Improvement of ferrocene acylation. Conventional vs. microwave heating for scandium-catalyzed reaction in alkylmethylimidazolium-based ionic liquids

Article abstract of DOI:10.1016/j.jorganchem.2008.06.022

Acylation of ferrocene-catalyzed by scandium triflate was thoroughly investigated. Reaction time and temperature are important parameters, in that prolonged treatment causes substrate decomposition and/or catalyst deactivation. Hydrophobic alkylmethylimid

Full text of DOI:10.1016/j.jorganchem.2008.06.022

Journal of Organometallic Chemistry 693 (2008) 3015–3020
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Improvement of ferrocene acylation. Conventional vs. microwave heating for
scandium-catalyzed reaction in alkylmethylimidazolium-based ionic liquids
*
Serena Berardi, Valeria Conte, Giulia Fiorani, Barbara Floris , Pierluca Galloni
Dipartimento di Scienze e Tecnologie Chimiche, Università di Roma ‘‘Tor Vergata’’, Via della Ricerca Scientifica, 00133 Roma, Italy
Dipartimento di Scienze Chimiche, Università di Padova, Via Marzolo 1, 35131 Padova, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 22 May 2008
Received in revised form 13 June 2008
Accepted 13 June 2008
Available online 25 June 2008
Acylation of ferrocene-catalyzed by scandium triflate was thoroughly investigated. Reaction time and
temperature are important parameters, in that prolonged treatment causes substrate decomposition
and/or catalyst deactivation. Hydrophobic alkylmethylimidazolium-based solvents gave best results, in
particular 1-butyl-3-methylimidazolium bis(trifluorometansulfonyl) imide. Anhydrides behave as supe-
rior acylating reagents with respect to acyl chlorides. MW irradiation allowed to quantitatively acylate
ferrocene within minutes.
Keywords:
Ferrocene
Ó 2008 Elsevier B.V. All rights reserved.
Acylation
Ionic liquids
Scandium
Metal catalysis
1. Introduction
can be reused [9]. In the first ferrocene acylation performed in an
ionic liquid, 1-ethyl-3-methylimidazolium chloroaluminate was
Ferrocene derivatives find an ever growing application in many
fields, from chiral catalysis [1] to medicinal chemistry [2], to mate-
rial science [3]. A number of reasons may be enumerated to explain
the success of ferrocenes, among which, its unusual stability for an
organometallic species: ferrocenes can be handled in the air when
solid and often in solution also. Ferrocenes are reactive aromatic
compounds, but they are sensitive to oxidizing agents – as most
electrophilic reagents are – as well as to acids. Thus, Friedel–Crafts
acylation is useful in functionalizing ferrocenes, because oxidation
can be easily avoided, reaction conditions are mild, and ferrocenyl
ketones may be transformed into a variety of functional groups.
However, Friedel–Crafts acylation needs more than stoichiometric
amounts of AlCl3, thus making it necessary to dispose large
amounts of material. Alternative procedure for ferrocene acylation
were devised to overcome the problem of Lewis acid disposal and,
possibly, to reuse it. Among them, we can mention the in situ gen-
erated mixed trifluoroacetic anhydride [4], the heterogeneous sys-
tem with alumina [5] or zeolites [6], and the electrochemical
catalysis [7]. In recent years a growing number of organic reactions
– especially catalytic ones – has been studied in different solvent
systems, such as ionic liquids (ILs) [8]. The advantages of using io-
nic liquids are numerous: they are non-volatile solvents, dissolve a
variety of compounds, can be tailor made, and in several cases they
used [10]. This pioneering method, however, presents the draw-
back that haloaluminate ILs are difficult to handle and are very
water sensitive. Moreover, AlCl3 was still needed, carboxylic anhy-
drides were not much reactive, and diacylated products were also
formed. Later, better results were achieved using N-butylpyridini-
um tetrafluoroborate as solvent [11]. Reactions were still carried
out under dry nitrogen atmosphere, and high yields of monoacyl
ferrocenes were reported with 10% molar ytterbium triflate,
although not consistent with the ‘‘considerable abundance of oxi-
dation” observed.
