced by the larger bis[(trifluormethyl)sulfonyl]amide anion.17
Diminished electroactivities attributed to ionꢀpairing effects have also been reported for
redoxꢀdoped polymer films, including electropolymerized viologens72, polypyrroles containing
redoxꢀactive octahedral transition metal complexes73, viologen siloxanes immobilized on indium
tin oxide74, ferrocenylꢀfunctionalized polypyrroles75, and electropolymerized iron
phenanthrolines.76 These and other papers77ꢀ81 addressing the transient vs steadyꢀstate currents for
CMEs have focused on the relative roles of ionꢀtransport in the film and electronꢀtransfer
hopping between redox sites on the surface layer. Lyon et al. have argued that siteꢀtoꢀsite
electron hopping is not likely to be slow in the top layer of a metallopolymer film.76 Similar
reasoning would suggest that slow electronꢀhopping is unlikely to be responsible for the
diminished transient currents in the EꢀBFD monolayer. This premise was probed by additional
16
ACS Paragon Plus Environment

Page 17 of 42
Langmuir
1
2
3
4
5
6
7
8
9
SWV measurements on the surfaceꢀbound EꢀBFD0/+ and EꢀBFD+/2+ couples. Square wave
voltammetry offers advantages over cyclic voltammetry for mechanistic analysis of electroactive
monolayers owing to its superior ability to discriminate between faradaic current and charging
current.50,57 SWV has, in fact, been effective for the determination of E.T. rates of surfaceꢀ
confined species.54, 55, 82ꢀ85 The SWV analysis required inspection of the forward and reverse
components of the SW currents (Ifwd and Irev, respectively) which are normally added to give the
customary ꢀI yꢀaxis of SWV plots such as those shown in Figures 6 and 9.50,57 The separation,
ꢀΕpk, between the peak potentials of the forward and reverse currents (Efwd and Erev,
respectively) is sensitive to the standard rate constant, ks, of the system,48,50,53,54,85ꢀ88 analogous
to how ꢀEp values are sensitive to that quantity in cyclic voltammetry scans.89
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 10 is representative of the data obtained for the forward and reverse components of
SW scans in either largeꢀanion or smallerꢀanion electrolytes. For SW frequencies from 40 Hz to
300 Hz, the values of ꢀEpk are never more than 54 mV for either the EꢀBFD0/+ or EꢀBFD+/2+
couple (Table S1 of supporting information), indicating that both E.T. reactions, in either
electrolyte medium, fall into the relatively fast quasiꢀreversible category,50,57,59 ks > 2 sꢀ1.,54 Thus,
the decreased anodic current for the second oxidation of surfaceꢀbound EꢀBFD cannot be
ascribed to a chargeꢀtransfer process for EꢀBFD+/2+ that is either inherently slow or subject to an
innerꢀsphere interaction between EꢀBFD+ and the counterꢀanion that would contribute to the
electronꢀtransfer barrier.
Rather, the diminution of transient current under certain conditions very likely arises from a
transport process that is sterically encumbered for large anions. One model for this effect
assumes that there is an optimum (i.e., lowest energy) structure for the 1:1 ion pair ([Eꢀ
BFD]+:[anion]ꢀ) (see Scheme 3). Diffusivity of the anion in and around the surface layer would
involve “normal” free diffusion as well as transport of an anion into the more interior optimum
ionꢀpairing position. The latter could limit the approach to steadyꢀstate if transport of the anion to
the interior position is sterically encumbered. This explanation has analogies to the mixed
diffusivity model described by Dalton and Murray as being responsible for diminished currents
in polyꢀviologen films.72 In this framework, the inner diffusivity may be considered to have a
migration component, being sensitive not only to size and spatial factors, but also to the
electrostatics of the charged surface once the [EꢀBFD]+ monocation is formed.
17
ACS Paragon Plus Environment

Langmuir
Page 18 of 42
1
2
3
4
5
6
Scheme 3. Schematic representations of an internal ionꢀtransport process which results (on the
right) in the thermodynamically most stable form of a 1:1 ion pair between EꢀBFD+ and a large
anion.
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
The time dependence of the relative contributions of the two types of diffusivity were probed
by measuring the effect of SW frequency on the ratio of peak currents, ꢀI2/ꢀI1, for the two
oxidation waves. For an EE system experiencing little or no counterion obstruction, the ꢀI2/ꢀI1
value is expected to be close to unity and essentially independent of SW frequency. This was,
indeed, observed for the EꢀBFD modified electrodes in all smallꢀanion electrolytes, but not in
those containing either of the large anions. Figure 11 illustrates this frequency effect for a
[B(C6F5)4]ꢀ solution, in which ꢀI2/ꢀI1 (the current of the second oxidation relative to the first) is
highly diminished when the SW frequency is increased from 20 Hz to 60 Hz.
