8
3
9
1
-Ethyl-2,3-dimethylimidazolium trifluoromethanesulfonate
2 H, H ), 0.93 (t, J = 7.4 Hz, 3 H, H ); δ (acetone-d , 150.87
C 6
ipso 2 ortho
(
2). Similar to the preparation of 1, reaction of 1,2-dimethyl-
MHz) 155.4 (m, C(phenyl) ), 139.2 (C ), 135.9 (C(phenyl) ),
4 meta para
imidazole (15.5 g, 161.2 mmol) with ethyl trifluoromethane-
sulfonate (25.0 g, 140.3 mmol) in 1,1,1-trichloroethane (60 mL)
gave after workup 30.9 g (80%) of a white solid, mp 113 ЊC
128.2 (C ), 128.0 (C(phenyl) ), 125.9 (C(phenyl) ), 121.5
5 6 7 8 9
(C ), 49.5 (C ), 33.9 (C ), 20.7 (C ), 14.3 (C ); m/z (FAB, positive
ϩ
ion) 366 (M , 5%), 289 (M Ϫ phenyl, 100%).
(
Found: C, 35.23; H, 4.90; N, 10.15; S, 11.47%. C H N O SF
8 13 2 3 3
requires C, 35.04; H, 4.78; N, 10.21; S, 11.69%; M 274.26);
ν/cm 3146s, 2983m, 2296w, 1641w, 1594m, 1548m, 1456w,
General procedure for hydroformylation experiments
Ϫ1
¯
In a nitrogen glovebox the catalyst precursor (0.15 mmol), a
magnetic stir bar and the solid reaction medium (5 g) or toluene
1
7
426m, 1392m, 1266s, 1227s, 1153s, 1088w, 1033s, 954m, 806m,
73s, 757m, 727m, 712w, 671m, 639s, 573s, 518s (KBr);
(
20 mL) were placed in a glass lined 300 ml stainless steel auto-
3
4/5
δ (acetone-d , 300 MHz) 7.68 (d, J = 2.1 Hz, 1 H, H ), 7.63
H
6
clave. The apparatus was sealed and 30 mmol of the freshly
distilled substrate was added. The vessel was placed in a
thermostatically controlled oil bath at the chosen temperature
and charged with a 1 : 1 hydrogen/carbon monoxide mixture
3
4/5
3
6
(
(
d, J = 2.1 Hz, 1 H, H ), 4.33 (q, J = 7.4 Hz, 2 H, H ), 3.94
9 8 3 7
s, 3 H, H ), 2.77 (s, 3 H, H ), 1.47 (t, J = 7.4 Hz, 3 H, H );
2
4/5
δC (acetone-d , 75.41 MHz) 136.7 (C ), 114.5 (C ), 113.3
6
1
4/5
6
9
(
(
q, J = 323.4 Hz, CF ), 112.4 (C ), 35.0 (C ), 26.3 (C ), 6.1
CF 3
7 8
(
Afrox) at 80 bar. After a reaction time of 6 h the apparatus was
C ), 0.4 (C ); δ (acetone-d , 283.65 MHz) Ϫ81.0; m/z (FAB,
F
6
cooled, opened, and the organic layer decanted off. The now
solid reaction medium was further extracted with chloroform
ϩ
Ϫ
positive ion) 125 (M , 100%); m/z (FAB, negative ion) 149 (M ,
00%).
1
(
10 mL). A sample of the combined organic fractions was
taken to determine the conversion and the linear : branched
ratio by H NMR. The reported results represent the average of
experiments carried out in duplicate.
1
-Butyl-3-methylimidazolium tetraphenylborate (3). To
1
ammonium tetraphenylborate (17.8 g, 52.8 mmol) was slowly
added at 25 ЊC a suspension of silver() oxide (6.1 g, 26.4 mmol)
in degassed H O (50 mL) under light protection. The grey
mixture was stirred for 1 h and a solution of 1-butyl-3-methyl-
2
X-Ray structure determination
imidazolium chloride (10.0 g, 57.3 mmol) in degassed H O
X-Ray structures were obtained for 1, 2, 3 and 4, however,
2
18
(
100 mL) was added. The suspension was stirred for 30 h and
recently the structure of 3 was published by Dupont et al. The
details of the data collection and structure solution and refine-
ment for 3 are, therefore, not included in this paper, although
they have been deposited with the Cambridge Crystallographic
Data Centre (CCDC deposition number 169679). However, in
the previous publication the authors were not able to determine
absolute structure, possibly as a result of twinning, since this
was also found to be present in the current determination.
In addition, and in contrast to the current determination, no
disorder was observed. As shown in the discussion, disorder
observed in a crystal structure is related to the melting point of
that crystal, therefore it is of great importance. Details of the
modelling of the disorder, as well as of the twinning in 3 are
thus included in this paper, and the discussion refers to the
current structural determination.