Metal triflates have been used as catalysts for acylation of other
aromatic compounds, such as electron rich methoxybenzenes [12]
and also methyl- and halobenzenes [13], in alkylmethylimidazoli-
um ILs. Alkylation also has been performed in the same type of ILs
[14]. As to metal triflates, the best performances were given by lan-
thanides and scandium. In fact, scandium triflate is a Lewis acid not
decomposed by water [15] and was used as a Friedel–Crafts cata-
lyst in molecular solvents also [16]. The use of lanthanides in ionic
liquids has been recently reviewed [17]. In the last few years we
were interested in exploring how metal-catalyzed reactions be-
have in ILs [18]. This and our long lasting interest in ferrocene
chemistry [19] prompted us to investigate how Friedel–Crafts acyl-
ation can be performed in imidazolium-based ILs – more versatile
than pyridium-based ones – with scandium triflate, less expensive
than lanthanides, as catalyst. For comparison purposes, the reac-
tion was performed also in molecular solvents. The screening
* Corresponding author. Tel.: +39 06 7259 4351; fax: +39 06 7259 4328.
E-mail address: floris@uniroma2.it (B. Floris).
0022-328X/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2008.06.022

3016
S. Berardi et al. / Journal of Organometallic Chemistry 693 (2008) 3015–3020
was made exploring other reaction parameters also, such as reac-
tants/catalyst ratio, catalyst nature, acylating reagent, tempera-
ture, added acid or base, workup procedure, recycle.
and interferes with scandium triflate dissociation, thus inhibiting
the catalyst.
In hydrophobic bmimPF6, 1-butyl-3-methylimidazolium hexa-
fluorophosphate, interesting conversions were obtained (around
80%, Table 1, entries 2–4). Isolated yields of acetylferrocene were
up to 64% on the basis of reacted ferrocene (47% vs. initial FcH). Ex-
cess Ac2O (Table 1, entry 3) did not give any disubstituted product,
thus allowing to reduce reaction time and, consequently, the ferro-
cene decomposition due to prolonged exposure to acid [20]. Re-
duced decomposition was observed also raising the temperature
from 25 °C to 50 °C, with resulting shorter reaction time (Table 1,
entries 2 and 3).
2. Results and discussion
2.1. Effect of solvent
The study began performing ferrocene (FcH) acetylation in
molecular solvents, both with acetyl chloride (AcCl) and acetic
anhydride (Ac2O). 0.1 mmol Sc(OTf)3 were used with 1 mmol FcH
and 1 mmol acetylating reagent. In nitromethane, FcH conversion
was quite low with AcCl, (10% at 25 °C, 37% at 50 °C) and isolated
yields of acetylferrocene (FcAc) were even lower, most likely be-
cause FcH disappearance is partly due to decomposition. With
Ac2O, some better results were obtained (at 25 °C, only 10% conver-
sion in dichloromethane, but 42% in MeNO2 and 21% in MeCN).
Increasing the temperature led to an improvement of conversion
(38% in MeCN at 50 °C) and also of isolated FcAc, up to 30% on
the basis of initial FcH.
These disappointing results notwithstanding, investigation was
extended to Ionic Liquids as solvents, where improved reactivity is
often observed [9,18]. Hydrophobic and hydrophilic imidazolium-
based ILs were tested. The terms hydrophilic and hydrophobic
are normally used with ionic liquids [9a], the former for ILs that
form a single phase with water, the latter for ILs that dissolve very
small quantities of water, thus forming two-phase systems. A list
of ionic liquids used and relative abbreviations are in Scheme 1.
Hydrophilic 1-butyl-3-methylimidazolium triflate, bmimOTf,
dissolves well Sc(OTf)3, but almost no reaction occurred after days
(1% acetylferrocene). Most likely, TfOꢀ exerts a common ion effect
Even better results were obtained in bmimNTf2, 1-butyl-3-
methyl-imidazolium
bis(trifluoromethanesulfonyl)imide,
the
other hydrophobic IL investigated (Table 1, entry 4), with 93%
FcH conversion and 78% FcAc isolated yield (73% vs. initial FcH).
However, ferrocene is not much soluble in bmimNTf2 and therefore
other bis(trifluoromethanesulfonyl) imides were tested, with
cations that could ensure an improved solubility, such as N-buty-
lpyridinium (bupyNTf2, Table 2, entry 1) and 1-hexyl-3-methylimi-
dazolium (hmimNTf2, Table 2, entry 2). It is clear that bmimNTf2 is
the solvent where FcH conversion is higher and decomposition is
lower. A reaction performed in Ac2O as solvent under the same
conditions – apart from the Ac2O excess – gave 66% conversion
and a FcH:FcAc ratio 48:52, with 20% decomposition. Therefore,
for subsequent throughout investigation, the following reaction
conditions were chosen: 1 mmol FcH in 1 mL bmimNTf2 with ex-
cess Ac2O was heated at 50 °C, added with catalyst, and let to react
for 2.5 h, unless otherwise stated.
