Halophilic reactions of pentafluorohalobenzenes with transition-metal carbonyl anions

Article abstract of DOI:10.1016/S0022-328X(99)00598-7

In the present work we widen the scope of the halophilic mechanism of nucleophilic aromatic substitution, which we found earlier for the reaction of pentafluorochlorobenzene with [CpFe(CO)₂]^- anion, to reactions of pentafluorohalobenzenes (Hal=Cl, Br, I) with various metal carbonyl anions [Re(CO)₅]^-, [Mn(CO)₅]^- and [CpFe(CO)₂]^-. Nucleophilic aromatic substitution with the [CpFe(CO)₂]^- anion yields C₆F₅Fe(CO)₂Cp, while with [Re(CO)₅]^- and [Mn(CO)₅]^- anions the halo(acyl)metallates cis-[C₆F₅(CO)M(CO)₄Hal]Na (M=Mn, Re) are obtained.

Full text of DOI:10.1016/S0022-328X(99)00598-7

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 597 (2000) 77–86
Halophilic reactions of pentafluorohalobenzenes with
transition-metal carbonyl anions
V.A. Ivushkin, P.K. Sazonov, G.A. Artamkina, I.P. Beletskaya *
Chemistry Department, Moscow State Uni6ersity, B-234, Vorob’yo6y Gory, 119899 Moscow, Russia
Received 6 August 1999; received in revised form 25 September 1999
Abstract
In the present work we widen the scope of the halophilic mechanism of nucleophilic aromatic substitution, which we found
earlier for the reaction of pentafluorochlorobenzene with [CpFe(CO)2]− anion, to reactions of pentafluorohalobenzenes (Hal=Cl,
Br, I) with various metal carbonyl anions [Re(CO)5]−, [Mn(CO)5]− and [CpFe(CO)2]−. Nucleophilic aromatic substitution with
the [CpFe(CO)2]− anion yields C6F5Fe(CO)2Cp, while with [Re(CO)5]− and [Mn(CO)5]− anions the halo(acyl)metallates
cis-[C6F5(CO)M(CO)4Hal]Na (M=Mn, Re) are obtained. © 2000 Elsevier Science S.A. All rights reserved.
Keywords: Reaction mechanisms; Carbonylmetallates; Polyfluorinated aryl halides; Halogen–metal exchange; Nucleophilic aromatic substitution;
Metal acyl anions
1. Introduction
CH- or OH-acids. In the presence of these compounds
the formation of the NS product was suppressed com-
pletely, and pentafluorobenzene and related products
were formed.
The question now emerges as to how general the
proposed mechanism is, and whether it is applicable to
the whole range of metal carbonyl anions and halogen
leaving groups (Cl, Br, I). In the present work we
studied the reactions of pentafluorohalobenzenes
C6F5Hal (Hal=Cl, Br, I) with several carbonylmetal-
lates, and proved the halophilic mechanism [10]
(Scheme 1) for all of the reactions under study.
Nucleophilic aromatic substitution in perfluoroarenes
with transition-metal carbonyl anions is a convenient
method for the syntheses of s-aryl transition-metal
complexes [1–3]. We [4] and other authors [1,5–7] have
found that a heavy halogen (Hal=Cl, Br, I) is prefer-
entially substituted in polyfluorinated aryl halides,
which are also much more reactive with carbonylmetal-
lates than perfluorinated analogues. In several cases this
fact allowed one to obtain nucleophilic substitution
(NS) products, which could not be prepared from the
corresponding perfluoroarene [5,8], though the yield of
NS products is often lowered by various redox side
reactions [6,7,9]. In spite of all facts mentioned, these
reactions were long considered in the framework of the
classic addition–elimination mechanism. Recently we
have shown that the reaction of pentafluorochloroben-
zene with the cyclopentadienyldicarbonylferrate anion
( [CpFe(CO)2]−) actually involves halogen–metal ex-
change (HME) as the key step (Scheme 1) [10]. Evi-
dence of a halophilic mechanism was obtained via
trapping the intermediate [C6F5]− anion with certain
2. Results and discussion
2.1. Reactions of [CpFe(CO)2]− anion with
pentafluorohalobenzenes
When a slight excess of C6F5Hal (Hal=Cl, Br, I) is
added to [CpFe(CO)2]K solution in THF at room
temperature an instantaneous reaction takes place [11]1.
As illustrated in Eq. (1), quite a large number of
1 The reported [11] rate constants for the reactions of
* Corresponding author.
E-mail address: beletska@org.chem.msu.ru (I.P. Beletskaya)
[CpFe(CO)2]K with C6F5Cl and C6F5Br in DMSO were 2.79×104
mol s−1 and 5.89×104 l mol s−1, respectively.
l
0022-328X/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(99)00598-7

78
V.A. I6ushkin et al. / Journal of Organometallic Chemistry 597 (2000) 77–86
Scheme 1. Halophilic mechanism for the reaction of C6F5Cl with [CpFe(CO)2]− anion.
Table 1
Product yields in the reaction of [CpFe(CO)2]K (FpK) with C6F5Hal, THF, r.t.
