The redox properties of ferrocenyl-substituted aryl azines

Article abstract of DOI:10.1139/v02-019

A series of ferrocenyl substituted azines (1-Fc/Ar, where Ar = 4-NO₂C₆H₄, 4-CNC₆H₄, 4-OCH₃C₆H₄, C₅H₄N, ferrocene, anthracene, and pyrene) were investigated by electrochemical and photochemical techniques. All the 1-Fc/Ar exhibited oxidation waves within 60 mV of each other, consistent with the expected oxidation of the ferrocene moiety. The reduction properties of 1-Fc/Ar is governed by the nature of the Ar substituent. The standard reduction potentials suggest that ferrocene has comparable electron donating abilities as a 4-methoxyphenyl and 4-dimethyl-aminophenyl group. The anthracenyl azines exhibit one-electron reversible reduction followed by dimerization of the radical anion resulting in dimerization rate constants (k_d) between 5.1 × 10⁴ and 1.5 × 10⁵ M^-1 s^-1. 1-Fc/Anth and related azines undergo photochemical E/Z isomerization of the C=N bonds to produce E/Z and Z/Z isomers from the thermodynamically most stable E/E form. Fluorescence at 77 K, was observed for these compounds only after long-wavelength irradiation to produce a mixture of E/E, E/Z, and Z/Z isomers. Fluorescence quantum yields of 0.042, 0.090, and 0.176 were determined for 2-Anth/H, 2-Anth/Anth, and 1-Fc/Anth, respectively. The electrochemical, photochemical, and X-ray data suggest that the azine unit is a conjugation limiter and may be a general characteristic of azine molecules.

Full text of DOI:10.1139/v02-019

250
The redox properties of ferrocenyl-substituted
aryl azines
Vittorio A. Sauro and Mark S. Workentin
Abstract: A series of ferrocenyl substituted azines (1-Fc/Ar, where Ar = 4-NO2C6H4, 4-CNC6H4, 4-OCH3C6H4,
C5H4N, ferrocene, anthracene, and pyrene) were investigated by electrochemical and photochemical techniques. All the
1-Fc/Ar exhibited oxidation waves within 60 mV of each other, consistent with the expected oxidation of the ferrocene
moiety. The reduction properties of 1-Fc/Ar is governed by the nature of the Ar substituent. The standard reduction
potentials suggest that ferrocene has comparable electron donating abilities as a 4-methoxyphenyl and 4-dimethyl-
aminophenyl group. The anthracenyl azines exhibit one-electron reversible reduction followed by dimerization of the
radical anion resulting in dimerization rate constants (kd) between 5.1 × 104 and 1.5 × 105 M–1 s–1. 1-Fc/Anth and re-
lated azines undergo photochemical E/Z isomerization of the C=N bonds to produce E/Z and Z/Z isomers from the
thermodynamically most stable E/E form. Fluorescence at 77 K, was observed for these compounds only after long-
wavelength irradiation to produce a mixture of E/E, E/Z, and Z/Z isomers. Fluorescence quantum yields of 0.042,
0.090, and 0.176 were determined for 2-Anth/H, 2-Anth/Anth, and 1-Fc/Anth, respectively. The electrochemical,
photochemical, and X-ray data suggest that the azine unit is a conjugation “limiter” and may be a general characteristic
of azine molecules.
Key words: azines, electrochemistry, cyclic voltammetry, reduction, electron transfer.
Résumé : Faisant appel à des techniques électrochimiques ainsi que photochimiques, on a étudié une série d’azines de
ferrocényles substitués (1-Fc/Ar dans lesquels Ar = 4-NO2C6H4, 4-CNC6H4, 4-OCH3C6H4, C5H4N, ferrocène, anthra-
cène et pyrène). Tous les 1-Fc/Ar présentent des vagues d’oxydation à moins de 60 mV les uns des autres, ce à quoi
on peut s’attendre pour l’oxydation de la partie ferrocène. Les propriétés de réduction des 1-Fc/Ar sont définies par la
nature du substituant Ar. Les potentiels de réduction standards suggèrent que les propriétés du ferrocène en ce qui a
trait à ses possibilités de donner des électrons sont semblables à celles des groupes 4-méthoxyphényle et 4-
diméthylaminophényle. Les azines dérivés de l’anthracène donnent lieu à une réduction réversible à un électron suivie
d’une dimérisation de l’anion radical avec des constantes de vitesse de dimérisation (kd) qui vont de 5,1 × 104 à 1,5 ×
105 M–1 s–1. La 1-Fc/Anth et les azines apparentées subissent des isomérisations photochimiques E/Z des liaisons C=N
qui conduisent à des mélanges d’isomères E/Z et Z/Z à partir de la forme E/E, thermodynamiquement la plus stable.
Ces composés ne donnent lieu à de la fluorescence à 77 K qu’après avoir été soumis à une irradiation à grande lon-
gueur d’onde qui permet de produire un mélange des isomères E/E, E/Z et Z/Z. On a déterminé des rendements quanti-
ques de fluorescence de 0,042, 0,090 et 0,176 respectivement pour les 2-Anth/H, 2-Anth/Anth et 1-Fc/Anth. Les
données électrochimiques, photochimiques ainsi que celles relatives à la diffraction des rayons X suggèrent que l’unité
azine « limite » la conjugaison et que cette propriété est peut-être une caractéristique générale pour les molécules
d’azine.
Mots clés : azines, électrochimie, voltampérométrie cyclique, réduction, transfert d’électron.
[Traduit par la Rédaction] Sauro and Workentin 262
Introduction
stereochemical, and electrochemical properties (1) of ferrocene,
there is great interest in ferrocene based ligands for the pro-
duction of multi-metal containing organometallic complexes
that can be used for a variety of applications (9, 10).
Ferrocene-containing compounds have also been investi-
gated for their second-order non-linear optical (NLO) prop-
erties (3, 11–13).
Ferrocene and its derivatives have been investigated for
their use in a variety of applications, including in organic
synthesis (1), in applications tied to their electrochemical
properties (1–5), in biological systems (6), as molecular
wires and in other nanoelectronic devices (7), and in anion
recognition applications (8). Due to the special chemical,
Ferrocene and its derivatives have been studied exten-
sively using electrochemical techniques (1) and generally
show a single one-electron reversible oxidation whose po-
tential varies depending on the substituent attached to the
cyclopentadienyl ring (3, 5, 10, 13, 14). The redox proper-
ties of the ferrocene group have been exploited to produce
molecules that may have potential applications as molecular
switches and molecular sensors (15, 16).
Received 30 July 2001. Published on the NRC Research
Press Web site at http://canjchem.nrc.ca on 19 March 2002.
V.A. Sauro and M.S. Workentin.1 Department of Chemistry,
The University of Western Ontario, London, ON N6A 5B7
Canada.
1Corresponding author (e-mail: mworkent@uwo.ca).
Can. J. Chem. 80: 250–262 (2002)
DOI: 10.1139/V02-019
© 2002 NRC Canada

