1-Sila-[1]ferrocenophanes, structure and reactivity Ring-opening by reactions with boron halides

Article abstract of DOI:10.1016/S0022-328X(03)00780-0

1-Sila-[1]ferrocenophanes (1-5) were studied with respect to their NMR spectroscopic parameters, in particular the coupling constants involving the carbon atom C(1) [¹ J (¹³ C(1),¹³C(2,5)), ¹ J (²⁹Si, ¹³C(1)) and ¹ J (⁵⁷Fe,¹³C)]. The disiloxane 6, containing two 1-sila-[1]ferrocenophane units, possesses a linear Si-O-Si group, as shown by X-ray structural analysis, being different from the bent structure of the same compound that has been reported in the literature. Ring-opening reactions of the complexes 2-5 with some boron halides (PhBCl₂, BCl₃, BBr₃ and BI₃) gave the ferrocene derivatives 7-12, in which the cyclopentadienyl rings bear silyl and boryl groups, respectively. Two of these complexes (9a and 10) were treated with lithium diethylamide in a 1:1 molar ratio, and in both cases selective substitution at the boron atom was observed (13, 14). All products were characterised by their ¹H-, ¹¹B-, ¹³C- and ²⁹Si-NMR data.

Full text of DOI:10.1016/S0022-328X(03)00780-0

Journal of Organometallic Chemistry 682 (2003) 180ꢀ187
/
www.elsevier.com/locate/jorganchem
1-Sila-[1]ferrocenophanes, structure and reactivity
Ring-opening by reactions with boron halides
Bernd Wrackmeyer *, Anahid Ayazi, Wolfgang Milius, Max Herberhold *
Laboratorium fur Anorganische Chemie der Universitat Bayreuth, Lehrstuhl f. Anorganische Chemie II, Universitactsstr. 30, NW1, D-95440 Bayreuth,
¨ ¨
Germany
Received 23 April 2003; received in revised form 24 June 2003; accepted 24 June 2003
Dedicated to Professor Hans Bock on the occasion of his 75th birthday
Abstract
1-Sila-[1]ferrocenophanes (1ꢀ
constants involving the carbon atom C(1) [1J(13C(1),13C(2,5)), 1J(29Si,13C(1)) and 1J(57Fe,13C)]. The disiloxane 6, containing two 1-
sila-[1]ferrocenophane units, possesses a linear SiÃOÃSi group, as shown by X-ray structural analysis, being different from the bent
structure of the same compound that has been reported in the literature. Ring-opening reactions of the complexes 2ꢀ5 with some
boron halides (PhBCl2, BCl3, BBr3 and BI3) gave the ferrocene derivatives 7ꢀ12, in which the cyclopentadienyl rings bear silyl and
/
5) were studied with respect to their NMR spectroscopic parameters, in particular the coupling
/
/
/
/
boryl groups, respectively. Two of these complexes (9a and 10) were treated with lithium diethylamide in a 1:1 molar ratio, and in
1
both cases selective substitution at the boron atom was observed (13, 14). All products were characterised by their H-, 11B-, 13C-
and 29Si-NMR data.
# 2003 Elsevier B.V. All rights reserved.
Keywords: Ferrocenes; 1,1?-Disubstituted; [1]Ferrocenophanes; Boron; Silicon; NMR; X-ray
1. Introduction
recently for the [1]ferrocenophanes 1 and 2 [8], the
general scope of the reaction with boron halides has not
been explored. In the present work, we discuss the
molecular structure of the disiloxane 6 containing two 1-
sila-[1]ferrocenpahane units, we report missing coupling
The severe distortion of the sandwich structure in
[1]ferrocenophanes [1ꢀ
in ring-opening reactions [4ꢀ
/
3] is mirrored by their reactivity
6], and also by unusual
/
constants [1J(13C,13C), nJ(29Si,13C) (nꢁ
1J(57Fe,13C)] for some of the 1-sila-[1]ferrocenophanes
1ꢀ5 (Scheme 1), and we describe the reactions of 2ꢀ
with some boron halides (BX3 with XꢁCl, Br, I, and
/1,2,3) and
NMR parameters. For the investigation of such effects,
1-sila-[1]ferrocenophanes are particularly suited because
they are readily accessible [3,4b,5a,4f], and can be stored
for prolonged periods without decomposition. Our
current interest in ferrocene derivatives containing
heteroelements [7] prompted us to use the ring-opening
reaction of 1-sila-[1]ferrocenophanes with boron halides
for the synthesis of ferrocene derivatives bearing differ-
ent substituents in 1,1?-positions. Although the princi-
ples of this particular ring-opening have been reported
/
/
5
/
PhBCl2) together with first attempts to make use of the
ring-opened products.
2. Results and discussion
2.1. Measurement of coupling constants for rare spin-1/2
nuclei in the 1-sila-[1]ferrocenophanes 1ꢀ5
/
* Corresponding authors. Tel.: ꢂ
/
49-921-552542; fax: ꢂ49-921-
/
552157.
E-mail addresses: b.wrack@uni-bayreuth.de (B. Wrackmeyer),
max.herberhold@uni-bayreuth.de (M. Herberhold).