2.2. Effect of concentration
Increase of FcH concentration (Table 3, entry 2 vs. 1) accelerated
conversion, but not yield. With reduced amounts of Ac2O (Table 3,
entries 3 and 4), conversion was very high, almost quantitative, but
some decomposition of FcH occurred, because isolated FcAc
reached 76% yield only.
N
N
N
+
N
+
N
-
+
N
PF6
-
(CF3SO2)2N-
CF3SO3
2.3. Effect of reaction conditions
Under all examined conditions, there is a loss of product, be-
cause isolated FcAc and recovered FcH did not sum up the initial
amount of ferrocene, but no other product is formed. The reaction
was performed also under inert atmosphere (nitrogen), with ILs
degassed and dehydrated under vacuum at 60 °C 2–3 h. Results
were the same, within experimental error, as those obtained in
the presence of air. Therefore, decomposition is likely due to the
acid formed as by-product from the acetylating agent. In fact, using
acetyl chloride, that produces HCl, the reaction yielded mainly
decomposition of FcH. To verify this hypothesis, experiments were
performed, under different conditions. First, the reaction was per-
bmim Tf2N
bmim OTf
bmim PF6
N
+
N
(CF3SO2)2N-
+
(CF3SO2)2N-
N
bupy Tf2N
hmim Tf2N
Scheme 1. Ionic liquids used as solvents in ferrocene acylation.
Table 1
Scandium-catalyzed acetylation of ferrocene in 1-butyl-3-methylimidazolium saltsa
Entry
1
Anion
TfO
Temp. (°C)
Time (h)
96
Molar ratio FcH:Ac2O:Sc(OTf)3
1:1:0.1
Conversionb (%)
1
Relative ratio FcH:FcAcc
99:1
Yieldd (%)
<1
25
2
3
4
PF6
25
50
50
4
3
2.5
1:1:0.1
1:1:0.1
1:5:0.1
79
67
73
75:25
20:80
32:68
21
57
64
5
Tf2N
50
2.5
1:5:0.1
94
7:93
78
a
FcH 1 M.
Calculated from recovered FcH.
GC analysis with tetradecane as internal standard.
Isolated FcAc, yield relative to conversion.
b
c
d

S. Berardi et al. / Journal of Organometallic Chemistry 693 (2008) 3015–3020
3017
Table 2
Scandium-catalyzed acetylation of ferrocene in bis(trifluoromethanesulfonyl)imides at 50 °Ca
Entry
Cation
Time (h)
Molar ratio FcH:Ac2O:Sc(OTf)3
Conversionb (%)
Relative ratio FcH:FcAcc
Yieldd (%)
1
2
3
bupy
hmim
bmim
2.5
2.5
2.5
1.0:1.3:0.10
1.0:1.3:0.10
1.0:5.0:0.10
89
17
94
15:85
93:3
7:93
75
2
78
a
FcH 1 M.
Calculated from recovered FcH.
Analysis with tetradecane as internal standard.
Isolated FcAc, yield relative to conversion.
b
c
d
Table 3
Scandium-catalyzed acetylation of ferrocene in bmimNTf2, at 50 °C, with different Ac2O excess
b
Entry
Time (h)
[FcH] (M)
Molar ratio FcH:Ac2O:Sc(OTf)3
Conversiona (%)
Relative ratio FcH:FcAc
Yieldc (%)
1
2
3
4
2.5
2.0
3.5
3.5
1.0
2.0
1.0
2.0
1.0:5.0:0.10
1.0:5.0:0.10
1.0:2.5:0.10
1.0:2.5:0.10
94
7:93
1:99
0:100
22:78
78
76
66
76
ca. 100
100
79
a
Calculated from recovered FcH.
Analysis with tetradecane as internal standard.