Systematic measurements of current ratios [either ꢀI2/ꢀI1 or (I2/I1)fwd] as a function of SW
frequency were made on an electrode containing 2.0 x 10ꢀ10 mol cmꢀ2 αEꢀBFD. This degree of
surface coverage is on the lower end of the range,1.4ꢀ5.5 x 10ꢀ10 mol cmꢀ2, obtained under
different deposition conditions and clearly corresponds to only a partial monolayer of coverage.
As given in Table 5, the ratio remains close to unity in a [PF6]ꢀ or [ClO4]ꢀ solution, but falls
sharply for [B(C6F5)4]ꢀ, from a value of about 0.6 at 10 Hz to an apparent low limit of 0.22 at
18
ACS Paragon Plus Environment

Page 19 of 42
Langmuir
1
2
3
4
5
6
7
8
9
frequencies of 60 Hz and above. Low current ratios (down to 0.14) were also observed for
measurements made in a [B(C6H3(CF3)2)4]ꢀ electrolyte. Based on these data, the time frame of
the internal diffusivity effect may be estimated for this particular modified electrode. Using a
time parameter, tp, for the SW scan of tp = 1/2f, where f is the SW frequency in sꢀ1,50,90 the data
suggest that the internal ionꢀtransport process is insufficiently rapid to allow for the full the
oxidation of [EꢀBFD]+ to take place over 0.05 s when a [B(C6F5)4]ꢀ counterion is involved, and is
perhaps even slower for the somewhat larger [B(C6H3(CF3)2)4]ꢀ anion. In contrast, the ionꢀ
transport process is complete in less than 0.01 s when a small counterion is involved. It must be
kept in mind that these time estimates, while empirically useful, relate to the properties of ions of
specific size and charge at a particular solution/electrode interface. They are expected to be
highly sensitive to the degree and consistency of the surface coverage, which is more ambiguous
in the present case than in that of a wellꢀdefined selfꢀassembled monolayer. The large anions
employed in the present study should find utility for the study of counterꢀanion effects in SAMs
and other wellꢀdefined surfaces containing multiꢀpositive redox sites and products.
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
IV. CONCLUSIONS
Two isomers of ethynylꢀderivatized bis(fulvalene)diiron (EꢀBFD) were electrochemically
grafted onto glassy carbon electrodes by anodic attachment methods. One method involved oneꢀ
electron oxidation of the lithioethynyl activated complex and gave surface coverages of up to Γ =
5.5 x 10ꢀ10 mol cmꢀ2 (roughly a monolayer of material) after several CV scans. The other required
scanning to the third anodic wave of EꢀBFD, which involves oxidation of its ethynyl group. In
that case, surface coverages approaching 4 x 10ꢀ10 mol cmꢀ2 were observed. When measured in
pure electrolytes, the CV responses of the modified electrodes do not diminish appreciably with
multiple scans of the [EꢀBFD]0/+ couple, confirming that the EꢀBFD moieties are strongly (most
likely covalently) attached to the electrode surface through its ethynyl group.
The voltammetric responses of the chemicallyꢀmodified electrodes were sensitive to the
molecular size of the supporting electrolyte anion. The current passed for the second oxidation of
surfaceꢀbound EꢀBFD was essentially the same as that passed for the first oxidation when the
anion was relatively small (e.g., [ClO4]ꢀ), but became significantly less when the anion was large
(e.g. [B(C6F5)4]ꢀ). Based on recently published calculations of spaceꢀfilling models,91 the anionic
volumes of [B(C6F5)4]ꢀ and [B(C6H3(CF3)2)4]ꢀ are between five and eight times greater than those
of the traditional smaller anions [PF6]ꢀ and [ClO4]ꢀ (Figure 1). The largeꢀanion effect is attributed
19
ACS Paragon Plus Environment

Langmuir
Page 20 of 42
1
2
3
4
5
6
7
8
9
to the occurrence of mixed diffusivity of the anion at the electrode/surface interface: free
diffusion along with sterically encumbered transport into a more interior position that allows
optimum ionꢀpairing with the [EꢀBFD]+ monocation. Under conditions of higher surface
coverage or shorter voltammetric time scales, the latter introduces a blockage factor for the
oxidation of [EꢀBFD]+ to the corresponding dication, analogous to the idea that the positive
charge in the monolayer is not fully “screened” by the electrolyte anion.92, 93
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 present results contribute to the family of reports on mediumꢀbased diminished
electroactivity for CMEs, and appear to be the first exploring this phenomenon through
systematic and sizeable variations in the volumes of pseudoꢀspherical electrolyte anions.
The fact that glassy carbon surfaces are irregularly shaped,94,95 and that the EꢀBFD modified
electrodes have been characterized only by electrochemical methods, limits the degree to which
the largeꢀanion effect can be geometrically modeled by our findings. However, these results
suggest a new approach to the systematic inquiry of ionꢀpairing effects in more definitivelyꢀ
organized electrode surfaces (e.g. SAMs) and show that square wave voltammetry may be a
useful tool for studying the timeꢀdependence of current blockage effects.