Although 1H-, 13C-, 29Si- [9] and even 57Fe-NMR data
[10] of 1-sila-[1]ferrocenophanes have been reported,
0022-328X/03/$ - see front matter # 2003 Elsevier B.V. All rights reserved.
doi:10.1016/S0022-328X(03)00780-0

B. Wrackmeyer et al. / Journal of Organometallic Chemistry 682 (2003) 180ꢀ
/
187
181
In contrast to J(13C(2),13C(3)), the magnitude of the
coupling constants 1J(13C(1),13C(2,5)), determined for 1
(31.5 Hz) and 3 (32.1 Hz), is markedly reduced when
compared with that of 1,1?-bis(trimethylsilyl)ferrocene
[13] (36.3 Hz). In the case of 3, all ring 13C resonances
are different, and the magnitude of 1J(13C(3),13C(4))
could also be measured (46.0 Hz) which is almost
1
Scheme 1. 1-Sila-[1]ferrocenophanes studied by NMR spectroscopy,
X-ray crystallography (6), and by ring-opening reactions (2ꢀ5) with
boron halides.
/
1
identical with that of J(13C(2),13C(3)) (45.0 Hz) and
1J(13C(4),13C(5)) (45.2 Hz). Since the magnitude of
1J(29Si,13C(1)) (57.8 Hz) in 1 is also much smaller than
in 1,1?-bis(trimethylsilyl)ferrocene [13] (73.6 Hz), the
electron distribution around C(1) must be similar to that
of sp3 hybridised carbon atoms. This is in agreement
with the chemical shifts, d13C(1), and suggests a rather
different bonding situation for C(1) when compared
with the other carbon atoms of the cyclopentadienyl
ring. On the other hand, the fact that the magnitude of
1J(57Fe,13C(1)) in 1-sila-[1]ferrocenophanes is not sig-
1
coupling constants involving 13C such as J(13C,13C),
1
nJ(29Si,13C) (nꢁ
/
1,2,3) and J(57Fe,13C) are completely
missing in the literature. These data are of interest with
regard to the unusual bonding situation of the carbon
atom C(1), as shown for example by the marked shift of
the 13C(1) resonance to low frequencies, typical of
[1]ferrocenophanes. In routine 13C-NMR spectra, the
29Si satellites belonging to the 13C(1) resonance are
difficult to detect, since these 13C nuclei possess long
relaxation times and usually give rise to notoriously
weak signals. However, application of the refocused
INEPT pulse sequence [11], based on coupling constants
1
nificantly different from that of J(57Fe,13C(2,3,4,5)) or
of 1J(57Fe,13C) in unstrained ferrocene derivatives
nJ(13C(1),1H) (nꢁ
2,3), permits to overcome signal-to-
/
[13,14] seems to indicate that the nature of FeÃ
/C(1)
bonding has not changed much. However, this assump-
tion is only valid if the sign of 1J(57Fe,13C(1)) in
[1]ferrocenophanes is the same as in other ferrocenes,
and experimental evidence on this respect is not avail-
able as yet. There is also the possibility of a compensa-
tion of counteracting influences, considering the
complex relationship between magnitude and sign of
coupling constants and the electronic structure, in
particular if transition metal nuclei are involved.
noise problems, as has been suggested for other ferro-
cene and metallocene derivatives [12]. Thus, various
coupling constants can be determined, including
1J(13C,13C), within reasonable time (see Fig. 1 and
Table 1). The refocused INEPT pulse sequence, based
on 1J(13C,1H) can also be successfully applied to
enhance the intensities of the 13C(2,3,4,5) NMR signals
in order to facilitate the detection of satellites (see Fig.
2).
Fig. 1. 62.9 MHz 13C{1H}-NMR spectrum of Fe(C5H4)2[SiMe2] (1) CDCl3 (ca. 5%), showing the 13C(1) resonance, recorded by using the refocused
INEPT pulse sequence [11], based on 2,3J(13C,1H)ꢁ7.5 Hz (result of 3000 transients; acquisition time 3.5 s; repetition time 3 s); 29Si, 57Fe and 13C
/
satellites are marked. Distortion of phase and wrong intensities for the 13C satellites arise because the effects of homonuclear decoupling are not
refocused in the INEPT pulse sequence.