Sauro and Workentin
251
Scheme 1.
Azines (>C=N-N=C<) appear to have the structural ele-
ments necessary to allow for complete conjugation between
the substituents used to make up its structure, although there
is now general agreement that the azine unit stops, or at the
very least limits, conjugation. The question of conjugation
through the azine unit has been investigated both theoreti-
cally and experimentally (17–29). Glaser and co-workers
(20–29) examined the solid-state structures of a large num-
ber of structurally related aryl azines, paying particular at-
tention to the structural consequences of conjugative
interactions that manifest themselves in the solid-state struc-
tures. As a result of their extensive X-ray analyses, including
the conformational properties about the N–N and Ar–C
bonds, bond length analysis, and theoretical studies, Glaser
and co-workers (25, 27) conclude that the azine unit is a
“conjugation stopper,” at least in the solid state.
N
R
Fe
N
Ar
1-Fc/Ar
NO2
CN
,
OCH3
N
R = CH3, Ar =
,
,
,
Fe
R = H,
Ar =
,
Ar1
H
Early and recent electrochemical investigations (2, 30–38)
of azines have suggested that they do not exhibit extensive
electronic communication through the azine unit. In a more
recent study, we reported (39) the redox behavior of an ex-
tensive series of substituted acetophenone azines with em-
phasis on their extent of conjugation. The standard reduction
potentials (E°) were found to vary as a function of
substituent, with electron donating substituents causing the
molecules to reduce at more negative potentials than those
containing electron withdrawing groups. The azines were
found to preferentially reduce at the C=N bond closest to the
aryl group substituted with the most electron withdrawing
substituent. Based on our extensive substituent-effect study,
in this investigation (39) the azine unit was found to impede
conjugation, and the limited electronic effects measured
could have been purley inductive, supporting the notion of
Glaser (20–29).
The electrochemical behavior of ferrocenyl azines has
been previously investigated in a few cases (2, 4). Xu-Bing
et al. (4) report the redox behavior of ferrocene coupled to
bipyridine by an azine linkage, and the cyclic voltammetry
(CV) plot shows the expected one-electron reversible oxida-
tion typical of ferrocene-containing molecules. In addition,
two-reduction processes were reported by the authors and at-
tributed to the reduction of the -CHNNC(C10H6N2) group,
which is consistent with the reduction of other types of az-
ines (30, 35–39). Osborne et al. (2) report the electrochemis-
try of several azines and related compounds that contain one
or two ferrocene groups. The single ferrocene compounds
exhibit the expected one-electron reversible oxidation with
standard oxidation potentials in the range of +0.38 to +0.67 V
vs. SCE in 0.2 M tetrabutylammonium perchlorate – CH2Cl2.
For compounds containing two ferrocenyl groups, including
the azine 1-Fc/Fc in Scheme 1, the CV show two almost
overlapping reversible one-electron oxidations. This indi-
cates that the electronic communication between the two
ferrocene subunits through the azine moiety is small. No re-
duction processes were observed up to –0.5 V vs. SCE in
0.2 M tetrabutylammonium perchlorate – CH2Cl2 solution.
To extend our study into the electronic nature of azines, a
series of azines containing a ferrocene group were investi-
gated by cyclic voltammetry (CV). Ferrocene was chosen as
a substituent because it is an excellent electron donor (1) and
when coupled to a substituted aryl azine containing an elec-
tron acceptor, the resultant compound could potentially ex-
Ar1
H
N
CH3
N
H
N
N
Ar1
2-Ar1/H
2-Ar1/Ar1
Ar1
=
,
hibit the NLO response seen in other azines (24, 25) and
ferrocene-containing molecules (11–13). Each ferrocene
containing molecule investigated for the various applications
listed above took advantage of the same general strategy,
mainly, to produce molecules where electronic communica-
tion between ferrocene and a second group was realized usu-
ally by the use of a conjugated bridge. For this reason, the
azine unit is synthetically simple to incorporate and could
serve as an ideal linkage to allow for the electronic commu-
nication between ferrocene and an aryl group; results with
other azines, however, show that the azine unit will not al-
low for conjugative interaction. In the absence of any
conjugative effects, the versatility of the azine unit as a
linker and the ease in which azines can be synthesized
makes them ideal synthons for coupling electron donating
and acceptor groups for many applications. An important as-
pect of any application is an understanding of their elec-
tronic properties and that is what we describe here. The
various compounds examined are illustrated in Scheme 1,
along with a shorter notation used within the text to quickly
identify each molecule. The terms Fc, Ar, Anth, Py, and
Pyrd refer to ferrocene, aryl, anthracene, pyrene, and
pyridine, respectively. In addition, nitro, cyano, and methoxy
substituted phenyl rings will be denoted by 4-NO2, 4-CN,
and 4-OCH3, respectively, where the number four refers to
para substitution of the phenyl rings. An unsubstituted
phenyl ring will be denoted by H (e.g., 2-Anth/H).
Results and discussion
The electrochemical behavior of the series of ferrocenyl
azines (1-Fc/Ar) was studied in N,N-dimethylformamide
(DMF), using cyclic voltammetry (CV). The electrochemical
characteristics depend on the structure of the aryl
© 2002 NRC Canada

252
Can. J. Chem. Vol. 80, 2002
Scheme 2.
2
+ e_
- e_
+ e_
- e_
+ e_
- e_
N
R
N
R
N
R
N
R
Fe
Fe
Fe
Fe
N
N
N
N
Ar
Ar
Ar
Ar
2
1-Fc/Ar
1-Fc/Ar
1-Fc/Ar
1-Fc/Ar
Scheme 3.
2
N
N
N
N
e-
e-
2H+
Fe
Fe
Fe
N
HN
CN
CN
CN
substituent. Ferrocenyl azines that contain a para-substituted
aryl group exhibit electrochemical behavior similar to that
observed for the acetophenone azines (39). Figures 1a–e
shows a series of cyclic voltammograms for acetylferrocene-
4-cyanoacetophenone azine (1-Fc/4-CN) in 0.1 M TEAP–
DMF using various scan directions and switching potentials.
A characteristic voltammetric feature is a single electron an-
odic wave at positive potentials that is reversible at all scan
rates investigated (the wave labeled A in Fig. 1e) and is
common to all of the 1-Fc/Ar investigated. The anodic
waves are due to formation of the azine radical cation (1-
in Table 1. The reduction potentials of 1-Fc/Ar exhibit a sig-
nificant substituent effect with values differing by more than
1.2 V between 1-Fc/4-NO2 and 1-Fc/4-OCH3. The more
electron withdrawing the substituent the easier the azine is
°
to reduce and the more positive the E° value. The E0/•− are
comparable to the E° of the similarly substituted
acetophenone azines but with a 4-methoxyphenyl or 4-
dimethylaminophenyl group in place of the ferrocene unit
(39). This illustrates that ferrocene has similar electron do-
nating characteristics as the two aryl groups, as has been
previously suggested by other researchers (10, 12). The re-
ductions of 1-Fc/Ar are single electron based on the ip / ν
values in comparison with the FeII/FeIII couple.
C+
Fc /Ar) and the position of the oxidation is typical for
ferrocene-containing molecules, representing the oxidation
of the iron core from the FeII to the FeIII state and vice versa
(2). Other voltammetric features include a cathodic wave at
negative potentials that is reversible at all scan rates investi-
gated (wave B in Fig. 1e), followed at more negative poten-
tials by a second single electron cathodic wave (wave C in
Fig. 1e). The two reduction waves represent the successive
reduction of the corresponding azine to its radical anion (1-
The dianions of 1-Fc/4-CN, 1-Fc/Pyrd, 1-Fc/4-OCH3, and
1-Fc/4-NO2 are not reversible on the CV timescale up to
50 V s–1 in DMF. Typical CVs at low scan rates are also il-
lustrated in Figs. 1e–h. The potentials (EpC–/2–) for the sec-
ond reduction follow the same trend with substituent
donating and withdrawing ability (see Table 1) and become
easier to reduce when electron withdrawing substituents act
to reduce the electron density from the radical anion. The
substituent effect is slightly smaller for the second wave
with potentials differing by 0.8 V through the series 1-Fc/4-
NO2 to 1-Fc/4-OCH3. Once again, the EpC–/2– values indicate
that ferrocene has comparable electron donor capabilities as
4-methoxyphenyl and (or) 4-dimethylaminophenyl in com-
parison to the appropriate acetophenone azines (39). The
second reduction wave observed for 1-Fc/4-OCH3, however,
may not be due to dianion formation but may result from re-
duction of the azomethine bond adjacent to the ferrocenyl
group since the EpC–/2– is comparable to that observed for the
C–
Fc/Ar ) and then its dianion (1-Fc/Ar2–), respectively. This
series of events is shown schematically in Scheme 2.
When the CV is scanned in the negative potential direc-
tion and goes through the formation of dianion, minor oxida-
tion waves are observed on the reverse scan. These processes
are evident only when the scan goes through the formation
of 1-Fc/Ar2– and result from the protonation or some other
reaction of 1-Fc/Ar2–. The dianion is sufficiently basic to be
protonated by solvent molecules or, more likely, residual
H2O. Reduction and protonation most likely occur at the
C=N moiety adjacent to the substituted aryl group to give a
hydrazone (Scheme 3) similar to that observed in the reac-
tions of acetophenone azine radical anions and dianions with
trifluoroethanol (TFE) (39). Reduction of the azomethine
bond adjacent to the ferrocenyl group occurs at much more
negative potential as evidenced by the irreversible reduction
at –2.303 V vs. SCE for 1-Fc/Fc (see below). The identity of
the minor products were not determined.
C–
first reduction of 1-Fc/Fc to 1-Fc/Fc . The exact nature of
this reduction process could not be investigated because it
occurs at very negative potentials and into the solvent dis-
charge.
The reduction of the nitro containing ferrocenyl azine
warrants further discussion. We previously reported (39) that
the reduction of all the nitro containing acetophenone azines
occurs at almost identical potentials (~ –1 V vs. SCE). This
was attributed to the fact that the reduction appears to be
that of azine substituted nitrobenzene molecules and not an
aryl-nitro substituted azine. Similarly, reduction of 1-Fc/4-
NO2 occurs at –0.956 V vs. SCE and supports our earlier
suggestion that reduction is occurring at the Ar-NO2 group
and not at one of the azine C=N imine bonds. The reduction
Azines 1-Fc/Pyrd, 1-Fc/4-OCH3, and 1-Fc/4-NO2 showed
similar voltammetric features as 1-Fc/4-CN described above.
In the cases of 1-Fc/4-CN, 1-Fc/Pyrd, 1-Fc/4-OCH3, and 1-
C–
Fc/4-NO2 (Figs. 1e–h) the formation of 1-Fc/Ar is com-
pletely chemically reversible (dashed line). For this series,
°
standard reduction potentials (E0/•−) vary as a function of the
electronic properties of the substituent and are summarized
© 2002 NRC Canada