Isolated FcAc, yield relative to conversion.
b
c
Table 4
Scandium-catalyzed acetylation of ferrocene in bmimNTf2, under different conditionsa
Entry
IL
Temp. (°C)
Time (h)
Molar ratio FcH:Ac2O:Sc(OTf)3
Conversionb (%)
Relative ratio FcH:FcAcc
Yieldd (%)
f
1e
2e
3
4
5
bmimNTf2
bmimNTf2
bmimNTf2 + Na2CO3
bmimNTf2 + AcOHh
bmimNTf2 + TfOHh
60
2.5
2.5
2.5
2.5
2.5
1:2.5:0.05
1:2.5:0.05
1:3.0:0.10
1:2.5:0.05
1:2.5:0.05
85
96
17
95
100
7:93
10:90
94:6
6:94
0:100
61
34
57
78
10
g
60
50
50
50
a
FcH 1 M.
b
c
d
e
f
Calculated from recovered FcH.
GC analysis with tetradecane as internal standard.
Isolated FcAc, yield relative to conversion.
Workup: sublimation.
Temperature was increased with the aim to completely transform ferrocene.
Recycle of run 1.
g
h
Equimolar with FcH.
formed at higher temperature (with half catalyst) and repeated on
the same batch of IL/catalyst, removing by sublimation FcH and
FcAc (Table 4, entry 1) and recharging with ferrocene and Ac2O (Ta-
ble 4, entry 2). After the same reaction time, FcH conversion was
higher in the second run, but FcAc was obtained in much lower
yield, thus indicating an higher extent of decomposition, no other
product being present. Then, the reaction was performed in the
presence of a base, such as sodium carbonate, not completely dis-
solved, in order to neutralized the acid formed. However, in this
conditions deactivation of the catalyst occurs, since a very low con-
version was achieved (Table 4, entry 3). Finally, the reaction was
carried on with added AcOH (Table 4, entry 5), with no effect on
either conversion or FcAc yield. These experiments suggest that
the prolonged contact with acids (that in ILs should have an in-
creased acidity) decomposes ferrocene and, likely, the catalyst. To
verify the hypothesis, the reaction was made in the presence of
an equimolar amount of triflic acid, as suggested by one referee.
However, extensive decomposition resulted, with very low amount
of acetylferrocene recovered (Table 4, entry 5). This results seems
to indicate that catalysis by free acid, that is at work in metal tri-
flate-catalyzed benzoylation of anisole in ILs [21], is not effective
with ferrocene with which it is advisable to perform the reaction
under conditions that ensure
decomposition at the minimum.
a fast transformation, keeping
2.4. Effect of catalyst triflate
Different catalysts were also tested, under the best reaction
conditions (FcH 1 M, 50 °C, 2.5 h, 10% molar catalyst, fivefold ex-
cess Ac2O). After the same reaction time, scandium triflate resulted
more efficient than yttrium and ytterbium salts (Table 5, entries 5,
1 and 3), producing the highest isolated yield of FcAc, whereas FcH
conversion was slightly lower. However, longer reaction times,
with yttrium catalyst caused quantitative disappearance of FcH,
without any increase of FcAc yield (Table 5, entry 2).
As to the ytterbium catalyst, FcH conversion is the same as that
with scandium, but yields are definitely lower (Table 5, entry 4).
Thus, the less expensive Sc(OTf)3 was a better Friedel–Crafts cata-
lyst for ferrocene acetylation than lanthanides.
2.5. Effect of catalyst amount
Although 10% molar for the Lewis acid is an improvement with
respect of the stoichiometric AlCl3, it is still a considerable amount

3018
S. Berardi et al. / Journal of Organometallic Chemistry 693 (2008) 3015–3020
Table 5
Acetylation of ferrocene in bmimNTf2, at 50 °C, with different metal triflatesa
Entry
Catalyst
Y(OTf)3
Time (h)
Molar ratio FcH:Ac2O:Sc(OTf)3
Conversionb (%)
Relative ratio FcH:FcAcc
Yieldd (%)
1
2
3
4
5
2.5
8.0
2.5
5.5
2.5
1:5:0.10
1:5:0.10
1:5:0.10
1:5:0.10
1:5:0.10
94
100
95
95
94
12:88
0:100
17:83
5:95
66
67
59
64
78
Yb(OTf)3
Sc(OTf)3
7:93
a
FcH 1 M.
Calculated from recovered FcH.
Analysis with tetradecane as internal standard.