182
B. Wrackmeyer et al. / Journal of Organometallic Chemistry 682 (2003) 180ꢀ187
/
Table 1
13C- and 29Si-NMR data a of 1-sila-[1]ferrocenophanes (1ꢀ
/
5) and of two 1,1?-bis(silyl)ferrocenes for comparison
13C(3,4) 13C(SiR)
Compound
d
13C(1)
d
13C(2,5)
d
d
d
29Si
(1) [Fe(C5H4)2]SiMe2
32.9 [57.8] {4.3} 74.9 {5.0}
31.8
77.0 {3.9}
ꢃ
/
3.5 [56.0]
ꢃ
/
4.7
(2) [Fe(C5H4)2]Si(Ph)Me 32.1 [50.6] {4.4} 75.9, 76.0 {5.2} {5.2}
33.9
77.9, 78.0 {4.0}
{4.0}
ꢃ
/
2.5 [56.4] 135.7 [73.7] (i), 134.4 (o), 128.8
ꢃ
/
7.7
(m), 131.5 (p)
0.1 [62.3]
(3) [Fe(C5H4)2]Si(Cl)Me 33.2 [73.4] {3.5} 73.3, 75.4 [8.7] {5.4} {4.8} 78.0, 78.3 {5.4}
{4.8}
5.6
(4) [Fe(C5H4)2]Si(Cl)Ph 32.4 [77.0] {4.4} 74.6, 76.1 [8.2] [8.5] {4.9} 78.2, 78.7 {5.4}
{5.0}
132.1 (i), 133.9 (o), 128.5 (m), 131.9 (p)
ꢃ
13.9
/
{4.7}
(5) [Fe(C5H4)2]SiCl2
(6)
36.1 [99.1] {3.9} 74.8 [8.6] {4.9}
74.0, 75.5
79.5 [6.7] {3.6}
77.3, 77.9
ꢀ
ꢃ
/
ꢃ
ꢃ/
/
4.8
38.4
/
0.7
{[Fe(C5H4)]Si(Me)}2O
b
13.4
4.0
23.2
Fe(C5H4Ã
/
SiMe3)2
Si(Cl)Me2]2
72.7 [73.6] 73.3 [5.5] {4.8}
69.0 [83.1] {5.2} 73.1 [6.7] {4.2}
71.2 [4.4] {c}
72.3 {5.2}
0.2 [52.6]
2.7 [60.3]
ꢃ
/
Fe[C5H4Ã
/
a
b
c
At 239
Ref. [13], in [d6]acetone; 1J(13C(1),13C(2,5))ꢁ
Ref. [13]; 57Fe satellites overlap with 29Si satellites.
/
1 8C; coupling constants J(29Si,13C) in brackets, and 1J(57Fe,13C) in braces (all9
36.3 Hz; 1J(13C(2,5),13C(3,4)) 44.8 Hz.
/0.2 Hz).
/
2.2. Formation and molecular structure of the disiloxane
6
bent arrangement of the SiÃ
Si 143.25(12)8). In contrast, the X-ray structural analysis
carried out for crystals of 6, isolated from solutions of 3
/
OÃ
/
Si handle (angle SiÃ
/
OÃ
/
Hydrolysis of 3 under mild conditions with traces of
water in chloroform gives in low yield the disiloxane 6,
in which the [1]ferrocenophane rings are still present.
The formation of compound 6 in trace quantities has
also been reported to take place in hexane solutions of 3,
and its molecular structure has been determined [15].
The complex isolated from hexane solutions shows a
in chloroform, revealed a structure with a linear SiÃ
Si unit as shown in Fig. 3. Being aware of the bent SiÃ
OÃSi bridge in 6 when it was crystallised from hexane
and the data set was measured at low temperature, we
solved and refined the crystal structure of our sample of
6 very carefully. However, there have been no hints
/
OÃ
/
/
/
pointing to a nonlinear SiÃ
/
OÃ/Si unit. The systematic
Fig. 2. 62.9 MHz 13C{1H}-NMR spectrum of Fe(C5H4)2[Si(Cl)Me] (3) in CDCl3 (ca. 5%), showing the 13C(2) resonance, recorded by using the
refocused INEPT pulse sequence [11], based on 1J(13C,1H)ꢁ175 Hz (result of 3000 transients; acquisition time 3.5 s; repetition time 3 s); 29Si, 57Fe
and 13C satellites are marked.
/

B. Wrackmeyer et al. / Journal of Organometallic Chemistry 682 (2003) 180ꢀ
/
187
183
Fig. 3. ORTEP plot (50% probability; hydrogen atoms are omitted for
clarity) of the molecular structure of the siloxane 6 (see Table 2).
extinctions reveal clearly the orthorhombic space group
Pccn. The comparison with the published space group
P212121 [15] indicates that our sample is a completely
different modification of 6.
Important structural features of 6 are collected in
Table 2 in order to compare with corresponding data of
the isomer with the bent structure [15]. Most geometrical
parameters concerning the [1]ferrocenophane units are
similar for the two different crystalline phases of 6
except of the angle SiÃ
/
OÃ
/
Si. It is well known that this
bond angle is very sensitive to subtle effects [16,17].
However, this appears to be the first case of a single
disiloxane in which structural analyses have revealed a
substantial difference in the bond angle SiÃ
distantly related case concerns the molecular structures
of diastereomers with bond angles SiÃOÃSi of 180 and
172.8(2)8 [18]. The variation in the bond angle SiÃOÃSi
has been the subject of numerous experimental and
theoretical studies [19,20]. The present findings empha-
size once more the extreme sensitivity of this parameter
to apparently subtle, unintentional influences.
/
OÃ
/
Si. A
Scheme 2. Ring-opening reactions of the 1-sila-[1]ferrocenophanes (2ꢀ
6) with some boron halides.