Entry
Reactants
Hal
Relative yield (%) a
C (FpK),
C6F5Hal/FpK
ratio
C6F5Fp
Fp2
C6F5(C6F4)nC6F4Hal
Fp(C6F4)nFp
C6F5H
(mol l−1
)
(n=0–3)
(n=2–5)
1
2
3
4
5
6
Cl
Cl
Cl
Br
Br
I
0.087
0.043
0.053
0.075
0.077
0.070
1.25
1.47
1.19
1.33
1.48
1.14
58 (43)
(49)
50
54 (49)
61 (38)
77 (53)
39 (34)
(30)
43.5
40 (19)
37 (28)
23 (14)
6.5 (10)
(5.5)
4
7 (15)
8 (13)
(2.5)
3 (4.5)
2.5
6.5
6 (3.5)
2
0
2
–
4
2
4
10
a The yields of the products are based on direct observation of the reaction mixtures by 19F- and 1H-NMR spectroscopy, and are normalized
to the 100% total yield of Fp-containing products. The isolated yields are given in brackets.
products are formed. The yields in Table 1 indicate
that the principal compounds are C6F5Fe(CO)2Cp,
[CpFe(CO)2]2 and haloperfluoropolyphenyls. The yield
of NS product is substantially improved over that
which was previously obtained in this reaction by Bruce
and Stone [6a], and is higher for C6F5I (77%) than for
C6F5Br(Cl) (50–60%).
Such products as haloperfluoropolyphenyls C6F5-
(C6F4)nC6F4Hal (n=0–3) are particularly important,
as they can be formed from C6F5Hal only by a series of
NS reactions with [C6F5]− carbanion (Scheme 2) [12].
We have isolated three individual bromoperfluoro-
polyphenyls, their linear structure being evident from
the 19F-NMR spectrum. A small amount of pen-
tafluorobenzene could result from trapping of the
[C6F5]− anion by impurities in THF.
According to the halophilic mechanism outlined in
Scheme 1, C6F5Fe(CO)2Cp results from the interaction
of [C6F5]− and CpFe(CO)2Hal intermediates. The role
of the [C6F5]− carbanion in the NS reaction is clearly
demonstrated by trapping experiments summarized in
Table 2 as was done earlier for the reaction of C6F5Cl
with [CpFe(CO)2]K [10]. A sufficiently strong CH-acid2,
such as PhCH(Et)CN, added to the solution of
(1)
The interaction of [CpFe(CO)2]K with C6F5Br and
C6F5I results in HME generating the [C6F5]K interme-
diate, which can manifest itself in many different ways.
2 The acidity of PhCH(Et)CN (estimated pKa\21) lies between
those of C6F5H (pKa=25) and CpFe(CO)2H (pKa=19.2) [13].

V.A. I6ushkin et al. / Journal of Organometallic Chemistry 597 (2000) 77–86
79
Scheme 2. Reaction sequence producing perfluoropolyphenyls.
Table 2
The effect of the anion and radical traps on the course of the reaction of C6F5Hal with [CpFe(CO)2]K (FpK), THF, r.t.
Entry Reactants
Hal Trap
Relative yields (%) a
Yield of other products (%)
C (FpK) C6F5Hal/FpK
C6F5Fp Fp2 C6F5(C6F4)nC6F4Hal C6F5H
(mol l−1
)
ratio
(n=0–3)
1
Cl Ph(Et)CHCN 3,5
0.061
1.17
0
97
0
25
p-ClC6F4C(Et)(Ph)CN (10)
p-HC6F4C(Et)(Ph)CN (8.5)
p-HC6F4Fp (3)
equiv.
2
3
Cl Cy2PH 7,6 equiv.
Br Ph(Et)CHCN 3,9
equiv.
0.043
0.049
1.17
1.29
63
0
35 10
97
12
33
Fp(C6F4)nFp, n=2–5 (2)
p-HC6F4Fp (3)
0
4
Br Cy2PH 6,0 equiv.
0.042
1.31
54
41 5.5
48 23
12.5
Fp(C6F4)nFp, n=2–5 (2)
b
C6F5Fe(CO)nLm (3)
5
6
Br C6F5Br c
0.073
0.064
13.1
1.24
40
0
5
65
CpFe(C5H4C6F5) d (12)
I
Ph(Et)CHCN 6,6
equiv.
90
0
p-HC6F4Fp (10)
7
I
Cy2PH 7,1 equiv.
0.041
1.23
65
35
58
a
a See footnote to Table 1.
b Iron carbonyl s-C6F5-derivative of unknown structure (signals in 19F-NMR (THF, lf): −103.6 (m, 2F), −162, 98 (t, 1F) and −164, 68 (m,
2F) relative to CFCl3).
c The solution of FpK was added dropwise to the solution of C6F5Br.
d Obtained only after separation of the reaction mixture on silica gel.
[CpFe(CO)2]K prior to the reaction, acts as an anionic
trap, quantitatively intercepting the [C6F5]K intermedi-
ate, completely precluding the formation of NS-substi-
tution product C6F5Fe(CO)2Cp and haloperfluoro-
polyphenyls (Table 2). In anion trapping conditions
only one half of the carbonylmetallate is consumed in
the reaction with the starting aryl halide. The other half
is supposed to react with CpFe(CO)2Hal to give the
[CpFe(CO)2]2 dimer, which is formed almost quantita-
tively (Table 2). Consequently, the maximum possible
yield of C6F5H (together with Cp(CO)2FeC6F4H) re-
sulting from protonation of the [C6F5]− carbanion is
twice as low (50%) as the amount of [CpFe(CO)2]K
entering the reaction. The amount of C6F5H formed
from C6F5I in the presence PhCH(Et)CN (entry 6 Table
2) exceeds this stoichiometry. Presumably, the carban-
ion resulting from CH-acid, [PhC(Et)CN]−, can also
act as a halophilic agent towards C6F5I producing an
additional amount of [C6F5]−3. However, the same
[PhC(Et)CN]− carbanion displaces the p-fluorine in
C6F5Cl, as illustrated in Scheme 3.