Acknowledgements This work was supported by the National Science Foundation (CHEꢀ
1212339 and CHEꢀ1565541) and by the Deutsche Forschungsgemeinschaft (DFG FOR 516). We
thank a reviewer for suggesting a comparison with the results in reference 92. The supporting
information (five CV and SWV scans; one table) is available free of charge on the ACS
Publications website.
20
ACS Paragon Plus Environment

Page 21 of 42
Langmuir
1
2
3
4
5
6
Table 1. E1/2 values in dichloromethane (volt vs. ferrocene/ferrocenium) for the oxidation of
bis(fulvalene)diiron (BFD) and the α and β isomers of ethynylbis(fulvalene)diiron (EꢀBFD).
CME = chemically modified electrode.
7
8
9
Compound Electrolyte
Aniona
Medium
E1/2(1)
(V)
E1/2(2)
(V)
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
BFD
BFD
[PF6]ꢀ
Homogeneous b
Homogeneous c
Homogeneous
CME
ꢀ0.29
ꢀ0.33
ꢀ0.20
ꢀ0.20
+0.38
+0.64
+0.41
+0.43
[B(C6F5)4]ꢀ
[PF6]ꢀ
αEꢀBFD
αEꢀBFD
αEꢀBFD
αEꢀBFD
βEꢀBFD
βEꢀBFD
βEꢀBFD
βEꢀBFD
[PF6]ꢀ
[B(C6F5)4]ꢀ
[B(C6F5)4]ꢀ
[PF6]ꢀ
Homogeneous
CME
ꢀ0.22
ꢀ0.10
ꢀ0.21
ꢀ0.23
ꢀ0.21
ꢀ0.11
+0.73
+0.75
+0.39
+0.34
+0.71
+0.71
Homogeneous
CME
[PF6]ꢀ
[B(C6F5)4]ꢀ
[B(C6F5)4]ꢀ
Homogeneous
CME
a Supporting electrolyte was either 0.1 M [NBu4][PF6] or 0.05 M [NBu4][B(C6F5)4]. Decamethylferrocene
was employed as the experimental inꢀsitu standard. In converting the experimental values to the potentials
values vs ferrocene listed in this table, 0.56 V was subtracted from the experimental value for [PF6]ꢀ
solutions, and 0.61 V was subtracted for [B(C6F5)4]ꢀ solutions. b Taken from ref 35. c Taken from ref 34.
21
ACS Paragon Plus Environment

Langmuir
Page 22 of 42
1
2
3
4
5
6
Table 2. E1/2 values (volt vs. ferrocene/ferrocenium) for the oxidation of the β isomer of
ethynylbis(fulvalene)diiron (βEꢀBFD) in THF and in nitromethane. CME = chemically modified
electrode.
7
8
9
Compound Solvent
Electrolyte
Aniona
Medium
E1/2(1)
(V)
E1/2(2)
(V)
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
THF
[PF6]ꢀ
Homogeneous
CME
ꢀ0.19
ꢀ0.22
ꢀ0.20
ꢀ0.16
ꢀ0.16
ꢀ0.14
ꢀ0.30
ꢀ0.23
+0.37
+0.38
+0.52
+0.61
+0.48
+0.47
+0.36
+0.46
βEꢀBFD
βEꢀBFD
βEꢀBFD
βEꢀBFD
βEꢀBFD
βEꢀBFD
βEꢀBFD
βEꢀBFD
THF
[PF6]ꢀ
THF
[B(C6F5)4]ꢀ Homogeneous
[B(C6F5)4]ꢀ
THF
CME
nitromethane [PF6]ꢀ
Homogeneous
CME
nitromethane [PF6]ꢀ
nitromethane [B(C6F5)4]ꢀ
nitromethane [B(C6F5)4]ꢀ
Homogeneous
CME
a Supporting electrolyte was either 0.1 M [NBu4][PF6] or 0.05 M [NBu4][B(C6F5)4]. Decamethylferrocene
was employed as the experimental inꢀsitu standard. In converting the experimental values to the potentials
values vs ferrocene listed in this table, the following subtractions were made: 0.45 V for THF/[PF6]ꢀ; 0.48
V for THF/[B(C6F5)4]ꢀ; 0.51 V for MeNO2/[PF6]ꢀ; 0.52 V for MeNO2/[B(C6F5)4]ꢀ; see Barrière, F.; Geiger,
W.E. J. Am. Chem. Soc. 2006, 128, 3980.
Table 3. Values of ꢀE1/2 (= E1/2(2) ꢀ E1/2(1)) measured for α and β isomers of
ethynylbis(fulvalene)diiron and those reported for bis(fulvalene)dinickel, Fv2Ni2 (Barrière, F.;
Geiger, W.E. J. Am. Chem. Soc. 2006, 128, 3980). ꢀꢀE1/2 gives the increase in ꢀE1/2 for a given
solvent as the supporting electrolyte is changed from 0.10 M [NBu4][PF6] to 0.05 M
[NBu4][B(C6F5)4].