Isolated FcAc, yield relative to conversion.
b
c
d
Table 6
Acetylation of ferrocene in bmimNTf2, at 50 °C, with different amounts of Sc(OTf)3
Entry
Catalyst (mol%)
[FcH] (M)
Time (h)
Molar ratio FcH:Ac2O:Sc(OTf)3
Conversiona (%)
Relative ratio FcH:FcAcb
Yieldc (%)
1
2
3
4
5
6
10
5
5
5
1
1.0
1.0
1.0
1.0
1.0
1.0
2.5
2.5
9.0
2.5 (60 °C)
2.5
29.5
1:5:0.1
94
89
94
100
32
95
7:93
76
43
55
60
20
37
1:5:0.05
1:5:0.05
1:5:0.05
1:5:0.01
1:2.5:0.01
32:69
14:86
0:100
90:10
83:17
1
a
Calculated from recovered FcH.
GC analysis with tetradecane as internal standard.
Isolated FcAc, yield relative to conversion.
b
c
Table 7
Sc(OTf)3-catalyzed acylation of ferrocene with propanoic anhydride under different conditions
O
FcCOEt
FcH +
O
O
EtCO2H
b
Run
IL
Fc:(EtCO)2O:cat
Time (h)
Temp. (°C)
Workup
Conversiona (%)
FcCOEt (%)
1
2
3
4
5
6
7
8
9e
bmimNTf2
bmimNTf2
bmimNTf2
bmimNTf2
bmimNTf2
bmimNTf2
bmimNTf2
bupyNTf2
bupyNTf2
1:3:0.1
1:3:0.1
1:1:0.1
1:1:0.1
1:1:0.1
1:2:0.1
1:2:0.1
1:2:0.1
1:2:0.1
2.5
2
50
60
50
50
50
60
60
50
50
Extraction
Extraction
Extraction
Distillation
Distillation
Extraction
Distillation
Extraction
Distillation
78
71
57
57
38
98
97
78
83
63
71
52
52
6
72
2.4
31
36
48
48
96
17
17
56
56
c
d
a
b
c
Calculated from recovered FcH.
Isolated FcCOEt, yield relative to conversion.
Recycle of run 4.
Recycle of run 6.
Recycle of run 8.
d
e
of a precious catalyst. Therefore, experiments were performed
reducing the catalyst. Five percent molar Sc(OTf)3 is still efficient,
but it is advisable to increase the temperature up to 60 °C, to avoid
prolonged reaction times (Table 6, entries 2–4). With 1% molar
Sc(OTf)3, efficiency decreased too much and increasing reaction
time did not affect significantly FcH conversion, nor isolated quan-
tity of ketone (Table 6, entries 5 and 6).
To summarize results of acetylation, acetylferrocene can be ob-
tained in good yields using scandium triflate as Lewis acid and
bmimNTf2 as solvent. Excess Ac2O is advisable, because it helps
in solubilizing FcH and speeding up the reaction; moreover, no
diacetylferrocene was ever found. No oxidation of ferrocene oc-
curred in bmimNTf2, at odd from what reported for the reaction
in bupyBF4 [11].
2.6. Effect of different acylating reagents
In order to establish the scope of the reaction, different acylat-
ing reagents were used. Experiments with acyl chlorides were
abandoned after some attempts that gave disappointing results,
likely because of the formation of HCl, that in IL at 50 °C causes
extensive decomposition of ferrocene. Propanoic anhydride (Table

S. Berardi et al. / Journal of Organometallic Chemistry 693 (2008) 3015–3020
3019
Table 8
Acylation of ferrocene in bmimTf2N-catalyzed by Sc(OTf)3 with MW irradiationa
c
Entry
1
Ab
FcH:A:cat molar ratio
Tbulk (°C)
Time (min)
Conversion (%)
GC yieldc (%)
Ac2O
Ac2O
Ac2O
Ac2O
Ac2O
1:5:0.05
1:10:0.05
1:5:0.05
1:5:0.05
1:5:0.05
80
3.5
3.5
7.0
4.5
1.5
80
94
88
100
100
77
93
79
100
100
d
2
105–110
105–110
95
3d,e
f
4
5
115
g
6
7
Ac2O
Ac2O
1:5:0.05
1:5:0.05
72–74
65
2.0
5.0
30
38
g
h
i
8
9
(EtCO)2O
(EtCO)2O
1:5:0.05
1:5:0.05
116–120
92–96
4.5
2.0
92
>99
84
86
j
k
10
(PhCO)2O
1:5:0.05
127–155
4.5
98
42
a
1 mmol FcH in 1 mL IL, previously dried; MW irradiation at W = 5 W (10 W in entry 1, 40 W in entry 7), under stirring and simultaneous cooling with compressed air at
30–40 (or 40–50) psi.
b
Acylating reagent.
c
Conversion determined by GC analysis with dodecane as internal standard.
