Metal-templated hydrogen bond donors as "organocatalysts" for carbon-carbon bond forming reactions: Syntheses, structures, and reactivities of 2-guanidinobenzimidazole cyclopentadienyl ruthenium complexes

Article abstract of DOI:10.1021/om500704u

The reaction of 2-guanidinobenzimidazole (GBI) and (η⁵-C₅H₅)Ru(PPh₃)₂(Cl) in refluxing toluene gives the chelate [(η⁵-C₅H₅)Ru(PPh₃)(GBI)]⁺Cl^- (1⁺Cl^-; 96%). Subsequent anion metatheses yield the BF₄^-, PF₆^-, and BAr_f^- (B(3,5-C₆H₃(CF₃)₂)₄^-) salts (77-85%). Reactions with CO give the carbonyl complexes [(η⁵-C₅H₅)Ru(CO)(GBI)]⁺X^- (2⁺X^-; X^- = Cl^-, BF₄^-, PF₆^-, BAr_f^-; 87-92%). The last three salts can also be obtained by anion metatheses of 2⁺Cl^- (77-87%), as can one with the chiral enantiopure anion P(o-C₆Cl₄O₂)₃^- ((δ)-TRISPHAT^-; 81%). The reaction of [(η⁵-C₅H₅)Ru(CO)(NCCH₃)₂]⁺PF₆^- and GBI also gives 2⁺PF₆^- (81%). The pentamethylcyclopentadienyl analogues [(η⁵-C₅Me₅)Ru(CO)(GBI)]⁺X^- (3⁺X^-; X^- = Cl^-, BF₄^-, PF₆^-, BAr_f^-; 61-84%) are prepared from (η⁵-C₅Me₅)Ru(PPh₃)₂(Cl), GBI, and CO followed (for the last three) by anion metatheses. An indenyl complex [(η⁵-C₉H₇)Ru(PPh₃)(GBI)]⁺Cl^- (96%) is prepared from (η⁵-C₉H₇)Ru(PPh₃)₂(Cl) and GBI. All complexes are characterized by NMR (¹H, ¹³C, ³¹P, ¹⁹F, ¹¹B), with 2D spectra aiding assignments. Crystal structures of 1⁺PF₆^-·CH₂Cl₂ and 1⁺BAr_f^-·CH₂Cl₂ are determined; the anion is hydrogen bonded to the cation in the former. Complexes 1-3⁺X^- are evaluated as catalysts (10 mol %, RT) for condensations of indoles and trans-β-nitrostyrene. The chloride salts are ineffective (0-5% yields, 48-60 h), but the BAr_f^- salts exhibit excellent reactivities (97-46% yields, 1-48 h), with the BF₄^- and PF₆^- salts intermediate. Evidence for hydrogen bonding of the nitro group to the GBI ligand is presented. GBI shows no catalytic activity; a BAr_f^- salt of methylated GBI is active, but much less so than 2-3⁺BAr_f^-.

Full text of DOI:10.1021/om500704u

Article
pubs.acs.org/Organometallics
Metal-Templated Hydrogen Bond Donors as “Organocatalysts”
for Carbon−Carbon Bond Forming Reactions: Syntheses, Structures,
and Reactivities of 2‑Guanidinobenzimidazole Cyclopentadienyl
Ruthenium Complexes
Alexander Scherer,† Tathagata Mukherjee,‡ Frank Hampel,† and John A. Gladysz*,†,‡
†Institut fur Organische Chemie and Interdisciplinary Center for Molecular Materials, Friedrich-Alexander-Universitat
̈
̈
Erlangen-Nurnberg, Henkestraße 42, 91054 Erlangen, Germany
̈
‡Department of Chemistry, Texas A&M University, PO Box 30012, College Station, Texas 77842-3012, United States
S
* Supporting Information
ABSTRACT: The reaction of 2-guanidinobenzimidazole (GBI) and
(η5-C5H5)Ru(PPh3)2(Cl) in refluxing toluene gives the chelate [(η5-
C5H5)Ru(PPh3)(GBI)]+Cl− (1+Cl−; 96%). Subsequent anion metatheses
−
−
−
yield the BF4 , PF6 , and BArf− (B(3,5-C6H3(CF3)2)4 ) salts (77−85%).
Reactions with CO give the carbonyl complexes [(η5-C5H5)Ru(CO)-
(GBI)]+X− (2+X−; X− = Cl−, BF4 , PF6 , BArf−; 87−92%). The last three
salts can also be obtained by anion metatheses of 2+Cl− (77−87%), as can
−
−
−
one with the chiral enantiopure anion P(o-C6Cl4O2)3 ((Δ)-TRI-
SPHAT−; 81%). The reaction of [(η5-C5H5)Ru(CO)(NCCH3)2]+PF6
−
and GBI also gives 2+PF6 (81%). The pentamethylcyclopentadienyl
−
analogues [(η5-C5Me5)Ru(CO)(GBI)]+X− (3+X−; X− = Cl−, BF4 , PF6 ,
−
−
BArf−; 61−84%) are prepared from (η5-C5Me5)Ru(PPh3)2(Cl), GBI, and
CO followed (for the last three) by anion metatheses. An indenyl complex
[(η5-C9H7)Ru(PPh3)(GBI)]+Cl− (96%) is prepared from (η5-C9H7)Ru-
(PPh3)2(Cl) and GBI. All complexes are characterized by NMR (1H, 13C, 31P, 19F, 11B), with 2D spectra aiding assignments.
Crystal structures of 1+PF6 ·CH2Cl2 and 1+BArf−·CH2Cl2 are determined; the anion is hydrogen bonded to the cation in the
−
former. Complexes 1−3+X− are evaluated as catalysts (10 mol %, RT) for condensations of indoles and trans-β-nitrostyrene. The
chloride salts are ineffective (0−5% yields, 48−60 h), but the BArf− salts exhibit excellent reactivities (97−46% yields, 1−48 h),
with the BF4− and PF6− salts intermediate. Evidence for hydrogen bonding of the nitro group to the GBI ligand is presented. GBI
shows no catalytic activity; a BArf− salt of methylated GBI is active, but much less so than 2−3+BArf−.
INTRODUCTION
■
Hydrogen bonding1 is a ubiquitous component of numerous
recognition2 and reactivity3,4 phenomena. In recent years, a great
deal of attention has been focused on developing small-molecule
hydrogen bond donors capable of catalyzing organic trans-
formations.3 Macromolecular hydrogen bond donor catalysts are
of course well known, as exemplified by enzyme active sites in
which peptidic NH or OH linkages activate substrates containing
carbonyl groups toward nucleophilic attack.2c,5
In designing small-molecule hydrogen bond donor catalysts,
modular systems that can be sterically or electronically fine-tuned
are advantageous.2e,6 In this regard, ureas and thioureas have
seen extensive use, often in conjunction with chiral substituents
and/or auxiliary functionality.3,6d This emphasis has been
prompted in part by the pioneering studies of Etter, who defined
a variety of hydrogen-bonding motifs in the solid state,7 such as
the two 1:1 adducts of simple carbonyl and nitro compounds
shown in Figure 1 (I, II).
Figure 1. Representative crystallographically characterized adducts of
urea hydrogen bond donors and Lewis bases.
transformations and manifest the diversity and modularity in-
herent in organometallic and coordination compounds. In work
to date, we have established that the inexpensive, air-stable, and
We have sought to develop families of metal-containing
hydrogen bond donors that are capable of catalyzing organic
Received: July 8, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/om500704u | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
easily synthesized cobalt(III) trication8 [Co(en)3]3+ and sub-
stituted analogues are effective catalysts for Michael additions of
malonate anions in the presence of amine bases.9 Good evidence
has been obtained for hydrogen bonding of the Michael accep-
tors to the NH hydrogen atoms, which would be expected to
become more acidic upon coordination to a tricationic spectator
metal. Furthermore, NH···X hydrogen bonds are evident in all
crystal structures of the salts [Co(en)3]3+yXz− (y/z = 3:1, 1.5:2,
1:3) that have been reported to date.10
enantioselective transformations are reported in the following
companion paper.15
RESULTS
■
1. Syntheses of Cyclopentadienyl GBI Complexes. The
ruthenium bis(phosphine) complex (η5-C5H5)Ru(PPh3)2(Cl)
was synthesized by a literature method16 or via a new microwave-
mediated procedure (Supporting Information). As shown in
Scheme 2, (η5-C5H5)Ru(PPh3)2(Cl) and GBI were reacted in
An allied approach would involve catalysts with NH hydrogen
atoms at positions remote from, as opposed to coordinated to,
the metal. Toward this end, our attention was drawn to a
chelate ligand that features five NH linkages, is commercially
available, and can be synthesized in a single step, namely,
2-guanidinobenzimidazole (GBI; Scheme 1, top).11 This species
Scheme 2. Syntheses of Cyclopentadienyl Ruthenium GBI
Complexes
Scheme 1. Structure of GBI and Chelation (Top) and
Preorganization Effects in Binding of Li+ to Cyclic Polyether
Hosts (Bottom)
has a well-established coordination chemistry.12 However, to our
knowledge there have never been any applications of its adducts
of transition metals, main group elements, or other electrophiles
in catalysis. Importantly, coordination reduces the number of
conformational degrees of freedom, thereby preorganizing13 the
hydrogen bond donor per Scheme 1 (top). The beneficial effect
of preorganization with respect to binding affinities has been
demonstrated for complexes of crown ethers and spherands with
cations, as illustrated for Li+ in Scheme 1 (bottom).14 There
would presumably be analogous effects upon reaction rates and
catalytic activity.