Sauro and Workentin
253
Fig. 1. Cyclic voltammograms of selected azines measured in
0.1 M TEAP–DMF at 0.2 V s–1. Arrows indicate the scan direc-
tion. Parts a–e are of 1-Fc/4-CN; (a) CV starting at 0 potential
and switching after the appearance of wave B. (b) CV starting at
0 potential and switching after the appearance of wave C. (c) CV
starting at 0 potential and switching after the appearance of wave
A. (d) CV starting at 0 potential and switching after the appear-
ance of wave B and continuing into the positive potential region
and switching again after the appearance of wave A. (e) CV start-
ing at 0 potential and switching after the appearance of wave C
and continuing into the positive potential region and switching
again after the appearance of wave A. Parts f–h are the CVs of 1-
Fc/Pyrd, 1-Fc/4-OCH3, and 1-Fc/4-NO2, respectively. Solid lines
are the CV going through radical anion and dianion formation and
continuing into positive potentials where the radical cation is
formed. Dashed lines are the CV when the potential is reversed
after formation of the radical anion and continuing into positive
potentials where the radical cation is formed.
dianion formation in Fig. 1. Scanning to even more negative
potentials (past the formation of dianion) produces two
multi-electron waves that are due to further reduction of the
protonated 1-Fc/4-NO2 dianion to ultimately form amines.
1-Fc/Py shows a one-electron reversible oxidation due to
formation of a radical cation, similar to the other ferrocenyl
azines discussed thus far (A in Fig. 2a). However, 1-Fc/Py
was the only ferrocenyl azine that exhibits a reversible cath-
odic reduction for the formation of the both the radical anion
°
and dianion and provides values for E0/•− = –1.549 V vs.
°
SCE and E•−/ 2− = –1.995 V vs. SCE, respectively. This is
typical for pyrene-containing molecules (41). Reversing the
potential after the formation of the radical anion (B) results
in a completely reversible wave. Continuing into the positive
potential region results in the CV showing two one-electron
reversible waves for the reduction and oxidation of 1-Fc/Py
(dashed line in Fig. 2a). No oxidation waves were observed for
2-Py/H or 1-pyrenecarboxaldehyde up to +1 V whose electro-
chemical behavior were examined for comparison to 1-
Fc/Py. Formation of the azine (i.e., 1-Fc/Py and 2-Py/H) re-
sults in a negative potential shift for the formation of the
pyrene radical anion in comparison to 1-pyrenecarbox-
aldehyde. It must be noted that in the case of the azines, it is
the C=N bond that is reduced, but the C=O bond is reduced
in 1-pyrenecarboxaldehyde. In general, the azomethine
group is more easily reduced than the carbonyl group from
which it is derived (40). Thus a shift to more positive reduc-
tion potentials is expected by replacing the oxygen of a C=O
bond with a nitrogen. In the case of the azines, the observed
shifts relative to the parent carbonyl compounds is to more
negative potentials, suggesting that the effects of replacing
the C=O with C=N are offset by the conjugative or inductive
effects of the azine unit and its second substituent. For 1-
Fc/Py, the negative shift in reduction potential indicates that
there is some electronic communication through the azine
moiety with the electron donating ferrocenyl group of 1-
Fc/Py having a larger effect (shift of –0.2 V) than the phenyl
ring of 2-Py/H (shift of –0.12 V). Reduction to the dianion
shows only a small effect due to azine formation.
An additional process is also observed in the CVs of the
pyrene-containing molecules. The wave for the formation of
the dianion is followed very closely by a third reversible reduc-
tion wave that can be clearly identified in the CV of 2-Py/H
(Fig. 2b). In the cases of 1-Fc/Py and 1-pyrenecarboxaldehyde,
the formation of dianion is coincidental with the third reduc-
tion process. This results in a wave that appears as a single
reversible reduction wave (at approximately –1.9 V), but it is
not completely reversible since the cathodic and anodic
peak-current ratios are greater than unity. At faster scan
rates, the peak-current ratio does approach unity, indicating
that the rate of the third reduction process is slow enough to
be outrun by the scan rate. It was noticed that the appear-
ance of the third reduction process was dependant on the
quality of the solvent used for the experiment (i.e., [H2O]).
In meticulously dried solvent, the third reduction process for
2-Py/H was almost not observable. As well, addition of an
acid (TFE) to the electrochemical solution of 2-Py/H, caused
an increase in the current for the third reduction. This third
reduction process is due to the reduction of the product de-
rived from protonation of the dianion by solvent molecules
or any residual water present. The formation of this
potentials of the nitro containing azines all occur within 44
mV of each other and their values are all approximately
equal to –1 V vs. SCE, roughly the same potential as
nitrobenzene in the same solvent (–1.1 V vs. SCE). 1-Fc/4-
NO2 also shows a second less well-defined cathodic wave
(Fig. 1h) that is due to further reduction of the aryl nitro
group, presumably to form hydroxylamines, which are
known to result from reduction of aryl-NO2 compounds
(40). The second cathodic wave is not completely reversible
and includes other electrochemical processes that make it
appear oddly shaped in comparison with other waves due to
© 2002 NRC Canada

254
Can. J. Chem. Vol. 80, 2002
Table 1. Electrochemical data for the oxidation and reduction of ferrocenyl and related azines in
0.1 M TEAP–DMF solution.
Azine
X
Y
E
(V)a,b
E 0/ − (V)a
(V)a
º
º
º
E
−/2−
+ /0
Fc
Fc
Anth
Anth
9-Anthraldehyde
Fc
Py
1-Pyrenecarboxaldehyde
Fc
Fc
Fc
4-NO2
Anth
H
0.612
0.583
–0.956
–1.437c
–1.403c
–1.232c
–1.207c
–1.549
–1.462
–1.338
–1.645
–1.767
–2.158
–2.303g
–2.262
–1.642d
f
—
—
e
f
—
—
—
e
Anth
–1.822
e
f
—
Py
H
0.576
–1.995
–1.859
–1.929
–2.029d
–2.115d
–2.439d
e
—
—
e
4-CN
Pyrd
4-OCH3
Fc
0.577
0.579
0.553
0.572
0.691
f
Fc
—
—
f
Acetylferrocene
aReported in V vs SCE.
bOxidation potential of the FeII/FeIII couple.
cReversible only at faster scan rates.
dPeak potential (EpC–/2–) at 0.1 V s–1 scan rate.
eNo oxidation process was observed up to +1 V.
fNo further reduction processes were observed.
gPeak potential at 0.1 V s–1; see text.
Fig. 2. Cyclic voltammograms of selected compounds measured
in 0.1 M TEAP–DMF at 0.2 V s–1: (a) 1-Fc/Py; (b) 2-Py/H; and
(c) 1-pyrenecarboxaldehyde. Arrows indicate scan direction.
Dashed lines are the CV when the potential is reversed after
radical anion formation.
Fig. 3. Cyclic voltammograms of selected compounds measured
in 0.1 M TEAP–DMF at 0.2 V s–1. Arrows indicate scan
direction: (a) 1-Fc/Anth; (b) 2-Anth/H; (c) 2-Anth/Anth; and
(d) 9-anthraldehyde.
dianion. Protonation would interrupt any conjugation that
may exist and result in a minor ferrocenyl-containing prod-
uct that is easier to oxidize than the original azine.
The anthracenyl-containing molecules exhibit different
electrochemical behavior compared with the others. Fig. 3a
shows the CV of 1-Fc/Anth, which features a completely re-
versible one-electron wave (A) at positive potentials, as well
as an irreversible one-electron cathodic wave (B) at slow
protonated product is also responsible for the minor anodic
peaks on the return scan (between –0.5 and –0.8 V, depend-
ing on scan rate) due to oxidation of the formed conjugate
base for each of the pyrene containing azines. As well, the
protonated product is also responsible for the minor peak
preceding wave A for 1-Fc/Py (Fig. 2a), which is not ob-
served unless the scan goes through the formation of a
© 2002 NRC Canada

