Comparison of the bis(ferrocenylethynyl)phenylmethylium cation with bis(ferrocenylethenyl)methylium analogues

Article abstract of DOI:10.1016/j.jorganchem.2006.04.002

Several new ferrocenylethynyl derivatives, (FcCC)₂CHOH, (FcCC)₂CH₂, (FcCC)₂CPhOH have been synthesised from ethynylferrocene. Attempts to synthesise the corresponding bis(ferrocenylethynyl)-stabilised carbocatio

Full text of DOI:10.1016/j.jorganchem.2006.04.002

Journal of Organometallic Chemistry 691 (2006) 3285–3292
www.elsevier.com/locate/jorganchem
Comparison of the bis(ferrocenylethynyl)phenylmethylium cation
with bis(ferrocenylethenyl)methylium analogues
a
a
a,b,
*
Christofer Arisandy , Elizabeth Fullam , Stephen Barlow
a
Inorganic Chemistry Laboratory, University of Oxford, South Parks Road, Oxford OX1 3QR, UK
School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, GA 30332-0400, USA
b
Received 21 December 2005; received in revised form 3 April 2006; accepted 3 April 2006
Available online 15 April 2006
Abstract
Several new ferrocenylethynyl derivatives, (FcCC)2CHOH, (FcCC)2CH2, (FcCC)2CPhOH have been synthesised from ethynylferro-
cene. Attempts to synthesise the corresponding bis(ferrocenylethynyl)-stabilised carbocations by hydroxyl or hydride abstraction from
the bridging group was unsuccessful for (FcCC)2CHOH and (FcCC)2CH2, respectively. In the case of (FcCC)2CPhOH, the
1
[(FcCC)2CPh]+ cation could be observed by H and 13C NMR, and by UV–Vis–NIR spectroscopy, but was too unstable for isolation
or prolonged study in solution. The UV–Vis–NIR spectrum of [(FcCC)2CPh]+ is compared to that of the considerably more stable cat-
ions, [Fc(CHCH)2CH]+ and [(FcCHCH)2CPh]+.
ꢀ 2006 Elsevier B.V. All rights reserved.
Keywords: Ferrocene; Polymethine; Alkyne
1. Introduction
ested in whether localisation could also be induced in rela-
tively short polymethine-like species by reduction of
coupling through the conjugated bridge.
Organic polymethines, or cyanines, are predicted to
undergo localisation at sufficiently long chain lengths [1–
3]; this has been demonstrated for a couple of all-organic
examples [4,5], the first example being the species shown
in Fig. 1 with n = 6 [4]. We have recently shown that the
choice of the charge-stabilising end group in polymethines
can affect the chain length at which localisation occurs; we
have found that 1,3-bis(ruthenocenyl)allylium cation is
localised, whereas its ferrocenyl analogue is a symmetrical
species (Fig. 1) [6,7]. This result can be rationalised by anal-
ogy with mixed-valence chemistry [4,8]; the localisation in
the ruthenium system can be attributed to the large reorga-
nisation energy of the ruthenocene/(g5-cyclopentadienyl-
g6-fulvene)ruthenium(II) cation system. We were inter-
Several studies of mixed-valence compounds have
revealed weaker coupling in alkyne-bridged mixed-valence
compounds than in the corresponding alkenes [9–15]. Hence,
we identified species in which the pentamethine bridge is
replaced with the AC„CCRC„CA bridge (Fig. 2) as inter-
esting targets and here we describe some synthetic studies
aimed at obtaining cations of the form [(FcCC)2CR]+.
Related AC„CCRC„CAbridged cations with the
Cp(PR3)2RuA end group have been reported; these have
been shown to be symmetrical in solution on the NMR time-
scale and, in one case, in the solid state by X-ray crystallog-
raphy [16]. Several reactions of (PhCC)2CHOH under acid
conditions, including conversion to PhC„CCH@C@
CPhBr and PhC„CCH@CHC(O)Ph, presumably involve
[(PhCC)2CH]+ as an intermediate [17]. However, attempts
to isolate cations such as [(4-Me2NC6H4CC)2CR]+ (R =
Ar, CCAr) from treatment of the corresponding alcohol with
perchloric acid gave the cyclised pyrylium species [18].
*
Corresponding author. Tel.: +1 404 385 6053; fax: +1 404 894 5909.
E-mail address: stephen.barlow@chemistry.gatech.edu (S. Barlow).
0022-328X/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2006.04.002

3286
C. Arisandy et al. / Journal of Organometallic Chemistry 691 (2006) 3285–3292
a solution of FcC„CHCHO (1.13 g, 4.8 mmol) in THF
or
(60 mL) was added dropwise to the stirred solution. After
45 min at ꢀ78 ꢁC the solution was allowed to warm to
room temperature over 1 h and left to stir for a further
1 h, by which time the reaction appeared complete by
TLC. The solution was then poured into an ice–water mix-
ture (100 mL). THF was removed under reduced pressure,
before extracting with diethyl ether (3 · 100 mL). The com-
bined ether extracts were then washed with water
(5 · 100 mL), washed with a saturated solution of NaCl
(2 · 100 mL), dried over potassium carbonate, filtered,
and evaporated under reduced pressure. The resulting
brown-yellow oil was recrystallised from a hot dichloro-
n
n
N
N
N
N
(a)
+
+
Rc
Rc
Rc
Fc
Fc
Fc
M
M
+
+
Rc
Fc
(b)
(c)
(d)
Fig. 1. (a) Polymethines studied by Tolbert and Zhao [4]; for n = 4 these
two structures are related by the resonance arrow, whereas for n = 6 they
are related by the equilibrium arrow. (b) Structure representing
a
termethine with group 8 metallocene end groups; (c) shows that this
represents one of two resonance structures where M = Fe {Fc = ferroce-
nyl} whereas (d) shows that this represents one of two equilibrating
structures where M = Ru {Rc = ruthenocenyl} [6,7].
methane/hexane to give
a red-brown solid (1.51 g,
3.4 mmol, 71%). Anal. Calc. for C25H20OFe2: C, 67.01;
H, 4.50; Fe, 24.92. Found: C, 66.70; H, 4.76; Fe, 24.16%.
1H NMR (chloroform-d, 300 Mz)
d 5.35 (1H, d,
J = 7.3 Hz, CHOH), 4.47 (4H, m, two overlapping
C5H4), 4.24 (10H, s, C5H5), 4.21 (4H, m, C5H4), 2.23
(1H, d, J = 7.3 Hz, OH). 13C{1H} (chloroform-d,
75 MHz) d 83.4 (alkyne), 83.0 (alkyne), 71.6 (C5H4CH),
71.5 (C5H4CH), 70.0 (C5H4CH), 68.9 (C5H5), 63.7 (C5H4
quat.), 53.5 (CHOH). EI MS m/z 448 (M+). IR (KBr) m
+
+
Fc
Fc
Fc
Fc
R
R
+
[3] , R = H
[4] , R = Ph
[5] , R = Me
+
+
[1] , R = H
[2] , R = Ph
+
+
3408 (–O–H), 2292 (alkyne) cmꢀ1
.