An aryl halide can itself act as a [C6F5]− carbanion
trap: higher yields of bromoperfluoropolyphenyls and
[CpFe(CO)2]2 are obtained when [CpFe(CO)2]K solu-
tion is slowly added to an excess of C6F5Br in THF
(entry 5, Table 2).
The effect of a secondary phosphine Cy2PH, which
could act as a radical trap [15], is somewhat more
ambiguous, though the yields of NS product and dimer
do not change significantly, at least such is the case for
C6F5Cl and C6F5Br (entries 2, 4, Table 2). In the case
of C6F5I a large amount of C6F5H apparently results
from a secondary reaction induced by Cy2PH4.
2.2. Reactions of [Re(CO5]− and [Mn(CO)5]− anions
with pentafluorohalobenzenes
The reaction of C6F5Hal with [Re(CO)5]Na (Hal=
Cl, Br, I) and [Mn(CO)5]K (Hal=Br, I) in THF at
room temperature takes place within the time of the
reactant mixing, resulting in a clear light-yellow (Re) or
3 We have shown that C6F5I is immediately and quantitatively
converted to C6F5H when treated with [Ph2CCN]Na and an excess of
Ph2CHCN in THF. Halophilic reactions of activated aryl halides
with good nucleophiles, such as carbanions or thiolates, are not
usually regarded feasible. However, there are examples of such reac-
tions with C6F5I and o-NO2C6H4I [14], which are good positive
halogen atom donors.
4 Even without carbonylate C6F5I is slowly reduced with Cy2PH in
THF producing C6F5H (24% after 24 h).

80
V.A. I6ushkin et al. / Journal of Organometallic Chemistry 597 (2000) 77–86
Scheme 3. Trapping of [C6F5]− carbanion by CH-acid.
can be recorded. Within a few minutes the precipitate
begins to form, and after 15 min only 31% of the
haloacylmetallate is left according to 19F-NMR spec-
troscopy. The final yield (30 min) of C6F5Mn(CO)5
(80%, Table 3, entries 5, 6) is higher than obtained
earlier in this reaction [5]. Much slower rearrangement
of [C6F5(CO)Mn(CO)4I]K takes 35 h to complete and
yields along with C6F5Mn(CO)5 (62%) two unidentified
s-Arf complexes of Mn (Table 3, entry 7).
dark-red (Mn) solution. In the case of [Re(CO)5]Na
reactions and the reaction of [Mn(CO)5]K with C6F5I,
the resulting solutions stay unchanged long enough to
record the 19F-NMR spectrum, which shows that the
only product formed almost quantitatively is the haloa-
cylmetallate complex [C6F5(CO)M(CO)4Hal]−M%+ (Eq.
(2), Table 3, entries 1–3, 7). The structure of haloacyl-
metallates was also confirmed by IR and 13C-NMR
spectroscopy.
Several points should be kept in mind while dis-
cussing the mechanism of haloacylmetallate formation.
(a) Migratory insertion of CꢀO in the MnꢁC6F5 bond
of C6F5Mn(CO)5 induced by KHal is hardly possible
because C6F5 as a strongly electron-withdrawing group
group does not participate in the carbonylation reac-
tion [16]. Moreover [C6F5(CO)M(CO)4Hal]−M%+ are
decomposed irreversibly with the elimination of KHal
poorly soluble in THF. (b) The mechanism involving
single-electron transfer is also scarcely possible in the
reactions of C6F5Hal (Hal=Cl, Br, I) with [Re-
(CO)5]Na and [Mn(CO)5]K, because of the large differ-
ence between the redox potentials of these reactants
[11,17]6.
(2)
Rhenium haloacylmetallates are relatively stable in
solution, slowly decomposing to a complex mixture of
s-polyfluoroaryl rhenium complexes of unknown
structure5. In the solid state all haloacylmetallates are
unstable and decompose on evaporation of their solu-
tions. Notably less stable than Re complexes are man-
ganese haloacylmetallates, but their decomposition is a
selective one, yielding C6F5Mn(CO)5 as the main
product (Scheme 4). Although [C6F5(CO)Mn(CO)4Br]K
is supposed to be formed with quantitative yield, it
rearranges to a large extent before the NMR spectrum
There is little doubt that reactions of C6F5Hal with
[Re(CO)5]Na and [Mn(CO)5]K follow the same
halophilic mechanism found for the reactions with
[CpFe(CO)2]K, and it can easily be demonstrated by
the same anion-trapping technique (Table 4). In the
5 Complex [C6F5(CO)Re(CO)4Cl]Na decomposes completely in 7
days; the Br-derivative cis-[C6F5(CO)Re(CO)4Br]Na is more stable
and 80% of it decomposes after 45 days.
6 The same is true even for [CpFe(CO)2]K, the strongest reducing
agent among carbonylmetallates.

V.A. I6ushkin et al. / Journal of Organometallic Chemistry 597 (2000) 77–86
81
Table 3
Product yields in the reaction of [Re(CO)5]Na and [Mn(CO)5]K with C6F5Hal, THF, r.t.