Sauro and Workentin
255
Scheme 4.
Table 2. Electrochemical data and dimerization rate constants for
anthracene containing azines.
R2
Ep (at 0.2 V s–1)a
Eo a
kd (M–1 s–1)
N
N
R1
Compound
H
H
-
H
H
O-
1-Fc/Anth
2-Anth/H
2-Anth/Anth
9-Anthraldehyde
aIn V vs. SCE.
–1.445
–1.397
–1.224
–1.203
–1.437
–1.403
–1.232
–1.207
5.1 × 104
1.5 × 105
3.3 × 105
1.8 × 105
R1 = CH3
,
,
R2
=
II
Fe
H
H
H
R1 = H ,
R2
R2
=
=
III
IV
R1 = CH3
-
N
H
O-
for dimerization (kd) can be calculated using known proce-
dures (43) and appear in Table 2. The kd value for 1-Fc/Anth
suggests that the electron donating abilities of the ferrocenyl
group act to hinder dimer formation. The rates are approxi-
mately equal for the remaining anthracenyl azines.
R1
N
I
R2
scan rates (i.e., 0.1–0.5 V s–1) that becomes quasi-reversible
at faster scan rates (i.e., 15–65 V s–1). The Ep shifts to more
negative potentials with increasing scan rate and does not
vary linearly with the logarithm of the scan rate (log ν). Also
present in the CV for 1-Fc/Anth is an anodic wave (C) with
an Ep at approximately –0.9 V whose intensity decreases
relative to the peak current of the irreversible wave as the
scan rate is increased. The Ep of wave C shifts towards more
positive values with increasing scan rate. This suggests that
the anodic wave C is due to a product formed during the re-
duction process at wave B. The decrease of relative intensity
of wave C compared with wave B with increasing scan rate
indicates that the formation of the new species is outrun at
faster scan rates. Reversing the scan direction towards more
negative potentials after the appearance of oxidation wave C
results in a CV that is identical to the one obtained on the
initial scan, indicating that the product formed upon azine
reduction is oxidized back to the original azine. This electro-
chemical behavior of 1-Fc/Anth is consistent with electro-
chemically reversible reduction followed by rapid
dimerization. The formed dimer can then be oxidized back
to starting azine. Digital simulation of the CV of a reversible
reduction followed by dimerization produces similar features
(i.e., Ep shifts to more negative values and non-linear Ep vs.
log ν) as are observed for the experimental data. Similar
voltammetric features are observed in the CVs for other
anthracenyl-containing model compounds (Figs. 3b–d) 2-
Anth/H, 2-Anth/Anth, and 9-anthraldehyde, which do not
show any oxidation processes up to +1 V.
The reduction of 9-anthraldehyde (used as a model com-
pound to examine substituted anthracene reduction, Fig. 3d)
was examined in 0.1 M TEAP–DMF solution by Lasia and
Rami (42). It exhibits similar voltammetric features upon re-
duction as 1-Fc/Anth and the other anthracenyl azines. The
one-electron cathodic wave results from the formation of the
9-anthraldehyde radical anion that quickly couples (kd = 1 ×
105 M–1 s–1) (42) with a second radical anion to form a
dianionic dimer (I in Scheme 4) and results in the observed
one-electron irreversible reduction wave. The anodic wave at
~ –0.5 V on the reverse scan results from the oxidation of
the dimer back to 9-anthraldehyde (42) and is not reversible
upon switching the scan direction once again. In the case of
1-Fc/Anth, 2-Anth/Anth, and 2-Anth/H, the corresponding
dimers II, III, and IV (Scheme 4) are proposed to account
for the oxidation waves observed at –0.927, –0.614, and
–0.772 V vs. SCE at 0.1 V s–1, respectively. Rate constants
For 1-Fc/Anth, the electrochemical behavior is governed by
two independent moieties, namely the ferrocene and the
anthracene groups. The ferrocene moiety has the effect of mak-
ing the anthracenyl unit more difficult to reduce as evidenced
by a shift of 0.24 V to more negative potentials relative to 9-
anthraldehyde. This indicates that some electronic communica-
tion must be occurring through the azine unit. A similar situa-
tion is observed for 2-Anth/H. The reduction of 2-Anth/Anth
also shows evidence for electronic communication. Although
the reduction potential of 2-Anth/Anth occurs at the roughly
same potential as 9-anthraldehyde, only one electron is in-
volved in the process. One might expect a two-electron wave
for 2-Anth/Anth if the azine moiety prevented any communica-
tion between the two anthracene units (i.e., reduction of both
anthracene units simultaneously, similar to the situation previ-
ously observed for 4,4′-dinitroacetophenone azine (39)).
The electrochemical behavior of acetylferrocene and
acetylferrocene azine (1-Fc/Fc) was also examined for com-
parison with the results obtained for the other 1-Fc/Ar. 1-
Fc/Fc shows an anodic wave that consists of two one-
electron reversible oxidations that have slightly different oxi-
dation potentials (Fig. 4b). This results in an oxidation wave
that is approximately twice the magnitude of the other one-
electron oxidations (Fig. 5) and appears to have a shoulder
on the leading edges of the waves. This is consistent with
the electrochemistry observed for 1-Fc/Fc by Osborne et al.
(2) who estimates the oxidation potentials of the two
ferrocene units to be +0.47 V and +0.56 V vs. SCE in 0.2 M
TBAP–CH2Cl2. 1-Fc/Fc also shows a reduction wave at very
negative potential caused by the reduction of the C=N bond
of the azine unit. This reduction occurs into the solvent dis-
charge, which makes determining the standard potential dif-
ficult. Reduction of a C=N bond at such negative potentials
results in the formation of minor electroactive species,
which are oxidized on the reverse scan and have similar
structure to the parent azine. These products most likely re-
sult from reaction of the azine radical anion with H2O or
electrolyte or some reduced form of the electrolyte. Figure 4
illustrates the CVs of 1-Fc/Fc as well as the reference com-
pound, acetylferrocene. The reduction that occurs in the CV
of acetylferrocene is the one-electron reduction of the carbonyl
group. All electrochemical data is summarized in Table 1.
The FeII/FeIII couple in the 1-Fc/Y is subject to a
substituent effect as indicated by the oxidation potentials of
the two model compounds, ferrocene and acetylferrocene
(+0.470 V and +0.691 V vs. SCE in DMF, respectively). The
© 2002 NRC Canada

256
Can. J. Chem. Vol. 80, 2002
Fig. 4. Cyclic voltammograms of (a) acetylferrocene and (b) 1-
Fc/Fc measured in 0.1 M TEAP–DMF at 0.2 V s–1. Arrows indi-
cate scan direction. Dashed line is the CV in the positive
potential region only.
Fig. 5. Cyclic voltammograms of the reversible oxidation waves
for the ferrocenyl azines measured in 0.1 M TEAP–DMF at 0.2
V s–1 scan rate: (a) 1-Fc/Fc; (b) 1-Fc/4-NO2; (c) 1-Fc/Anth;
(d) 1-Fc/Py; (e) 1-Fc/4-CN; (f) 1-Fc/Pyrd; and (g) 1-Fc/4-OCH3.
be determined in this way because they do not exhibit suit-
able reversible waves, a requirement for the calculation.
The 1-Fc/Ar compounds offer the opportunity to calculate
the HOMO–LUMO energy gap (listed in Table 3). The en-
ergy difference between oxidation and reduction potentials is
the HOMO–LUMO energy gap and is an indication of the
electron donor–acceptor properties of the compound. The
variation of ∆EHOMO–LUMO as a function of the substituent
(values differ by more than 30 kcal mol–1) illustrates the
“fine tuning” possible for azine electronic properties simply
by changing the substituent(s) attached to the azine unit.
To further our examination of the electronic nature of az-
ines, the photochemical behavior of a series of azines con-
taining a chromophoric group (specifically anthracene) was
investigated. This allowed for the evaluation of the effect of
the ferrocene group on the photochemical characteristics of
these azines. Azine photochemistry is governed by E/Z
isomerization of the azomethine bonds (45–51), which has
been shown to be a general property of the 2,3-diaza-1,3-
butadienes. The photochemistry and photophysics of 2-
Anth/Anth has been thoroughly investigated by Appenroth et
al. (48–51). 2-Anth/Anth can exist as three isomers, E/E,
E/Z, and Z/Z, where the E/Z and Z/Z isomers can be selec-
tively produced, starting from the thermodynamically stable
E/E form, by irradiation with wavelengths λ = 488 and
436 nm, respectively. It was found that neither of the three
isomers fluoresce at room temperature and only the Z/Z iso-
mer fluoresced at 77 K (51). The fluorescence spectrum of
(Z/Z)-2-Anth/Anth contained the same structural features as
anthracene and its quantum yield of fluorescence in ethanol
was determined to be Φf = 0.090 using an excitation wave-
length of 350 nm at 77 K.
oxidation potentials of the ferrocene unit can be used to probe
the effects of adding an azine unit and any possible contribu-
tion from the aryl group to the electrochemical behavior of
the iron core. The CVs showing the reversible oxidation of 1-
C+
Fc/Ar to 1-Fc/Ar at 0.2 V s–1 are shown in Fig. 5. As ex-
pected, the ferrocenyl unit becomes easier to oxidize (less
positive potentials) with more electron donating substituents
°
and vice versa. All E•+/ 0 are within 59 mV of each other, in-
dicating only a modest substituent effect on the oxidation.
This reinforces the notion of little to no conjugation through
the azine unit, producing only small substituent effects, par-
ticularly on the Fe core. A more illustrative example of the
extent of communication through the azine unit is the oxida-
tion of 1-Fc/Fc that shows two oxidation processes that occur
at slightly different potentials. If there was no conjugation
through the azine unit, we would expect that the oxidation po-
tentials would be coincidental and that the observed wave
would appear as if it were a completely reversible two-
electron reduction process. That is, both ferrocene units
would be oxidized simultaneously. Thus the oxidation of 1-
Fc/Fc would be similar to the simultaneous double one-
electron reduction of 4,4′-dinitroacetophenone azine (39).
Since the oxidation of 1-Fc/Fc appears as two processes at
slightly different potentials, there must be some electronic
communication through the azine moiety that results in the
second oxidation being slightly harder to accomplish (i.e., oc-
curs at more positive potential) than the first.
The heterogeneous rate constants (khet) for electron trans-
fer can be calculated (43, 44) for the compounds investi-
gated and are listed in Table 3. The calculated khet are within
the expected range (10–10–3 cm s–1) for reversible or quasi-
reversible electrode kinetics. Rate constants for compounds 1-
Fc/Fc, 2-Anth/Anth, 2-Anth/H, and 9-anthraldehyde could not
To investigate the luminescent properties, the fluorescence
spectra of the anthracenyl-containing azines were recorded
© 2002 NRC Canada