(a)
(b)
Fig. 2. (a) [(FcCC)2CR]+ cations targeted in this work, with (b) their
polymethine analogues, [(FcCHCH)2CR]+, shown for comparison.
2.3. (FcCC)2CH2 (II)
A solution of I (107 mg, 0.24 mmol) in diethyl ether
(30 mL) was added dropwise to a mixture of AlCl3
(64 mg, 0.48 mmol) and LiAlH4 (18 mg, 0.47 mmol) in
diethyl ether (50 mL). After stirring at room temperature
for 2 h, the reaction was quenched by the dropwise addi-
tion of deoxygenated water (125 mL). The solution was
extracted with diethyl ether (3 · 100 mL); the combined
ether extracts were combined, dried over magnesium sul-
fate, filtered, and evaporated under reduced pressure. The
resulting red/brown solid was recrystallised from a mixture
of dichloromethane and pentane at ꢀ80 ꢁC (80 mg,
2 mmol, 77%). Anal. Calc. for C25H20Fe2: C, 69.49; H,
2. Experimental
2.1. General
Electrochemical data were acquired at 298 K using a
BAS potentiostat, a glassy-carbon working electrode (cir-
cular, 3 mm diameter), a platinum-wire auxiliary electrode,
and a silver wire (coated with AgCl by anodising in 1 M
KCl solution) pseudo-reference electrode. Cyclic voltam-
metry (CV) data were acquired in deoxygenated dry dichlo-
romethane solutions 0.1 M in [nBu4N]+[PF6]ꢀ and ca.
10ꢀ4–10ꢀ3 M in analyte using a scan rate of ca. 50 mV sꢀ1
.
4.66; Fe, 25.85. Found: C, 68.38; H, 4.92; Fe, 24.29%. H
1
Potentials were referenced to ferrocenium/ferrocene by
addition of 1,2,3,4,10,20,30,40-octamethylferrocene (the
potential of which was measured as ꢀ445 mV vs. ferrocene
[19]) to the cell; the error in the reported potentials is esti-
mated from comparison of successive scans to be ca.
10 mV. FcC„CH [20], FcC„CCHO [21] and
FcCH@CHBr [7] were synthesised according to literature
procedures (in our hands, the 0.5 M aqueous NaOH spec-
ified in the literature did not successfully effect the conver-
sion of FcCHCl@CHCHO to FcC„CH, but we found
1.3 M aqueous NaOH to be effective).
NMR (chloroform-d, 300 MHz) d 4.40 (4H, m, C5H4),
4.21 (10H, s, C5H5), 4.14 (4H, m, C5H4), 3.29 (2H, s,
CH2). 13C{1H} (chloroform-d, 75 MHz) d 80.1 (alkyne),
78.8 (alkyne), 70.0 (C5H4CH), 69.8 (C5H5), 68.4
(C5H4CH), 65.4 (C5H4 quat.), 11.5 (CH2). EI MS m/z
432 (M+). IR (KBr) m 2230 (alkyne) cmꢀ1
.
2.4. (FcCC)2CPhOH (III)
n-Butyllithium (3.56 mL of 2.5 M solution in hexane,
8.9 mmol) was added dropwise to a solution of FcC„CH
(1.74 g, 8.3 mmol) in THF (20 mL) at ꢀ78 ꢁC. After
30 min, methyl benzoate (0.6 mL, 4.2 mmol, dried by distil-
lation from CaH2) was added dropwise to the stirred solu-
tion. After 1 h at ꢀ78 ꢁC, the solution was allowed to warm
to room temperature over 1 h and then poured into ice–
water (100 mL). THF was removed using a rotary evapora-
2.2. (FcCC)2CHOH (I)
n-Butyllithium (2.1 mL of a 2.5 M solution in hexane,
3.2 mmol) was added dropwise to a solution of FcC„CH
(1.00 g, 4.8 mmol) in THF (30 mL) at ꢀ78 ꢁC. After 30 min

C. Arisandy et al. / Journal of Organometallic Chemistry 691 (2006) 3285–3292
3287
tor before extracting with diethyl ether (4 · 100 mL). The
combined ether extracts were then washed with water
(3 · 100 mL), washed with saturated aqueous NaCl
(2 · 150 mL), dried over potassium carbonate, filtered,
and evaporated under reduced pressure. The product was
purified by passing through a short silica gel plug; impuri-
ties were removed by elution with petroleum ether (b.p. 40–
60 ꢁC) and with petroleum ether/ethyl acetate (100:1); the
product was eluted with ethyl acetate and recrystallised
from dichloromethane/pentane at ꢀ35 ꢁC to give a red-
brown solid (1.05 g, 2.0 mmol, 48% yield). Anal. Calc. for
C31H26OFe2: C, 71.03; H, 4.61; Fe, 21.31; O, 3.05. Found:
zene-d6, 500 MHz) d 6.54 (dd, J = 15.4, 10.7 Hz, 1H), 6.19
(dd, J = 15.4, 10.7 Hz, 1H), 6.18 (d, J = 15.4 Hz, 1H), 5.84
(dt, J = 15.4, 6.8 Hz, 1H), 4.24 (apparent t, J = ca. 1.8 Hz,
2H), 4.04 (m, 4H), 4.00 (s, 5H), 3.98 (apparent t, J = ca.
1.8 Hz, 2H), 3.96 (s, 5H), 3.02 (d, J = 6.8 Hz, 2H). 13C
(C6D6, 125 MHz) d 131.8 (vinyl), 131.4 (vinyl), 129.5 (vinyl),
127.3 (vinyl), 87.5 (quat.), 83.9 (quat.), 69.5 (C5H5), 69.1
(C5H4CH), 68.8 (C5H5), 68.5 (C5H4CH), 67.8 (C5H4CH),
67.0 (C5H4CH), 33.1 (CH2).
2.6. (FcCH@CH)2CHOH (V)
1
C, 70.99; H, 4.97; Fe, 20.78%. H NMR (chloroform-d,
FcCH@CHBr (isomeric mixture, 0.58 g, 2 mmol) was
dissolved in a 1:1:4 mixture of pentane, diethyl ether, and
THF (15 mL). The solution was cooled to ꢀ98 ꢁC
300 MHz) d 7.91 (2H, dd, Ph), 7.36–7.46 (overlapping
peaks, 3H, Ph), 4.48 (4H, m, 2 overlapping C5H4), 4.22
(10H, s, C5H5), 4.20 (4H, m, C5H4), 2.87 (1H, s, OH).