filtered using a d4-frit. After washing twice with degassed H O
2
Ϫ2
(
50 mL), the solvent was removed under vacuum (10 mbar)
and the remaining off-white solid was extracted with dichloro-
methane (50 mL). The organic fraction was dried with calcium
Ϫ2
chloride and the solvent was removed under vacuum (10
mbar) leaving 15.4 g (63%) of a white solid, mp 131 ЊC
(
decomp.) (Found: C, 83.75; H, 7.27; N, 6.00%. C H N B
32 35 2
Ϫ1
¯
requires C, 83.84; H, 7.69; N, 6.11%; M 458.45); ν/cm 3438m,
3
1
1
1
162m, 3133m, 3100s, 3082s, 3051s, 2994m, 2954s, 2869m,
941w, 1888w, 1824w, 1766w, 1678w, 1608w, 1579m, 1561s,
479s, 1449m, 1426s, 1379m, 1338w, 1267m, 1248s, 1159s,
143s, 1032m, 841s, 741s, 708s, 646m, 622m, 611s, 484m, 472w
(
KBr); δ (chloroform-d , 300 MHz) 7.53 (m, 8 H, CH(phen-
H 1
ortho meta
yl) ), 6.97 (m, 8 H, CH(phenyl) ), 6.78 (m, 4 H, CH(phen-
yl) ), 5.94 (t, J = 2.0 Hz, 1 H, H ), 5.76 (t, J = 2.0 Hz, 1 H,
H ), 4.67 (s, 1 H, H ), 3.15 (t, J = 7.6 Hz, 2 H, H ), 2.74 (s, 3H,
H ), 1.32 (m, 2 H, H ), 1.14 (m, 2 H, H ), 0.89 (t, J = 7.2 Hz,
para
3
5
3
Data were collected on a Nonius Kappa CCD diffract-
4
2
3
6
19
ometer using monochomated Mo-Kα radiation (λ = 0.71073
10
7
8
3
2
Å). Refinement was based on F . All non-hydrogen atoms were
9
1
3
H, H ); δ (chloroform-d , 75.41 MHz) 164.4 (q, J = 49.4
refined anisotropically. H atoms could in many cases be identi-
fied from Fourier difference maps, but were placed in calculated
positions using a riding model. Disorder was observed in the
structures of 2, 3 (this disorder had not been modelled in a
C
1
CB
2
ipso
ortho
Hz, C(phenyl) ), 135.9 (C(phenyl) ), 134.9 (C ), 126.1
meta
5
para
4
(
(
C(phenyl) ), 122.6 (C ), 122.2 (C(phenyl) ), 120.5 (C ), 48.9
6 10 7 8 9
C ), 35.3 (C ), 31.5 (C ), 19.2 (C ), 13.2 (C ); m/z (FAB, posi-
ϩ
Ϫ
18
tive ion) 139 (M , 100%); m/z (FAB, negative ion) 319 (M ,
00%).
previous structure determination ) and 4. Each structure was
1
dealt with in the same way: the positions of the disordered
atoms were identified from the Fourier difference map; all
atoms in the same disordered group were refined anisotropically
with common site occupancies and the total site occupancy for
the two disordered groups was constrained to 1; if necessary the
ISOR, SADI and/or SIMU restraint commands were used to
obtain sensible structures and displacement parameters. In
addition to the large amount of static disorder there is also a
considerable amount of dynamic disorder that is evidenced by
large anisotropic displacement ellipsoids. Structures for both 2
and 3 were found to be twinned. The space group in 3 is non-
(
3-Butylimidazole)triphenylboron (4). To a suspension of
ammonium tetraphenylborate (17.6 g, 52.2 mmol) in
acetonitrile (50 mL) was added 1-butylimidazole (15.0 mL,
1
14.1 mmol) at 20 ЊC. The white suspension was heated for 70 h
to 81 ЊC while the system was continuously purged with argon.
The mixture was filtered using a d4-frit and the residue was
washed twice with acetonitrile (10 mL). The solvent was
Ϫ2
removed under vacuum (10 mbar) and the remaining solid
was recrystallized from ethylacetate yielding 16.5 g (86%) of a
white solid, mp 152 ЊC (Found: C, 81.60; H, 7.75; N, 7.49%.
C H N B requires C, 81.97; H, 7.43; N, 7.65%; M 366.31);
20
centrosymmetric, and the Flack test suggested the presence of
a racemic twin. Refinement revealed a racemic twin ratio of
0.86 : 0.14. Identification of twinning in 2 was more compli-
cated—systematic absences observed were not consistent with
any space group, although the unit cell was found to have all
angles approximately equal to 90Њ, and thus data were collected
in P1. Careful inspection of the systematic absences and atomic
positions identified from an initial direct methods solution
2
5
27
Ϫ1
2
¯
ν/cm 3426m, 3145m, 3125s, 3062s, 3000s, 2966s, 2930s,
2
1
1
8
877m, 1961w, 1889w, 1829w, 1775w, 1725w, 1644w, 1614m,
586m, 1537s, 1484m, 1455s, 1428s, 1386w, 1361m, 1309w,
261m, 1241m, 1162s, 1108s, 1069m, 1027m, 998m, 882w,
65m, 834s, 799m, 751s, 727s, 706s, 672s, 648s, 620m, 608m
4
2
(
KBr); δ (acetone-d , 600 MHz) 8.01 (t, J = 1.5 Hz, 1 H, H ),
H
6
5
ortho
7
.39 (m, 1 H, H ), 7.20 (m, 6 H, CH(phenyl) ), 7.14 (m, 6 H,
revealed that the space group was P2 /c with twinning across a
1
meta
para
CH(phenyl) ), 7.07 (m, 3 H, CH(phenyl) ), 7.02 (m, 1 H,
H ), 4.19 (t, J = 7.6 Hz, 2 H, H ), 1.84 (m, 2 H, H ), 1.33 (m,
mirror plane perpendicular to the c-axis. In addition it was
found that the asymmetric unit consisted of two chemically
4
3
6
7
J. Chem. Soc., Dalton Trans., 2002, 1132–1138
1133