Sauro and Workentin
Table 3. Heterogeneous rate constants and HOMO–LUMO energy gap for ferrocenyl and related
257
azines.
∆EHOMO–LUMO
khet (cm s–1)b
E 0/ − (V)a
Compound
(kcal mol–1)c
º
º
E
(V)a
+ /0
1-Fc/4-NO2
0.612
–0.956
4.9 × 10–2 d
5.0 × 10–2 e
3.1 × 10–2 e
1.9 × 10–2 d
2.5 × 10–2 e
2.4 × 10–2 d
2.7 × 10–2 d
1.9 × 10–2 d
3.1 × 10–2 e
1.4 × 10–2 d
1.6 × 10–2 e
1.1 × 10–2 d
2.0 × 10–2 e
—
36.2
1-Fc/Anth
1-Fc/Py
0.583
0.576
–1.437
–1.549
46.6
49.0
2-Py/H
1-Pyrenecarboxaldehyde
1-Fc/4-CN
—
—
0.577
–1.462
–1.338
–1.645
—
—
51.2
1-Fc/Pyrd
0.579
0.553
–1.767
–2.158
54.1
62.5
1-Fc/4-OCH3
1-Fc/Fc
Acetylferrocene
0.572
0.691
–2.303
–2.262
66.3
68.1
2.4 × 10–2 d
1.3 × 10–2 e
aReported in V vs SCE.
bAverage value for scan rates between 0.05 and 0.5 V s–1.
cEnergy difference between the HOMO and LUMO energy levels determined from electrochemical data.
dValue calculated for a reduction process.
eValue calculated for an oxidation process.
the E/E isomer was shown to be non-fluorescent. The sam-
ples were then exposed to radiation of wavelength 410–
490 nm to induce isomerization to the Z/Z isomer and the
fluorescence was reexamined. Once again, little to no emis-
sion was observed at room temperature. In contrast, the
emission intensity at 77 K was dramatically increased, ap-
proximately 12–70 times relative to the samples before irra-
diation. The low-emission intensity at room temperature can
be attributed to efficient energy dissipation in the excited
state via E/Z isomerization of the azomethine bonds. This
isomerization process becomes less important at 77 K with a
concomitant enhancement of fluorescence since the azine
molecules are now contained in the solid glass of the frozen
EPA solution. The emission observed for 1-Fc/Anth, 2-
Anth/H, and 2-Anth/Anth contains similar structural features
as that of anthracene (52).
Fig. 6. The emission spectra of 1-Fc/Anth, 2-Anth/H, and 2-
Anth/Anth in EPA at 77 K using a 350 nm excitation
wavelength: (a) 2-Anth/H; (b) (dashed line) 2-Anth/Anth; (c) 1-
Fc/Anth.
Appenroth et al. (51), attributed the observed emission of
(Z/Z)-2-Anth/Anth to the anthracene chromophore that is
twisted out of the molecular plane of the azine moiety in the
Z/Z isomer (53, 54). The twisting results from steric interac-
tions between the azine nitrogen and the anthracene unit.
This results in an interruption of any conjugation that may
be occurring. The E/Z isomer of 2-Anth/Anth did not show
any emission at either temperature despite twisting of one of
the anthracene units. The situation is very similar for 1-
Fc/Anth and 2-Anth/H. The need for irradiation at 410–
490 nm prior to the observation of any emission suggests
that the E/E isomers of 1-Fc/Anth and 2-Anth/H, the ther-
modynamically most stable and initial structures, do not flu-
oresce and any fluorescence must come from one of the
other two isomers. Similar to 2-Anth/Anth, long-wavelength
irradiation induces isomerization of the C=N bonds and
results in fluorescence not observed for the E/E isomers.
Emission from the E/Z or Z/Z isomers of 1-Fc/Anth and 2-
in EPA both at room and low temperature (77 K). Figure 6
shows the fluorescence spectra of 1-Fc/Anth, 2-Anth/H, and
2-Anth/Anth at 77 K. None of the azines showed fluores-
cence at room temperature, and only weak fluorescence at
low temperature when an excitation wavelength of 350 nm
was used. Such behavior was observed previously for sev-
eral different azines including 2-Anth/Anth (48, 51), where
© 2002 NRC Canada

258
Can. J. Chem. Vol. 80, 2002
Table 4. Quantum yields (Φf ) and total normalized fluorescence
intensities of 1-Fc/Anth, 2-Anth/H and 2-Anth/Anth in EPA at
room temperature and 77 K.
Scheme 5.
N
N
O
N
H
N+
O–
N+
O
H
N
O–
Fe
Fe
Before irradiation
After irradiation
a
Azine
Φf
r.t.
77 K
r.t.
77 K
1-Fc/4-NO2
1-Fc/Anth
2-Anth/H
2-Anth/Anth
0.176
0.042
0.090
0.3
0.08
0.14
3.4
5.1
7.2
20
1
1.7
242.1
58.4
123.7
0.0002–0.0025 Å) and parallel to each other (angle between
rings of 0.9–3.5°). The C—C of the cyclopentadienyl rings,
N—N, and C=N bond lengths are similar to those deter-
mined for other ferrocenyl azines (2, 4, 12).
Note: Fluorescence intensities are normalized relative to the area under
the room temperature emission curve of 2-Anth/H after irradiation at 410–
490 nm.
aQuantum yields in EPA at 77 K using 350 nm excitation wavelength
relative to that of 2-Anth/Anth (51). The concentrations of the fluorescence
samples were adjusted so that absorption at 350 nm was equivalent.
One striking feature of the determined 1-Fc/Ar structures
is the twisting of the molecules through the azine moiety as
indicated by >C=N-N=C< dihedral angles of 99.1–160.9°
(see Supplementary material). This is in complete contrast
with the structure of ferrocenealdehyde-4-nitrobenzaldehyde
azine, reported by Houlton and co-workers (12), illustrated
in Scheme 5, whose X-ray structure shows an almost com-
plete planarity of the C5H4CHNNCHC6H4NO2 fragment to
produce an extended π system. The similar azine 1-Fc/4-
NO2 (Scheme 5), which differs only by the addition of two
methyl groups, produces a significantly different orientation.
This suggests that 1-Fc/Ar should not exhibit any conjuga-
tion through the azine unit. The electrochemical results sug-
gest, however, that there is significant conjugative
interactions in the 1-Fc/Ar in solution. The twisting of the
molecules observed in the solid state, therefore, may not be
a good reflection of their orientation in solution and may be
caused by packing effects in the solid. The same argument
was proposed by Glaser and co-workers (22, 25) to explain
the nonplanarity of the acetophenone azines in the solid
state.
Anth/H cannot, however, be distinguished since it is unclear
whether both C=N bonds need to be in the Z configuration
or only the C=N bond nearest the anthracenyl group.
The fluorescence quantum yields (Φf ) of 1-Fc/Anth and 2-
Anth/H were determined by comparison to the Φf previously
determined for 2-Anth/Anth (51) under similar experimental
conditions. The determined quantum yields at 77 K are
listed in Table 4 along with the normalized fluorescence in-
tensities. Table 4 shows that, in frozen solution where the
isomerization is prevented, the fluorescence of 1-Fc/Anth is
twice that of 2-Anth/Anth and four times that of 2-Anth/H.
The emission intensity of 2-Anth/Anth is approximately
twice that of 2-Anth/H under all conditions owing to the two
anthracenyl groups in 2-Anth/Anth compared with one in 2-
Anth/H. The fluorescence intensity of 1-Fc/Anth at room
temperature after irradiation was significantly higher than
the other two azines. This suggests that radiationless deacti-
vation processes are less favoured for 1-Fc/Anth than for 2-
Anth/H and 2-Anth/Anth in fluid solution, and that the
ferrocene group may in fact hinder the E/Z isomerization
that is the main deactivation mechanism at room tempera-
ture. In addition, fluorescence quenching by ferrocene does
not appear to dominate, either at room temperature or 77 K,
in comparison to the observed fluorescence of 2-Anth/H and
2-Anth/Anth. This suggests that (Z/Z)-1-Fc/Anth (or (E/Z)-1-
Fc/Anth) does not exhibit significant electronic communica-
tion between the two azine substituents either by through-
bond or through-space interactions since ferrocene is known
to be a good fluorescence quencher (15, 55). Any fluores-
cence observed before irradiation presumably results from
minor amounts of E/Z or Z/Z isomers that may be present in
the initial solution, formed by thermal processes.
To extend the solid-state analysis of azines, crystals of 1-
Fc/4-NO2, 1-Fc/4-CN, 1-Fc/Pyrd, and 1-Fc/Anth suitable for
X-ray analysis were grown by slow evaporation of a solution
of the azine in an appropriate solvent. All C=N bonds are in
their E configurations as expected. The average Fe—C bond
distances of the ferrocene moiety are all similar (2.030–
2.040 Å) and fall within the range commonly found for
ferrocene compounds (2, 4, 56) but are shorter than those
found in ferrocene itself (2.045 Å) (57). This may be due to
any of the conjugative effects of the attached azine, which
pulls electron density from the cyclopentadienyl ring and re-
sults in the shortening of the average Fe—C bond lengths.
All the cyclopentadienyl rings are planar (mean deviation of
Conclusions
The electrochemical behavior of the 1-Fc/Ar is governed
by the ferrocenyl and aryl group attached to the azine moi-
ety. All the 1-Fc/Ar showed a reversible oxidation between
+0.62 and +0.55 V vs. SCE as a function of the aryl
substituent. Electron withdrawing groups resulted in more
positive oxidation potentials with only a slight substituent
effect on the FeII/FeIII oxidation, as evidenced by the small
variation (59 mV) in potentials. The reduction potentials of
1-Fc/Ar were determined and electron donating groups
showed more negative reduction potentials, whereas less
negative potentials were observed for electron withdrawing
substituents. For 1-Fc/Ar, where the aryl group was a substi-
tuted benzene ring, the effect of introducing an electron do-
nating ferrocenyl group on azine reduction results in the
ferrocene having similar electron donating characteristics as
a methoxyphenyl or dimethylaminophenyl group compared
with the reduction potentials determined for similar systems
(39). The reduction and oxidation potentials determined for
1-Fc/Ar suggest that there is some electronic communication
between the two substituents of the azine moiety in compari-
son with model compounds. Therefore, we suggest that the
azine unit does allow for limited conjugation in the 1-Fc/Ar
as it did for the acetophenone azines (39). The nonplanarity
of the 1-Fc/Ar observed in the solid-state structures, suggest-
© 2002 NRC Canada