/
/
/
/
/
2.3. Reactions of the 1-sila-[1]ferrocenophanes 2ꢀ5 with
/
boron halides
The results of the ring-opening reactions are sum-
marised in Scheme 2. The reactions proceed smoothly
Table 2
Selected geometrical data of the disiloxane 6 with linear and bent structure a
Bond lengths (pm) (linear SiÃ
OÃSi)
/
Bond lengths (pm) (bent SiÃ
OÃSi)
/
Bond angles (8) (linear SiÃ
OÃSi)
/
Bond angles (8) (lbent SiÃ
OÃSi)
/
/
/
/
/
SiÃ
/
O
160.59(11)
184.7(4)
163.08(18)
163.77(19)
185.1(3)
180
143.25(12)
96.82(11)
SiÃ
/
OÃ
/
Si
SiÃ
/
C(Me)
95.78(17)
C(1)Ã
C(6)
/
SiÃ
/
184.7(3)
188.2(3)
SiÃ
/
186.9(4)
186.2(4)
C(1,6)) 200.5(4)
111.78(13)
111.18(11)
OÃ
/
SiÃ
/
C(1)
C(6)
C(1,6)(fc)
188.5(3)
201.0(2)
201.6(2)
109.47(14)
21.5
111.21(11)
19.7, 20.2
OÃ
/
SiÃ
/
FeÃ
/
a b
b c
200.2(4)
35.5 (at C(1))
32.6 (at C(6))
Not given
a
b
c
Data for the bent structure taken from Ref. [15]; data collected at 173 K.
The angle a describes the angle between the least-square planes of the cyclopentadienyl rings.
The angle b describes the deviation of the C(1)Ã
/
Si and C(6)ÃSi bond vectors from the respective cyclopentadienyl ring plane.
/

184
B. Wrackmeyer et al. / Journal of Organometallic Chemistry 682 (2003) 180ꢀ187
/
under mild conditions, and afford the pure products 7ꢀ
12 in quantitative yield. In the case of BBr3 and BI3
(Scheme 2, b2), cooling to ꢃ78 8C is required before the
boron halides are added in order to suppress side
reactions. In all cases studied, 11B- and 29Si-NMR
spectra of the reaction solutions give the desired
information on the progress of the reactions and on
/
aspect of further transformations concerns the synthesis
of [n]ferrocenophanes (n ]2). Therefore, we have
studied the reaction of both 9a and 10 with lithium
diethylamide. As shown in Scheme 3 and by the NMR
data in Table 3, this reaction leads selectively to the
/
/
compounds 13 and 14, in which the SiÃ
/
Cl bonds are
unchanged. The 13C-NMR data of the NEt2 group
the number of products. The compounds 7ꢀ
/
12 are
indicates that there is restricted rotation about the BÃ
/
N
extremely sensitive towards traces of moisture. How-
ever, they could be characterised by the consistent set of
1H- (experimental), 11B-, 13C-, and 29Si-NMR data
(Table 3). Typically, the presence of a chiral centre
gives rise to eight 13CH signals of the cyclopentadienyl
rings (Fig. 4). In some cases (e.g. 9a,c), the assignments
in Table 3 were confirmed by 2D 1H/1H NOESY
experiments [21] which indicate the spatial proximity
bond as in many other aminoboranes [22].
3. Conclusions
The relatively small magnitude of the coupling con-
1
1
stants J(29Si,13C(1)) and J(13C(1)C(2)) in 1-sila-[1]fer-
1
of the H2,5 nuclei to the organic substituent at silicon.
rocenophanes such as 1ꢀ5 mirrors the changes in the
/
bonding situation for carbon atom C(1) when compared
with the other cyclopentadienyl carbon atoms. The
molecular structure of the disiloxane 6 in the solid state
appears to be affected by extremely subtle influences
such as the solvent from which the crystals are grown. It
appears that all boron halides can be used for ring-
opening reactions with 1-sila-[1]ferrocenophanes bear-
ing alkyl-, aryl- or chloro-substituents at silicon. The 1-
2.4. Substitution of a Bꢀ
diethylamino group
/
Cl function in 9a and 10 by the
Further use of the ring-opened products such as 7ꢀ12
depends a great deal on the functional groups at the
boron atom. In the cases of 9, 10 or 12, the replacement
/
of one of the BꢀCl functions by a B-dialkylamino
/
function is desirable, since the stabilising effect of the
amino group is expected to help later on in selective
silyl-1?-boryl-substituted ferrocenes 7ꢀ
further transformations, and in this context the selective
substitution of one BꢀCl against a BꢀNEt2 function was
demonstrated in the cases of 13 and 14.
/
12 invite for
reactions of the remaining Bꢀ/Cl function. Considering
/
/
the presence of SiꢀCl functions in these molecules, one
/
Table 3
11B-, 13C- and 29Si-NMR data a of the 1-silyl-1?-boryl-ferrocenes (7ꢀ
/
14)
Compound Fe(C5H4Ã
/
R)(C5H4Ã/R?)
