84 Organometallics, Vol. 23, No. 1, 2004
Hill et al.
fortuitous consequence that, in the data collection, the
separation of reflections overlapped because of twinning
was reasonably constant for any particular separation
along c*. The twin plane is an n glide perpendicular to
parameter TLX model27 was used for the group of atoms
(C1-C3, C8-C11, O1). All other non-hydrogen atoms
were refined as individual anisotropic atoms. Hydrogen
atoms were reincluded after each refinement cycle in
sensible calculated positions and given atomic displace-
ment parameters determined by the atoms to which
they are attached. The methyl hydrogens were included
in two orientations of half-occupancy.
1
c* at z ) /2.
For a reflection H ) ha * + kb* + lc* of the first twin
component, the corresponding reflection for the second
twin component is located at H′ ) ha ′* + kb′* + lc′*,
where a ′* ) a * - 2|a*/c*|(cos â*)c* and b ′* ) b* -
2|b*/c*|(cos R*)c*. Note that a ′*‚c′* ) a *‚c*, b′*‚c′* )
b*‚c*, and c′* ) -c*. The nearest twin component for
the first component is then H′′ ) ha * + kb* + l′′c*,
where H′′ is the nearest reflection to H′ ) ha * + kb* -
(l + 2h|a*/c*|(cos â*) + 2k|b*/c*|(cos R*))c*; i.e., l′′ is the
integer nearest to -(l - 0.3965h - 0.4993k) and the
separation in reciprocal space is (l′′ + l - 0.3965h -
0.4993k)c*. When the mirror is repeated, the h,k,l"
reflection of the second component is the reflection
nearest to the h,k,l reflection of the first component with
the same separation in reciprocal space. For convenience
we will say l′′ is -(l - 0.3965h - 0.5k), allowing values
for the separation in reciprocal space of reflections from
different twin components to be simply a function of h
and whether k is odd or even. To a first approximation,
assuming 0.3965 is exactly 2/5, almost total overlap
exists for h ) 5n when k is even, partial overlap exists
for h ) 5n ( 1 when k is odd, and even less partial
overlap exists for h ) 5n ( 2 when k is even; i.e., these
are the instances when |l′′ + l - 0.3965h - 0.5k| < 0.25.
We thus set j ) 0 for the first twin component and pj )
|h| for the second component, which only exists for the
appropriate parity of k.
Con clu d in g Rem a r k s
The macrocyclization of TTDD by [Ru(CO)2(PPh3)3]
is remarkable for the facility with which it proceeds
under ambient conditions. This we attribute, in part,
to the labilization of coligands by alkynes coordinated
to 16-electron metal centers (2- vs 4-electron alkyne
coordination).20 However, the failure of [Ru(CO)2(PPh3)3]
to cyclize 2,8-decadiyne indicates that thioether coor-
dination also plays a crucial role. While the synthesis
of 2 is illustrative of a potential route to functionalized
macrocyclic thioethers, the tenacity with which buta-
diene linkages bind to d8-M0L3 fragments suggests that
processes catalytic in metal will require extension of the
principle to other metal centers.
Exp er im en ta l Section
Syn th esis of [Ru (CO)(P P h 3)({η4-S(C2H4SCCMe)2CO-K-
S)] (2). A mixture of [Ru(CO)2(PPh3)3]11 (2.00 g, 2.12 mmol)
and 4,7,10-trithiatrideca-2,11-diyne (TTDD; 0.49 g, 2.12 mmol)
was stirred in dry toluene under nitrogen at room temperature
overnight. The pale yellow supernatant was decanted, and the
yellow product was dried in vacuo. Yield: 1.00 g (75%). Mp:
226-228 °C. Anal. Calcd for C30H29O2PRuS3: C, 55.45; H, 4.50;
P, 4.77; S, 14.80. Found: C, 55.72; H, 4.23; P, 4.80; S, 14.71.