0.5 mmol FcH.
Catalyst added in two portions.
0.25 mmol FcH.
d
e
f
g
In MeCN.
h
Three irradiation cycles, 1.5 min each.
After column chromatography, 72% propanoylferrocene was isolated, together with 18% disubstituted ketones.
Single irradiation cycle.
i
j
k
After column chromatography, 84% propanoylferrocene was recovered, together with 13% disubstituted ketones.
7) gave good yields of propanoylferrocene, that was isolated either
by extraction with diethyl ether, or by direct distillation under vac-
uum from the reaction mixture. After the latter workup, a recycle
was attempted, re-loading with FcH and propanoic anhydride.
Unfortunately, even after prolonged reaction times, only a small
amount of propanoylferrocene was obtained (Table 7, entry 5).
The reaction was performed also in a pyridinium based IL, i.e., 1-
butylpyridinium bis(trifluoromethanesulfonyl) imide, bupyNTf2.
The results were not as good as in bmimNTf2 (Table 7, entries 8
and 9). Attempts to recycle IL/catalyst confirmed that prolonged
exposure to acidic medium is detrimental for FcH and/or the cata-
lyst. No product at all was obtained with butanedioic or chloroeth-
anoic anhydride and only traces of trifluoroacetylferrocene were
detected. As to aromatic anhydrides, benzoic anhydride was scar-
cely reactive, since 24 h heating produced, at best, only 25% iso-
lated benzoylferrocene.
gether with 13% 1,10-dipropanoylferrocene. Total isolated products
account for 97% of initial ferrocene, thus confirming that decompo-
sition is due to prolonged heating. Also benzoylferrocene yield im-
proved from 25% (conventional heating) to 42% with MW
irradiation.
3. Conclusion
In conclusion, an extensive investigation of ferrocene acylation
was performed. In our hands, imidazolium-based ionic liquids
were superior to pyridinium based ones as solvents for ferrocene
acylation, and scandium triflate as catalyst was better than yttrium
or ytterbium triflates. A major improvement was realized using
microwaves as heating source, where, in the best performance,
quantitative GC and isolated yields of ferrocenylketones were
reached within minutes. Thus, the results presented above clearly
show that the protocol IL-MW-Sc(OTf)3 is, at present, the most effi-
cient in ferrocene acylation.
2.7. Effect of microwave activation
To complete the exploration of reaction parameters, we decided
to investigate the effect of microwaves, that should be readily ab-
sorbed by the ionic solvent. When microwave irradiation was used
to heat the reaction mixture (Table 8), it was possible to drastically
reduce reaction times, from hours to minutes. Different runs were
performed, varying the initial concentration of ferrocene, the acetic
anhydride excess, and the number of irradiation cycles. Addition of
the catalyst in two portions did not improve the yield (Table 8, en-
try 3). Under best conditions, 5% molar catalyst yielded quantita-
tively acetylferrocene in 1.5 min (Table 8, entry 5). On the other
hand, the same yields as with conventional heating were achieved
in acetonitrile (Table 8, entries 6 and 7). It is worth mentioning that
Tbulk in MeCN remained lower than in IL, despite the different irra-
diation power used (40 W in MeCN, 5 W in IL). This is due to the
more efficient MW absorption by ionic species in comparison with
the molecular solvent [22].
4. Experimental section
4.1. General
Mass spectra were recorded with a GC/MS Shimadzu CP 6000
instrument. GC analyses were performed with a Varian CP 3900,
equipped with a Supelco SPB-1701 capillary column 30 m ꢁ
0.25 mm, coated with a 0.25 mm film of methylsilicone.
1H NMR spectra were recorded with a Bruker AV 300 spectrom-
eter, with CDCl3 as solvent and using residual signal solvent as ref-
erence (7.28 ppm).
Commercial grade solvents were used. Ionic liquids were syn-
thesized according to general methods [23], preparing bromides
first and then exchanging the anion with the appropriate salt
(Tf2NLi, CF3SO3K, and KPF6).