Sauro and Workentin
259
ing no conjugation, may not be an accurate reflection of
their orientation in solution.
reference electrode was a silver wire housed in a glass tube
sealed with a porous ceramic tip and filled with a 0.1 M solu-
tion of tetraethylammonium perchlorate (TEAP). The same
reference electrode was used in all experiments. Concentra-
tions of 2 mM of azine in 0.1 M TEAP in DMF were used.
Ferrocene, tetracyanoethylene, or 4-nitroacetophenone azine
(39) were used in all experiments as internal references to a
saturated calomel electrode (E° = 0.470, –0.319, and –0.973
V vs. SCE in DMF, respectively). Positive feedback IR com-
pensation was used in all experiments to minimize the effects
of uncompensated solution resistance.
The photochemistry of 1-Fc/Anth, 2-Anth/Anth, and 2-
Anth/H is governed by E/Z isomerization processes and no
emission is observed for these compounds either at room
temperature or at 77 K. Fluorescence at 77 K is observed af-
ter long wavelength irradiation where the thermodynami-
cally most stable E/E isomer is converted to the E/Z and Z/Z
isomers from which emission occurs. Any conjugative inter-
actions in the E/E isomer is interrupted upon formation of
the E/Z or Z/Z isomer, allowing for fluorescence of the
anthracenyl group. The ferrocenyl group in 1-Fc/Anth has
only a slight effect on the anthracenyl fluorescence, enhanc-
ing it relative to 2-Anth/Anth and 2-Anth/H, and no evi-
dence for intramolecular or inductive quenching by
ferrocene was observed under the experimental conditions.
The donor–acceptor properties of ferrocenyl azines, however,
can be easily manipulated by altering the aryl substituent on
the azine unit, making them good candidates for materials
applications.
Fluorescence
Fluorescence spectra were recorded on an ISA Spex 3
Fluorimeter interfaced to a personal computer. EPA (2:5:5
mixture of ethanol – diethyl ether – isopentane) was prepared
from spectroscopic grade solvents and stored in a tightly
sealed bottle. Azine solutions were prepared by dissolving the
appropriate amount of azine in EPA so that the UV–vis
absorbance at 350 nm was the same for all samples and kept
below 0.07. Samples were kept in the dark as much as possi-
ble during preparation, transport, and storage. Azine solutions
were then placed in quartz tubes, purged with argon gas to re-
move oxygen and sealed with a rubber septa. The samples
were placed in the fluorimeter inside a low-temperature sam-
ple holder filled with liquid nitrogen (the same samples and
apparatus was used for room temperature spectra but without
the liquid nitrogen). The samples were checked for any cracks
in the frozen solvent and the fluorescence or excitation spec-
tra were recorded. Samples were removed from the liquid ni-
trogen and allowed to warm to room temperature between
runs. Any condensation that accumulated on the tubes during
the warming process was removed before immersing into the
liquid nitrogen again. The liquid nitrogen was frequently
changed in the sample holder to remove interference from ice
particles which form at the nitrogen-air interface. Photolysis
with light of wavelength 410–490 nm was performed using
the setup described below for the long wavelength
photolysis. The fluorescence samples were irradiated in the
same quartz tubes used for recording the spectra and placed
back in the fluorimeter. The samples were irradiated until no
increase in the cold emission intensity was observed, ap-
proximately 4 h.
Experimental
Materials
All chemicals used were purchased from Aldrich and used
as received unless otherwise stated. For electrochemical ex-
periments only, acetylferrocene, 9-anthraldehyde, and 1-
pyrenecarboxaldehyde were recrystallized from suitable sol-
vents prior to use. N,N-dimethylformamide(DMF) was dis-
tilled under reduced pressure over calcium hydride prior to
use. Argon was flushed over the DMF during the transfers
into the electrochemical cells. Tetraethyl ammonium per-
chlorate (Kodak) was recrystallized three times and dried
under vacuum at 60°C prior to use and stored in a vacuum
dessicator. Ferrocene was sublimed once before use. NMR
spectra, recorded on a Varian Gemini 300MHz (300.075 MHz
for 1H and 75.462 MHz for 13C) or Varian Mercury
400 MHz instrument (400.0854801 MHz for 1H and
100.6015500 MHz for 13C), are herein referred to as 300 or
1
400 MHz for H and 75 or 100 MHz for 13C. All chemical
shifts are reported in parts per million downfield from
tetramethylsilane. Melting points were recorded using a
Gallenkamp melting point apparatus and are uncorrected. IR
data was recorded on a PerkinElmer System 2000 FT-IR.
Mass spectra analysis was performed with a Finnigan MAT
8200 mass-data system. Gas chromatography analysis was
performed on a Hewlett–Packard 5890 Series II gas
chromatograph equipped with a 10 m HP5 column and an
FID detector. UV–vis data were collected on a Varian Cary
100 spectrometer. Fluorescence spectra were recorded using
an ISA Spex 3 Fluorimeter.
Long wavelength photolysis
Photolysis with light of wavelength 410–490 nm was per-
formed using a high-pressure mercury lamp surrounded by a
filter solution of CuSO4·5H2O (2.7 M NH3) and NaNO2 (58)
in a quartz reaction chamber. Photolysis was performed on
the fluorescence samples whose preparation is described above.
Synthesis of acetylferrocene azines (1-Fc/Ar)
Electrochemistry — cyclic voltammetry
Experiments were performed using either a Princeton Ap-
plied Research (PAR) 283 or 263 potentiostat interfaced to a
personal computer. PAR M270 software was used to conduct
the experiments as well as analyze the derived data. Measure-
ments were conducted in an all-glass cell kept at 25°C. All
glassware was predried in an oven and assembled, hot while
flushing with argon. The working electrode was glassy carbon
(Tokai) and the counter electrode was a platinum flag. The
The symmetric azines 1-Fc/Fc and 2-Anth/Anth were syn-
thesized according to a prescribed procedure and the spectral
characterization and other physical data matched that previ-
ously reported (2, 59). The asymmetrical azines were syn-
thesized by adding the appropriate substituted ketone or
aldehyde (1 equiv) to the corresponding acetylferrocene
(diethoxyphosphinyl) hydrazone (1 equiv) in benzene–NaH
(1.3 equiv). Both the preparation of the acetylferrocene
© 2002 NRC Canada