ꢀE1/2 ([PF6]ꢀ) (mV)
ꢀE1/2 (TFAB) (mV)
ꢀꢀE1/2 (mV)
Compound
Fv2Ni2
αEꢀBFD
βEꢀBFD
Fv2Ni2
βEꢀBFD
Fv2Ni2
βEꢀBFD
Solvent
dichloromethane
dichloromethane
dichloromethane
THF
THF
nitromethane
nitromethane
480
610
600
407
560
485
640
753
950
920
595
720
500
660
273
340
320
188
160
15
20
22
ACS Paragon Plus Environment

Page 23 of 42
Langmuir
1
2
3
4
5
6
7
8
Table 4. Comparison of ꢀE1/2 (= E1/2(2) ꢀ E1/2(1)) values for EꢀBFD isomers in solution vs on a
chemically modified electrode (CME). The solvent was dichloromethane and the electrolyte was
either 0.1 M [NBu4][PF6] or 0.05 M [NBu4][B(C6F5)4]. E1/2 values were taken from the data in
Table 1.
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
ꢀE1/2, [PF6]ꢀ ꢀE1/2, B(C6F5)4]ꢀ
Medium
ꢀꢀE1/2 ([B(C6F5)4]ꢀ ꢀ [PF6]ꢀ)
Compound
(mV)
(mV)
970
950
850
920
820
(mV)
300
340
220
320
250
BFD
670
610
630
600
570
homogeneous
homogeneous
CME
homogeneous
CME
αEꢀBFD
αEꢀBFD
βEꢀBFD
βEꢀBFD
Table 5. Ratios, I2/I1, of the peak currents for the forward components (Ifwd) of square wave
voltammograms recorded for an αEꢀBFD chemically modified electrode that was modified by
the direct oxidation method. The scans were recorded at different SW frequencies on
dichloromethane solutions contain ning a [NBu4][A] supporting electrolyte: A = [PF6]ꢀ or [ClO4]ꢀ
, 0.1 M; A = [B(C6H3(CF3)2)4]ꢀ or [B(C6F5)4]ꢀ, 0.05 M. Manual baseꢀline corrections were used
to obtain the peak current values.
[ClO4]ꢀ
1.0
[B(C6H3(CF3)2)4]ꢀ
SW freq (Hz)
[PF6]ꢀ
1.2
1.2
1.1
1.0
b
[B(C6F5)4]ꢀ
0.6a
10
20
40
0.19a
0.18
0.16
0.14
b
0.80
0.83
0.83
b
0.50
0.25
0.22
0.22
60
150 or 300
a. Low currents introduce more uncertainty into the backgroundꢀsubtracted value of I2.
b. Not measured.
23
ACS Paragon Plus Environment

Langmuir
Page 24 of 42
1
2
3
4
References
5
6
7
8
9
1. (a) Elliott, C.M.; Redepenning, J.G.; Balk, E.M. Electronic conductivity of poly [tris (5, 5'ꢀbis
[(3ꢀacrylylꢀ1ꢀpropoxy) carbonyl]ꢀ2, 2'ꢀbipyridine) ruthenium (0)]. J. Am. Chem. Soc. 1985, 107,
8302ꢀ8304 (b) Elliott, C.M.; Kopelove, A.B.; Albery, W.J. Chen, Z. Nonaqueous
electrochemistry of polypyrrole/polystyrenesulfonate composite films: voltammetric,
coulometric, EPR, and ac impedance studies. J. Phys. Chem. 1991, 95, 1743ꢀ1747.
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. Smith, C.P.; White, H.S. Theory of the Interfacial Potential Distribution and Reversible
Voltammetric Response of Electrodes Coated with Electroactive Molecular Films. Anal. Chem.
1992, 64, 2398ꢀ2405.
3. De Long, H.C.; Buttry, D.A. Ionic interactions play a major role in determining the
electrochemical behavior of selfꢀassembling viologen monolayers. Langmuir 1990, 6, 1319ꢀ
1322.
4. Sheridan, M.V.; Sherman, B.D.; Wee, K.ꢀR.; Marquard, S.L.; Gold, A.S.; Meyer, T.J.
Nonaqueous electrocatalytic water oxidation by a surfaceꢀbound Ru (bda)(L) 2 complex. Dalton
Trans. 2016, 45, 6324ꢀ6328.
5. Filippini, G.; Goujon, F.; Bonal, C.; Malfreyt, P. Environment effect on the redox properties of
SelfꢀAssembled Monolayers: a theoretical investigation of the nature of the supporting
electrolyte. Soft Matter 2011, 7, 8961ꢀ8968.
6. Valincius, G.; Niaura, G.; Kazakevičienė, B.; Talaikytė, Z; Kažemėkaitė, M.; Butkus, E.;
Razumas, V. Anion effect on mediated electron transfer through ferroceneꢀterminated selfꢀ
assembled monolayers. Langmuir 2004, 20, 6631ꢀ6638.
7. Nguyen, K.ꢀL.; Dionne, E.R.; Badia, A. Redoxꢀcontrolled ionꢀpairing association of anionic
surfactant to ferroceneꢀterminated selfꢀassembled monolayers. Langmuir 2015, 31, 6385ꢀ6394.