Entry
Reactants
Time
Yield (%)
a
Hal
C [M(CO)5]M%
(mol l−1
C6F5Hal/[M(CO)5]M%
ratio
[C6F5(CO)M(CO)4-
Hal]M%
C6F5Mn(CO)5 C6F5H ArfM(CO)nLm
)
1
2
Cl
Br
[Re(CO)5]Na
(0.165)
[Re(CO)5]Na
(0.312)
[Re(CO)5]Na (1.38) 1.12
[Mn(CO)5]K
(0.360)
[Mn(CO)5]K
(0.070)
[Mn(CO)5]K
(0.420)
1.53
2.86
95
90
0
0
Trace
Trace
3
4
I
80 b
0
1.5
0.6
Cl c
1.24
1.09
1.19
10 h
4.6
51
44
12
3.6
9
5
6
Br
Br
30 min
15 min
0
5
0
31
30 min
18 min
0
91
80
2
0
9
3
1
7
I
[Mn(CO)5]K
(0.083)
1.31
10 h
35 h
32.5
0
46
62
9
9
20
23.5
a In each of the experiments with [Mn(CO)5]K several s-polyfluoroaryl complexes of Mn of general formulae ArfMn(CO)nLm are formed; L
denotes an unknown ligand. The overall yield of these complexes is given in this column.
b Re(CO)5I was also formed (20%).
c Main reaction products were Mn2(CO)10 (80%) and C6F5(C6F4)nC6F4Cl (n=0–3) (27%).
Scheme 4. Rearrangement of haloacylmanganates.
presence of anion traps, such as Ph2CHCN or BuOH7,
the formation of acyl complexes [C6F5(CO)Mn(CO)4-
Hal]K in the reaction of C6F5Hal (Hal=Br, I) with
[Mn(CO)5]K is completely eliminated. Trapping of the
[C6F5]− anion produces C6F5H, which under these con-
ditions is the only product resulting from C6F5Hal.
Proton transfer to the [C6F5]− anion should also gener-
ate the conjugated bases of CH- or OH-acid —
[Ph2CCN]− and [tBuO]−. Their fate is of certain inter-
est in connection with the unexplainably large amount
of C6F5H observed in these reactions (Table 4, entries
2, 3). With C6F5I it exceeds the maximum stoichiomet-
ric yield of 100%, possible if all the [Mn(CO)5]K is
spent in the reaction with the aryl halide, but not with
the Mn(CO)5Hal intermediate.
C6F5H originates from a halophilic reaction of
t
[Ph2CCN]K with C6F5I. Yet we must propose a differ-
ent explanation for the excessive yield of C6F5H in the
t
t
experiment with BuOH, considering that BuOK as a
‘hard’ nucleophile replaces fluoride in C6F5I to give
p-(tBuO)C6F4I (70%)8. To our surprise, this product
was not observed in the reaction of C6F5I with
t
[Mn(CO)5]K in the presence of BuOH. Instead we
t
found that BuOK is trapped with Mn(CO)5I to give
another haloacylmanganate, [tBuO(CO)Mn(CO)4I]K.
This fact allows one to explain the absence of
Mn2(CO)10 among the reaction products. The complex
[tBuO(CO)Mn(CO)4I]K was independently synthesized
t
from BuOK and Mn(CO)5I in THF. Its dark-red solu-
tion decomposes in two days at room temperature
giving Mn(CO)5I as the only identified Mn product. We
suppose that some reactive Mn anion (not necessarily
[Mn(CO)5]K) is also generated in the decomposition
process, and this anion abstracts halogen from the
In the case of Ph2CHCN the excessive yield of C6F5H
can be explained in the same way as it was for the
reaction of [CpFe(CO)2]K with C6F5I, that additional
7 Both Ph2CHCN (pKa=18) and BuOH (pKa=20 for neat liquid)
are weaker acids than Mn(CO)5H (pKa=15) [13].
t
8 Shown in a separate blank experiment.

82
V.A. I6ushkin et al. / Journal of Organometallic Chemistry 597 (2000) 77–86
Table 4
The effect of the anion and radical traps on the course of the reaction of C6F5Hal with [Mn(CO)5]K, THF, r.t.
Entry
Reactants
Trap, RH
Yield (%)
Other products
Hal
C ([Mn(CO)5]K)
C6F5Hal/[Mn(CO)5]K
ratio
[C6F5(CO)-
Mn(CO)4Hal]K
C6F5H
(mol l−1
)
1
2
3
4
Br
I
Ph2CHCN, 4.9
equiv.
0.054
1.20
1.14
1.27
1.26
0
0
100
127
114
68
Mn2(CO)10
Mn2(CO)10
Ph2CHCN, 4.1
equiv.
0.056
I
tBuOH, 5.2
equiv.
0.066
0
[tBuO(CO)Mn(CO)4I]K
Mn(CO)5I
I
Cy2PH, 3.8
equiv.
0.053
51
Scheme 5. Trapping of [C6F5]− anion in the reaction of [Mn(CO)5]K with C6F5Hal.
starting C6F5I, which actually accounts for the addi-
tional amount of C6F5H formed in the reaction of
C6F5I with [Mn(CO)5]K in the presence of tBuOH
(Scheme 5).
The effect of Cy2PH as a radical trap in the reaction
of C6F5I with [Mn(CO)5]K was also examined, but
results were complicated because of the secondary
reactions9 (entry 4, Table 4).