13C{1H} NMR (chloroform-d, 75 MHz) d 142.6 (Ph quat),
128.6 (PhCH), 128.5 (PhCH), 126.1 (PhCH), 85.9 (alkyne),
84.2 (alkyne), 71.7 (C5H4CH), 71.6 (C5H4CH), 70.0
(C5H5), 69.0 (C5H4CH), 66.3 (CPhOH), 63.9 (C5H4 quat).
EI MS m/z 525 (M+). IR (KBr) m 3540 (–O–H), 2228
t
(MeOH/1 Æ N2) and treated with BuLi (1.7 M in pentane,
4 mmol). After 30 min, the reaction was allowed to warm
to ꢀ78 ꢁC over another 30 min. The solution was then trea-
ted with ethyl formate (0.2 mL) in THF; after 15 min at
ꢀ78 ꢁC, the reaction mixture was allowed to warm to room
temperature and treated with water (20 mL) and diethyl
ether (20 mL). The layers were separated and the aqueous
portion was extracted with diethyl ether (3 · 20 mL).
Finally, the combined extracts were dried over K2CO3, fil-
tered and evaporated under reduced pressure. Attempted
purification of the products by column chromatography
led to the decomposition of the product; this material
was used without further purification for the in situ gener-
ation of [3]+.
(alkyne) cmꢀ1
.
2.5. (FcCH@CH)2CH2 (IVa) and Fc(CH@CH)2CH2Fc
(IVb)
According to the scheme described by Tolbert and co-
workers [22], n-butyllithium (3.4 mL of a 2.5 M hexane solu-
tion, 8.4 mmol) was added dropwise to a slurry of
[(Ph3PCH2)2CH2]2+(Brꢀ)2 (2.9 g, 4 mmol) in THF (30 mL)
at ꢀ78 ꢁC. After 30 min, the reaction mixture was allowed
to warm to room temperature over 1 h; the reaction mixture
was then recooled to ꢀ78 ꢁC and a solution of FcCHO
(4 mmol) in THF (15 mL) added dropwise. The reaction
mixture was allowed to warm to room temperature and then
refluxed for 12 h. Water (20 mL) and diethyl ether (20 mL)
was added and the mixture transferred to a separating fun-
nel. The aqueous layer was extracted with more ether and
the combined organics were dried over magnesium sulfate,
filtered and evaporated under reduced pressure. The product
was purified by eluting a short silica gel plug with 1:10 diethyl
ether/pentane, and evaporating the first fraction to give an
orange oil, which was an ca. 1:2:1 E,E:E,Z:Z,Z mixture of
2.7. (FcCH@CH)2CPhOH (VI)
VI was obtained as a deep-red oil in an analogous way
to V using ethyl benzoate in place of ethyl formate. Yield:
1
0.25 g (1 mmol, 45%). H NMR (chloroform-d, 300 MHz)
d 8.10 (2H, d, J = 8 Hz), 7.40–7.70 (overlapping peaks),
6.40 (1H, br, dd), 5.95 (2H, d, J = 15 Hz), 5.60 (1H, br,
dd), 4.42 (4H, t, J = 1.8 Hz), 4.05–4.15 (4H, overlapping
peaks), 4.03 (10H, s). TOF-EI-MS m/z 528 (46%, M+),
510 (M+ꢀOH, 100%). Anal. Calc. for C31H28Fe2O: C,
70.49; H, 5.34. Found: C, 69.20; H, 5.74%.
1
the isomers of (FcCH@CH)2CH2. Yield: 60%. H NMR
2.8. [(PPh3CH2)2CHMe]2+(Brꢀ)2 (VII)
(benzene-d6, 500 MHz) d 6.26–6.16 (m, overlapping vinyl
CHs), 5.92–5.83 (m, overlapping vinyl CHs), 5.52–5.56 (m,
overlapping vinyl CHs), 4.28-4.22 (overlapping Cp CHs),
4.04–3.96 (overlapping Cp Chs), 3.29 (t, J = 7.3 Hz, minor
isomer CH2), 3.00 (dd, J = 6.1, 7.6 Hz, major E,Z isomer,
CH2), 2.77 (t, J = 6.5 Hz, minor isomer CH2). 13C NMR
(benzene-d6, 125 MHz) d 127.2 (vinyl CH), 126.6 (vinyl
CH), 125.9 (vinyl CH), 125.8 (other vinyl CH resonances
presumably obscured by solvent), 84.3 (quat.), 84.2 (quat.),
82.3 (quat.), 82.2 (quat.), 69.6–66.8 (overlapping Fc CHs),
33.2 (CH2), 32.7 (CH2), 33.4 (CH2). Use of KOtBu as a base
in place of nBuLi (as described in detail for VII) resulted in
1,3-Dibromo-2-methylpropane (6.93 g, 32 mmol), tri-
phenylphosphine (25 g, 96 mmol), and ethanol (30 mL)
were deoxygenated in a rotaflo ampoule and then heated
to 100 ꢁC for 5 d. The ampoule’s contents were transferred
to a large conical flask and allowed to cool with stirring;
addition of diethyl ether (500 mL) caused a white mixture
of oil and solid to separate. After stirring for an additional
1 h, the product fully solidified and was collected on a frit
and washed with diethyl ether. After drying in vacuum, the
product was obtained as a white powder (20 g, 27 mmol,
84%). 1H NMR (500 MHz) d 7.89 (m, 9 H, ortho- and
1
the isolation of E,E-Fc(CH@CH)2CH2Fc. H NMR (ben-

3288
C. Arisandy et al. / Journal of Organometallic Chemistry 691 (2006) 3285–3292
para-CH), 7.76 (apparent t, J = 7.0 Hz, 6 H, Ph meta-CH),
4.20 (m, 2H, CH2), 3.99 (m, 2H, CH2), 2.26 (m, 1H, CH),
0.99 (d, 3H, J = 6.3 Hz, CH3). 13C{1H} NMR (125 MHz) d
136.4 (s, Cpara), 134.8 (d, JPC = 10 Hz, Cmeta), 131.7 (d,
JPC = 12 Hz, Cortho), 119.4 (d, JPC = 87 Hz, Cipso), 30.5
(dd, JPC = 52, 16 Hz, CH2), 27.0 (s, CH), 22.1 (s, CH3).
31P{1H} (122 MHz) d 22.9. ES-MS (MeOH) m/z 661
(3%, [C40H38P2Br]+), 290 (100%, [C40H38P2]2+).
C5H5 of this isomer), 3.24 (d, J = 7.3 Hz, 2H, CH2), 1.83
(s, 3H, CH3). TOF-EI-MS m/z 450 (100%, M+).