8. Rowe, G.K.; Creager, S.E. Redox and ionꢀpairing thermodynamics in selfꢀassembled
monolayers. Langmuir 1991, 7, 2307ꢀ2312.
9. Cruañes, M.T.; Drickamer, H.G.; Faulkner, L.R. Characterization of ChargeꢀTransfer
Processes in SelfꢀAssembled Monolayers by HighꢀPressure Electrochemical Techniques.
Langmuir 1995, 11, 4089ꢀ4097.
10. Ju, H.; Leech, D. Effect of electrolytes on the electrochemical behaviour of 11ꢀ
(ferrocenylcarbonyloxy) undecanethiol SAMs on gold disk electrodes. Phys. Chem. Chem. Phys.
1999, 1, 1549ꢀ1554.
11. Rowe, G.K.; Creager, S.E. Interfacial solvation and doubleꢀlayer effects on redox reactions in
organized assemblies. J. Phys. Chem. 1994, 98, 5500ꢀ5507.
24
ACS Paragon Plus Environment

Page 25 of 42
Langmuir
1
2
3
4
5
6
12. Yokota, Y.; Yamada, T.; Kawai, M. Ionꢀpair formation between ferroceneꢀterminated selfꢀ
assembled monolayers and counteranions studied by force measurements. J. Phys. Chem. C
2011, 115, 6775ꢀ6781.
7
8
9
13. Tian, H.; Dai, Y.; Shao, H.; Yu, H.ꢀZ. Modulated intermolecular interactions in
ferrocenylalkanethiolate selfꢀassembled monolayers on gold. J. Phys. Chem. C 2013, 117, 1006ꢀ
1012.
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
14. Norman, L.L.; Badia, A. Microcantilevers modified with ferroceneꢀterminated selfꢀ
assembled monolayers: effect of molecular structure and electrolyte anion on the redoxꢀinduced
surface stress. J. Phys. Chem. C 2011, 115, 1985ꢀ1995.
15. Creager, S.E.; Rowe, G.K. Redox properties of ferrocenylalkane thiols coadsorbed with
linear nꢀalkanethiols on polycrystalline bulk gold electrodes. Anal. Chim. Acta 1991, 246, 233ꢀ
239.
16. Dionne, E.R.; Sultana, T.; Norman, L.L.; Toader, V.; Badia, A. Redoxꢀinduced ion pairing of
anionic surfactants with ferroceneꢀterminated selfꢀassembled monolayers: faradaic
electrochemistry and surfactant aggregation at the monolayer/liquid interface. J. Am. Chem. Soc.
2013, 135, 17457ꢀ17468.
17. Sun, Q.ꢀW.; Murase, K.; Ichii, T.; Sugimura, H. Anionic effect of ionic liquids electrolyte on
electrochemical behavior of ferrocenylthiol/alkanethiol binary SAMs. J. Electroanal. Chem.
2010, 643, 58ꢀ66.
18. Jalkh, J.; Leroux, Y.R.; Lagrost, C.; Hapiot, P. Comparative electrochemical investigations in
ionic liquids and molecular solvents of a carbon surface modified by a redox monolayer. J. Phys.
Chem. C 2014, 118, 28640ꢀ28646.
19. Hedberg, F.L.; Rosenberg, H. Synthesis of 1, 1'ꢀbiferrocenylene. J. Am. Chem. Soc. 1969, 91,
1258ꢀ1259.
20. Rausch, M.D.; Kovar, R.F.; Kraihanzel, C.S. Organometallic. pi. complexes of fulvalene. J.
Am. Chem. Soc. 1969, 91, 1259ꢀ1261.
21. This compound has also been referred to as 1,1’ꢀbiferrocenylene.
22. LeVanda, C.; Bechgaard, K.; Cowan, D.O.; MuellerꢀWesterhoff, U.T.; Eilbracht, P.;
Candela, G.A.; Collins, R.L. Bis (fulvalene) diiron, its monoꢀand dications. Intramolecular
exchange interactions in a rigid system. J. Am. Chem. Soc. 1976, 98, 3181ꢀ3187.
23. Breuer, R.; Schmittel, M. 1, 1′ꢀBiferrocenylenes The More Redox Stable Ferrocenes! New
Derivatives, Corrected NMR Assignments, Redox Behavior, and Spectroelectrochemistry.
Organometallics 2012, 31, 1870ꢀ1878.
24. Warratz, R.; Aboulfadl, H.; Bally, T.; Tuczek, F. Electronic Structure and Absorption Spectra
of Biferrocenyl and Bisfulvalenide Diiron Radical Cations: Detection and Assignment of New
Low Energy Transitions. Chem. Eur. J. 2009, 15, 1604ꢀ1617 and references therein.
25
ACS Paragon Plus Environment

Langmuir
Page 26 of 42
1
2
3
4
5
6
7
8
9
25. (a) Hillman, M.; Kvick, A. Structural consequences of oxidation of ferrocene derivatives. 1.