260
Can. J. Chem. Vol. 80, 2002
(diethoxyphosphinyl) hydrazone and azine were prepared
according to the published procedures (22, 39).
15.99, 14.81. MS m/z (%): 369 (M+, 100), 304 (15), 185
(57), 129 (29), 121 (37). Anal. calcd. for C21H19N3Fe:
369.0928; found: 369.0825.
Preparation of diethyl phosphorohydrazidate
Acetylferrocene-4-methoxyacetophenone azine (1-Fc/4-
OCH3)
Hydrazine hydrate (20 equiv) was added dropwise to a
stirred solution of anhydrous K2CO3 (15 equiv) and triethyl-
benzyl ammonium chloride (0.1 equiv) in CCl4–CH2Cl2
(4:7) at 20–25°C. A solution of diethylphosphite (1 equiv) in
CH2Cl2 was then added at 20–30°C. The stirring was contin-
ued for 6 h after which the K2CO3 crystals were filtered off
and rinsed with CH2Cl2. The solvents were removed and the
residue was kept under high vacuum for several hours. The
resultant pale yellow liquid was used without further purifi-
cation.
This compound was prepared by adding acetylferrocene
(1 equiv) to the corresponding p-methoxyacetophenone
(diethoxyphosphinyl)hydrazone (1 equiv) (22, 39) in
refluxing benzene containing NaH (1.3 equiv). A small
amount of product was then isolated by preparative HPLC.
1H NMR (300 MHz, CDCl3) δ: 2.18 (s, 3H), 2.27 (s, 3H),
3.84 (s, 3H), 4.22 (s, 5H), 4.40 (br s, 2H), 4.77 (br s, 2H),
6.92 (d, 2H), 7.83 (d, 2H). MS m/z (%): 379 (M+, 100), 309
(28), 185 (11), 121 (18). Anal. calcd. for C21H22N2OFe:
347.1081; found: 347.1084.
Preparation of acetylferrocene(diethoxyphosphinyl)-
hydrazone
Acetylferrocene-9-anthraldehyde azine (1-Fc/Anth)
mp 169–170°C. UV–vis (DMF) λ max (nm) (ε, M–1 cm–1):
266 (59 070). FT-IR (CCl4, cm–1): 3087, 3054, 2926, 2854,
Diethyl phosphorohydrazidate (1 equiv) and acetyl-
ferrocene were dissolved in benzene and heated at reflux for
5 h. The benzene was removed by rotary evaporation to
yield a crystalline sample. The solid was then purified by
recystallization from hexanes–benzene (2:1). The purity of
the sample was verified by GC and NMR.
1
1606, 1474, 1384, 1362, 1292, 1107, 1003, 909. H NMR
(300 MHz, CDCl3) δ: 2.54 (s, 3H), 4.46 (s, 5H), 4.47 (t, 2H),
4.84 (t, 2H), 7.53 (m, 4H), 8.03 (d, 2H), 8.52 (s, 1H), 8.79
(d, 2H), 9.66 (s, 1H). 13C NMR (75 MHz, CDCl3) δ: 156.63,
131.44, 130.64, 130.08, 128.93, 126.86, 125.43, 125.32,
70.80, 70.63, 69.56, 68.11, 16.76. MS m/z (%): 430 (M+,
100), 365 (66), 324 (20), 203 (31), 185 (29), 121 (64). Anal.
calcd. for C27H22N2Fe: 430.1132; found: 430.1137.
Acetylferrocene azine (1-Fc/Fc) (2)
1
mp 234–235°C. H NMR (300 MHz, CDCl3) δ: 2.16 (s,
6H), 4.17 (s, 10H), 4.34 (t, 4H), 4.71 (t, 4H). Anal. calcd.
for xxxx: 452.0637; found: 452.0632.
Acetylferrocene-1-pyrenecarboxaldehyde azine (1-Fc/Py)
mp 178–179°C. UV–vis (DMF) λ max (nm) (ε, M–1 cm–1):
380 (34 460). FT-IR (CCl4, cm–1): 3099, 3049, 2984, 2962,
9-Anthraldehyde azine (2-Anth/Anth) (60)
1
mp 273–275°C. H NMR (300 MHz, CDCl3) δ: 7.60 (m,
4H), 8.09 (d, 2H), 8.60 (s, 1H), 8.80 (br d, 2H), 10.15 (br s,
1H). Anal. calcd. for C30H20N2: 408.1626; found: 408.1621.
The following are, to the best of our knowledge new com-
pounds or the complete spectral data has not yet been re-
ported. X-ray crystallographic data is also available for some
1-Fc/Ar.
1
2927, 1742, 1607, 1475, 1361, 1291, 1242, 1107. H NMR
(300 MHz, CDCl3) δ: 2.58 (s, 3H), 4.30 (s, 5H), 4.47 (br s,
2H), 4.84 (br s, 2H), 8.15 (m, 7H), 8.71 (d, 1H), 8.93 (d,
1H), 9.49 (br s, 1H). 13C NMR (75 MHz, CDCl3) δ: 156.48,
148.01, 132.86, 131.30, 130.66, 130.07, 128.69, 128.56,
127.63, 127.48, 126.22, 126.16, 125.88, 125.68, 125.00,
124.62, 123.06, 70.84, 69.58, 68.06, 16.55. MS m/z (%): 454
(M+, 100), 389 (31), 348 (12), 227 (27), 201 (18), 185 (18),
129 (21), 121 (40). Anal. calcd. for C29H22N2Fe: 454.1132;
found: 454.1139.
Acetylferrocene(diethylphosphinyl)hydrazone
1H NMR (300 MHz, CDCl3) δ: 1.36 (dt, 6H), 2.05 (s, 3H),
4.18 (m, 11H), 4.55 (m, 2H), 6.35 (br d, 1H).
Acetylferrocene-4-nitroacetophenone azine (1-Fc/4-NO2)
mp 178–180°C. UV–vis (DMF) λ max (nm) (ε, M–1 cm–1):
323 (21 170). FT-IR (CCl4, cm–1): 3105, 3080, 2920, 1592,
Acetylferrocene-4-acetylpyridine azine (1-Fc/Pyrd)
mp 123–124°C. UV–vis (DMF) λ max (nm) (ε, M–1 cm–1):
285 (21 700). FT-IR (CCl4, cm–1): 3100, 3089, 3038, 2959,
2923, 1608, 1475, 1409, 1364, 1291, 1122, 1107, 1065,
1
1512, 1342, 855, 821. H NMR (300 MHz, CDCl3) δ: 2.23
(s, 3H), 2.38 (s, 3H), 4.23 (s, 5H), 4.45 (t, 2H), 4.81 (t, 2H),
8.05 (d, 2H), 8.27 (d, 2H). 13C NMR (75 MHz, CDCl3) δ:
161.75, 155.95, 148.15, 144.58, 127.26, 123.54, 82.56,
70.55, 70.45, 69.55, 69.25, 67.75, 67.65, 15.95, 14.95. MS
m/z (%): 389 (M+, 100), 324 (9), 270 (18), 185 (31), 121
(31). Anal. calcd. for C20H19N3O2Fe: 389.0826; found:
389.0827.
1
1019, 1002, 889. H NMR (300 MHz, CDCl3) δ: 2.19 (s,
3H), 2.38 (s, 3H), 4.19 (s, 5H), 4.40 (br s, 2H), 4.74 (br s,
2H), 7.72 (br s, 2H), 8.66 (br s, 2H). 13C NMR (75 MHz,
CDCl3) δ: 161.26, 155.75, 150.08, 145.67, 120.63, 71.04,
70.44, 69.39, 67.67, 15.98, 14.44. MS m/z (%): 345 (M+,
100), 280 (18), 185 (38), 121 (20). Anal. calcd. for
C19H19N3Fe: 345.0928; found: 345.0932.
Acetylferrocene-4-cyanoacetophenone azine (1-Fc/4-CN)
mp 156–157°C. UV–vis (DMF) λ max (nm) (ε, M–1 cm–1):
296 (25 050). FT-IR (CCl4, cm–1): 3100, 2960, 2924, 1603,
Acetophenone-9-anthraldehyde azine (2-Anth/H)
This compound was prepared by adding 9-anthraldehyde
(1 equiv) to acetophenone (diethoxyphosphinyl)hydrazone (1
equiv) (39) in refluxing benzene containing NaH (1.3 equiv).
mp 95–97°C. UV–vis (EPA) λ max (nm) (ε, M–1 cm–1): 257
(105 700). FT-IR (CCl4, cm–1): 3086, 3055, 3031, 1624,
1
1573, 1563, 1476, 1364, 1292, 1131, 1106, 1017. H NMR
(300 MHz, CDCl3) δ: 2.20 (s, 3H), 2.31 (s, 3H), 4.19 (s,
5H), 4.40 (t, 2H), 4.74 (t, 2H), 7.68 (d, 2H), 7.97 (d, 2H).
13C NMR (75 MHz, CDCl3) δ: 161.50, 156.15, 142.82,
132.09, 121.04, 118.81, 112.60, 82.89, 70.57, 69.62, 67.88,
1
1611, 1519, 1453, 1444, 1364, 1294, 1021, 889. H NMR
© 2002 NRC Canada