Excellent results were obtained by MW heating also with prop-
anoic anhydride (Table 8, entries 8 and 9). In fact, fast quantitative
conversion of ferrocene was accompanied by high yield of propa-
noylferrocene. The reaction mixture was also chromatographed
and isolated yields were similar to GC ones. For example, from
run of entry 9 propanoylferrocene was isolated in 84% yield, to-
4.2. Procedure for conventionally heated acylation
In a typical experiment, 1 mmol FcH was weighed in a round-
bottomed 10 mL flask, added with known amounts of solvent
(either molecular or ionic) and acylating reagent. The resulting

3020
S. Berardi et al. / Journal of Organometallic Chemistry 693 (2008) 3015–3020
mixture was heated in a thermostatted oil bath and, finally, the
weighed amount of catalyst was introduced. The reaction course
was followed by TLC. In the workup, different procedures were fol-
lowed. For reactions performed in molecular solvents, the solvent
was removed under vacuum and the residue chromatographed.
For reactions performed in ILs, solvent extraction or distillation
was used. Alternatively, the reaction mixture was directly chro-
matographed on a silica gel column.
References
[1] (a) R. Gómez Arrayás, J. Adrio, J.C. Carretero, Angew. Chem., Int. Ed. 45 (2006)
7674–7715;
(b) B.F. Bonini, M. Fochi, A. Ricci, Synlett (2007) 360–373.
[2] M.F.R. Fouda, M.M. Abd-Elzaher, R.A. Abdelsamaia, A.A. Labib, Appl.
Organometal. Chem. 21 (2007) 613–625.
[3] A. Togni, T. Hayashi (Eds.), Ferrocenes, Wiley-VCH, Weinheim, 1995.
_
[4] D. Plazuk, J. Zakrewski, Synth. Commun. 34 (2004) 99–107.
[5] B.C. Ranu, U. Jana, A. Majee, Green Chem. 1 (1999) 33–34.
[6] M. Bejblová, S.I. Zones, J. Cejka, Appl. Catal. A: Gen. 327 (2007) 255–260.
ˇ
´
´
´
´
´
´
´
[7] M.D. Vukicevic, Z.R. Ratkovic, A.V. Teodorovic, G.S. Stojanovic, R.D. Vukicevic,
Tetrahedron 58 (2002) 9001–9006.
4.2.1. Extraction
The reaction mixture was repeatedly extracted with diethyl
ether, until colourless organic phase and acylferrocenes were puri-
fied by column chromatography.
[8] V.I. Pârvulescu, C. Hardacre, Chem. Rev. 107 (2007) 2615–2665.
[9] (a) P. Wasserscheid, T. Welton (Eds.), Ionic Liquids in Synthesis, 2nd ed.,
Wiley-VCH, Weinheim, 2008;
(b) C. Chiappe, D. Pieraccini, J. Phys. Org. Chem. 18 (2005) 275–297.
[10] A. Stark, B.L. MacLean, R.D. Singer, J. Chem. Soc., Dalton Trans. (1999) 63–66.
[11] J. Li, W. Su, J. Lin, M. Chen, J. Li, Synth. Commun. 35 (2005) 1929–1937.
[12] J. Ross, J. Xiao, Green Chem. 4 (2002) 129–133.
[13] M.J. Earle, U. Hakala, B.J. McAuley, M. Nieuwenhuyzen, A. Ramani, K.R. Seddon,
Chem. Commun. (2004) 1368–1369.
[14] C.E. Song, W.H. Shim, E.J. Roh, J.H. Choi, Chem. Commun. (2000) 1695–1696.
[15] S. Kobayashi, Eur. J. Org. Chem. (1999) 15–27.
[16] (a) K. Ishihara, M. Kubota, H. Kurihara, H. Yamamoto, J. Am. Chem. Soc. 117
(1995) 4413–4414;
4.2.2. Distillation
The reaction flask was introduced into a Kugelrohr distillation
apparatus and distilled under reduced pressure. When volatile
acids formed as by-products, pure acylferrocenes were obtained.
With less volatile acids, an extraction with aqueous NaHCO3 or a
column chromatography yielded pure acylferrocenes.
Purity and identity of products was checked by GC/MS and 1H
NMR analyses and by comparison with authentic samples.
(b) A. Kawada, S. Mitamura, S. Kobayashi, Synlett (1994) 545–546.