2.10. [2]+[CF3CO2]ꢀ, [2]+[BF4]ꢀ, [3]+[CF3CO2]ꢀ and
[4]+[CF3CO2]ꢀ
[2]+[CF3CO2]ꢀ and [2]+[BF4]ꢀ were generated in situ for
UV–Vis–NIR studies by addition of a drop of CF3CO2H
(0.1 mL) or HBF4 (0.1 mL, 85% in diethyl ether) respec-
tively to ca. 0.1 mM solutions of III in dry dichlorometh-
ane. UV–Vis–NIR (CH2Cl2) kmax (emax) 355 (9100), 455
(14000), 522 (20000), 1127 (11000) nm (Mꢀ1 cmꢀ1). Solu-
tions for NMR studies were generated in the same way by
addition of TFAD or TFAH by syringe to a dichlorometh-
ane-d2 solution of III. 1H NMR (dichloromethane-d2/
CF3CO2D, 300 MHz) d 8.38 (d, J = 8.0 Hz, 2H, Ph o-H),
8.17 (t, J = 8.0 Hz, 1H, Ph p-H), 7.50 (t, J = 8.0 Hz, 2H,
Ph m-H), 5.81 (t, J = 1.8 Hz, 4H, C5H4), 5.33 (t,
J = 1.8 Hz, 4H, C5H4), 4.75 (s, 10H, C5H5). 13C{1H}
NMR (dichloromethane-d2/CF3CO2H, ꢀ25 ꢁC, 125 MHz;
assigned using DEPT-135 and GHSQC experiments) d
176.5 (quat.), 143.1 (quat.), 132.4 (Phm–CH), 131.2 (Php–
CH), 126.7 (Pho–CH), 120.1 (quat.), 114.4 (quat.), 82.8
2.9. FcCH@CHCHMe@CHCH2Fc (VIII)
A solution of KOtBu (7.0 g, 62 mmol) in THF (20 mL)
was added dropwise to a suspension of VII (3.00 g,
4.05 mmol) in THF (25 mL) at ꢀ78 ꢁC. The resulting
orange reaction mixture was allowed to warm to room tem-
perature, before recooling to ꢀ78 ꢁC and dropwise addi-
tion of a solution of FcCHO (1.733 g, 8.1 mmol) in THF
(10 mL). The reaction mixture was allowed to warm to
room temperature and heated to reflux until TLC showed
no further change (38 h). The reaction mixture was poured
into water and the organics were extracted with several
portions of diethyl ether. The combined extracts were dried
on MgSO4, filtered, and evaporated under reduced pressure
to afford an orange solid, which was redissolved in dichlo-
romethane and absorbed onto silica gel. The absorbed
product was then transferred to a short silica gel plug
which was eluted with pentane to yield a small unidentified
yellow fraction, followed by a second orange-red fraction
(60 mg, 0.24 mmol, 6%), identified as E-1-ferrocenyl-3-
methyl-butadiene, FcCH@CHC(Me)@CH2. 1H NMR
(benzene-d6, 300 MHz) d 6.63 (d, J = 16 Hz, 1H, vinyl
CH), 6.31 (d, J = 16 Hz, 1H, vinyl CH), 5.03 (s, 1H,
CH2), 4.96 (s, 1H, CH2), 4.26 (apparent t, J = ca. 1.8 Hz,
2H, C5H4), 4.06 (apparent t, J = ca. 1.8 Hz, 2H, C5H4),
3.97 (s, 5H, C5H5). 13C NMR (C6D6, 75 MHz) d 142.5
(vinyl quat.), 129.7 (vinyl CH), 127.4 (vinyl CH), 115.1
(vinyl CH2), 83.7 (C5H4 quat.), 69.5 (C5H5CH), 69.2
(C5H4CH), 67.2 (C5H4CH), 18.7 (CH3). TOF-EI-MS m/z
504 (8%, 2M+), 252 (100%, M+). Elution with a 1:10 mix-
ture of diethyl ether and pentane gave an orange-red oily
solid (orange solid after recrystallisation from a mixture
of hexane and dichloromethane), shown to be a mixture
of the E,E and E,Z isomers of FcCH@CHC(Me)@
CHCH2Fc (1.177 g, 2.61 mmol, 64%). Major and minor
isomers were found in the approximate ratio 2:1. Major
isomer 1H NMR (benzene-d6, 300 MHz) d 6.68 (d,
J = 16.0 Hz, 1H, CH@CH), 6.34 (d, J = 16.0 Hz, 1H,
CH@CH), 5.80 (t, J = 7.3 Hz, 1H, CMe@CH), ca. 4.32
(m, 2H, C5H4), ca. 4.08 (m, 2H, C5H4), 4.03 (s, 5H,
C5H5), ca. 4.00 (s, 7H, C5H5 and C5H4), ca. 3.97 (m,
C5H4), 3.11 (d, J = 7.3 Hz, 2H, CH2), 1.83 (s, 3H, CH3).
1
(C5H4CH corresponding to 5.81 ppm H resonance), 77.0
1
(C5H5), 76.9 (C5H4CH corresponding to 5.33 ppm H res-
onance), 70.9 (quat.).
Solutions containing [3]+[CF3CO2]ꢀ and [4]+[CF3CO2]ꢀ
were obtained in the same way from V and VI, respectively,
and were considerably more stable than those of
[2]+[CF3CO2]ꢀ.
2.11. [3]+[BF4]ꢀ
A
solution of trityl tetrafluoroborate (185 mg,
0.56 mmol) in dichloromethane (10 mL) was added drop-
wise to a stirred solution of IVa (260 mg, 0.60 mmol) in
dichloromethane (10 mL); the reaction mixture instantly
darkened. After 30 min, the solution was filtered and diethyl
ether (150 mL) added. The resulting precipitate was washed
with diethyl ether (2 · 20 mL) and dried in vacuo to afford
the salt as a dark microcrystalline powder (217 mg,
0.42 mmol, 75%). 1H NMR (dichloromethane-d2,
300 MHz) d 8.17 (d, 2H, J = 14.2 Hz, FcCH), 8.07 (t, 1H,
J = 12.6 Hz, FcCHCHCH), 6.60 (apparent t, 2H,
J = 13.3 Hz, FcCHCH), 5.47 (m, 4H, C5H4), 5.02 (m, 4H,
C5H4), 4.51 (s, 10H, C5H5). 13C{1H} NMR (dichlorometh-
ane-d2, 75 MHz) d 155.6 (FcCH), 154.9 (FcCHCHCH),
130.2 (FcCHCH), 88.5 (C5H4 quat.), 81.9 (C5H4CH), 75.5
(C5H4CH), 73.3 (C5H5). UV–Vis–NIR (CH2Cl2) kmax (emax
)
256 (11000), 506 (23000), 904 (16000) nm (Mꢀ1 cmꢀ1). UV–
Vis–NIR (MeCN) kmax 491, 832 nm. UV–Vis–NIR
(Me2CO) kmax 493, 842 nm. UV–Vis–NIR (DMSO) kmax
862 nm. UV–Vis–NIR (DMF) kmax 892 nm. IR (KBr)
3094, 1548 (s), 1494 (s), 1421, 1194 (s), 1149, 1060 (s, br),
1034 (s, br), 927, 834, 644, 513, 479, 440, 427 cmꢀ1. Anal.