In this paper, we describe (1) the syntheses, structures, and
physical characterization of cationic organoruthenium deriva-
tives of GBI, (2) applications of these adducts as catalysts for
condensations of indoles with nitroalkenes, and (3) data that
establish the critical importance of hydrogen bonding in the
transition-state assemblies and thereby “second coordination
sphere” mechanisms. The complexes disclosed herein are
chiral but racemic. Related adducts that can be accessed in
enantiomerically pure form and applied as catalysts in highly
refluxing toluene. Workup gave the racemic “chiral-at-metal”
cationic monophosphine complex [(η5-C5H5)Ru(PPh3)-
(GBI)]+Cl− (1+Cl−) as a yellow powder in 96% yield. The salt
was insoluble in benzene and toluene, slightly soluble in CH2Cl2,
and soluble in polar solvents such as MeOH, EtOH, and DMSO.
Like most new complexes below, 1+Cl− was characterized
by NMR (1H, 13C, 31P), IR, and UV−visible spectroscopy, as
summarized in Table 1 and the Experimental Section. The mass
spectrum showed a strong ion for the cation 1+. A satisfactory
microanalysis was obtained. Together with literature data,12c,d
1
2D NMR experiments (1H,1H COSY and H,13C HETCOR)
enabled all of the GBI proton and carbon signals to be un-
ambiguously assigned (Supporting Information, Tables s1, s2).
These and other data supported the coordination of the benz-
imidazole CNAr and guanidine CNH groups, as verified
crystallographically below.
Next, as shown in Scheme 2 (step A1), simple metatheses
allowed the chloride anion of 1+Cl− to be replaced by the more
B
dx.doi.org/10.1021/om500704u | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
a
Table 1. C5H5 and C5Me5 H NMR Signals of 1−3+X− and IR
1
state; powders showed no noticeable decomposition after five
years. Curiously, microanalyses gave consistently low nitrogen
values, as summarized in the Experimental Section.
b
νCO Values (parentheses) for 2−3+X−
cation
2. Syntheses of Substituted Cyclopentadienyl GBI
Complexes. Another possibility for fine-tuning catalytic activity
would be to modify the cyclopentadienyl ligands. Accordingly,
analogues with bulkier, more electron donating, and more lipo-
philic pentamethylcyclopentadienyl ligands were sought. The
precursor (η5-C5Me5)Ru(PPh3)2(Cl)21 was readily available by a
new one-pot synthesis using pentamethylcyclopentadiene
(Experimental Section). This gave better yields than the two-
step literature procedures21 and avoided handling sensitive
ruthenium oligomers.
anion
Cl−
1+
2+
3+
4.41
4.43
4.61
5.02
5.19 (1938)
5.19 (1938)
5.20 (1942)
5.30 (1961)
1.58 (1915)
1.58 (1915)
1.58 (1922)
−
BF4
c
−
PF6
BArf−
1.56 (1931)
a
b
c
δ, DMSO-d6, 300 or 400 MHz, ppm. cm−1
.
Data for
2+(Δ)-TRISPHAT−: 5.18 ppm and 1945 cm−1.
weakly coordinating anions BF4 , PF6 , and BArf−.8,17 The new
salts 1+X− were isolated in 77−85% yields as slightly air sensitive
yellow powders with progressively increasing solubilities in
CH2Cl2. They were similarly characterized, including 19F and 11B
NMR spectra. The cyclopentadienyl 1H NMR signals exhibited
progressively downfield chemical shifts (Table 1), suggesting the
ruthenium center in 1+BArf− to have more cationic character
−
−
Reactions of (η5-C5Me5)Ru(PPh3)2(Cl) and GBI were carried
out under conditions similar to those used for the cyclo-
pentadienyl analogue 1+Cl− in Scheme 2. However, as detailed
elsewhere,22 workups did not give the target molecule [(η5-
C5Me5)Ru(PPh3)(GBI)]+Cl−. Thus, the crude product was
treated with CO. As shown in Scheme 3 (top), the carbonyl
18
than that in 1+Cl−. IR bands associated with the BF4− and PF6
anions are presented in Table 2, and key features are interpreted
below.
−
Scheme 3. Syntheses of Pentamethylcyclopentadienyl (Top)
and Indenyl (Bottom) Ruthenium GBI Complexes
−
−
Table 2. IR ν(BF4 ) and ν(PF6 ) Values for Selected BF4− and
a
PF6− Salts
anion
−
−
cation
BF4
PF6
b
Na+
1+
1015
806
1089, 1078, 1011
1069, 1015
880, 862, 841
b
2+
837
b
3+
1093, 1089, 1023, 997
842
a
b
Powder film measurements (ATR); values are in cm−1. Asymmetric
absorption with unresolved shoulders.
In general, electron-withdrawing substituents lead to stronger
hydrogen bond donors. Thus, in the interest of fine-tuning
catalyst activity (see below), it was sought to replace the PPh3
ligand by a more weakly donating or stronger π-accepting ligand.
As shown in Scheme 2 (step B1), a solution of 1+Cl− was
aspirated with a stream of CO or stirred under a static CO
atmosphere. Workups gave the substitution product [(η5-
C5H5)Ru(CO)(GBI)]+Cl− (2+Cl−) as an off-white powder in
91−93% yields. Analogous carbonylations were conducted with
1+BF4 , 1+PF6 , and 1+BArf− (step A2). These afforded the
corresponding salts 2+X− as yellow powders in 87−92% yields.
complex [(η5-C5Me5)Ru(CO)(GBI)]+Cl− (3+Cl−) was isolated
as an off-white powder in 77% yield. Subsequent anion ex-
change reactions analogous to those in Scheme 2 gave the salts
−
−
−
−
Alternatively, 2+BF4 , 2+PF6 , and 2+BArf− could be accessed
in 77−88% yields by exchange of the chloride ion of 2+Cl−, as
shown in Scheme 2 (step B2). Both overall routes from 1+Cl− to
2+X−, “A” and “B” (Scheme 2), have been repeated by several
generations of co-workers, and “B” has been found to be more
easily reproducible.19 Another refinement involves an alternative
starting material, the cationic bis(acetonitrile) complex [(η5-
3+BF4 , 3+PF6 , and 3+BArf− as yellow powders in 61−84%
yields. These complexes were air stable in solution and the solid
state, were readily soluble in CH2Cl2 and other solvents of
moderate polarity, and gave correct microanalyses. The IR νCO
values were lower than those of 2+X− (1915−1931 vs 1938−
1961 cm−1; Table 1), indicative of increased electron density at
ruthenium.
−
−
C5H5)Ru(CO)(NCCH3)2]+PF6 employed in Scheme 2, step
−
Cl. As with the starting material (η5-C5H5)Ru(PPh3)2(Cl), this
educt is easily prepared in one step from a commercially available
precursor.20 Addition of GBI directly affords the hexafluor-
ophosphate salt 2+PF6− in 81% yield, saving two steps.
The synthesis of an indenyl analogue of 1+Cl− was also
investigated. As shown in Scheme 3 (bottom), a reaction of
(η5-C9H7)Ru(PPh3)2(Cl)23 and GBI similar to that used for
1+Cl− gave the target complex [(η5-C9H7)Ru(PPh3)(GBI)]+Cl−
(4+Cl−) as an orange powder in 72% yield. However, 4+Cl− de-
composed over the course of several days in CH2Cl2, as assayed
The cyclopentadienyl 1H NMR chemical shifts of 2+X− were
downfield of those of 1+X− (δ 5.19−5.30 vs 4.41−5.02; Table 1),
suggesting reduced electron density at ruthenium.18 Accordingly,
2+X− exhibited good air stability both in solution and in the solid
1
by H and 31P NMR. Furthermore, anion exchange and PPh3/
CO substitution reactions were unsuccessful.22
C
dx.doi.org/10.1021/om500704u | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
3. Hydration, H/D Exchange, and Nonracemic Com-
The new salt was characterized by NMR and IR spectroscopy
plexes. Aqueous conditions (biphasic CH2Cl2/water) were
and mass spectrometry, analogously to 2+X− above. Only
−
1
used when preparing the tetrafluoroborate salts 1−3+BF4 by
one broad cyclopentadienyl H NMR signal was observed in
anion exchange, and hydrates (2.0−0.5 H2O) were isolated in all
cases. The salt Na+BArf− is commonly obtained as a hydrate,24
and 1−3+BArf− all contained low levels of water (2.0−1.0 H2O).