[17] K. Binnemans, Chem. Rev. 107 (2007) 2592–2614.
[18] (a) V. Conte, B. Floris, A. Silvagni, ACS Symp. Ser. 974 (2007) 28–37;
(b) V. Conte, B. Floris, P. Galloni, V. Mirruzzo, A. Scarso, D. Sordi, G. Strukul,
Green Chem. 7 (2005) 262–266;
4.3. Procedure for microwave heated acylation
(c) V. Conte, B. Floris, P. Galloni, A. Silvagni, Adv. Synth. Catal. 347 (2005)
1341–1344.
[19] (a) F. Caporossi, B. Floris, P. Galloni, E. Gatto, M. Venanzi, Eur. J. Org. Chem.
(2006) 4362–4366;
Microwave experiments were performed with use of a mono-
mode microwave apparatus operating at 2.45 GHz with continu-
ous irradiation power. In a typical experiment, FcH (1 mmol)
was weighed in a vessel, added with known amounts of solvent,
the acylating reagent (5 mmol) and, finally, the weighed amount
of catalyst (0.05 mmol). The reaction vessel was placed in the sin-
gle-mode cavity of the instrument and irradiated for the required
time with power at 5 W (with ionic liquid) or 40 W (with aceto-
nitrile) under simultaneous stirring and cooling by a stream of
compressed air. The reaction was sampled by quantitative GC
analysis after dilution in CH2Cl2 containing dodecane as internal
standard. Blank experiments with ferrocene and Ac2O in IL with-
out Sc(OTf)3 gave no product with both conventional and MW
heating.
(b) S. Fiorentini, B. Floris, P. Galloni, F. Grepioni, M. Polito, P. Tagliatesta, Eur. J.
Org. Chem. (2006) 1726–1732;
(c) T. Brunetti, M. Diddoro, M.L. Di Vona, B. Floris, P. Galloni, S. Licoccia, Eur. J.
Org. Chem. (2004) 521–526;
(d) B. Floris, P. Galloni, R. Seraglia, P. Tagliatesta, J. Organomet. Chem. 679
(2003) 202–207.
[20] (a) J. Silver, J. Chem. Soc., Dalton Trans. (1990) 3513–3516;
(b) M. Castagnola, B. Floris, G. Illuminati, G. Ortaggi, J. Organomet. Chem. 60
(1973) C17–C18;
(c) M.A.R. Bramley, J. Upadhyay, A. Wassermann, P. Woolliams, Chem.
Commun. (1965) 404–406.
[21] P. Goodrich, C. Hardacre, H. Mehdi, P. Nancarrow, D.W. Rooney, Ing. Eng. Chem.
Res. 45 (2006) 6640–6647.
[22] (a) N.E. Leadbeater, H.M. Torenius, H. Tye, Comb. Chem. High Through. Screen.
7 (2004) 511–528;
(b) C.O. Kappe, Am. Lab. 33 (2001) 13–19.
[23] (a) P. Bonhôte, A.P. Dias, N. Papageorgiou, K. Kalyanasundaram, M. Grätzel,
Inorg. Chem. 35 (1996) 1168–1178;
Acknowledgments
(b) P.A.Z. Suarez, J.E.L. Dullius, S. Einloft, R.F. de Souza, J. Dupont, Polyhedron
15 (1996) 1217–1219;
(c) P.A.Z. Suarez, J.E.L. Dullius, R.F. de Souza, J. Dupont, J. Chim. Phys. Phys.–
Chim. Biol. 95 (1998) 1626–1639;
(d) K.R. Seddon, A. Stark, M. Torres, Pure Appl. Chem. 72 (2000) 2275–2287;
(e) L. Cammarata, S.G. Kazarian, P.A. Salter, T. Welton, Phys. Chem. Chem.
Phys. 3 (2001) 5192–5200;
Experimental contribution by undergraduate students Valenti-
na Armuzza and Sara Lentini is acknowledged. Marcella Bonchio
and Gianfranco Scorrano (ITM-CNR and Department of Chemical
Sciences, Padova) are gratefully thanked for hospitality in MW
experiments.
(f) V. Farmer, T. Welton, Green Chem. 4 (2002) 97–102;
(g) G.S. Owens, M.M. Abu Omar, J. Mol. Catal. A: Chem. 187 (2002) 215–225.