d
13C(1) 13C(2,5)
d
d
13C(3,4)
d
13C(SiR) (BPh)
d
29Si
d
11B
(7) Rꢁ
R?ꢁB(Ph)Cl
(8) RꢁSi(Me)Cl2
R?ꢁB(Ph)Cl
(9a) RꢁSi(Me)Cl2
R?ꢁBCl2
(9b) RꢁSi(Me)(Cl)Br
R?ꢁBBr2
(9c) RꢁSi(Me)(Cl)I
R?ꢁBI2
(10) Rꢁ
RꢁBCl2
(11) Rꢁ
R?ꢁ
(12) Rꢁ
R?ꢁ
(13) Rꢁ
R?ꢁ
(14) Rꢁ
R?ꢁB(NEt2)Cl
/
Si(Me)(Ph)Cl
69.3
74.6, 74.8
76.1, 76.6
74.5
73.7, 73.8 0.5
b
77.1, 77.2
13.3
18.0
22.2
16.9
/
n.m. (br)
68.6 [100.6]
n.m. (br)
68.2 [99.6]
69.9 (br) c
70.4
55.4
55.4
51.4
49.1
29.8
52.0
54.6
52.6
/
74.6
77.8
6.0 [72.9]
136.2 (i), 136.8 (o), 128.1 (m), 135.1 (p)
6.5 [72.8]
/
77.6
/
74.96 [8.1] 74.98
/
76.8
78.1
/
75.3, 75.4
77.8, 77.9
75.5, 75.8
79.6, 79.7
75.2
75.8, 75.8 8.2 [70.1]
79.3, 79.3
/
69.7 (br) c
/
72.0
68.2 (br) c
68.9
76.7, 76.8 10.8 [66.5]
80.6, 80.6
ꢃ/4.2
/
/Si(Ph)Cl2
75.5
78.4
75.0
78.2
132.9 (i), 132.3 (o), 128.3 (m), 131.7 (p)
10.3
2.2
/
n.m. (br)
69.8
76.9
74.6
/SiCl3
136.2 (i), 136.7 (o), 128.1 (m), 134.9 (p)
/
B(Ph)Cl
n.m. (br)
68.2 [127.8]
69.8 (br) c
66.9 [101.8]
68.5 (br) c
65.7
77.9
/SiCl3
75.0 [10.1] 75.5 [7.3]
ꢀ/
2.0
/
BCl2
77.3
73.2
73.9
73.5
74.4
78.6
73.7
75.9
74.1
76.3
/Si(Me)Cl2
8.5 [72.8]
22.5
11.0
/
B(NEt2)Cl
Si(Ph)Cl2
43.5, 44.2; 15.0, 15.6
133.7 (i), 133.4 (o), 128.2 (m), 131.4 (p)
43.5, 44.2, 15.2, 15.8
36.5
36.4
/
/
n.m. (br)
a
At 239
/
1 8C; coupling constants J(29Si,13C) in brackets (90.2 Hz); n.m., not measured; (br) denotes the broad 13C-NMR signal of a carbon
/
atom linked to boron [25].
b
13C (SiPh, BPh)ꢁ
Measured at 235 K.
/
127.5, 127.6, 131.5, 130.6, 132.9, 134.3, 135.2, 135.0 (without assignment).
c

B. Wrackmeyer et al. / Journal of Organometallic Chemistry 682 (2003) 180ꢀ
/
187
185
Fig. 4. 62.9 MHz 13C{1H}-NMR spectrum of the ring-opened product 7, showing the range of the 13CH(cyclopentadienyl) resonances. The silicon is
a chiral centre, and therefore, all 13CH signals are different.
4.2. 1,1-Chloro(phenyl)-1-sila-[1]ferrocenophane (4)
Trichloro(phenyl)silane (1.5 ml, 9.4 mmol), dissolved
in hexane (50 ml) was added dropwise at ꢃ78 8C to a
/
suspension of fcLi2(tmeda) (2.7 g, 8.6 mmol) in hexane
(100 ml). The reaction mixture was kept stirring for 12 h,
then Na2SO4 (3 g) was added, and all insoluble material
was filtered off. The solvent was almost completely
removed in vacuo, the remaining concentrated solution
Scheme 3. Reactions of the 1-silyl-1?-dichloroboryl-ferrocenes 9a and
10 with lithium diethylamide.
was cooled to ꢃ
crystalline product (1.8 g; 66%; m.p. 79 8C) 4: 1H-NMR
(250.1 MHz; CDCl3): d1Hꢁ
3.83, 4.36, 4.43, 4.47 (all m,
8H, C5H4), 7.17ꢀ7.26, 7.97ꢀ8.0 (m, m, 3H, 2H, SiPh).
/
80 8C, and 4 precipitated as an orange
4. Experimental
/
/
/
4.1. General
4.3. Isolation of the disiloxane derivative 6
All preparative work and handling of the compounds
were carried out in an atmosphere of argon, using oven-
dried glassware, and carefully dried solvents (hexane
Hydrolysis of 0.05 g (0.1 mmol) fc[SiMeCl] 3 in
CHCl3 (5 ml) containing traces of water gave, after
partial slow evaporation of the solvent and storage for
and toluene). The 1-sila-[1]ferrocenophanes 1ꢀ3 and 5
/
several days at ꢃ
low yield, m.p. 150 8C (decomp.) 6: 1H-NMR (250.1
MHz; CDCl3): d1Hꢁ
4.56, 4.52, 4.32, 4.03 (all m, 16H,
C5H4), 0.68 (s, 6H, SiMe).
/
30 8C, a red crystalline product in a
[3,4b,5a,4f] were prepared by literature methods, using
dilithiated ferrocene fcLi2(tmeda) [23]. The chlorosi-
lanes, ferrocene, butyl lithium (1.6 M in hexane), and the
boron halides were used as commercial products with-
out further purification.