Calc. for C25H23BF4Fe2: C, 57.53; H, 4.44. Found: C,
56.98; H, 4.77%.
1
Minor isomer H NMR (benzene-d6, 300 MHz) d 7.07 (d,
J = 15.6 Hz, 1H, CH@CH), 6.45 (d, J = 15.6 Hz, 1H,
CH@CH), 5.63 (t, J = 7.3 Hz, 1H, CMe@CH), 4.32–3.94
(C5H5 and C5H4 resonances overlapping with those of
the major isomer, but s at 4.05 may be attributed to

C. Arisandy et al. / Journal of Organometallic Chemistry 691 (2006) 3285–3292
3289
2.12. [3]+[PF6]ꢀ
more group 8 metallocenes from alcohol precursors
[7,23–36]. Indeed, [Fc2CCCFc]+ – related to our target cat-
ions in featuring a ferrocenylalkynyl group – was obtained
from protonation of Fc2C(OH)C„CFc, itself obtained
from the reaction of FcC„CLi and Fc2CO [32,37]. Alter-
natively, hydride can be abstracted from methylene units
in the bridging group using reagents including [Ph3C]+
[22,38,39], DDQ [40–45], and [Cp2RuHal]+ [46,47]. Hence
we identified bis(ferrocenylethynyl)methanol (I) and
bis(ferrocenylethynyl)methane (II) as potential precursors
to our target cations. We synthesised I as shown in
Scheme 1 from the reaction of lithiated ethynylferrocene
[20] with FcC„CCHO, which was itself obtained by the
reaction of FcC„CLi and DMF according to a literature
procedure [21]. I is a new compound, although an isomer,
FcC„CC(O)CH@CHFc has previously been reported
[48]. The alcohol I was readily converted to II by reduction
with LiAlH4/AlCl3. Due to our inability to characterise the
[1]+ cation (vide infra), we synthesised III as a precursor to
[2]+ through the reaction of FcC„CLi with ethyl benzoate
(Scheme 1). Compounds I–III are oxidized at potentials of
+0.14, +0.11 and +0.14 V, respectively, vs. ferrocenium/
ferrocene in CH2Cl2/0.1 M [nBu4N]+[PF6]ꢀ. No separation
was resolved between first and second oxidation; this is not
surprising in view of the length of bridge and the lack of
complete conjugation between the two ferrocene groups
in these species. The positive values of these potentials
are consistent with the alkynyl fragment exerting an elec-
tron-withdrawing effect on the iron centers, as has previ-
ously been observed for, example, FcC„CH, for which a
value of +0.14 V vs. ferrocenium/ferrocene (CH2Cl2/
0.1 M [nBu4N]+[PF6]ꢀ) has been reported [49].
This salt was obtained in an analogous fashion to its
[BF4]ꢀ analogue, using [Ph3C]+[PF6]ꢀ in place of
[Ph3C]+[BF4]ꢀ. UV (CH2Cl2) kmax 505, 900 nm. Anal.
Calc. for C25H23F6Fe2P: C, 51.76; H, 4.00. Found: C,
52.13; H, 4.46%.
2.13. [4]+[BF4]ꢀ
This salt (0.27 g, 0.45 mmol, 49%) was obtained from VI
using the method we have previously described for the syn-
thesis of [Mc(CH)3Mc]+[BF4]ꢀ salts from McCH@CHCH-
OHMc [7]. 1H NMR (dichloromethane-d2, 300 MHz) d
7.77 (2H, d, J = 13.2 Hz, FcCH), 7.66 (1H, t, J = 5.0 Hz,
Ph), 7.58 (2H, t, J = 5.0 Hz, Ph), 7.34 (1H, d, J = 5.0 Hz,
Ph), 6.83 (2H, d, J = 13.2 Hz, CPhCH), 5.54 (4H, apparent
s, C5H4), 5.08 (4H, apparent s, substituted C5H4), 4.52
(10H, s, C5H5). 13C{1H} NMR (dichloromethane-d2,
75 MHz) d 161.5 (CPh), 153.5 (CPhCH), 136.8 (Ph quat.),
132.4 (Ph CH), 130.4 (FcCH), 129.3 (Ph CH), 129.2 (Ph
CH), 89.4 (C5H4 quat.), 82.2 (C5H4 CH), 75.5 (C5H5),
73.4 (C5H4 CH). ES-MS (MeOH) m/z 510 (100%, M+).
UV–Vis (CH2Cl2) kmax(emax) 515 (21000), 926 (14000)
nm (Mꢀ1 cmꢀ1). Anal. Calc. for C31H27F4Fe2B: C, 62.26;
H, 4.55; B, 1.81. Found: C, 61.03; H, 4.66; B, 1.88 (1.81)%.
2.14. [5]+[BF4]ꢀ
This salt was obtained from VII using [Ph3C]+[BF4]ꢀ in
the same way as [3]+[BF4]ꢀ. H NMR (dichloromethane-
1
d2, 300 MHz) d 8.29 (d, J = 14.2 Hz, 2H, FcCH), 6.47 (d,
J = 14.2 Hz, 2H, FcCHCH), 5.56 (m, 4H, C5H4), 5.11
(m, 4H, C5H4), 4.56 (s, 10H, C5H5), 1.84 (s, 3H, CH3).
13C{1H} NMR (dichloromethane-d2, 300 MHz) d 159.4
(CMe), 149.4 (FcCH), 132.5 (FcCHCH), 89.8 (C5H4 quat),
82.1 (C5H4CH), 75.7 (C5H5), 73.5 (C5H4CH), 16.6 (CH3).
ES-MS (MeOH) m/z 449 (100%, M+). UV–Vis (CH2Cl2)
kmax 255, 514, 922 nm. Anal. Calc. for C26H25F4Fe2B: C,
58.27; H, 4.70. Found: C, 56.53; H, 4.44%.
3.2. Synthesis of [(FcCC)2CR]+ and [(FcCHCH)2CR]+
cations
We attempted to obtain salts of [(FcCC)2CH]+, [1]+,
using the methods we and others have used for synthesis
of [Fc(CH)nFc]+ cations, i.e. reaction of I with Brønsted
and Lewis acids (ethereal HBF4, aqueous HPF6,
[Ph3C]+[BF4]ꢀ) and reaction of II with [Ph3C]+[BF4]ꢀ.
Invariably the isolable products were dark NMR-silent
paramagnetic species [50]. Whilst attempts to isolate salts
of [(FcCC)2CPh]+, [2]+, from III in the ways attempted
for [1]+ from I were also unsuccessful, we were able to
2.15. [5]+[PF6]ꢀ
This salt was obtained from VII using [Ph3C]+[PF6]ꢀ in
the same way as [3]+[PF6]ꢀ. IR (KBr) 3112, 1520 (s), 1494
(s), 1412, 1368, 1351, 1239 (s), 1106, 1050, 1015, 964, 928,
834 (s, br), 642, 558, 497, 467 cmꢀ1. Anal. Calc. for
C26H25F6Fe2P: C, 52.56; H, 4.24. Found: C, 52.69; H,
4.26%.