The experiments with anionic traps clearly demon-
strate the intermediacy of the [C6F5]− anion in the
reactions of Mn and Re carbonylmetallates with
C6F5Hal. The [C6F5]− and M(CO)5Hal intermediates
result from HME between the reactants, which is the
fist step in the general reaction mechanism (Scheme 6).
Formation of such products as halo(acyl)metallates in
the second step may be considered as ‘autotrapping’ of
these intermediates, and as such is probably the most
impressive evidence of the halophilic character of these
reactions. In the reaction of C6F5I the first step may
actually involve the formation of ate-complex [18],
which transforms to haloacylmetallate via intramolecu-
lar rearrangement (Scheme 6)10. It is difficult to distin-
guish between these two pathways by anion-trapping
experiments, because the C6F5 group in the ate-complex
can be protonated just as the free [C6F5]− anion [20].
The choice between these alternatives is not really
important for our study, because in both cases the key
reaction involves the attack of carbonylmetallate on the
halogen.
We have modelled the second stage in the proposed
mechanism — the addition of the intermediate carban-
9 In the reaction with [Mn(CO)5]K in the presence of Cy2PH an
excess of C6F5I (relative to [Mn(CO)5]K) is consumed completely to
give [C6F5(CO)Mn(CO)4I]K and C6F5H; their total yield exceeds
100%.
10 Stable iodine at complexes [(C6F5)2I]Li are known [19].

V.A. I6ushkin et al. / Journal of Organometallic Chemistry 597 (2000) 77–86
83
ion at CꢀO group of M(CO)5Hal. The attack of differ-
ent nucleophiles at the CꢀO group is typical of various
transition-metal carbonyls [16,21,22], but only a few
cases of such reactions with metal carbonyl halides are
known. The addition of MeLi to Re(CO)5Hal (Hal=
Cl, Br, I) and Mn(CO)5Br was reported to give stable
haloacylmetallates [Me(CO)M(CO)4Hal]Li [23], and in
the reaction of Re(CO)4(PPh3)Br with the MeLi com-
plex [CH3(CO)Re(CO)3(PPh3)Br]Li was observed as an
intermediate, which turns thereafter into the s-complex
MeRe(CO)4(PPh3) [24]. We have found that reaction of
C6F5Li with M(CO)5Hal allows one to obtain Li salts
of the corresponding haloacylmetallates (Eq. (3)). The
complexes are relatively stable at −60°C but decom-
pose even faster than Na or K salts at room tempera-
ture. For example, [C6F5(CO)Mn(CO)4I]Li is com-
pletely decomposed in 30 min at room temperature
giving Mn(CO)5I and a set of s-polyfluoroaryl com-
plexes of Mn, including C6F5Mn(CO)5.
(4)
In summary, we have shown that HME is a general
mechanism for the reactions of pentafluorohaloben-
zenes with a series of carbonylmetallates [CpFe(CO)2]−,
[Re(CO)5]− and [Mn(CO)5]− covering a wide range of
metal nucleophilicity.
3. Experimental
All air-sensitive reagents and reaction products were
handled in vacuum or in a purified argon atmosphere
using vacuum line and Schlenk techniques. Argon was
further purified by passage through an oxygen scav-
,
enger and 4 A molecular sieve columns.
3.1. Reagents and sol6ents
All solvents were refluxed over the appropriate dry-
ing agent and distilled prior to use. THF and THF-d8
were vacuum transferred from sodium benzophenone
ketyl. (Me3Si)2O, C6F5Br and PhCF3 were fractionally
distilled. C6F5I was extracted with concentrated H2SO4
to remove any C6F5NH2 impurity, washed with
aqueous K2CO3 and water, dried with CaCl2 and P2O5
and distilled under reduced pressure. The carbonylfer-
rate dimer [CpFe(CO)2]2 was purified as described pre-
viously [25]. Ph2CHCN was recrystallized from methyl
alcohol.
(3)
The reaction of C6F5Cl with [Mn(CO)5]K is rather
sluggish. It takes 10 h to complete and leads to a
complex mixture of products (Eq. (4), Table 3, entry 4),
the major one being Mn2(CO)10 (80%). Very weak
signals observed in the 19F-NMR spectrum correspond,
with a high degree of probability, to a trace amount of
haloacylmetallate [C6F5(CO)Mn(CO)4Cl]K. A substan-
tial amount of cloroperfluoropolyphenyls (27%) is
formed, as it was the case in the reaction of C6F5Cl
with [CpFe(CO)2]K (Reaction (4)). It is obvious that
this reaction obeys the same HME mechanism, but
because of the slow reaction between [C6F5]− and
Mn(CO)5Cl, these intermediates are intercepted by
C6F5Cl and [Mn(CO)5]K, respectively.
Volatile reagents and standards were dried over the
appropriate drying agent and vacuum transferred:
C6F5Hal, (Me3Si)2O and PhCF3 over the sodium mir-
t
t
ror, Cy2PH over P2O5 and BuOH over BuOK.
Metal carbonyl salts [CpFe(CO)2]K and [Mn(CO)5]K
were obtained quantitatively (95–98%) [3,26] by reduc-
tive cleavage of the dimers [CpFe(CO)2]2 and
Scheme 6. Alternative mechanisms of haloacylmetallate formation.