/
NMR spectra (1H-, 11B-, 13C-, 29Si-NMR): Bruker
ARX 250 spectrometer; pulse lengths for the INEPT
experiments [11,24] (13C- and 29Si-NMR) were carefully
calibrated, and the decoupler power and offset were
optimised in order to achieve signals as sharp as possible
and to avoid dielectric heating of the samples. The
careful preparation of the samples for these experiments
4.4. Synthesis of the compounds 7ꢀ12
/
4.4.1. General procedure
The respective 1-sila-[1]ferrocenophane (3 mmol) was
dissolved in hexane (40 ml), and the solution was added
dropwise at 0 8C to a solution of the respective boron
chloride (BCl3, PhBCl2; 15 mmol) in hexane (40 ml).
After stirring the reaction mixture at room temperature
(r.t.) for 3 h, the volatile materials were removed in
vacuo. The residue was taken up in hexane, the solution
filtered over Na2SO4, and then hexane was removed to
1
was controlled by observing highly resolved H-NMR
spectra. This guaranteed the best results in 13C-NMR
spectra using INEPT experiments. Mass spectra (EI, 70
eV): Finnigan MAT 8500 with direct inlet; the m/z data
1
refer to the isotopes H, 11B, 12C, 28Si, 56Fe. Melting
leave the pure (1H-NMR:
ꢀ95%) products. The
/
points (m.p.) (uncorrected): Buchi 510 melting point
¨
apparatus.
compounds 9b and 9c were obtained in a similar way,
except that BBr3 or BI3 (5 mmol) dissolved in hexane

186
B. Wrackmeyer et al. / Journal of Organometallic Chemistry 682 (2003) 180ꢀ187
/
(120 ml) were added very slowly under vigorous stirring
to a cold (ꢃ78 8C) solution of 3 (5 mmol) in hexane (100
ml).
Compound 7: red oil; yield 80%; 1H-NMR (250.1
MHz; CDCl3): d1Hꢁ
4.23, 4.30, 4.43, 4.69, 4.73, 4.77,
4.82 (all m, 8H, C5H4), 0.75 (s, 3H, SiMe), 7.38ꢀ7.63,
8.03ꢀ8.07 (m, m, 10H, SiPh, BPh).
Compound 8: red oil; yield 98%; 1H-NMR (250.1
MHz; CDCl3): d1Hꢁ
4.94, 4.85, 4.51, 4.36 (all m, 8H,
C5H4), 0.87 (s, 3H, SiMe), 7.49ꢀ7.73, 8.10ꢀ8.23 (m, m,
3H, 2H, BPh).
Compound 9a: red oil; yield 93%; H-NMR (250.1
MHz; CDCl3): d1Hꢁ
4.85, 4.61, 4.51, 4.36 (all m, 8H,
C5H4), 0.97 (s, 3H, SiMe). C11H11BCl4FeSi; EI-MS: m/
eꢁ
380 [42%, Mꢂ].
Compound 9b: dark red oil; yield 60%; 1H-NMR
(250.1 MHz; CDCl3): d1Hꢁ
4.97, 4.62, 4.55, 4.38 (all m,
8H, C5H4), 1.12 (a, 3H, SiMe). C11H11BBr3ClFeSi; EI-
MS: m/eꢁ
513 [100%, Mꢂ].
Compound 9c: dark red oil; yield 66%; 1H-NMR
(250.1 MHz; CDCl3): d1Hꢁ
4.65, 4.66, 4.21, 4.04, 3.98
(all m, 8H, C5H4), 0.89 (s, 3H, SiMe). C11H11BClFeI3Si;
EI-MS: m/eꢁ
653 [4%, Mꢂ].
Compound 10: red oil; yield 88.5%; H-NMR (250.1
MHz; CDCl3): d1Hꢁ
4.73, 4.60, 4.48, 4.39 (all m, 8H,
C5H4), 7.38ꢀ7.49, 7.72ꢀ7.76 (m, m, 3H, 2H, SiPh).
Compound 11: red oil; yield 97%; H-NMR (250.1
MHz; CDCl3): d1Hꢁ
4.80, 4.75, 4.40, 4.29 (all m, 8H,
C5H4), 7.32ꢀ7.56, 7.95ꢀ8.06 (m, m, 3H, 2H, BPh).
Compound 12: red oil; yield 89.1%; H-NMR (250.1
MHz; CDCl3): d1Hꢁ
4.78, 4.66, 4.57, 4.46 (all m, 8H,
C5H4).
4.6. Crystal structure determination of 6
/
The reflection intensities were collected on a Siemens
P4 diffractometer (Moꢀ
/
Ka-radiation, lꢁ71.073 pm,
/
/
graphite monochromated). Structure solution and re-
finement were carried out using the program package
SHELXTL-PLUS V.5.1. Measuring temperature for the
structure determination was 296 K. All non-hydrogen
atoms were refined with anisotropic temperature fac-
tors. The hydrogen atoms are on calculated positions.
All hydrogen atoms were refined applying the riding
model with fixed isotropic temperature factors.