1H NMR (CD2Cl2, 298 K, 300 MHz): δ 7.25 (m, 15 H, C6H5),
4.10 (ddd, 1 H, J HH ) 13.7, 13.5, 2.5 Hz, on C δ 28.6), 3.14
(ddd, 1 H, J HH ) 14.8, 3.0, 3.0 Hz, on C δ 40.3), 2.81 (ddd, 1 H,
J HH ) 15, 4.4, 2.8 Hz, on C δ 28.6), 2.68 (ddd, 1 H, J HH ) 13.3,
2.8, 2.7 Hz, on C δ 27.3), 2.30 (ddd, 1 H, J HH ) 14.8, 4.4, 2.4
Hz, on C δ 20.7), 2.12 (ddd, 1 H, coupling not resolved, on C δ
40.3), 2.01 (d, 3 H, J HP ) 1.4 Hz, CH3 on C δ 9.9, pseudo-trans
to CO), 2.00 (ddd, 1 H, coupling not resolved, on C δ 27.3),
1.13 (ddd, 1 H, J HH ) 14.9, 12.1, 2.9 Hz, on C δ 20.7), 0.96 (d,
3 H, J HP ) 4.1 Hz, CH3 on C δ8.8, pseudo-trans to PPh3).
The procedure used works well and is assisted by the
following. (i) When the overlapping reflection is much
more intense than the reference reflection, the back-
ground error increases, increasing the standard error
of the resulting intensity and thus reducing the weight-
ing for such reflections in the refinement. Refinement
only includes those reflections with I > 3σ(I), and some
of the most troublesome reflections get excluded from
refinement. Reflections for which Icalcd < 2.5σ(I) were
also excluded from the refinement. (ii) It was found that
excluding reflections with (sin θ)/λ < 0.2 was beneficial.
The limited number of reflections in this range are
better resolved than higher angle reflections. Their
exclusion highlights this and allows the pj values to
decrease as a function of |l′′ + l - 0.3965h - 0.5k| in a
sensible fashion. The values for p1 to p12 were found to
be 0.775(15), 0.210(13), 0.298(14), 0.836(15), 0.947(15),
0.655(15), 0.152(19), 0.409(20), 0.822(25), 0.909(30),
0.469(47), and 0.066(85) with values for |l′′ + l -
0.3965h - 0.5k| of 0.1035, 0.2070, 0.1895, 0.0860,
0.0175, 0.1210, 0.2245, 0.1720, 0.0685, 0.0350, 0.1385,
and 0.2420.
13C{1H} NMR (CD2Cl2, 298 K, 75 MHz): δ 208.3 (br, RuCO),
1
155.0 (CCO), 134.6 (d, J CP ) 45, C1(C6H5)), 133.9 (d, J CP
)
)
12, C2,3,5,6(C6H5)), 130.5 (d, 4J CP ) 2, C4(C6H5)), 128.7 (d, J CP
10 Hz, C2,3,5,6(C6H5)), 97.8, 89.7, 80.8, 80.2 (CdC × 4), 40.3
(CH2, R to C δ 27.3), 28.6 (CH2, R to C δ 20.7), 27.3 (CH2, R to
C δ 40.3), 20.7 (CH2, R to C δ 28.6), 9.9 (CH3, pseudo-trans to
CO), 8.8 (CH3, pseudo-trans to P). 31P{1H} NMR (CD2Cl2, 298
K, 121 MHz): δ 48.5 (s). IR (Nujol): 1918 (νRuCO), 1587, 1576,
1565 (νCO) cm-1. IR (CH2Cl2): 1929 (νRuCO), 1561 (νCO) cm-1
.
MS (FAB, positive ion, nba matrix): m/z 651 [M + H]+, 623
[M - CO + H]+ (isotopic distributions confirmed by simula-
tion). Crystal data: C30H29O2PRuS3, Mr ) 649.78, triclinic, P1h
(No. 2), a ) 9.4529(2) Å, b ) 11.2806(3) Å, c ) 13.1159(3) Å,
R ) 77.595(1)°, â ) 81.762(1)°, γ ) 89.895(1)°, V ) 1351.25(6)
Å3, Z ) 2, Fcalcd ) 1.597 g cm-3, T ) 200 K, yellow plate, 6144
independent measured reflections, R1 ) 0.052, wR2 ) 0.063,
4147 absorption-corrected reflections (I > 3σ(I), 2θ e 55°), 185
parameters.
Syn th esis of [Ru (CO)(d p p e){η4-S(C2H4SCCMe)2CO}]
(6). A solution of 2 (0.13 g, 0.20 mmol) and dppe (0.076 g, 0.2
mmol) in toluene (30 mL) was heated under reflux for 2 h.
The resulting pale yellow solution was cooled, filtered, and
The anisotropic refinement used a number of con-
straints. The three phenyl rings were constrained to
have a common refinable planar geometry of mm2
symmetry relative to the P atom to which they are
attached. All other non-hydrogen atom positional pa-
rameters were unconstrained. The atomic displacement
parameters for the triphenylphosphine was modeled
using TL rigid body models, with the T common to all
atoms (6 parameters) and the librations (6 parameters
per phenyl group) centered on the P atom. A 15-
(27) Rae, A. D. Acta Crystallogr., Sect. A 1975, A31, 570-574.