The water could be removed by crystallization, as illustrated by
an X-ray structure below.
DMSO-d6. In contrast, in the much less polar solvent C6D6, two
closely spaced signals of equal intensity were apparent, presum-
ably one for each of the diastereomeric salts (Scheme 4). How-
ever, all attempts to separate the diastereomers by crystallization
were unsuccessful.
When DMSO-d6 or CDCl3 solutions of 1−4+X− were treated
with MeOH-d4 (6 equiv), the NH protons underwent rapid H/D
exchange. As shown in the Supporting Information (Figure s1),
the NH signals disappeared or coalesced into a single peak
and the total integration diminished. Some of the aromatic CH
signals became sharper. A variety of cationic coordination com-
pounds of GBI have been quantitatively deprotonated by weak
bases such as pyridine, NaOMe, and Na2CO3.12b−d,25 Hence, it is
not surprising that rapid exchange can be observed in the absence
of added base. Also, the GBI ligand is in principle capable of
numerous prototropic equilibria, some of which entail formal
1,3-shifts of protons from the noncoordinating NH/NH2 moie-
ties to the coordinating CNAr/CNH moieties. These may
participate in the exchange process, and examples are illustrated
in the Supporting Information (Scheme s1).
The second approach involved the reversible derivatization
of one of the NH or NH2 moieties by the enantiopure chiral
cyclopentadienyl rhenium cation [(η5-C5H5)Re(PPh3)(NO)-
(CO)]+.27 Although a series of stable adducts could be prepared,
as described elsewhere,22 the spectroscopic data did not un-
ambiguously reveal which NH or NH2 group had reacted.
Furthermore, only one singlet was observed for each of the
ruthenium and rhenium cyclopentadienyl ligands, as opposed to
doubled singlets diagnostic of diastereomeric adducts. All attempts
to separate the putative diastereomers were unsuccessful.
4. Crystallographic Characterization. During the course
of the above syntheses, single crystals of the solvate 1+PF6 ·
−
CH2Cl2 were obtained. X-ray data were collected and refined as
described in the Supporting Information (Table s3) and
Experimental Section. The resulting structure is shown in
Figure 2. Key metrical parameters are summarized in Table 3.
The cation is formally octahedral, with the cyclopentadienyl
ligand occupying three coordination sites, as evidenced by
P−Ru−N and N−Ru−N bond angles of ca. 90°. The GBI ligand
is nearly planar, as reflected by many torsion angles with values
near 0° or 180°. The bond lengths of the coordinated CNH
(C22−N21) and CNAr (C24−N32) linkages (1.280(4) and
1.314(4) Å) are shorter than the other four carbon−nitrogen
bonds about C22 and C24 (1.349(4)−1.385(4) Å). Alternative
tautomers of the GBI ligand (Scheme s1) would afford different
bond length patterns.
In order to test the preceding chiral ruthenium complexes
as enantioselective catalysts, nonracemic variants would be
required. Two possible routes to enantiopure complexes were
investigated. The first involved the nonracemic chiral anion (Δ)-
TRISPHAT−,8 the structure of which is depicted in Scheme 4.
26
The chloride salt 2+Cl− and (n-Bu)3NH+(Δ)-TRISPHAT−
were combined in CH2Cl2, and a biphasic workup gave the target
salt 2+(Δ)-TRISPHAT− as a white powder of ca. 95% purity in
81% yield.
Scheme 4. Reaction of 2+Cl− and (Δ)-TRISPHAT− (Top): 1H
NMR Spectra of the Cyclopentadienyl Signal in DMSO-d6
(br s) and C6D6 (2 × br s) (Bottom)
−
As illustrated in Figure 2, each PF6 anion exhibits multiple
hydrogen bonding to each of two neighboring cations. All of the
NH linkages except for N21−H21 participate. Although the
motif about each PF6− anion is unsymmetrical, the two cations
are related by an inversion center, resulting in identical patterns
of bonds to the anions. Curiously, we are only able to locate
structures of two other complexes that exhibit hydrogen bonding
between NH and PF6 moieties.28 In both cases, the PF6 unit
is covalently bound via a M−F−P linkage to the metal fragment.
In any case, the F···H, F···N, and P···N distances, which are
summarized in Table 4, are in typical ranges for hydrogen
bonds.1b,28
A CH2Cl2 monosolvate of 1+BArf− could also be crystallo-
graphically characterized. The resulting structure is shown in
Figure 3. One CF3 group was disordered over two positions, the
occupancies of which could be refined to a 62:38 ratio. Key
metrical data are summarized in Table 3. In this case the GBI
ligand is noticeably puckered, with torsion angles that deviate
more from 0° or 180°. However, the carbon−nitrogen bond
lengthsexhibit similar patterns. Nohydrogen-bonding interactions
are evident, consistent with the poor acceptor properties of
BArf−.29 The ruthenium−phosphorus distance (2.3154(10) Å) is
slightly longer than that in 1+PF6 ·CH2Cl2 (2.302(3) Å), sug-
−
gesting that the electron density on ruthenium and back-bonding
are enhanced when the anion can engage in hydrogen bonding.
5. Reactions Involving Nitroalkenes. Condensations of
indoles (5) and trans-β-nitrostyrene (6) to give 3-substituted
indoles (7)sometimes termed Friedel−Crafts alkylations
are often used as benchmarks for hydrogen bond donor
D
dx.doi.org/10.1021/om500704u | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
−
Figure 2. Thermal ellipsoid diagram (50% probability level) showing the structures of two molecules of 1+PF6 ·CH2Cl2 with solvate molecules omitted
and hydrogen bonding between cations and anions.
a
−
Table 3. Key Bond Lengths [Å], Torsion Angles [deg], and Bond Angles [deg] for 1+PF6 ·CH2Cl2 and 1+BArf−·CH2Cl2
1+PF6−·CH2Cl2
1+BArf−·CH2Cl2
1+PF6−·CH2Cl2
1+BArf−·CH2Cl2
Rul−N21
Rul−N32
Rul−Pl
N21−C22
C22−N21A
C22−N23
N23−C24
C24−N25
C24−N32
N25−C26
C26−C31
C31−N32
Pl−Rul−N21−C22
Pl−Rul−N32−C24
Rul−N21−C22−N23
Rul−N21−C22−N21A
Rul−N32−C24−N23
N21−C22−N23−C24
C22−N23−C24−N32
2.104(2)
2.116(2)
2.302(3)
1.280(4)
1.349(4)
1.385(4)
1.379(4)
1.358(4)
1.314(4)
1.394(4)
1.394(4)
1.405(4)
104.2(7)
−99.0(4)
−7.2(6)
171.2(4)
6.2(1)
2.109(4)
2.101(3)
2.3154(10)
1.261(6)
1.370(6)
1.376(7)
1.371(6)
1.365(6)
1.302(6)
1.360(7)
1.395(7)
1.403(5)
116.4(8)
−105.9(7)
−13.5(8)
167.6(4)
6.6(7)
Pl−Rul−N21
Pl−Rul−N32
90.49(7)
94.04(6)
83.27(9)
133.9(2)
121.2(3)
125.1(3)
126.5(3)
127.1(3)
119.2(3)
106.43(9)
105.7(2)
109.4(3)
127.14(19)
127.40(19)
104.9(2)
87.28(11)
92.64(9)
82.87(14)
131.0(3)
122.1(4)
125.2(5)
125.3(4)
126.9(4)
120.4(4)
107.6(4)
105.4(4)
109.3(4)
126.0(3)
128.6(3)
104.9(4)
N21−Rul−N32
Rul−N21−C22
N21−C22−N23
N21−C22−N21A
C22−N23−C24
N23−C24−N32
N23−C24−N25
C24−N25−C26
N25−C26−C31
C26−C31−N32
C24−N32−Rul
C31−N32−Rul
C24−N32−C31
−1.0(7)
1.2(6)
−10.9(9)
14.9(9)
a
For atom numbering, see Figure 2.
catalysts.30 Thus, the ruthenium salts 1−3+X− were screened for
activity. Indole or 1-methylindole (5a or 5b; 2.0 equiv) and 6
(1.0 equiv) were combined in CD2Cl2 in NMR tubes in the pre-
sence of a salt (0.10 equiv; 10 mol %) and an internal standard.