3. Results and discussion
3.1. Synthesis of (FcCC)2CROH and (FcCC)2CH2
Two principal strategies can be identified for the synthe-
sis of bis(ferrocenyl) cations. Brønsted or Lewis acids have
been used to generate carbocations stabilised by one or
Scheme 1. Reagents and conditions: (a) nBuLi, THF, ꢀ78 ꢁC, then
FcC„CCHO; (b) LiAlH4, AlCl3, Et2O; (c) nBuLi, THF, ꢀ78 ꢁC, then
PhCO2Et; (d) CF3CO2H or HBF4, CH2Cl2.

3290
C. Arisandy et al. / Journal of Organometallic Chemistry 691 (2006) 3285–3292
observe a room temperature 1H NMR spectrum consistent
with the proposed structure in dichloromethane-d2 on the
addition of deuterated trifluoroacetic acid to III
(CF3CO2H has previously been used by others to generate
ferrocenyl carbocations [51–53]). The UV–Vis–NIR spec-
trum (vide infra) obtained under similar conditions can,
therefore, be identified with that of [2]+. Both NMR and
UV–Vis–NIR spectra (Fig. 3) are found to change over
time, clearly indicating the instability of this species in solu-
tion, consistent with our inability to isolate pure salts of
this cation. The peak observed at ca. 750 nm in the decom-
position product (Fig. 3) could potentially be due to a ferr-
ocenium-containing product (kmax for unsubstituted
ferrocenium and for [FcCCH]+ are 628 and 698 nm,
respectively, in dichloromethane [49]); this is consistent
with NMR spectra which reveal appearance of a paramag-
netic species. The same UV–Vis–NIR spectrum is obtained
on addition on ethereal tetrafluoroboric acid to III in
dichloromethane, but decays more rapidly than that gener-
ated using CF3CO2H. The formation of ferrocenium prod-
ucts is reminiscent of that observed for many mononuclear
ferrocenyl carbocations where the reactivity can be under-
stood in terms of a redox tautomer in which the ferrocene is
oxidised and the ‘‘carbocation’’ center is a radical; for
example, the dimagnetic FeII species [FcCH2]+ has been
shown to dimerise to the paramagnetic FeIII dimer
[FcCH2CH2Fc]2+ [54].
While solutions were sufficiently unstable to permit
acquisition of 13C spectra at room temperature, we were
able to acquire such spectra at ꢀ25 ꢁC if the sample was
inserted into the pre-cooled spectrometer immediately after
preparation [55]. The 13C resonance corresponding to the
central carbon of the p bridge is presumably that at
176.5 ppm; this is considerably upfield from that of the cen-
tral carbon of trityl cation (211.6 ppm in H2SO4 [56]) sug-
gesting the FcCC groups allow for considerable more
delocalisation of positive charge than phenyl groups. How-
ever, this resonance is somewhat downfield from that of the
corresponding resonance in [4]+ (161.5 ppm in CD2Cl2),
suggesting FcCC is a less effective donor than FcCHCH.
We also synthesised the pentamethine cations [3]+, [4]+,
and [5]+ for comparison (Scheme 2) with [1]+ and [2]+ and
with the 1,3-ferrocenylallylium cation. We were able to
obtain [3]+[X]ꢀ {X = BF4, PF6} using the previously
reported hydride abstraction with [Ph3C]+[X]ꢀ from 1,5-
bis(ferrocenyl)-1,4-pentadiene (IVa) which was obtained
as a mixture of E,E, E,Z, and Z,Z isomers from the Wittig
reaction of FcCHO and [(Ph3PCH2)2CH2]2+(Brꢀ)2 using
stoichiometric n-butyllithium as a base [22]. Use of excess
potassium tert-butoxide as base in the Wittig reaction leads
to the pure E,E isomer of 1,5-bis(ferrocenyl)-1,3-pentadi-
ene (IVb) which is presumably the thermodynamic product
and which is presumably formed due to equilibration via
[3]ꢀ. We were also able to obtain 1,5-bis(ferrocenyl)-3-
methyl-1,3-pentadiene (VIII) as a mixture of E,E and E,Z
isomers in the same way as IVb and convert it to
[5]+[X]ꢀ {X = BF4, PF6} using trityl salts. In addition,
[3]+ can be obtained by the action of various acids on
(FcCH@CH)2CHOH (V) which was obtained in turn, as
a mixture of geometric isomers, from FcCH@CHBr (iso-
mer mixture) using lithium-bromine exchange followed
by quenching with excess ethyl formate (the intermediate
FcCH@CHCHO is expected to be more reactive with
FcCH@CHLi than ethyl formate [57]). FcCH@CHCH-
OHFc and similar species have previously been reported
to be rather unstable with respect to oxidation to the cor-
responding ketones [7,32]; V is also rather unstable and
was, therefore, used without further purification for the
synthesis of [3]+. The precursor to [4]+, (FcCH@CH)2-
CPhOH (VI) was obtained in the same way as V, using
ethyl benzoate in place of ethyl formate, and is consider-
ably more stable.
3.3. Spectra of the cations
The isolated salts of [3]+–[5]+ all form stable solutions in
CH2Cl2 and were studied using UV–Vis–NIR spectroscopy
(see Table 1) [58]. For direct comparison to [2]+, solutions
of [3]+ and [4]+ were also generated in situ by treatment of
V and VI, respectively, with CF3CO2H; spectra obtained in
Fig. 3. UV–Vis–NIR spectra of [2]+ generated by addition of CF3CO2H
to III in dichloromethane. Above: spectrum with approximate absorptiv-
ity determined. Below: spectra taken at ca. 0, 30, 100 and 400 min showing
decomposition.

C. Arisandy et al. / Journal of Organometallic Chemistry 691 (2006) 3285–3292
3291
Scheme 2. Reagents and conditions: (a) nBuLi, THF, ꢀ78 ꢁC, then FcCHO; (b) excess KOtBu, THF, ꢀ78 ꢁC, then FcCHO; (c) 2 equiv. tBuLi, pentane/
t
diethyl ether/THF, ꢀ98 ꢁC, then HCO2Et, ꢀ78 ꢁC; (d) 2 equiv. BuLi, pentane/diethyl ether/THF, ꢀ98 ꢁC, then PhCO2Et, ꢀ78 ꢁC; (e) excess Ph3P,
EtOH, 100 ꢁC; (f) [Ph3C]+[X]ꢀ {X = BF4, PF6}, CH2Cl2; (g) HBF4 or CF3CO2H, CH2Cl2.
to an Fe ! p* charge transfer, in principle consisting of a
Table 1
Wavelengths and frequencies corresponding to absorption maxima for the
high-energy (HE) and low-energy (LE) transitions seen for bis(ferrocenyl)-
orbitals (xy, x2–y2 and z2) to p*; hence, the lineshape of this
terminated cations in CH2Cl2
superposition transitions from each of the three filled d
transition will depend to some extent on the energy differ-
Cation
kmax (nm)
mmax (103 cmꢀ1
)
ence between these transitions. The spectral differences may
possibly reflect a structure for [2]+ differing from that of
[3]+–[5]+ in that the charge is more localised on the central
carbon atom of the bridge, the p* orbital is less destabilised
through interaction with the metal, and in that the Fe orbi-
tals are less stabilised through interaction with the a-car-
bon. This possibility is consistent with 13C NMR
evidence (vide supra).