84
V.A. I6ushkin et al. / Journal of Organometallic Chemistry 597 (2000) 77–86
Table 5
The spectral data for haloperfluoropolyphenyls
a
Compound
19F-NMR Spectrum (THF, lf, ppm) b
Mass spectrum (m/z)
F1
F2
F3
F4
F5
F6
C6F5C6F4Br
−132.02
−131.87
−131.83
−137.92
−137.65
−137.64
−137.16
−136.87
−136.65
−161.13
−161.00
−160.97
−150.74
−150.36
−150.29
394, 396 [M+] (100%)
542, 544 [M+] (100%)
690, 692 [M+] (100%)
C6F5C6F4C6F4Br
C6F5(C6F4)2C6F4Br
C6F5(C6F4)3C6F4Br c
C6F5C6F4I d
−136.87
−136.65
838, 840 [M+
]
−120.04
−119.90
−138.35
−138.07
−137.31
−136.93
−161.73
−161.54
−151.12
−150.63
442 [M+] (100%)
590 [M+] (68)
C6F5C6F4C6F4I d
−137.49, −137.57
a Spectra of chloroperfluoropolyphenyls and CpFe(CO)2(C6F4)nFe(CO)2Cp (n=2–5) were published earlier [10].
b C6F6 (lf=−162.9 ppm) was used as an internal standard.
c In a mixture with C6F5(C6F4)2C6F4Br.
d In a mixture with other iodoperfluoropolyphenyls C6F5(C6F4)nC6F4I (n=0–3).
Mn2(CO)10 with excess of NaK2.8 alloy (0.20–0.25 ml
per 1 mmol of dimer) according to the procedure in the
literature [26]. [Re(CO)5]Na was prepared by the reduc-
tion of Re2(CO)10 with 0.5% NaHg [26] and purified by
the low-temperature crystallization from THF [27]. In
the preparation of Mn(CO)5I we slightly modified
the literature procedure [28], using potassium salt
[Mn(CO)5]K instead of sodium salt. The crude
Mn(CO)5I was recrystallized from hexane and further
purified by column chromatography on silica gel
(Merck, L 63/200). Re(CO)5Cl was synthesized accord-
ing to literature procedure [29].
with electron impact ionization, 70 eV ionizing energy,
using the direct inlet method, with a source temperature
of 150°C. The spectra of the reaction products are
presented in Tables 5 and 6.
3.3. The reaction of [CpFe(CO)2]K with C6F5Hal
(Hal=Br, I)
The general procedure for the reaction of
[CpFe(CO)2]K with C6F5Hal (Hal=Br, I) was almost
identical to that previously described for the reaction of
[CpFe(CO)2]K with C6F5Cl [10]. In the preparative
scale experiments 1.7–1.9 mmol of [CpFe(CO)2]K was
reacted with 1.9–2.8 mmol of C6F5Hal. The products
were isolated by column chromatography on silica gel.
When only NMR analysis was performed smaller
amounts of reactants were used: 0.16–0.18 mmol of
[CpFe(CO)2]K and 0.20–0.24 mmol of C6F5Hal. In all
of the experiments the aryl halide was in 15–50%
excess, the concentration of [CpFe(CO)2]K varied in the
range 0.04–0.08 mol l−1. In the experiment with the
reverse order of mixing, a solution of [CpFe(CO)2]K
(0.52 mmol) in THF (6 ml) was added dropwise to the
rapidly stirred solution of C6F5Br (1.68 g, 6.8 mmol) in
THF (1.1 ml).
3.2. Spectroscopic methods and reaction mixture
analysis
The 1H- (400 MHz), 19F- (376.3 MHz) and 13C-NMR
(100, 56 MHz) spectra were recorded on a Varian
VXR-4000 spectrometer at room temperature. The sig-
nals of the solvent were used as a chemical shift refer-
1
ence in H- and 13C-NMR spectra. 19F-NMR chemical
shifts are reported in ppm upfield from CFCl3 relative
to the signal of internal standard PhCF3 (−61.96
ppm). Qualitative and quantitative analysis of the reac-
tion mixtures was performed by 1H- and 19F-NMR
spectroscopy. The yields of the products were calcu-
lated from the ratio of the integral intensity of the
respective signals and the signals of the internal stan-
dard, a weighted amount of which was added to the
whole volume or an aliquot of the reaction mixture.
The IR spectra of the reaction mixtures and isolated
products were recorded in THF solution on an UR-20
spectrophotometer in a 0.2 mm CaF2 cell. Mass spectra
were recorded on a MS-890 spectrometer operating
3.4. The reaction of C6F5Hal (Hal=Cl, Br, I) with
[Re(CO)5]Na and [Mn(CO)5]K
The carbonylmetallate salt solution (0.25–0.56
mmol) in THF was transferred in vacuum from a
break-seal ampoule into an NMR sample tube, and
evaporated to a volume of 0.6–1.0 ml, to reach the
concentration of 0.25–0.8 mol l−1. The NMR tube was

V.A. I6ushkin et al. / Journal of Organometallic Chemistry 597 (2000) 77–86
85
Table 6
The spectral data for reaction products (THF, r.t.)