[0.0]Ferrocenophanium picrate hemihydroquinone. Organometallics 1983, 2, 1780ꢀ1785 (b)
Brüggeller, P.; Jaitner, P.; Schottenberger, H. Kristallographische Gegenüberstellung der
Monokationen von Bis(fulvalen)dieisien und Bis(fulvalen) eisenꢀcobalt mit identischem
Gegenion (PF6−). J. Organomet. Chem. 1991, 417, C53ꢀC58.
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
26. Churchill, M.R.; Wormald, J. Crystal and molecular structure of bis (fulvalene) diiron. Inorg.
Chem. 1969, 8, 1970ꢀ1974.
27. Morrison, W.H., Jr.; Krogsrud, S.; Hendrickson, D.N. Polarographic and magnetic
susceptibility study of various biferrocene compounds. Inorg. Chem. 1973, 12, 1998ꢀ2004.
28. Levanda, C.; Bechgaard, K.; Cowan, D.O. Mixed valence cations. Chemistry of. pi.ꢀbridged
analogues of biferrocene and biferrocenylene. J. Org. Chem. 1976, 41, 2700ꢀ2704.
29. Pittman Jr., C.U.; Surynarayanan, B. Synthesis and polymerization of 3ꢀ
vinylbisfulvalenediiron. Preparation and conductivity of its polymeric [FeIIFeIII](TCNQ) 2.ꢀ
salts. J. Am. Chem. Soc. 1974, 96, 7916ꢀ7919.
30. Bocarsly, A.B.; Walton, E.G.; Bradley, M.G.; Wrighton, M.S. Twoꢀelectron oxidations at
illuminated Nꢀtype semiconducting silicon electrodes: Use of chemically derivatized
photoelectrodes. J. Electroanal. Chem. 1979, 100, 283ꢀ306.
31. Nishihara, H.; Ohta, M.; Aramaki, K. Electrochemical and electrical properties of a mixedꢀ
valence bis(fulvalene)diiron salt in polyethylene glycol studied using an interdigitated array
microelectrode. J. Chem. Soc., Faraday Trans.1992, 88, 827ꢀ831.
32. Sakamoto, K.; Nishihara, H.; Aramaki, K. Synthesis and electrochemical characterization of
bis(fulvalene)diiron electrode films. J. Chem. Soc., Dalton Trans.1992, 1877ꢀ1881.
33. Jaitner, P.; Schottenberger, H.; Gamper, S.; Obendorf, D. Termetallocenes. J. Organomet.
Chem. 1994, 475, 113ꢀ120.
34. Camire, N.; MuellerꢀWesterhoff, U.T.; Geiger, W.E. Improved electrochemistry of multiꢀ
ferrocenyl compounds: investigation of biferrocene, terferrocene, bis (fulvalene) diiron and
diferrocenylethane in dichloromethane using [NBu 4][B (C 6 F 5) 4] as supporting electrolyte. J.
Organomet. Chem. 2001, 637-639, 823ꢀ826.
35. Breuer, R.; Schmittel, M. Redoxꢀstable SAMs in water (pH 0–12) from 1, 1′ꢀbiferrocenyleneꢀ
terminated thiols on gold. Organometallics 2012, 31, 6642ꢀ6651.
36. Barrière, F.; Geiger, W.E. Use of weakly coordinating anions to develop an integrated
approach to the tuning of ꢁ E 1/2 values by medium effects. J. Am. Chem. Soc. 2006, 128, 3980ꢀ
3989.
37. Shimazu, K.; Ye, S.; Sato, Y.; Uosaki, K. Simultaneous detection of structural change and
mass transport accompanying the redox of a ferrocenylundecanethiol monolayer with the novel
FTꢀIR reflection absorption spectroscopy/electrochemical quartz crystal microbalance combined
system. J. Electroanal. Chem. 1994, 375, 409ꢀ413.
26
ACS Paragon Plus Environment

Page 27 of 42
Langmuir
1
2
3
4
5
38. Quist, F.; TabardꢀCossa, V.; Badia, A. Nanomechanical cantilever motion generated by a
surfaceꢀconfined redox reaction. J. Phys. Chem. B 2003, 107, 10691ꢀ10695.
6
7
8
9
39. Uosaki, K.; Sato, Y.; Kita, H. A selfꢀassembled monolayer of ferrocenylalkane thiols on gold
as an electron mediator for the reduction of Fe (III)ꢀEDTA in solution. Electrochim. Acta 1991,
36, 1799ꢀ1801.
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
40. Uosaki, K.; Sato, Y.; Kita, H. Electrochemical characteristics of a gold electrode modified
with a selfꢀassembled monolayer of ferrocenylalkanethiols. Langmuir 1991, 7, 1510ꢀ1514.
41. LeSuer, R.J.; Buttolph, C.; Geiger, W.E. Comparison of the conductivity properties of the
tetrabutylammonium salt of tetrakis (pentafluorophenyl) borate anion with those of traditional
supporting electrolyte anions in nonaqueous solvents. Anal. Chem. 2004, 76, 6395ꢀ6401.