/
/
/
/
/
1
/
Compound 6: C22H22Fe2OSi2, a red plate with
dimensions 0.18ꢄ
orthorhombic space group Pccn with the lattice para-
meters aꢁ11.2485(7), bꢁ24.1939(14), cꢁ7.2724(5)
1979.1(2) 106 pm3, Zꢁ 1.596 mmꢃ1
pm, Vꢁ 4, mꢁ
2333 reflections collected in the range 352q 5508, 1732
/
/
0.15ꢄ/0.08 mm crystallised in the
/
/
/
/
/
/
/
/
;
/
/
/
reflections independent, 1311 assigned as observed [I ꢀ
2s(I)], full-matrix least-squares refinement against F2
with 125 parameters converged at R1/wR2-values of
0.039/0.088; empirical absorption correction (C-scans)
resulted in min./max. transmission factors of 0.388/
0.462; the max./min. residual electron density was 0.31/
/
/
1
/
ꢃ .
0.32 10ꢃ6 e pmꢃ3
/
/
/
1
/
5. Supplementary material
/
/
1
Crystallographic data (excluding structure factors) for
the structure reported in this paper have been deposited
with the Cambridge Crystallographic Data Centre as
supplementary publication no. CCDC 208088 for com-
pound 6. Copies of the data can be obtained free of
charge on application to The Director, CCDC, 12
Union Road, Cambridge CB2 1EZ, UK (Fax (inter-
/
4.5. Synthesis of the compounds 13, 14
nat.):
ac.uk or www: http://www.ccdc.cam.ac.uk).
ꢂ44-1223-336033; e-mail: deposit@ccdc.cam.
/
4.5.1. General procedure
A suspension of lithium diethylamide (10 mmol) in
toluene (40 ml) was added dropwise at 0 8C to a solution
of 9a or 10 (10 mmol) in toluene (70 ml). The reaction
mixture was kept stirring for 4 h at r.t., then Na2SO4 (2
g) was added, and insoluble materials were filtered off.
Evaporation of the toluene in vacuo left the products as
viscous orange oils.
Acknowledgements
This work was supported by the Deutsche For-
schungsgemeinschaft and the Fonds der Chemischen
Industrie. A.A. thanks the DAAD for a scholarship.
1
Compound 13: yield 96.5%; H-NMR (250.1 MHz;
CDCl3): d1Hꢁ
/
4.47, 4.44, 4.29 (m, m, m, 4H, 2H, 2H,
C5H4), 1.00 (s, 3H, SiMe), 3.27ꢀ3.39, 1.03ꢀ1.23 (m, m,
4H, 6H, NEt2). C15H21BCl3FeNSi; EI-MS: m/eꢁ416
[100%, Mꢂ].
Compound 14: yield 92.9%; H-NMR (250.1 MHz;
CDCl3): d1Hꢁ
4.48, 4.37 (m, m, 5H, 3H, C5H4), 7.40ꢀ
7.53, 7.75ꢀ7.81 (m, m, 3H, 2H, SiPh), 3.24ꢀ3.34, 1.14ꢀ
1.22 (m, m, 4H, 6H, NEt2). C20H23BCl3FeNSi; EI-MS:
m/eꢁ
478 [100%, Mꢂ].
References
/
/
/
[1] (a) M. Herberhold, Angew. Chem. 107 (1995) 1985; Angew.
Chem., Int. Ed. Engl. 34 (1995) 1837.;
1
(b) M. Herberhold, T. Bartl, Z. Naturforsch. Teil B 50 (1995)
¨
1692;
/
/
(c) M. Herberhold, F. Hertel, W. Milius, B. Wrackmeyer, J.
Organomet. Chem. 582 (1999) 352;
/
/
/
(d) S. Barlow, M.J. Drewitt, T. Dijkstra, J.C. Green, D. O?Hare,
C. Whittingham, H.H. Wynn, Organometallics 17 (1998) 2113.
/

B. Wrackmeyer et al. / Journal of Organometallic Chemistry 682 (2003) 180ꢀ
/
187
187
[2] K.H. Pannell, V.V. Dementiev, H.L.F. Cervantes-Lee, M.T.
Nguyen, A.F. Diaz, Organometallics 13 (1994) 3644.
[3] W. Finckh, B. Tang, D.A. Foucher, D.B. Zambie, R. Ziembinski,
A. Lough, I. Manners, Organometallics 12 (1993) 823.
[4] (a) D.A. Foucher, B.Z. Tang, I. Manners, J. Am. Chem. Soc. 114
(1992) 6246;
[13] F.H. Kohler, W.A. Geike, N. Hertkorn, J. Organomet. Chem. 334
¨
(1987) 359.
[14] A.A. Koridze, N.M. Astakhova, P.V. Petrovskii, J. Organomet.
Chem. 254 (1983) 345.
[15] A. Berenbaum, A.J. Lough, I. Manners, Acta Crystallogr. Sect. E
58 (2002) 562.
(b) D.A. Foucher, R. Ziembinski, R. Petersen, J. Pudelski, M.
Edwards, Y. Ni, J. Massey, C.R. Jaeger, G.J. Vancso, I. Manners,
Macromolecules 27 (1994) 3992;
[16] S. Shambayati, S.L. Schreiber, J.F. Blake, S.G. Wierschke, W.L.
Jorgensen, J. Am. Chem. Soc. 112 (1990) 697; see also ibid 112
(1990) 6155 for corrections.