Compound
19F-NMR spectrum (lf)
13C-NMR spectrum (lc)
CꢂO
IR spectrum (w, cm−1
)
CꢂO
CꢂO
CꢂO
CpFe(C5H4C6F5) a
−139.09, −159.32, −163.40
−122.17, −149.18
n-(tBuO)C6F4I
cis-[C6F5(CO)Re(CO)4Cl]Na −150.21, −160.26, −162.09 185.45 (1C), 187.65 (2C), 188.18 (1C) 256.00 1928 s, 1998 vs, 2100 m 1607 w
cis-[C6F5(CO)Re(CO)4Br]Na −149.66, −160.25, −161.95 187.69 (1C), 189.71 (3C) 256.60 1928 s, 1995 vs, 2099 m 1605 w
cis-[C6F5(CO)Re(CO)4I]Na −148.05, −160.58, −162.49 187.06 (1C), 188.29 (1C), 188.63 (2C) 253.29 1928 s, 1992 vs, 2093 m 1605 w
cis-[C6F5(CO)Re(CO)4Cl]Li b −149.41, −158.55, −161.42
1936 s, 2005 vs, 2104 m 1581 w
1938 s, 1999 vs, 2099 m 1581 w
cis-[C6F5(CO)Re(CO)4Br]-
−148.90, −158.87, −161.50
Li b
cis-[C6F5(CO)Re(CO)4I]Li b
cis-[C6F5(CO)Mn(CO)4Br]K −148.24, −160.98, −162.59
−146.43, −159.44, −161.99
1924 s, 1984 s, 2095 m 1575 w
1932 s, 1995 s, 2079 m 1612 w
cis-[C6F5(CO)Mn(CO)4I]K
−147.15, −161.42, −162.90 213.76 (2C), 214.61 (1C), 222.26 (1C) 270.55 1931 s, 1998 vs, 2073 s 1612 w
cis-[C6F5(CO)Mn(CO)4I]Li b −145.58, −160.06, −162.34
cis-[tBuO(CO)Mn(CO)4I]K
1940 s, 1990 s, 2074 m 1600 w
1940 vs, 1987 vs, 2069 m 1619 w
a 1H-NMR (l) 4.12 (s, 5 H, Cp), 4.42 (t, 2H, C5H4C6F5), 4.75 (t, 2H, C5H4C6F5). MS (m/z): 352 [M+] (100%), 121 [CpFe+] (32%), 56 [Fe+
]
(10%).
b In 1:1 THF–hexane mixed solvent.
then frozen in liquid nitrogen, and thin glass spheres
containing the substrate (0.29–0.97 mmol), THF-d8 (ca.
0.2 ml) and PhCF3 (NMR standard) were broken, their
contents being vacuum transferred into the NMR sam-
ple tube, which was sealed off by flame. The tube was
warmed to r.t., shaken to mix the reactants and imme-
diately the recording of the NMR spectra was started.
We failed to isolate the rhenium haloacylmetallates in
crystalline form. Their sodium salts immediately de-
compose when their THF solutions are evaporated to
dryness. The stability of haloacylmetallates increased
when the sodium cation was exchanged for the bulky
[(Ph)3PꢂNꢂP(Ph)3]+, but these salts precipitated as an
oil from THF–ether mixed solvent, and decomposed
slowly even at low temperature.
tion of hydrogen) followed. When the reaction was
complete (B5 min), C6F5I (0.127 g, 0.43 mmol) was
added to the obtained bright-yellow solution of
[Ph2CCN]Na/Ph2CHCN. The reaction mixture immedi-
ately turned bright orange, but then the colour slowly
disappeared. The reaction mixture was filtered into an
NMR sample tube and analysed by 19F-NMR spec-
troscopy.
t
3.7. The reaction of C6F5I with BuOK/tBuOH
Similarly, tBuOK(0.4 g, 3.6 mmol) and tBuOH (1.5 g,
20 mmol) were dissolved in THF (7.4 ml) to which
C6F5I (1.34 g, 4.6 mmol) dissolved in THF (10 ml) was
then added. The obtained light-yellow reaction mixture
was analysed by 19F-NMR spectroscopy.
3.5. The reaction of C6F5Li with Mn(CO)5I and
Re(CO)5Hal (Hal=Cl, Br, I)
3.8. The reaction of C6F5I with Cy2PH
The solution of C6F5H (0.134 g, 0.8 mmol) in THF (1
ml) was slowly added under Ar atmosphere to BuLi in
hexane (0.25 mol l−1, 2 ml) at −80°C [8]. The C6F5Li
solution obtained was quickly added to a solution of
M(CO)5Hal (M=Mn, Re) (0.5 mmol) in THF (2 ml) at
−80°C. The resulting solutions were analysed by 19F-
NMR and IR spectroscopy.
In a similar way the solutions of Cy2PH (0.27 g, 1.4
mmol) in THF (4 ml) and C6F5I (0.14 g, 0.46 mmol) in
THF (10 ml) were mixed, without any visual change.
The reaction mixture was analysed by 19F-NMR
spectroscopy.
t
3.9. The reaction of Mn(CO)5I with BuOK
3.6. The reaction of C6F5I with
[Ph2CCN]Na/Ph2CHCN
t
When a solution of BuOK (0.028 g, 0.25 mmol) in
THF (2.5 ml) was quickly added to the solution of
Mn(CO)5I (0.032 g, 0.1 mmol) in THF (2.5 ml) the
reaction mixture immediately turned ruby. The reaction
was monitored by IR spectroscopy.
THF (3 ml) was vacuum distilled to a mixture of
Ph2CHCN (0.13 g, 0.66 mmol) and NaH (0.008 g, 0.33
mmol), upon which a vigorous reaction (with the evolu-

86
V.A. I6ushkin et al. / Journal of Organometallic Chemistry 597 (2000) 77–86
[13] (a) O.A. Reutov, I.P.Beletskaya, K.P. Butin, CH-acids, Nauka,
Acknowledgements
Moscow, 1980, pp. 19–40 (in Russian). (b) J.P. Collman, L.S.