42. Barrière, F.; Geiger, W.E. Organometallic electrochemistry based on electrolytes containing
weaklyꢀcoordinating fluoroarylborate anions. Acc. Chem. Res. 2010, 43, 1030ꢀ1039.
43. Geiger, W.E. in Bard, A.J.; Zoski, C. (Eds) Electroanalytical Chemistry, CRC Press, Boca
Raton, 2014, Vol 25, pp 179ꢀ222.
44. Sheridan, M.V.; Lam, K.; Geiger, W.E. An anodic method for covalent attachment of
molecules to electrodes through an ethynyl linkage. J. Am. Chem. Soc. 2013, 135, 2939ꢀ2942.
45. Sheridan, M.V.; Lam, K.; Geiger, W.E. Covalent attachment of porphyrins and ferrocenes to
electrode surfaces through direct anodic oxidation of terminal ethynyl groups. Angew. Chem. Int.
Ed. 2013, 52, 12897ꢀ12900.
46. Sheridan, M.V.; Lam, K.; Sharafi, M.; Schneebeli, S.T.; Geiger, W.E. Anodic Methods for
Covalent Attachment of Ethynylferrocenes to Electrode Surfaces: Comparison of Ethynyl
Activation Processes. Langmuir 2016, 32, 1645ꢀ1657.
47. Stankovich, M.T.; Bard, A.J. The electrochemistry of proteins and related substances part III.
bovine serum albumin. J. Electroanal. Chem. 1978, 86, 189ꢀ199.
48. Heering, H.A.; Mondal, M.S.; Armstrong, F.A. Using the pulsed nature of staircase cyclic
voltammetry to determine interfacial electronꢀtransfer rates of adsorbed species. Anal. Chem.
1999, 71, 174ꢀ182.
49. Bard, A.J.; Faulkner, L.R. Electrochemical Methods, John Wiley & Sons, New York, 2nd Ed.,
2001, pp 286ꢀ293 (DPV) and 293ꢀ299 (SWV).
50. Osteryoung, J.; O’Dea, J.J. in Bard, A.J. (Ed) Electroanalytical Chemistry, Marcel Dekker,
Inc., New York, 1986, Vol 14, pp 209ꢀ308.
51. Lovrić, M.; KomorskyꢀLovric, Š. Squareꢀwave voltammetry of an adsorbed reactant. J.
Electroanal. Chem. 1988, 248, 239ꢀ253.
52. Ribes, A.J.; Osteryoung, J. Determination of 8ꢀchloroꢀ6ꢀ(2ꢀfluorophenyl)ꢀ1ꢀmethylꢀ4Hꢀ
imidazo [1, 5ꢀa][1, 4] benzodiazepine by adsorptive stripping with pulse voltammetry. Anal.
Chem.1990, 62, 2632ꢀ2636.
27
ACS Paragon Plus Environment

Langmuir
Page 28 of 42
1
2
3
4
5
53. O’Dea, J.J.; Ribes, A.; Osteryoung, J. G. Squareꢀwave voltammetry applied to the totally
irreversible reduction of adsorbate. J. Electroanal. Chem. 1993, 345, 287ꢀ301.
6
7
8
54. Reeves, J.H.; Song, S.; Bowden, E.F. Application of square wave voltammetry to strongly
adsorbed quasireversible redox molecules. Anal. Chem.1993, 65, 683ꢀ688.
9
55. Jeuken, L.J.C.; McEvoy, J.P.; Armstrong, F.A. Insights into gated electronꢀtransfer kinetics
at the electrode− protein interface: A square wave voltammetry study of the blue copper protein
azurin. J. Phys. Chem. B 2002, 106, 2304ꢀ2313.
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
56. Mirčeski, V.; Quentel, F.; L’Her, M.; Elleouet, C. Homogeneous versus heterogeneous
catalysis at electrodes modified with a thin organic layer: theoretical and experimental study
under conditions of squareꢀwave voltammetry. J. Phys. Chem. C 2007, 111, 8283ꢀ8290, and
references therein.
57. Š. KomorskyꢀLovrić; M. Lovrić in E.P. Vargus (Ed) New Research on Electrochemistry,
Nova Science Publishers, 2007, pp 1ꢀ44.
58. KomorskyꢀLovrić, Š.; Lovrić, M. Theory of squareꢀwave voltammetry of twoꢀstep electrode
reaction with kinetically stabilized intermediate. J. Electroanal. Chem. 2011, 660, 22ꢀ25.
59. Garay, F.; Lovrić, M. Squareꢀwave voltammetry of quasiꢀreversible electrode processes with
coupled homogeneous chemical reactions. J. Electroanal. Chem. 2002, 518, 91ꢀ102.
60. Laws, D.R.; Chong, D.; Nash, K.; Rheingold, A.L.; Geiger, W.E. Cymantrene Radical Cation
Family: Spectral and Structural Characterization of the HalfꢀSandwich Analogues of
Ferrocenium Ion. J. Am. Chem. Soc. 2008, 130, 9859ꢀ9870.