(c) J.K. Pudelski, I. Manners, J. Am. Chem. Soc. 117 (1995) 7265;
(d) I. Manners, Adv. Organomet. Chem. 37 (1995) 131;
(e) P. Nguyen, P. Go´mez-Elipe, I. Manners, Chem. Rev. 99 (1999)
1515;
[17] (a) P.D. Lickiss, S.A. Litster, A.D. Redhouse, C.J. Wisener,
Chem. Commun. (1991) 173;
(b) W. Wojnowski, B. Becker, K. Peters, E.M. Peters, H.G. von
Schnering, Z. Anorg. Allg. Chem. 562 (1988) 17;
(c) I.L. Karle, J.M. Karle, C.J. Nielson, Acta Crystallogr. Sect. C
42 (1986) 64;
(f) D.L. Zechel, D.A. Foucher, J.K. Pudelski, G.P.A. Yap, A.L.
Rheingold, I. Manners, J. Chem. Soc. Dalton Trans. (1995) 1893.
[5] (a) D.L. Zechel, K.C. Hultzsch, R. Rulkens, D. Balaishis, Y. Ni,
J.K. Pudelski, A.J. Lough, I. Manners, Organometallics 15 (1996)
1972;
(d) A.I. Gusev, M. Antipin, D.S. Yu. Yufit, Yu. T. Struchkov,
V.D. Sheludyakov, V.I. Zhun, S.D. Vlasenko, Zh. Strukt. Khim.
24 (1983) 178;
(b) R. Rulkens, A.J. Lough, I. Manners, S.R. Lovelace, C. Grant,
W.E. Geiger, J. Am. Chem. Soc. 118 (1996) 12683;
(c) P. Nguyen, G. Stojcevic, K. Kulbaba, M.J. MacLachlan, X.-
H. Liu, I. Manners, Macromolecules 31 (1998) 5977.
[6] (a) T. Mizuta, M. Onishi, K. Miyoshi, Organometallics 19 (2000)
5005;
(e) M.G. Voronkov, D. Yu. Nesterov, A.I. Gusev, N.V. Alekseev,
M.B. Lotarev, Dokl. Akad. Nauk SSSR 257 (1981) 1377.
[18] E. Lukevics, V. Dirnens, P. Arsenyan, J. Popelis, A. Kemme,
Main Group Met. Chem. 19 (1996) 167.
[19] (a) C.A. Ernst, A.L. Allred, M.A. Ratner, M.D. Newton, G.V.
Gibbs, J.W. Moskowitz, Chem. Phys. Lett. 81 (1981) 424;
(b) A.V. Belyakov, A. Haaland, D.E. Shorokhov, H.V. Volden,
R. West, Russ. J. Gen. Chem. 69 (1999) 235;
(b) T. Mizuta, Y. Imamura, K. Miyoshi, J. Am. Chem. Soc. 125
(2003) 2068;
(c) G. Calleja, F. Carre´, G. Cerveau, P. Labbe´, L. Coche-
Gue´rente, Organometallics 20 (2001) 4211.
(c) A.I. Ermakov, A.E. Merkulov, Zh. Strukt. Khim. 39 (1998)
624.
[7] M. Herberhold, in: A. Togni, T. Hayashi (Eds.), Ferrocenes.
Homogeneous Catalysis, Organic Synthesis, Materials Science,
[20] F. Tielens, F. de Proft, P. Geerlings, Theochem. 542 (2001) 227.
[21] J. Jeener, B.H. Meier, P- Bachmann, R.R. Ernst, J. Chem. Phys.
71 (1979) 4546.
VCH, Weinheim, 1995, pp. 219ꢀ278.
/
[8] F. Jakle, A. Berenbaum, A.J. Lough, I. Manners, Chem. Eur. J. 6
¨
(2000) 2762.
[22] H. Beall, C.H. Bushweller, Chem. Rev. 73 (1973) 465.
[23] M.D. Rausch, D.J. Ciappenelli, J. Organomet. Chem. 10 (1967)
127.
[9] M. Herberhold, A. Ayazi, W. Milius, B. Wrackmeyer, J.
Organomet. Chem. 656 (2002) 71.
[10] B. Wrackmeyer, O.L. Tok, A. Ayazi, F. Hertel, M. Herberhold,
Z. Naturforsch. Teil B 57 (2002) 305.
[24] J. Schraml, in: Z. Rappoport, Y. Apeloig (Eds.), The Chemistry
of Organic Silicon Compounds, vol. 3, Wiley, Chichester, 2001,
[11] (a) G.A. Morris, R. Freeman, J. Am. Chem. Soc. 101 (1979) 760;
(b) G.A. Morris, J. Am. Chem. Soc. 102 (1980) 428;
(c) G.A. Morris, J. Magn. Reson. 41 (1980) 185;
(d) D.P. Burum, R.R. Ernst, J. Magn. Reson. 39 (1980) 163.
[12] B. Wrackmeyer, M. Biersack, H.-D. Brendel, M. Herberhold, Z.
Naturforsch. Teil B 47 (1992) 1397.
pp. 223ꢀ
[25] (a) A. Abragam, The Principles of Nuclear Magnetism, University
Press, Oxford, 1961, pp. 333ꢀ338;
/
339.
/
(b) B. Wrackmeyer, Polyhedron 5 (1986) 1709;
(c) B. Wrackmeyer, Progr. NMR Spectrosc. 12 (1979) 227.