Reactions of 5a were stopped after 48 h, irrespective of the state
of completion. Results are summarized in Table 5, and selected
rate profiles are given in Figure 4.
With many salts, the 3-substitued indoles 7a,b shown in Table 5
cleanly formed. In all cases, 7b was produced faster, consistent
with an electron-donating effect of the N-methyl group. How-
ever, the slower rate profiles with 7a are illustrated in Figure 4, as
they allow cleaner reactivity comparisons. As shown in entries 1
and 2 of Table 5, no reactions were observed without catalyst or
in the presence of GBI alone. However, GBI is poorly soluble
in CH2Cl2, and a ruthenium-free system more comparable to
1−3+X− is described below.
The rates showed strong dependencies on the counteranions
of the salts. The chloride salts 1−3+Cl− (entries 3, 7, 11) did not
exhibit any significant activity. The indenyl complex 4+Cl− was
not tested due to its reduced stability in CH2Cl2 and the pos-
sibility of new mechanisms involving η3/η1 intermediates. The
tetrafluoroborate salts 1−3+BF4 (entries 4, 8, 12) gave only
−
poor yields of 7a (2−6%). More productive were the hexaflu-
orophosphate salts 1−3+PF6− (entries 5, 9, 13), which afforded
7a in yields up to 27%. The best results were obtained with
1−3+BArf−, which gave yields of 46−97% (entries 6, 10, 14).
E
dx.doi.org/10.1021/om500704u | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 4. Selected F···H, P···N, and F···N Distances [Å] in
Table 5. Friedel−Crafts Alkylations Catalyzed by 1−3+X−
−
1+PF6 ·CH2Cl2
F11···H21A′
F11···H21B′
F12···H23A
F12···H25A
F13···H23A
F13···H25A
F15···H21B
F15···H23A
F15···H21A′
F15···H21B′
F16···H21A′
F16···H21B′
2.769
3.251
2.789
3.801
2.545
2.450
2.195
2.283
2.863
2.795
2.286
3.324
P10···N21A
P10···N21A′
P10···N23
P10···N25
F11···N21A′
F12···N23
F13···N23
F13···N25
F15···N23
4.441
3.802
3.882
4.311
3.262
3.284
3.018
2.939
3.325
3.215
3.158
3.068
7a
7b
a
a
entry
catalyst
time
yield
time
yield
b
b
1
none
GBI
1+Cl−
1+BF4
1+PF6
1+BArf−
2+Cl−
2+BF4
2+PF6
48 h
48 h
48 h
48 h
48 h
48 h
48 h
48 h
48 h
48 h
48 h
48 h
48 h
48 h
−
48 h
48 h
−
b
b
F15···N21A
F15···N21A′
F16···N21A′
2
−
−
b
c
3
−
−
c
−
4
2%
−
−
5
9%
25 h
8 h
30%
53%
4%
6
46%
b
7
−
60 h
60 h
9.5 h
1.0 h
31 h
31 h
7 h
−
8
6%
20%
55%
97%
−
9
27%
94%
10
11
12
13
14
15
2+BArf−
3+Cl−
b
b
−
−
3+BF4
3%
14%
29%
97%
−
3+PF6
22%
84%
−
3+BArf−
1.5 h
3.5 h
c
c
d
[1-methylGBI-H]+BArf−
−
−
40%
a
Yields were determined by 1H NMR versus an internal standard (see
b
c
text). No formation of 7 was observed. This experiment was not
d
conducted. The yield after 1.0 h was 25%.
Figure 3. Thermal ellipsoid diagram (50% probability level) of the
molecular structure of 1+BArf−·CH2Cl2 with the solvate molecule
omitted.
Within each counteranion series, rates increased as the cations
were varied in the order 1+ < 3+ < 2+. Although these data are
further interpreted in the Discussion section, note that the
poorer hydrogen bond accepting anions29 and the less electron
rich cations give faster rates.
In order to help analyze the role of the ruthenium and chelate
ring in promoting catalysis, an organic cationic GBI derivative
was sought. Accordingly, as shown in Scheme 5, GBI was con-
verted to 1-methylGBI by a known deprotonation/methylation
sequence12b and treated with HCl to give the previously reported
hydrochloride salt [1-methylGBI-H]+Cl−.12b Subsequent anion
exchange using Na+BArf− gave the new salt [1-methylGBI-
H]+BArf− as a pale pink solid in 58% yield. The compound was
soluble in CH2Cl2 and other organic solvents of moderate
polarity and exhibited a single NH 1H NMR signal. As indicated
in entry 15 of Table 5, it showed some catalytic activity with the
more reactive indole 5b, but considerably less than the lead BArf−
salts, 2+BArf− (40% yield of 7b after 3.5 h vs 97% after 1.0 h) and
3+BArf− (97% after 1.5 h).
Figure 4. Rate profiles for the condensation of indole (2.0 equiv) and
trans-β-nitrostyrene (1.0 equiv) with different catalysts (10 mol %, rt,
+
−
+
−
◆
■
selected reactions from Table 5): (red ) 2 BArf− , (red ) 2 PF6−,
+
−
+
−
+
+
◆
■
▲
▲
(red ) 2 BF4 , (blue ) 3 BArf , (blue ) 3 PF6 , (blue ) 3 BF4 ,
+
−
◆
(green ) 1 BArf .
the magnitudes of the NH shifts (Figure 5, box), it is proposed
that 6 binds to the cation 2+ predominantly as shown in IIIa.
Downfield shifts of NH signals have also been observed when
carbonyl compound substrates have been added to urea-based
catalysts.31
DISCUSSION
Finally, possible interactions between trans-β-nitrostyrene and
■
1−3+X− (X− = BF4 , PF6 , BArf−) were probed by H NMR.
Spectra of equimolar mixtures of 6 and 1−3+X− were recorded in
CD2Cl2 and compared to those of 1−3+X− under identical
conditions. The four NH/NH2 signals of the GBI ligand can be
assigned (above), and in each case three shifted downfield. With
2+BArf−, which is illustrated in Figure 5, the shifts ranged from
0.09 to 0.02 ppm (top vs bottom spectrum). Similarly, the CH
CHNO2 proton of 6 shifted slightly downfield. On the basis of
1. Mechanism of Catalysis. The data in Table 5 and
Figure 4 validate the underlying hypothesis of this study, namely,
that by chelation-induced preorganization of the conformation-
ally flexible GBI ligand by “spectator” transition metal fragments
an otherwise unreactive species can be rendered an effective
hydrogen bond donor catalyst. The use of a cationic ruthenium
chelate was coincidental, simply reflecting the first adduct
successfully synthesized. The complexes 1−4+X− represent, to
−
−
1
F
dx.doi.org/10.1021/om500704u | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 5. Synthesis of [1-methylGBI-H]+BArf−
trans-β-nitrostyrene preferentially binds to the two synperiplanar
NH units not associated with the NH2 group, as depicted in IIIa.
Note that these two NH groups could adopt any number of
conformations in the free ligand, including one in which they
would be approximately anti.
In any event, preorganization can be an important aspect of
second coordination sphere binding to coordinated ligands.
However, since the ruthenium fragment is cationic, there remains
a question as to the effect of positive charge alone, as this should
also enhance NH acidities and hydrogen bond donor strengths.
Indeed, the ruthenium-free cationic GBI species [1-methylGBI-
H]+BArf− (Scheme 5) proved to be a more effective catalyst than
GBI (Table 5, entry 15 vs 2). Nonetheless, activities remain far
below those of the lead ruthenium catalysts 2+BArf− and 3+BArf−
(entries 10 and 14, Table 5), underscoring the role of conforma-
tional preorganization.
The counteranion also greatly affects the activities of the
ruthenium catalysts 1−3+X−. In each case, the same trend is
observed, Cl− < BF4− < PF6− < BArf− (Table 5). This parallels the
diminishing hydrogen bond accepting properties of the anions.29
In particular, chloride is an excellent hydrogen bond accept-
or,17d,32 and a single such anion effectively “poisons” the catalyst.
Accordingly, we suggest that (1) there is only one productive
substrate binding site that leads to turnover and (2) chloride
preferentially binds to the same two NH groups as the trans-β-
nitrostyrene in IIIa.