HE
LE
HE
LE
8.9
11.1
10.8
10.9
[2]+
[3]+
[4]+
[5]+
522
506
515
514
1127
904
926
922
19.2
19.8
19.4
19.5
this way were essentially identical to those obtained from
solutions of [BF4]ꢀ or [PF6]ꢀ salts. The spectra of [3]+–
[5]+ are qualitatively similar to one another and to the pre-
viously reported spectrum of [Fc(CH)3Fc]+ [7,32]. All show
two prominent transitions in the Vis–NIR region: we have
previously assigned the lower energy of these transitions to
a charge transfer from Fe filled d orbitals to the p* orbital
associated with the conjugated bridge, and the higher
energy feature to a p–p* transition [7,59]. Accordingly,
both transitions for [3]+ are red-shifted relative to those
for [Fc(CH)3Fc]+, but the high energy band shows more
bridge-length dependence (the energies of the high and
low-energy bands of [3]+, [Fc(CH)3Fc]+, and [Fc(CH)Fc]+
have been compared in Fig. 7 of Ref. [59]). Introduction of
a phenyl or methyl group into the central part of the bridge
leads to a slight red shift of both high and low-energy
bands.
Acknowledgement
We thank for the Royal Society for partial support and
Les Gelbaum for acquiring low-temperature NMR data.
References
[1] C. Kuhn, Synth. Met. 41–43 (1991) 3681.
[2] J.S. Craw, J.R. Reimers, G.B. Bacskay, A.T. Wong, N.S. Hush,
Chem. Phys. 167 (1992) 77.
[3] L.M. Tolbert, M.E. Ogle, Synth. Met. (1992) 391.
[4] L.M. Tolbert, X. Zhao, J. Am. Chem. Soc. 119 (1997) 3253.
[5] S.L. Lepkowicz, O.V. Przhonska, J.M. Hales, J. Fu, D.J. Hagan,
E.W. Van Stryland, M.V. Bondar, Y.L. Slominsky, A.D. Kachkov-
ski, Chem. Phys. 305 (2004) 259.
[6] S. Barlow, L.M. Henling, M.W. Day, S.R. Marder, Chem. Commun.
(1999) 1567.
[7] S. Barlow, L.M. Henling, M.W. Day, W.P. Schaefer, J.C. Green, T.
Hascall, S.R. Marder, J. Am. Chem. Soc. 124 (2002) 6285.
[8] S.F. Nelsen, H.Q. Tran, M.A. Nagy, J. Am. Chem. Soc. 120 (1998)
298.
[9] C. LeVanda, K. Bechgaard, D.O. Cowan, J. Org. Chem. 41 (1976)
2700.
[10] A.C. Ribou, J.-P. Launay, M.L. Sachtleben, H. Li, C.W. Spangler,
Inorg. Chem. 35 (1996) 3735.
The optical spectrum of the alkynyl species, [2]+ (Fig. 3),
shows some qualitative differences from those of [3]+–[5]+.
There is a much more prominent additional band on the
high-energy side of the higher energy band, although
[3]+–[5]+ also show weak shoulders in this region.
The low-energy band for [2]+ is considerably broader
(Dm1/2 = ca. 5300 cmꢀ1
)
than that for [3]+–[5]+ (ca.
4000 cmꢀ1). While this broadness might be interpreted as
evidence for symmetry breaking in the alkynyl cation (the
localised species referred to in the introduction show broad
low-energy absorptions [4–7]), in that case one would also
expect the high-energy band would also be expected to be
broad, whereas in [2]+ it has a similar bandwidth to [3]+–
[5]+. Moreover, we have attributed the low-energy band
[11] J.E. Sutton, H. Taube, Inorg. Chem. 20 (1981) 3125.
[12] S.V. Rosokha, D.-L. Sun, J.K. Kochi, J. Phys. Chem. A 106 (2002)
2283.
[13] Other donor–acceptor interactions have also been found to be more
strongly mediated through alkenes; for example, in E-
FcCH@CHB(mes)2 {Fc = ferrocenyl, mes = 2,4,6-trimethylphenyl}
shows a lower-energy and more intense donor–acceptor CT transition
than FcC„CB(mes)2 Z. Yuan, G. Stringer, I.R. Jobe, D. Kreller, K.

3292
C. Arisandy et al. / Journal of Organometallic Chemistry 691 (2006) 3285–3292
Scott, L. Koch, N.J. Taylor, T.B. Marder, J. Organomet. Chem. 452
(1993) 115.
[40] U.T. Mueller-Westerhoff, Angew. Chem., Int. Ed. Engl. 25 (1986)
702.
[14] S. Barlow, C. Risko, V. Coropceanu, N.M. Tucker, S.C. Jones, Z.
Levi, V.N. Khrustalev, M.Y. Antipin, T.L. Kinnibrugh, T. Timofe-
[41] U.T. Mueller-Westerhoff, A. Nazzal, W. Pro¨ssdorf, J.J. Meyerle,
R.L. Collins, Angew. Chem., Int. Ed. Eng. 21 (1982) 293.
[42] B.G. Gavini, T.J. Haas, K.L. Pluorde, U.T. Mueller-Westerhoff,
Synth. React. Inorg. Metal-Org. Chem. 22 (1992) 481.
[43] M.F. Moore, S.R. Wilson, D.N. Hendrickson, U.T. Mueller-West-
erhoff, Inorg. Chem. 23 (1984) 2918.
´
eva, S.R. Marder, J.L. Bredas, Chem. Commun. (2005) 764.
[15] S. Barlow, C. Risko, S.-J. Chung, N.M. Tucker, V. Coropceanu,
´
S.C. Jones, Z. Levi, J.L. Bredas, S.R. Marder, J. Am. Chem. Soc.
127 (2005) 16900.
[16] H.P. Xia, W.F. Wu, W.W. Ng, I.D. Williams, G. Jia, Organometallics
16 (1997) 2940.
[17] K.G. Migliorese, Y. Tanaka, S.I. Miller, J. Org. Chem. 39 (1974) 739.