Hegedus, J.R. Norton, R.G. Finke, Principles and Applications
of Organotransition Metal Chemistry, University Science Books,
Mill Valley, CA, 1987 (Chapter 3).
The financial support from the Russian Fundamental
Science Foundation (Grants 97-03-33161a and 96-15-
97484) and from the program ‘Integration of High
School with the Russian Academy of Sciences’ (Grant
234) is gratefully acknowledged.
[14] (a) M.R. Wiles, A.G. Massey, Tetrahedron Lett. (1967) 5137. (b)
B.J. Grady, D.C. Dittmer, J. Fluorine Chem. 50 (1990) 151. (c)
S. Montanari, C. Paradisi, G. Scorrano, J. Org. Chem. 58 (1993)
5628.
[15] (a) G.F. Smith, H.G. Kuivila, R. Simon, L. Sultan, J. Am.
Chem. Soc. 103 (1981) 833. (b) E.C. Ashby, R.N. De Priest,
Wei-Yang Su, Organometallics 3 (1984) 1718.
References
[16] J.P. Collman, L.S. Hegedus, J.R. Norton, R.G. Finke, Principles
and Applications of Organotransition Metal Chemistry, Univer-
sity Science Books, Mill Valley, CA, 1987 (Chapter 7).
[17] M. Tilset, V.D. Parker, J. Am. Chem. Soc. 111 (1989) 6711.
[18] (a) H.J. Reich, D.P. Green, M.H. Phillips, J. Am. Chem. Soc.
111 (1989) 3444. (b) H.J. Reich, D.P. Green, M.H. Phillips, J.
Am. Chem. Soc. 113 (1991) 1414.
[19] W.B. Farnham, J.C. Calabrese, J. Am. Chem. Soc. 108 (1986)
2449.
[20] V Schulze, M. Bro¨nstrup, V.P.W. Bo¨hm, P. Schwerdtfeger, M.
Schimeczek, R.W. Hoffmann, Angew. Chem. Int. Ed. 37 (1998)
824.
[1] (a) M.I. Bruce, F.G.A. Stone, Angew. Chem. B 80 (1968) 835.
(b) R.B. King, Acc. Chem. Res. 3 (1970) 417.
[2] (a) R.B. King, M.B. Bisnette, J. Organomet. Chem. 2 (1964) 38.
(b) M.I. Bruce, J. Organomet. Chem. 14 (1968) 461. (c) M.I.
Bruce, C.H. Davies, J. Chem. Soc. A. (1969) 1077. (d) I.P.
Beletskaya, G.A. Artamkina, A.Yu. Mil’chenko, P.K. Sazonov,
M.M. Shtern, J. Phys. Org. Chem. 9 (1996) 319.
[3] G.A. Artamkina, A. Yu. Mil’chenko, I.P. Beletskaya, O.A.
Reutov, J. Organomet. Chem. 311 (1986) 199.
[4] G.A. Artamkina, A. Yu. Mil’chenko, I.P. Beletskaya, O.A.
Reutov, Dokl. Akad. Nauk. 304 (1989) 616.
[21] M.Y. Darensbourg, Prog. Inorg. Chem. 33 (1985) 221.
[22] Houben-Weyl, Methoden der organischen Chemie, Thieme,
Stuttgart/New York, 1986. Bd. 13/9a, S. 7–160.
[23] K.P. Darst, C.M. Lukehart, J. Organomet. Chem. 171 (1979) 65.
[24] D.W. Parker, M. Marsi, J.A. Gladysz, J. Organomet. Chem. 194
(1980) C1.
[25] G.A. Artamkina, M.M. Shtern, P.K. Sazonov, I.P. Beletskaya,
Zh. Org. Khim. 32 (1996) 1319 [Russ. J. Org. Chem. 32 (1996)
1271 (Engl. Transl.)]
[26] J.E. Ellis, E.A. Flom, J. Organomet. Chem. 99 (1975) 263.
[27] P.K. Sazonov, M.M. Shtern, G.A. Artamkina, I.P. Beletskaya,
Zh. Org. Khim. 34 (1998) 1807.
[5] M.I. Bruce, D.N. Sharrocks, F.G.A. Stone, J. Chem. Soc. A
(1970) 680.
[6] (a) M.I. Bruce, F.G.A. Stone, J. Chem. Soc. A (1966) 1837. (b)
S.C. Cohen, J. Chem. Soc. Dalton Trans. (1973) 553.
[7] N. Ishikawa, S. Ibuki, Bull. Chem. Soc. Jpn. 47 (1974) 2621.
[8] D.E. Fenton, A.G. Massey, Tetrahedron 21 (1965) 3009.
[9] R. Chukwu, A.D. Hunter, B.D. Santarsiero, Organometallics 10
(1991) 2141.
[10] G.A. Artamkina, P.K. Sazonov, V.A. Ivushkin, I.P. Beletskaya,
Chem. Eur. J. 4 (1998) 1169.
[11] T.V. Magdesieva, I.I. Kukhareva, G.A. Artamkina, I.P. Belet-
skaya, K.P. Butin, Zh. Org. Khim. 30 (1994) 591.
[12] D.D. Callander, P.L. Coe, J.C. Tatlow, Tetrahedron 22 (1966)
419.
[28] M.H. Quick, R.L. Angelici, Inorg. Synth. (1979) 161.
[29] E.W. Abel, G. Wilkinson, J. Chem. Soc. (1959) 1501.
.
.