The tetrafluoroborate salts remain very poor catalysts, con-
sistent with the still appreciable coordinating17 and hydrogen-
bonding29 ability of this formally tetrahedral anion. In this
context, the IR data in Table 2 establish that the anion is
desymmetrized in the presence of cations 1−3+.17a The
hexafluorophosphate salts are distinctly more reactive, but the
−
crystal structure of 1+PF6 ·CH2Cl2 (Figure 2) demonstrates that
this formally octahedral anion remains a viable hydrogen bond
acceptor. Accordingly, the IR data in Table 2 show that this anion
is also desymmetrized by the cations 1−3+. Similar spectra are
obtained with the sodium salt, presumably due to hydration.33
Finally, there is also a marked dependence of catalyst activities
upon the cation (Table 5 and Figure 4). Since CO ligands are
weaker donors and stronger π-acceptors than PPh3 ligands, the
ruthenium should be less electron rich in 2+X− as compared to
1+X−, as reflected by the downfield shift of the cyclopentadienyl
1H NMR signals noted above (Table 1).18 This increases the
acidities of the NH units and likewise their hydrogen bond donor
strengths, leading to more active catalysts. In the same vein, the
pentamethylcyclopentadienyl ligand is more electron releasing
than the cyclopentadienyl ligand, as reflected by the decreased
νCO values noted above (Table 1). This decreases hydrogen bond
donor strengths, rendering 3+X− slightly less active than 2+X−.
It would be premature to propose a detailed transition state
model for the reactions in Table 5 based upon the present data.
For example, there are two possible orientations about the trans-
β-nitrostyrene CC linkages in IIIa,b (Figure 5), either of
which could react with the nucleophilic indole carbon atom.
Adding to the uncertainty, some computational groups have
advanced models in which the nucleophilic component pre-
ferentially interacts with the hydrogen bond donor.34 Additional
information can often be derived from the absolute config-
urations of products in enantioselective variants, a subject ex-
plored in the following paper.15
Figure 5. H NMR spectra (rt, 300 MHz, CD2Cl2) of 2+BArf− before
1
(above) and after addition of 1 equiv of 6 (below) and some possible
structures of 2+BArf−·6.
our knowledge, the first organometallic adducts of GBI, which
has an extensive inorganic coordination chemistry that has
remained largely unexploited.12
By analogy to ureas and thioureas (Figure 1), substrate acti-
vation would most logically involve two synperiplanar NH units.
As illustrated by the crystal structures in Figures 2 and 3, chela-
tion leads to a triad of three synperiplanar NH units and an
orthogonal dyad of two synperiplanar NH units. However,
there remains a residual conformational degree of freedom
about the NH2 group. The NMR data in Figure 5 suggest that
2. Scope of Catalysis and Future Directions. The new
complexes described herein represent a distinct conceptual
advance with respect to small-molecule hydrogen bond donor
G
dx.doi.org/10.1021/om500704u | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
catalysts, which to date have been dominated by organic
molecules and therefore have not exploited the diverse types of
hydrogen-bonding interactions associated with coordinated
ligands. Together with studies involving the chiral cobalt(III)
trication [Co(en)3]3+ and substituted derivatives,9 it is now
firmly established that both coordinated and noncoordinated
NH units can effectively serve as hydrogen bond donors. In an
example highlighted in the following paper, an octahedral
iridum(III) cation has been used as a template for a ligand
containing a NH(CX)NH moiety and two ligands containing
ArCH2OH moieties.35 This has been shown to be an excellent
catalyst for other types of additions to nitroalkenes, although in
contrast to 1−3+X− the NH linkages were not conformationally
constrained. The hydroxyl groups were proposed to activate the
nucleophile by hydrogen bonding.
These new families of catalysts offer a variety of parameters
that can be systematically varied, such as the metal, the charge,
the counteranion in the case of cations, the non-hydrogen-
bonding ancillary ligands, and substituents on the hydrogen-
bonding ligand. This represents an impressive number of
diversity elements and a high degree of modularity. Although
NH hydrogen bond donors dominate in the preceding examples,
it is likelyas suggested by the iridium(III) systemthat other
types of heteroatom−hydrogen bonds (e.g., PH, OH) can also be
exploited. Also, in the case of GBI, extensions to main group
elements may prove feasible, as chelates are known both with
boron and tin.12c,d The cobalt tris(GBI) complex 83+3BArf−
shown in Figure 6 has proven to be a particularly effective catalyst
cyclopentadienyl ligands remain in development.40 However, the
preparation of enantiopure bifunctional analogues, and their
successful application as highly enantioselective catalysts, is
described in the following paper.15
Since ferrocene-containing systems are often included in re-
views of “organocatalysis”,41 one could consider expanding this
term to accommodate any metal-containing catalyst in which the
metal does not directly participate in the any of the bond-
breaking or bond-forming steps.42 Alternatively, perhaps some-
one more creatively inclined will be able to suggest a catchy
new phrase for such systems. However, in our opinion this work
highlights the artificiality of such classifications, which can do a
disservice by obscuring common underlying mechanistic
principles.
In conclusion, the preceding results and related studies15,36
have established that cationic transition metal chelates of GBI are
effective hydrogen bond donors that can catalyze a variety of
organic transformations. Chelation removes a conformational
degree of freedom, thereby enhancing catalytic activity, and the
introduction of positive charge also has a beneficial effect. Com-
plementary data with other types of ligands point to a heretofore
unappreciated universe of modular metal-containing hydrogen
bond donor catalysts that effect “second coordination sphere
promoted catalysis”. These encompass but are not limited to the
many coordination and organometallic compounds with
coordinated and noncoordinated NH moieties.
EXPERIMENTAL SECTION
■
General Data. All reactions and workups were carried out under
nitrogen unless noted. Standard instrumentation and calibration
procedures were employed, as detailed in the Supporting Information.
Solvents were treated as described in the Supporting Information. The
2-guanidinobenzimidazole (GBI; 95%, Aldrich) and other commercial
−
chemicals were used as received, except for Na+PF6 (98.5%, Acros),
which was freshly washed with CH2Cl2 (5 mL) and dried by oil pump
vacuum.
(η5-C5H5)Ru(PPh3)2(Cl).16,43 A three-necked flask was charged with
PPh3 (14.458 g, 55.182 mmol) and ethanol (100 mL). The mixture was
refluxed with stirring. After 15 min, RuCl3·xH2O (3.581 g, 17.26 mmol
for x = 0; 30−40% Ru) in ethanol (40 mL) and then cyclopentadiene
(18 mL) were added. The brown solution was refluxed for 16 h, cooled
to room temperature, and stored in a freezer. After 24 h, an orange
precipitate was collected by filtration and washed with cold ethanol
(2 × 5 mL), water (2 × 10 mL), cold ethanol (1 × 5 mL), and hexanes
(2 × 15 mL). The residue was dried by oil pump vacuum to give the
product as a bright orange solid (8.105 g, 11.16 mmol, ca. 65%).44 Mp:
131−132 °C (capillary). NMR (δ, CDCl3): 1H (400 MHz) 7.70−6.94
(m, 30H, P(C6H5)3), 4.13 (s, 5H, C5H5); 13C{1H} (100 MHz) 138.5 (t,
Figure 6. Other relevant catalysts.
for lactide polymerizations,36 with activities somewhat higher
than the monocationic ruthenium complex 2+BArf−.
Some second coordination sphere properties of coordinated
NH moieties merit note in passing. First, a variety of ammonia
complexes [LnMNH3]z+ (z = 0, 1) have been found to afford
stable adducts with crown ethers.37,38 Second, MN−H bonds can
undergo cleavage in certain metal-catalyzed reactions. The best
known examples involve the Noyori ruthenium catalysts for
enantioselective hydrogenations of ketones, which proceed via
transition-state assemblies such as IV in Figure 6.39 In second
coordination sphere catalysis involving 1−3+X−, only supra-
molecular interactions are involved, whereas in the tandem first/
second coordination sphere catalysis embodied in IV, covalent
bonds are broken and formed.
2
1JCP = 19.6 Hz, i-C6H5), 133.8 (t, JCP = 5.1 Hz, o-C6H5), 128.7 (s,
p-C6H5), 127.5 (t, 3JCP = 2.7 Hz, m-C6H5); 31P{1H} (161 MHz) 39.3 (s).
IR (cm−1, powder film): 1478 (m), 1432 (s), 1181 (w), 1158 (w), 1085
(s), 995 (m), 829 (w), 806 (m), 745 (m), 690 (s); MS:45 726 (23) [M]+,
691 (71) [M − Cl]+, 464 (33) [M − PPh3]+, 429 (100) [M − Cl −
PPh3]+.