[18] S. Nakatsuji, K. Nakashima, K. Yamamura, S. Akiyama, Tetrahe-
dron Lett. 25 (1984) 5143.
[44] M.F. Moore, S.R. Wilson, M.J. Cohn, T.-Y. Dong, U.T. Mueller-
Westerhoff, D.N. Hendrickson, Inorg. Chem. 24 (1985) 4559.
[45] S. Ito, N. Morita, T. Asao, J. Org. Chem. 61 (1996) 5077.
[46] M. Watanabe, I. Motoyama, T. Takayama, Bull. Chem. Soc. Jpn. 69
(1996) 2877.
[19] A. Mendiratta, S. Barlow, M.W. Day, S.R. Marder, Organometallics
18 (1999) 454.
[47] M. Watanabe, I. Moyoyama, T. Takayama, J. Organomet. Chem
524 (1996) 9.
[20] M. Rosenblum, N. Brown, J. Papenmeir, N. Applebaum, J. Orga-
nomet. Chem. 6 (1966) 173.
[21] G. Doisneau, G. Balvoine, T. Fillebeen-Khan, J. Organomet. Chem.
94 (1992) 441.
[22] L.M. Tolbert, X. Zhao, Y. Ding, L.A. Bottomley, J. Am. Chem. Soc.
117 (1995) 12891.
[23] C.U. Pittman, Tetrahedron Lett. 37 (1967) 3619.
[24] V.I. Boev, A.V. Dombrovskii, Zh. Obshch. Khim. 50 (1980) 2520.
[25] V.I. Boev, A.V. Dombrovskii, Zh. Obshch. Khim. 55 (1985) 1173.
[26] E.I. Fedin, A.L. Blumenfeld, P.V. Petrovskii, A.Z. Kreindlin, S.S.
Fadeeva, M.I. Rybinskaya, J. Organomet. Chem. 292 (1985) 257.
[27] A.Z. Kreindlin, P.V. Petrovskii, M.I. Rybinskaya, A.I. Yanovskii,
Y.T. Struchkov, J. Organomet. Chem. 319 (1987) 229.
[28] M.I. Rybinskaya, A.Z. Kreindlin, S.S. Fadeeva, J. Organomet.
Chem. 358 (1988) 363.
[48] H. Egger, K. Schloegel, Monatsh. Chem. 95 (1964) 1750.
[49] H. Fink, N.J. Long, A.J. Martin, G. Opromolla, A.J.P. White, D.J.
Williams, P. Zanello, Organometallics 16 (1997) 2646.
[50] One possible product is the 2,6-bis(ferrocenyl)pyrylium cation, at
least in reactions where water is present (the formation of pyrylium
species is described in Ref. [18] for reaction of some organic analogues
of I on treatment with aqueous HClO4). However, the elemental
analysis and the paramagnetic NMR spectra rule this out; 2,6-
bis(ferrocenyl)pyrylium perchlorate has been synthesised by a differ-
ent method and is diamagnetic G.N. Dorofeenko, C.V.V. Krasnikov,
Zh. Org. Khim. 8 (1972) 2620;
Translated in G.N. Dorofeenko, C.V.V. Krasnikov, J. Org. Chem.
USSR 8 (1972) 267.
[51] R. Herrmann, I. Ugi, Tetrahedron 37 (1981) 1001.
[52] J.J. Dannenberg, Angew. Chem., Int. Ed. Engl. 14 (1975) 641.
[53] G.K.S. Prakesh, H.A. Buchholz, J.F.S. Vaughan, Q. Wang, G.A.
Olah, J. Braz. Chem. 10 (1999) 313.
[29] M.I. Rybinskaya, A.Z. Kreindlin, Y.T. Struchkov, A.I. Yanovsky, J.
Organomet. Chem. 359 (1989) 233.
[30] A.I. Yanovsky, Y.T. Struchkov, A.Z. Kreindlin, M.I. Rybinskaya, J.
Organomet. Chem. 369 (1989) 125.
[54] K.L. Rinehart, C.J. Michejda, P.A. Kittle, J. Am. Chem. Soc. 81
(1959) 3162.
[31] C. Zou, M.S. Wrighton, J. Am. Chem. Soc. 112 (1990) 7578.
[32] J. Lukasser, H. Angleitner, H. Schottenberger, H. Kopacka, M.
Schweiger, B. Bildstein, K.-H. Ongania, K. Wurst, Organometallics
14 (1995) 5566.
[55] Attempts to obtain spectra at lower temperatures were complicated
by precipitation.
[56] B. Abarca, G. Asensio, R. Ballesteros, T. Varea, J. Org. Chem. 56
(1991) 3224.
[33] B. Bildstein, P. Denifl, K. Wurst, J. Organomet. Chem. 496 (1995) 175.
[34] B. Bildstein, A. Hradsky, H. Kopacka, R. Malleier, K.-H. Ongania,
J. Organomet. Chem. 540 (1997) 127.
[35] A.Z. Kreindlin, F.M. Dolgushin, A.I. Yanovsky, Z.A. Kerzina, P.V.
Petrovskii, M.I. Rybinskaya, J. Organomet. Chem. 616 (2000) 106.
[36] S. Barlow, A.R. Cowley, J.C. Green, T.J. Brunker, T. Hascall,
Organometallics 20 (2001) 5351.
[37] [FcCCCFc2]+ forms stable crystalline salts; the greater stability
relative to that of our target cations is not too surprising since it can
be regarded as a FcCC-substituted variant of the stable [Fc2CH]+
cation.
[38] M.J.A. Habib, J. Park, W.E. Watts, J. Chem. Soc. C (1970) 2556.
[39] U.T. Mueller-Westerhoff, A. Nazzal, W. Pro¨ssdorf, J.J. Meyerle,
R.L. Collins, Angew. Chem. Suppl. (1982) 686.
[57] J. March, Advanced Organic Chemistry, Wiley, New York, 1985.
[58] We have also studied the electrochemistry of [3]+ in CH2Cl2/0.1 M
[nBu4N]+[PF6]ꢀ and found qualitatively similar behavior to that we
have previously reported for [Fc(CH)3Fc]+ and methylated analogues
in Ref. [6]. The [3]2+/+ couple is reversible and observed at E1/2
= +0.11 V vs. ferrocenium/ferrocene (+0.21 V for [Fc(CH)3Fc]+); the
[3]3+/2+ and [3]+/0 couples are EC-type processes and are observed at
+0.72 and ꢀ0.60 V, respectively (+0.82 and ꢀ0.64 V, respectively, for
[Fc(CH)3Fc]+). In both cases decomposition products, presumably
derived from the unstable trication or neutral species, are also
observed to grow in at close in potential to ferrocene; this could
account with the much lower value of 0.14 V reported for DE1/2
(E31þ=2=2þ ꢀ E2þ=þ) of [3]+ by others in Ref. [22].
1=2
[59] S. Barlow, S.R. Marder, Chem. Commun. (2000) 1555.