H. Kuniyasu et al. / Journal of Organometallic Chemistry 769 (2014) 34e37
35
clean conversion of 1 (
d
40.1) to CpRu(Br)(PPh3)2 (3) (
d
39.0) and
energy of TS2, which leads to the formation of 9, was 101.6 kJ/mol
higher than the reactants. On the other hand, the formation of 7
required 147.5 kJ/mol, which was clearly energetically more
demanding. Similarly, the energy of 8 (128.3 kJ/mol) was much
higher than that of TS5 (95.6 kJ/mol), which generates 3 by elimi-
nation 4. Therefore, we concluded that Mechanism 1 is unlikely. In
intermediate 9 (79.2 kJ/mol), one hydrogen of Me of the incoming
acetyl bromide is found in close proximity to the Cl atom, hydrogen
bond-like interaction as suggested by the Cl/H (2.53 Å). Two
MeC(O)Cl (4) was observed (6.3% of 3 after 0.5 h) [5c,8]. After 9 h, 3
was produced quantitatively. No intermediate was detected during
the course of the reaction. A similar treatment of 1 with MeC(O)I (5)
produced CpRu(I)(PPh3)2 (6) (d 38.0) after 2.5 h at room tempera-
ture. Next, preparative scale reactions were executed to demon-
strate the utility of the present reaction as a synthetic method.
Compounds 1 (1.0 mmol) and 2 (5.0 mmol) were added to C6H6
(50 mL) in a 100 mL flask fitted with a stirring bar in a glove box.
After the reaction mixture was stirred at 25 ꢀC for 3 h, the solvent,
excess 2 (b.p. 75e77 ꢀC), and 4 (b.p. 52 ꢀC) were removed in vacuo.
NMR spectra and elemental analysis showed that analytically pure
3 was obtained. A similar large scale reaction was conducted with 5
at 25 ꢀC for 4 h to quantitatively afford 6 [9].
pathways were considered from 9: The
s-bond metathesis that
provides 11 via the transition state TS1 (Mechanism 2), and the
oxidative addition of BreC bond, which yields Ru(IV) complex 10
via TS3 followed by the CleC bond-forming reductive elimination
to afford 11 via TS4 (Mechanism 3). The energies of TS1 and 11
amounted to 117.9 and 72.3 kJ/mol, respectively. The energies of
TS3, 10, and TS4 equaled 124.7, 93.6 and 117.5 kJ/mol, respectively.
The energy of TS1 was 6.8 kJ/mol lower than that for TS3, sug-
gesting that Mechanism 2 was more favorable than Mechanism 3.
After the formation of 11, in which the Br/H distance is 2.53 Å, the
associative elimination of 4 via TS5 (95.6 kJ/mol) produced 3 and 4.
A total energy change of ꢁ15.5 kJ/mol was calculated for the entire
reaction
(D
G
¼
ꢁ16.0 kJ/mol). In TS2, dihedral angles
:RueCleCeO and :RueBreCeO equaled ꢁ179.4ꢀ and 179.9ꢀ,
respectively. Likewise, in TS5, the dihedral angles :RueCleCeO
and :RueBreCeO equaled ꢁ179.9ꢀ and 177.8ꢀ, respectively. This
shows that Ru, Cl, C, O of the carbonyl group, and Br atoms are
nearly coplanar in TS2 and TS5. On the other hand, the dihedral
angles :RueCleBreC of TS2, 9, TS1, 11, and TS5 amounted to
179.8ꢀ, 116.0ꢀ, 137.9ꢀ, 104.0ꢀ, and 179.9ꢀ, respectively. Therefore, the
RueCleCeBr quadrangle bends downward from flat TS2 to 9, and
flips back to TS1 before bending downward again to 11 and flat-
tening again to TS5. The dihedral angle :RueCleBreC of TS3, 10,
and TS4 equaled 102.1ꢀ, 104.8ꢀ, and 93.6ꢀ in TS3, 10, and TS4,
respectively, suggesting that this quadrangle folded similarly to the
RueCleCeBr quadrangle. However, this movement in Mechanism
(2)
The mechanism of the reaction between 1 and 2 was theoreti-
cally investigated DFT using the M06 functional [10,11].
This computational method was shown to give reliable geom-
etries and energies in previous studies on the ligand exchange re-
actions of trans-M(Cl)[C(O)Ph](PPh3)2 (M ¼ Pt, Pd) and Au(Cl)(PPh3)
with RC(O)Br (R ¼ Ph, Me) [3,4]. Three possible reaction pathways,
Mechanism 1e3, initially were investigated (Scheme 1). In mech-
anism 1, the liberation of PPh3 produces the coordinately unsatu-
rated CpRu(Cl)(PPh3) (7). Next,
s-bond metathesis between the
RueCl bond of 7 and the CeBr bond of 2 affords TS1, and the sub-
sequent elimination of 4 yields CpRu(Br)(PPh3) (8), which can un-
dergo recoordination by PPh3 to form 3. The associative elimination
of PPh3 to form 9 before TS1 formation is considered in Mechanism
2. Mechanism 3 involves the oxidative addition of the BreC bond to
9 that produces Ru(IV) complex 10 [8d,12]. In addition to the spe-
cies shown in Scheme 1, the study identified transition states TS2,
TS3, TS4 and TS5. The energy diagram and optimized transition
state and intermediate structures are shown in Fig. 1 [13,14]. The
3 was less dynamic than in the of s-bond metathesis mechanism.
The C]O group gradually changed direction from right to left in
both mechanisms during this folding process (Fig. 1).
Besides those shown in Scheme 1, another possible reaction
mechanism involves the formation of the 18-electron cationic
complex 12 (L0 ¼ PPh3 or solvent), which reacts with 2 to provide 13,
and the subsequentelimination of L0 toform3 (Scheme 2)[5b,c,15]. If
the reaction proceeded via 12, the addition of free PPh3 or acetoni-
trile would accelerate the reaction [16a]. However, the trans-
formation was actually rather suppressed by the addition of PPh3
(0.1 equiv) or hardly affected by the addition of CD3CN (20 equiv)
under the same conditions described in Eq. (2), excluding the for-
mation of 12 and this mechanism. This retardation is consistent with
Mechanism 2, which involves the elimination of PPh3.
In agreement with the concerted
s-bond metathesis mecha-
nism, no significant effect was observed by the addition of 2,2,6,6-
tetramethylpiperidine 1-oxyl (TEMPO) and galvinoxyl during the
transformation of trans-Pt(Cl)[C(O)Ph](PPh3)2 to the corresponding
bromide [3]. In stark contrast, the radical and radical inhibitor
remarkably affected the present Ru-system. The reaction of 1 with 2
in the presence of 0.1 equiv of TEMPO under otherwise similar
conditions only gave 19% of 3 after 9 h (Eq. (3), compare this with
the result of Eq. (2)). On the other hand, the same reaction in the
presence of 1.0 equiv of 9,10-dihydroanthracene (DHA) produced 3
in 75% after 0.5 h and quantitatively within 2 h [16]. These results
clearly show the participation of a radical species in the reaction
mechanism. Reports have shown that Cp*Ru(Cl)(PPh3)2, an analog
of 1, subtracted a halogen from halogenated compounds during the
process of the atom transfer radical addition (ATRA) to alkenes [17].
A revised mechanism, Mechanism 4, is therefore proposed in
Scheme 1. Possible reaction pathways of the reaction between 1 and 2.