(η5-C5Me5)Ru(PPh3)2(Cl).21 In a new one pot synthesis, a Schlenk
flask was charged with RuCl3·xH2O (0.989 g, 4.82 mmol for x = 0; 30−
40% Ru), pentamethylcyclopentadiene (1.5 mL),46 and ethanol
(35 mL). The mixture was refluxed. After 3 h, PPh3 (3.157 g, 12.05
mmol) was added. After another 16 h, the mixture was cooled to room
temperature and stored in a freezer. After 24 h, the precipitate was
collected by filtration, washed with cold ethanol (15 mL) and hexanes
(3 × 25 mL), and dried by oil pump vacuum to give the product as an
orange powder (3.109 g, 3.905 mmol, ca. 81%).44 Mp: 94−95 °C
(capillary). Anal. Calcd (%) for C46H45ClP2Ru (797.03): C 69.38,
Although the ruthenium catalysts 1−3+X− are “chiral at metal”
species, they are so far available only in racemic form. For this
reason, we have not attempted to expand the scope of the organic
reactions studied at this stage. Indeed, with the stereocenter so
far removed from the substrate binding site, the prospects
for reasonably enantioselective catalysis would seem to be
remote, although some strategies with bulky pentasubstituted
1
H 5.70. Found: C 69.12, H 5.71. NMR (δ, CDCl3): H (400 MHz)
7.69−7.03 (m, 30H, P(C6H5)3), 1.01 (s, 15H, C5(CH3)5); 13C{1H}
H
dx.doi.org/10.1021/om500704u | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(100 MHz) 137.1 (t, JCP = 19.6 Hz, i-C6H5), 133.7 (t, 2JCP = 5.1 Hz,
1
(s, o-C6H5), 132.3 (s, NCCHCHCHCHCN), 130.1 (s, p-C6H5), 128.8 (s,
m-C6H5), 123.3 (s, NCCHCHCHCHCN), 122.7 (s, NCCHCHCHC-
HCN), 118.8 (s, NCCHCHCHCHCN), 111.5 (s, NCCHCHCH-
3
o-C6H5), 128.5 (t, JCP = 2.7 Hz, m-C6H5), 126.7 (s, p-C6H5), 88.9
(s, C5(CH3)5), 9.1 (s, C5(CH3)5); 31P{1H} (161 MHz) 41.5 (s). IR
(cm−1, powder film): 3053 (w), 3022 (w), 2988 (w), 2953 (w), 2903
(w), 1568 (m), 1482 (m), 1432 (s), 1185 (w), 1158 (w), 1089 (m),
1023 (w), 999 (w), 741 (m), 698 (m). MS:45 796 (13) [M]+, 761 (67)
[M − Cl]+, 534 (17) [M − PPh3]+, 499 (100) [M − Cl − PPh3]+.
CHCN), 75.2 (s, C5H5); 31P{1H} (161 MHz) 56.2 (s); 19F{1H}
(282 MHz) −150.3 (s); 11B{1H} (128 MHz) −1.03 (s). IR (cm–1,
powder film): 3381 (m), 3354 (m), 1683 (s), 1637 (w), 1586 (m), 1563
(s), 1494 (w), 1463 (m), 1436 (m), 1409 (m), 1239 (w), 1089 (s), 1078
(s), 1011 (s), 845 (w), 741 (s), 694 (s). MS:45 603 (51) [1]+, 341 (100)
[1 – PPh3]+. UV–visible (nm, 0.0011 M in DMSO (ε, M–1 cm–1)): 289
(4650), 310 (5260), 322 (5610), 342 (3360), 390 (1800).
20
−
[(η5-C5H5)Ru(CO)(NCCH3)2]+PF6
.
A round-bottom flask was
47
−
charged with [(η5-C5H5)Ru(NCCH3)3]+PF6 (0.504 g, 1.16 mmol)
and CH3CN (25 mL). A stream of CO was passed through the brown-
orange solution. After 40 min, the solvent was removed by oil pump
vacuum. The residue was chromatographed on a silica gel column (1 ×
20 cm, 3:1 v/v CH2Cl2/CH3CN). The solvent was removed from the
product-containing fractions to give the product as a golden yellow solid
(0.346 g, 0.823 mmol, 71%). 1H NMR (δ, CD3CN, 400 MHz):20 5.16
(s, 5H, C5H5), 2.36 (s, 6H, CH3CN). IR (cm−1, powder film): 3128 (w),
2949 (w), 2324 (w), 1978 (vs), 1415 (m), 1369 (m), 827 (vs), 556 (vs).
[(η5-C5H5)Ru(PPh3)(GBI)]+Cl− (1+Cl−). A Schlenk flask was charged
with (η5-C5H5)Ru(PPh3)2(Cl) (3.326 g, 4.580 mmol), GBI (0.842 g,
4.80 mmol), and toluene (15 mL) and fitted with a condenser. The
mixture was refluxed with stirring. After 24 h, the mixture was cooled to
room temperature. The solvent was decanted from a precipitate, which
was washed with toluene (4 × 5 mL) and hexanes (2 × 15 mL) and
dried by oil pump vacuum to give 1+Cl− as a yellow powder (2.798 g,
4.378 mmol, 96%). Dec pt: 247 °C (capillary). TGA: onset of the first
mass loss regime, Ti 196.07 °C, Tf 264.37 °C; onset of the second mass
loss regime, Ti 175.83 °C, Tf 413.65 °C. Anal. Calcd (%) for
C31H29ClN5PRu (639.09): C 58.26, H 4.57, N 10.96. Found: C 58.20,
H 4.78, N 11.26. NMR (δ, DMSO-d6): 1H (400 MHz)48 11.83 (br s, 1H,
NH), 10.19 (br s, 1H, NH), 7.32−7.09 (m, 17H, P(C6H5)3 and
−
[(η5-C5H5)Ru(PPh3)(GBI)]+PF6 (1+PF6−). A Schlenk flask was
charged with 1+Cl− (0.224 g, 0.350 mmol), Na+PF6− (0.295 g, 1.76 mmol),
and CH2Cl2 (5 mL). The mixture was stirred for 12 h and filtered
through a plug of Celite (1 × 1 cm), which was rinsed with CH2Cl2 (3 ×
5 mL). The filtrate was concentrated by oil pump vacuum (ca. 5 mL).
Hexanes (25 mL) was added, and the CH2Cl2 was removed by oil pump
vacuum. The solvent was decanted from the precipitate, which was
dissolved in CH2Cl2 (5 mL). The solution was added dropwise to stirred
hexanes (25 mL), and the CH2Cl2 was removed by oil pump vacuum.
The solvent was decanted from the precipitate, which was dried by oil
pump vacuum to give 1+PF6− as a yellow powder (0.218 g, 0.291 mmol,
83%).19 Dec pt: 237 °C (capillary). Anal. Calcd (%) for C31H29F6N5P2Ru
(749.09): C 49.74, H 3.90, N 9.36. Found: C 49.39, H 3.85, N 9.10. NMR
(δ, DMSO-d6): 1H (300 MHz) 12.13 (s, 1H, NH), 10.82 (s, 1H, NH),
7.53−7.20 (m, 19H, P(C6H5)3 and NCCH(CH)2CHCN), 6.63 (s, 2H,
NH2), 6.45 (s, 1H, NH), 4.61 (s, 5H, C5H5); 13C{1H} (75 MHz) 152.7 (s,
NHCNH2), 145.3 (s, NC(NH)2), 143.5 (s, NCCHCHCHCHCN),
1
136.8 (d, JCP = 39.2 Hz, i-C6H5), 133.8 (s, o-C6H5), 132.3 (s,
NCCH(CH)2CHCN), 7.00−6.99 (m, 2H, NCCH(CH)2CHCN), 6.28
(s, 2H, NH2), 6.12 (s, 1H, NH), 4.41 (s, 5H, C5H5); 13C{1H} (100
MHz) 154.1 (s, NHCNH2), 144.7 (s, NC(NH)2), 142.4 (s,
NCCHCHCHCHCN), 130.1 (s, p-C6H5), 128.7 (s, m-C6H5), 123.4
(s, NCCHCHCHCHCN), 122.8 (s, NCCHCHCHCHCN), 118.9
1
NCCHCHCHCHCN), 136.6 (d, JCP = 42.9 Hz, i-C6H5), 132.7 (d,
(s, NCCHCHCHCHCN), 111.4 (s, NCCHCHCHCHCN), 75.2 (s,
1
C5H5); 31P{1H} (161 MHz) 56.3 (s, P(C6H5)3), −142.9 (sep, JPF
=
2JCP = 13.2 Hz, o-C6H5), 131.6 (s, NCCHCHCHCHCN), 129.0 (s,
–
1
703.6 Hz, PF6 ). 19F{1H} (282 MHz) −71.6 (d, JFP = 707.3 Hz). IR
(cm–1, powder film): 3505 (w), 3435 (w), 3412 (w), 3377 (w), 1687 (s),
1637 (w), 1586 (m), 1567 (s), 1478 (w), 1436 (m), 1401 (w), 1254
(m), 1092 (m), 880 (s), 862 (s), 841 (s), 741 (s), 698 (s). MS:45 603
(85) [1]+, 341 (100) [1 – PPh3]+. UV–visible (nm, 0.0010 M in DMSO
(ε, M–1 cm–1)): 296 (5200), 302 (4830), 314 (5340), 328 (5440), 338
(3980), 349 (4020), 400 (1760).