61. Sheridan, M.V.; Lam, K.; Geiger, W.E. Spontaneous attachment of lithiumꢀactivated
ferrocenylalkynes to carbon and gold. Electrochem. Commun. 2015, 52, 63ꢀ66.
62. Doisneau, G.; Balavoine, G.; FillebeenꢀKhan, T. Synthesis and some reactions of
ferrocenylacetylenes. J. Organomet. Chem. 1992, 425, 113ꢀ117.
63. Pople, J; Beveridge, D. Approximate Molecular Orbital Theory, McGrawꢀHill, 1970.
64. Stewart, J.J.P. Optimization of parameters for semiempirical methods VI: more modifications
to the NDDO approximations and reꢀoptimization of parameters. J. Mol. Mod. 2013, 19, 1ꢀ32.
65. MOPAC2016, Version: 17.048M, Stewart, J.J.P., Stewart Computational Chemistry, web:
HTTP://OpenMOPAC.net.
66. Breuer, R.; Schmittel, M. Unsymmetrically Substituted 1, 1′ꢀBiferrocenylenes Maintain Class
III MixedꢀValence Character. Organometallics 2013, 32, 5980ꢀ5987.
67. The redox wave for electroꢀgenerated EꢀBFD is broader than normal owing in part to the
fact that a surface layer of EꢀBFD is already present as the potential becomes sufficiently
positive to oxidize the compound formed in the reaction layer.
28
ACS Paragon Plus Environment

Page 29 of 42
Langmuir
1
2
3
4
5
68. Bard, A.J.; Faulkner, L.R. Electrochemical Methods, John Wiley & Sons, New York, 2nd Ed.,
2001, pp 589ꢀ593.
6
7
8
69. Seo, K.; Jeon, I.C.; Yoo, D.J. Electrochemical characteristics of ferrocenecarboxylateꢀ
coupled aminoundecylthiol selfꢀassembled monolayers. Langmuir 2004, 20, 4147ꢀ4154.
9
70. Bélanger, D.; Pinson, J. Electrografting: a powerful method for surface modification. Chem.
Soc. Rev. 2011, 40, 3995ꢀ4048.
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
71. It should be kept in mind that although the size of the anions appear to play a major role in
the heterogeneous interactions described in the present work, it is the delocalization of negative
charge throughout the anion that is usually thought to play the larger influence on its interaction
with cations in homogeneous solutions. See (a) Strauss, S.H. The search for larger and more
weakly coordinating anions. Chem. Rev. 1993, 93, 927ꢀ942 (b) Krossing, I.; Raabe, I.
Noncoordinating anions—fact or fiction? A survey of likely candidates. Angew. Chem. Int. Ed.
2004, 43, 2066ꢀ2090.
72. Dalton, E.F.; Murray, R.W. Viologen (2+/1+) and viologen (1+/0) electronꢀselfꢀexchange
reactions in a redox polymer. J. Phys. Chem. 1991, 95, 6383ꢀ6389.
73. DunandꢀSauthier, M.ꢀN.C.; Deronzier, A.; Moutet, J.ꢀC.; Tingry, S. Electrosynthesis and coꢀ
ordination chemistry of polymer films coated on electrode surfaces from pyrrole polypyridyls. J.
Chem. Soc., Dalton Trans. 1996, 2503ꢀ2509.
74. Trammell, S.A.; Dressick, W.J.; Melde, B.J.; Moore, M. Photocurrents from the Direct
Irradiation of a Donor–Acceptor Complex Contained in a Thin Film on Indium Tin Oxide. J.
Phys. Chem. C 2011, 115, 13446ꢀ13461.
75. Berduque,A.; Herzog, G.; Watson, Y.E.; Arrigan, D.W.M.; Moutet, J.ꢀC.; Reynes, O.; Royal,
G.; SaintꢀAman, E. Development of Surface Modified Microelectrode Arrays for the
Electrochemical Detection of Dihydrogen Phosphate. Electroanalysis 2005, 17, 392ꢀ399.
76. Lyon, L.A.; Ratner, M.A.; Hupp, J.T. Ion modulated electroactivity in thinꢀfilm polymers
derived from bipyridyl and phenanthroline complexes of iron. J. Electroanal. Chem. 1995, 387,
109ꢀ113.
77. Andrieux, C.P.; Savéant, J.M. Electroneutrality coupling of electron hopping between
localized sites with electroinactive counterion displacement. 1. Potentialꢀstep plateau currents. J.
Phys. Chem. 1988, 92, 6761ꢀ6767.
78. Chidsey, C.E.D.; Murray, R.W. Redox capacity and direct current electron conductivity in
electroactive materials. J. Phys. Chem. 1986, 90, 1479ꢀ1484.
79. Mathias, M.F.; Haas, O. An alternating current impedance model including migration and
redoxꢀsite interactions at polymerꢀmodified electrodes. J. Phys. Chem. 1992, 96, 3174ꢀ3182.
29
ACS Paragon Plus Environment