Sauro and Workentin
261
(400 MHz, CDCl3) δ: 2.65 (s, 3H), 7.52 (m, 7H), 8.03 (m,
4H), 8.54 (s, 1H), 8.79 (d, 2H), 9.71 (s, 1H). 13C NMR
(100 MHz, CDCl3) δ: 165.47, 157.86, 138.27, 131.37,
130.65, 130.37, 130.26, 128.99, 128.51, 127.05, 127.00,
125.77, 125.36, 125.27, 15.95. MS m/z (%): M+, 322 (76),
204 (100), 177 (56), 118 (18), 77 (66). Anal. calcd. for
C23H18N2: 322.1472; found: 322.1470.
References
1. Ferrocenes: Homogeneous catalysis, organic synthesis, materi-
als science. Edited by A. Togni and T. Hayashi. VCH Pub-
lishers, New York. 1995.
2. A.G. Osborne, M.W d. Silva, M.B. Hursthouse, K.M.A. Malik,
G. Opromolla, and P. Zanello. J. Organomet. Chem. 516, 167
(1996).
3. K.N. Jayaprakash, P.C. Ray, I. Matsuoka, M.M. Bhadbhade,
V.G. Puranike, P.K. Das, H. Nishihara, and A. Sarkar. Organo-
metallics, 18, 3851 (1999).
4. X. Xu-Bing, D. Chun-Ying, Z. Long-Gen, Y. Xiao-Zeng, Y.
Yao, and W. Hua-Qin. Polyhedron, 11, 1917 (1992).
5. P. Zanello, G. Opromolla, F.F.d. Biani, A. Ceccanti, and G.
Giorgi. Inorg. Chim. Acta, 255, 47 (1997).
6. D. Osella, M. Farrali, P. Zanello, F. Laschi, M. Fontani, C.
Nervi, and G. Cavigiolio. Inorg. Chim. Acta, 306, 42 (2000).
7. B. Bildstein, M. Schweiger, H. Angleitner, H. Kopacka, K.
Wurst, K.H. Ongania, M. Fontani, and P. Zanello. Organo-
metallics, 18, 4286 (1999).
8. R. Altmann, O. Gausset, D. Horn, K. Jurkschat, M. Schürmann,
M. Fontani, and P. Zanello. Organometallics, 19, 430 (2000).
9. B. Neumann, U. Siemeling, H.-G. Stammler, U. Vorfeld, J.G.P.
Delis, P.W.N.M. van Leeuwen, K. Vrieze, J. Fraanje, K.
Goubitz, F.F.d. Biani, and P. Zanello. J. Chem. Soc. Dalton
Trans. 4705 (1997).
Acetophenone-1-pyrenecarboxaldehyde azine (2-Py/H)
This compound was prepared by adding 9-anthraldehyde
(1 equiv) to acetophenone (diethoxyphosphinyl)hydrazone (1
equiv) (39) in refluxing benzene containing NaH (1.3 equiv).
mp 115–117°C. UV–vis (DMF) λ max (nm) (ε, M–1 cm–1):
380 (41 500). FT-IR (CCl4, cm–1): 3084, 3050, 1610, 1580,
1496, 1444, 1364, 1294, 1235, 1185, 850. 1H NMR
(400 MHz, CDCl3) δ: 2.68 (s, 3H), 7.47 (m, 3H), 8.08 (m,
5H), 8.22 (m, 4H), 8.72 (d, 1H), 8.94 (d, 1H), 9.52 (br s,
1H). 13C NMR (100 MHz, CDCl3) δ: 165.30, 157.61,
138.39, 133.11, 131.25, 130.61, 130.22, 130.18, 128.86,
128.78, 128.49, 127.44, 127.19, 126.99, 126.42, 126.21,
126.03, 125.81, 124.99, 124.96, 124.59, 122.90, 15.77. MS
m/z (%): 346 (M+, 100), 227 (47), 201 (80), 145 (20), 119
(9), 77 (30). Anal. calcd. for C25H18N2: 346.1470; found:
346.1465.
10. B. Bildstein, M. Malaun, H. Kopacka, M. Fontani, and P.
Zanello. Inorg. Chim. Acta, 300–302, 16 (2000).
11. S. Barlow, H.E. Bunting, C. Ringham, J.C. Green, G.U. Bublitz,
S.G. Boxer, J.W. Perry, and S.R. Marder. J. Am. Chem. Soc.
121, 3715 (1999).
12. A. Houlton, N. Jasim, R.M.G. Roberts, J. Silver, D.
Cunningham, P. McArdle, and T. Higgins. J. Chem. Soc. Dal-
ton Trans. 2235 (1992).
13. H.K. Sharma, K.H. Pannell, I. Ledoux, J. Zyss, A. Ceccanti,
and P. Zanello. Organometallics, 19, 770 (2000).
14. M. Prato, M. Maggini, C. Giacometti, G. Scorrano, G. Sandona,
and G. Farnia. Tetrahedron, 52, 5221 (1996).
15. Z. Wang, K. Chen, and H. Tian. Chem. Lett. 423 (1999).
16. Z. Wang, J. Zhu, K. Chen, and H. Tian. J. Chem. Res. S, 438
(1999).
X-ray crystallography
Single crystals of 1-Fc/4-NO2, 1-Fc/4-CN, 1-Fc/Pyrd, and
1-Fc/Anth were grown by slow evaporation of a concen-
trated solution of the azine in an appropriate solvent. Crys-
tals were mounted on a glass fibre. Data were collected on a
Nonius Kappa-CCD diffractometer using COLLECT (Nonius,
1998) software. The unit cell parameters were calculated and
refined from the full data sets. Crystal cell refinement and
data reduction was carried out using the Nonius DENZO
package. The data were scaled using SCALEPACK (Nonius,
1998) and no other adsorption corrections were applied. The
SHELXTL 5.1 (Sheldrick, G.M., Madison, WI) program
package was used to solve the structures by direct methods,
followed by successive difference Fouriers. All non-
hydrogen atoms were refined with anisotropic thermal pa-
rameters. The hydrogen atoms were calculated geometrically
and were riding on their respective carbon atoms. The crys-
tal data for the structures determined in this report are re-
ported in the Supplementary material.2
17. R. Centore and C. Garzillo. J. Chem. Soc. Perkin Trans. 2, 79
(1997).
18. E.C. Kesslen, W.B. Euler, and B.M. Foxman. Chem. Mater.
11, 336 (1999).
19. R. Centore, B. Panunzi, and A. Tuzi. Z. Kristallog. 211, 31
(1996).
20. G.S. Chen, M. Anthamatten, C.L. Barnes, and R. Glaser. Angew.
Chem. Int. Ed. Engl. 34, 1081 (1994).
21. G.S. Chen, M. Anthamatten, C.L. Barnes, and R. Glaser. J.
Org. Chem. 59, 4336 (1994).
Acknowledgments
The financial support of the Natural Sciences and Engi-
neering Research Council of Canada (NSERC), Canadian
Foundation for Innovation, ORDCF, and the University of
Western Ontario (ADF) is appreciated. V.A.S. thanks the
OGS program for a scholarship. Thanks to Dr. M.C.
Jennings and Dr. J.F. Corrigan for crystal structure analyses
and structural determinations.
22. G.S. Chen, J.K. Wilbur, C.L. Barnes, and R. Glaser. J. Chem.
Soc. Perkin Trans. 2, 2311 (1995).
23. R. Glaser, G.S. Chen, and C.L. Barnes. J. Org. Chem. 58, 7446
(1993).
24. R. Glaser, G.S. Chen, M. Anthamatten, and C.L. Barnes. J.
Chem. Perkin Trans. 2, 1449 (1995).
25. R. Glaser and G.S. Chen. J. Comp. Chem. 19, 1130 (1998).
2 Supplementary material may be purchased from the Depository of Unpublished Data, Document Delivery, CISTI, National Research Council
Canada, Ottawa, ON K1A 0S2, Canada. For information on obtaining material electronically go to http://www.nrc.ca/cisti/irm/unpub_e.shtml.
Crystallographic information has also been deposited with the Cambridge Crystallographic Data Centre (CCDC Nos. 178709, 178710, 178711,
178712). Copies of the data can be obtained free of charge via www.ccdc.cam.ac.uk/conts/retrieving.html (or from the Cambridge Crystallo-
graphic Data Centre, 12, Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033; or deposit@ccdc.cam.ac.uk).
© 2002 NRC Canada

262
Can. J. Chem. Vol. 80, 2002
45. U.W. Grummt. J. Photochem. 31, 131 (1985).
46. S. Vettermann, K. Gustav, and J. Jungel. J. Mol. Struct. Theo-
chem 104 259 (1983).
47. K. Gustav, S. Vettermann, and P. Birner. J. Mol. Struct. Theo-
chem. 88, 249 (1982).
48. K. Appenroth, M. Reichenbacher, and R. Paetzold. J. Photo-
chem. 14, 39 (1980).
49. K. Appenroth, M. Reichenbacher, and R. Paetzold. J. Photo-
chem. 14, 51 (1980).
50. K. Appenroth, M. Reichenbacher, and R. Paetzold. Tetrahe-
dron, 37, 569 (1981).
26. R. Glaser, M. Lewis, and Z. Wu. J. Mol. Model 6, 86 (2000).
27. M. Lewis, C.L. Barnes, and R. Glaser. Can. J. Chem. 76, 1371
(1998); (b) M. Lewis and R. Glaser. J. Org. Chem. 67, 1441
(2002).
28. M. Lewis, C.L. Barnes, B.A. Hathaway, and R. Glaser. Acta.
Cryst. C55, 975 (1999).
29. M. Lewis, C.L. Barnes, and R. Glaser. Acta Cryst. C56, 393
(2000).
30. H. Lund. Acta Chem. Scand. 13, 249 (1959).
31. K.B. Wiberg and T.P. Lewis. J. Am. Chem. Soc. 92, 7154
(1970).
32. M.A.G. Nieto, M.D.L.d. Castro, and M. Valcarcel. Electrochim.
Acta 28, 1725 (1983).
51. K. Appenroth, E. Birckner, and M. Reichenbacher. J. Photo-
chem. 37, 117 (1987).
33. J.H. Barnes, F.M. Triebe, and M.D. Hawley. J. Electroanal.
Chem. 139, 395 (1982).
52. J.D. Coyle. Introduction to organic photochemistry. John Wiley
& Sons, New York. 1986.
34. K.L. Handoo and S.K. Handoo. Ind. J. Chem. 19B, 684 (1980).
35. J. Ludvík, F. Riedl, F. Liska, and P. Zuman. J. Electroanal.
Chem 457, 177 (1998).
36. F. Riedl, J. Ludvík, F. Liska, and P. Zuman. J. Heterocyclic
Chem. 33, 2063 (1996).
37. M. Privman and P. Zuman. J. Electroanal. Chem. 455, 235
(1998).
38. P. Zuman. Tetrahedron Lett. 41, 7851 (2000).
39. V.A. Sauro and M.S. Workentin. J. Org. Chem. 66, 831 (2001).
40. Organic electrochemistry. Edited by H. Lund and M.M. Baizer.
Marcel Dekker Inc., New York. 1991.
41. T. Gareis, O. Kothe, and J. Daub. Eur. J. Org. Chem. 1549
(1998).
42. A. Lasia and A. Rami. Can. J. Chem. 65, 744 (1987).
43. A.J. Bard and L.R. Faulkner. Electrochemical methods. Funda-
mentals and applications. John Wiley & Sons, New York. 1980.
44. R. Nicholson. Anal. Chem. 37, 1351 (1965).
53. R. Paetzold, M. Reichenbacher, and K. Appenroth. Z. Chem.
21, 421 (1981).
54. K. Gustav and S. Vettermann. Z. Chem. 18, 456 (1978).
55. R.E. Bozak. Photochemistry in the metallocenes. In Advances
in photochemistry. Vol. 8. Edited by N. Pitts, G.S. Hammond,
and W.A. Noyes. John Wiley & Sons Inc. New York. 1971.
56. R.M.G. Roberts, J. Silver, B.M. Yamin, M.G.B. Drew, and U.
Ebergardt. J. Chem. Soc. Dalton Trans. 1549 (1988).
57. J.D. Dunitz, L.E. Orgel, and A. Rich. Acta Cryst. 9 373 (1956).
58. Handbook of photochemistry. 2nd ed. Edited by S.L. Murov, I.
Carmichael, and G.L. Hug. Marcel Dekker Inc., New York.
1993.
59. S.G. Cohen, S.J. Groszos, and D.B. Sparrow. J. Am. Chem.
Soc. 72, 3947 (1950).
60. K. Appenroth, M. Reichenbacher and R. Paetzold. Z. Chem.
23, 414 (1983).
© 2002 NRC Canada