3
p-C6H5), 127.8 (d, JCP = 9.9 Hz, m-C6H5), 121.6 (s, NCCHCH-
CHCHCN), 121.2 (s, NCCHCHCHCHCN), 117.2 (s, NCCHC-
HCHCHCN), 110.5 (s, NCCHCHCHCHCN), 74.1 (s, C5H5);
31P{1H} (161 MHz) 55.9 (s). IR (cm–1, powder film): 3347 (m),
3254 (m), 3200 (w), 3103 (w), 3080 (w), 2798 (m), 2764 (m), 2729
(m), 1679 (s), 1640 (w), 1617 (m), 1590 (m), 1559 (s), 1463 (m), 1436
(m), 1417 (m), 1274 (w), 1251 (m), 1096 (m), 833 (m), 791 (m), 749
(s), 695 (s). MS:45 603 (39) [1]+, 585 (12) [1 – NH2]+, 341 (100) [1 –
PPh3]+. UV–visible (nm, 0.0010 M in DMSO (ε, M–1 cm–1)): 309
(2760), 324 (2560), 332 (2550), 345 (2380), 384 (1660).
[(η5-C5H5)Ru(PPh3)(GBI)]+BArf− (1+BArf−).8 A Schlenk flask was
charged with 1+Cl− (0.273 g, 0.427 mmol), Na+BArf− (0.415 g, 0.469
mmol),24 and CH2Cl2 (5 mL). The mixture was stirred for 12 h and
filtered through a plug of Celite (1 × 2.5 cm), which was rinsed with
CH2Cl2 (15 mL). The filtrate was concentrated by oil pump vacuum
(ca. 5 mL). Hexanes (25 mL) was added, and the solvent was decanted
from the precipitate, which was dissolved in CH2Cl2 (5 mL). The solu-
tion was added dropwise to stirred hexanes (25 mL), and the CH2Cl2
was removed by oil pump vacuum. The solvent was decanted from the
precipitate, which was dried by oil pump vacuum to give 1+BArf−·
(H2O)2 as a yellow powder (0.545 g, 0.363 mmol, 85%).19 Dec pt:
196 °C (capillary). Anal. Calcd (%) for C63H41BF24N5PRu·(H2O)2
(1502.88): C 50.96, H 2.92, N 4.72. Found: C 50.65, H 2.61, N 4.64.
NMR (δ, DMSO-d6): 1H (300 MHz) 11.75 (s, 1H, NH), 9.68 (s, 1H,
NH), 8.31−8.03 (m, 31H, B(C6H3(CF3)2)4, P(C6H5)3, and
−
[(η5-C5H5)Ru(PPh3)(GBI)]+BF4 (1+BF4−). A Schlenk flask was
−
charged with 1+Cl− (0.273 g, 0.427 mmol), Na+BF4 (0.051 g,
0.47 mmol), and CH2Cl2/water (7.5 mL, 2:1 v/v). The mixture was
stirred for 12 h. The organic phase was separated, and the aqueous phase
was extracted with CH2Cl2 (2 × 5 mL). The combined organic phases
were filtered through a plug of Na2SO4 (1 × 1 cm), which was rinsed with
CH2Cl2 (3 × 5 mL). The filtrate was concentrated by oil pump vacuum
(ca. 5 mL). Hexanes (25 mL) was added, and the CH2Cl2 was removed by
oil pump vacuum. The solvent was decanted from the precipitate, which
was dissolved in CH2Cl2 (5 mL). The solution was added dropwise to
stirred hexanes (25 mL), and the CH2Cl2 was removed by oil pump
vacuum. The solvent was decanted from the precipitate, which was dried
by oil pump vacuum to give 1+BF4−·(H2O)0.5 as a yellow powder (0.230 g,
0.329 mmol, 77%).19 Dec pt: 242 °C (capillary). Anal. Calcd (%) for
C31H29BF4N5PRu·(H2O)0.5 (700.13): C 53.23, H 4.32, N 10.01. Found: C
53.59, H 4.15, N 9.83. NMR (δ, DMSO-d6): 1H (400 MHz) 11.73 (s, 1H,
NH), 9.71 (s, 1H, NH), 7.36−7.10 (m, 17H, P(C6H5)3 and
NCCH(CH)2CHCN), 6.12 (s, 1H, NH), 6.03 (s, 2H, NH2), 5.02 (s,
5H, C5H5), 3.33 (s, H2O); 13C{1H} (75 MHz) 163.1 (q, 1JCB = 49.6 Hz,
i-C6H3(CF3)2), 152.6 (s, NHCNH2), 145.3 (s, NC(NH)2), 143.9
(s, NCCHCHCHCHCN), 137.1 (d, 1JCP = 27.9 Hz, i-C6H5), 135.2 (s,
o-C6H3(CF3)2), 134.9 (s, o-C6H5), 132.3 (s, NCCHCHCHCHCN),
2
130.1 (s, p-C6H5), 128.8 (s, m-C6H5), 129.5 (q, JCF = 31.2 Hz,
NCCH(CH)2CHCN), 7.04−7.01 (m, 2H, NCCH(CH)2CHCN), 6.10
(s, 1H, NH), 6.02 (s, 2H, NH2), 4.43 (s, 5H, C5H5), 3.32 (s, H2O);
13C{1H} (75 MHz) 152.6 (s, NHCNH2), 145.3 (s, NC(NH)2),
1
m-C6H3(CF3)2), 126.7 (q, JCF = 270.7 Hz, C6H3(CF3)2), 123.3
(s, NCCHCHCHCHCN), 122.7 (s, NCCHCHCHCHCN), 118.8
(s, NCCHCHCHCHCN), 117.9 (s, p-C6H3(CF3)2), 111.5
143.9 (s, NCCHCHCHCHCN), 137.1 (d, 1JCP = 27.9 Hz, i-C6H5), 134.9
I
dx.doi.org/10.1021/om500704u | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(br s, 2H, NH2), 6.33 (s, 1H, NH), 5.19 (s, 5H, C5H5), 3.32 (s, H2O);
13C{1H} (CD2Cl2, 75 MHz) 204.0 (s, CO), 153.3 (s, NHCNH2),
(s, NCCHCHCHCHCN), 74.7 (s, C5H5); 31P{1H} (161 MHz) 56.4
(s); 19F{1H} (282 MHz) −63.7 (s); 11B{1H} (128 MHz) −6.63 (s). IR
(cm–1, powder film): 3443 (w), 3405 (w), 1679 (m), 1586 (m), 1563
(m), 1459 (m), 1355 (s), 1274 (s), 1170 (s), 1119 (s), 1011 (w), 887
(m), 837 (m), 810 (m), 737 (m), 714 (m), 683 (m); MS:45 603 (49)
[1]+, 341 (100) [1 – PPh3]+; UV–visible (nm, 0.0010 M in DMSO
(ε, M–1 cm–1)): 295 (4170), 306 (5940), 320 (3160), 391 (1160).
[(η5-C5H5)Ru(CO)(GBI)]+Cl− (2+Cl−).49 A Schlenk flask was charged
with 1+Cl− (0.314 g, 0.491 mmol) and CHCl3 (25 mL). The solution
was aspirated with CO and monitored by 31P{1H} NMR. After 24 h, the
solution was concentrated by rotary evaporation (5 mL) and filtered
through a plug of Celite (5 × 1 cm), which was rinsed with CHCl3
(30 mL).50 The filtrate was concentrated by rotary evaporation (ca.
25 mL) and added dropwise to stirred n-pentane (150 mL). The solvent
was decanted from the precipitate, which was dissolved in CHCl3
(25 mL).51 The solution was added dropwise to stirred hexanes
(100 mL), and the solvent was decanted from the precipitate. This
sequence was repeated twice. The residue was triturated with benzene
and dried by oil pump vacuum to give 2+Cl− as an off-white powder
(0.181 g, 0.447 mmol, 91%). Dec pt: 252 °C (capillary). Anal. Calcd (%)
for C14H14ClN5ORu (404.99): C 41.54, H 3.49, N 17.30. Found: C
41.08, H 3.68, N 15.80.52 NMR (δ): 1H (CDCl3/MeOH-d4, 400
145.1 (s, NC(NH)2), 142.3 (s, NCCHCHCHCHCN), 131.3 (s,
NCCHCHCHCHCN), 122.9 (s, NCCHCHCHCHCN), 122.4 (s,
NCCHCHCHCHCN), 116.9 (s, NCCHCHCHCHCN), 111.1 (s,
NCCHCHCHCHCN), 81.6 (s, C5H5); 19F{1H} (CD2Cl2, 282 MHz)
−148.5 (s); 11B{1H} (DMSO-d6, 128 MHz) −1.13 (s). IR (cm–1, powder
film): 3331 (w), 3304 (w), 3242 (w), 3212 (w), 3188 (w), 3138 (w),
3100 (w), 3080 (w), 3011 (w), 2922 (w), 2154 (m), 1938 (s, νCO),
1679 (s), 1637 (m), 1586 (m), 1563 (s), 1463 (m), 1413 (w), 1254 (w),
1220 (w), 1089 (m), 1078 (w), 1011 (m), 829 (w), 802 (m), 741 (s),
690 (s). MS:45 371 (67) [2]+, 341 (100) [2–CO]+. UV–visible
(nm, 0.0010 M in DMSO (ε, M–1 cm–1)): 287 (2650), 300 (2870), 305
(3610), 430 (173).
−
[(η5-C5H5)Ru(CO)(GBI)]+PF6 (2+PF6−). Route A. A Schlenk flask
was charged with 1+PF6− (0.172 g, 0.229 mmol) and CH2Cl2 (5 mL).
The sample was saturated with CO, fitted with a balloon filled with CO,
and stirred. After 12 h, the mixture was filtered through a plug of Celite
(1 × 1 cm), which was rinsed with CH2Cl2 (3 × 5 mL).50 The filtrate was
concentrated by rotary evaporation (ca. 5 mL). Hexanes (25 mL) was
added, and the CH2Cl2 was removed by rotary evaporation. The solvent
was decanted from the precipitate, which was dissolved in CH2Cl2
(5 mL). The solution was added dropwise to stirred hexanes (25 mL),
and the CH2Cl2 was removed by rotary evaporation. The solvent was
decanted from the precipitate, which was dried by oil pump vacuum to
give 2+PF6− as a yellow powder (0.105 g, 0.204 mmol, 89%).19 Route B.
MHz)53 7.21−7.18 (m, 1H, NCCH(CH)2CHCN), 7.09−7.06 (m, 2H,
NCCH(CH)2CHCN), 6.99−6.96 (m, 1H, NCCH(CH)2CHCN), 4.87
(s, 5H, C5H5); H (DMSO-d6, 400 MHz) 11.42 (br s, 2H, NH),54
48
1
7.40−7.38 (m, 1H, NCCH(CH)2CHCN), 7.20−7.13 (m, 3H, NCCH-
−
A Schlenk flask was charged with 2+Cl− (0.218 g, 0.538 mmol), Na+PF6
(CH)2CHCN), 6.72 (br s, 2H, NH2), 6.39 (s, 1H, NH), 5.19 (s, 5H,
C5H5); 13C{1H} (CDCl3, 100 MHz)48 204.1 (s, CO), 153.6 (s, NH
(0.452 g, 2.69 mmol), and CH2Cl2 (5 mL). The mixture was stirred for
12 h and filtered through a plug of Celite (1 × 1 cm), which was rinsed
with CH2Cl2 (3 × 5 mL).50 The filtrate was concentrated by rotary
evaporation (ca. 5 mL). Hexanes (25 mL) was added, and the CH2Cl2
was removed by rotary evaporation. The solvent was decanted from the
precipitate, which was dissolved in CH2Cl2 (5 mL). The solution was
added dropwise to stirred hexanes (25 mL), and the CH2Cl2 was
removed by rotary evaporation. The solvent was decanted from the
CNH2), 145.4 (s, NC(NH)2), 142.5 (s, NCCHCHCHCHCN), 131.6
(s, NCCHCHCHCHCN), 123.0 (s, NCCHCHCHCHCN), 122.5 (s,
NCCHCHCHCHCN), 116.9 (s, NCCHCHCHCHCN), 111.5
(s, NCCHCHCHCHCN), 81.7 (s, C5H5). IR (cm–1, powder film):
3331 (w), 3266 (w), 3208 (m), 3138 (m), 3111 (w), 1938 (s, νCO), 1683
(s), 1652 (w), 1567 (s), 1494 (w), 1463 (m), 1420 (w), 1262 (m), 1220
(w), 1092 (w), 1015 (m), 972 (w), 934 (w), 806 (m), 741 (m), 694 (s),
667 (m). MS:45 371 (52) [2]+, 341 (100) [2 – CO]+, 325 (32) [2 – CO –
NH2]+, 299 (26) [2 – CO – NHCNH2]+; UV–visible (nm, 0.0010 M in
DMSO (ε, M–1 cm–1)): 289 (4180), 294 (3860), 304 (3030), 412 (200).
−
precipitate, which was dried by oil pump vacuum to give 2+PF6 as a
yellow powder (0.241 g, 0.468 mmol, 87%). Dec pt: 221 °C (capillary).
Anal. Calcd (%) for C14H14F6N5OPRu (514.99): C 32.69, H 2.74, N
13.62. Found: C 32.70, H 3.10, N 11.97.52 NMR (δ, DMSO-d6): H
1
(300 MHz) 12.48 (s, 1H, NH), 10.43 (s, 1H, NH), 7.43−7.39 (m, 1H,
−
[(η5-C5H5)Ru(CO)(GBI)]+BF4 (2+BF4−). Route A. A Schlenk flask
−
was charged with 1+BF4 ·(H2O)0.5 (0.143 g, 0.204 mmol) and CH2Cl2
NCCH(CH)2CHCN), 7.24−7.16 (m, 3H, NCCH(CH)2CHCN), 6.46
(s, 2H, NH2), 6.34 (s, 1H, NH), 5.20 (s, 5H, C5H5); 13C{1H} (75 MHz)
203.9 (s, CO), 152.9 (s, NHCNH2), 144.7 (s, NC(NH)2), 142.7
(5 mL). The sample was saturated with CO, fitted with a balloon filled
with CO, and stirred. After 24 h, the mixture was filtered through a
plug of Celite (1 × 1 cm), which was rinsed with CH2Cl2 (3 × 5 mL).50
The filtrate was concentrated by rotary evaporation (ca. 5 mL).
Hexanes (25 mL) was added, and the CH2Cl2 was removed by rotary
evaporation. The solvent was decanted from the precipitate, which was
dissolved in CH2Cl2 (5 mL). The solution was added dropwise to stirred
hexanes (25 mL), and the CH2Cl2 was removed by rotary evaporation.
The solvent was decanted from the precipitate, which was dried by oil
pump vacuum to give 2+BF4− as a yellow powder (0.081 g, 0.18 mmol,
87%).19 Route B. A Schlenk flask was charged with 2+Cl− (0.161 g,
(s, NCCHCHCHCHCN), 131.2 (s, NCCHCHCHCHCN), 124.3
(s, NCCHCHCHCHCN), 123.8 (s, NCCHCHCHCHCN), 117.9
(s, NCCHCHCHCHCN), 111.6 (s, NCCHCHCHCHCN), 82.0 (s,
C5H5); 31P{1H} (DMSO-d6, 121 MHz) −142.7 (sep, 1JPF = 710.3 Hz);
19F{1H} (282 MHz) −69.8 (d, 1JFP = 712.3 Hz). IR (cm–1, powder film):
2347 (m), 1942 (s, νCO), 1683 (m), 1652 (w), 1590 (m), 1567 (m),
1521 (w), 1494 (w), 1463 (m), 1243 (m), 1104 (m), 1061 (w), 1015
(w), 837 (s), 741 (m), 660 (w). MS:45 371 (52) [2]+, 341 (100) [2 –
CO]+. UV–visible (nm, 0.0010 M in DMSO (ε, M–1 cm–1)): 291 (2280),
295 (2400), 298 (2650), 305 (4120), 311 (2550), 419 (206).
−
0.402 mmol), Na+BF4 (0.218 g, 1.99 mmol), and CH2Cl2/water
(10 mL, 1:1 v/v) with stirring. After 12 h, the organic phase was
separated and dried (Na2SO4).50 The mixture was filtered through a
plug of Celite (1 × 1 cm), which was rinsed with CH2Cl2 (3 × 10 mL).
The filtrate was concentrated by rotary evaporation (ca. 5 mL). Hexanes
(25 mL) was added, and the solvent was decanted from the precipitate,
which was dissolved in CH2Cl2 (5 mL). The solution was added
dropwise to stirred hexanes (25 mL). The solvent was decanted from the
Route C. A round-bottom flask was charged with [(η5-C5H5)-
−
Ru(CO)(NCCH3)2]+PF6 (0.040 g, 0.095 mmol; see above), GBI
(0.016 g, 0.095 mmol), CH2Cl2 (2 mL), and MeOH (1 mL) with
stirring. After 2 d at room temperature, the solvent was removed by oil
pump vacuum and the residue was chromatographed on a silica gel
column (0.5 × 15 cm, 3:1 v/v CH2Cl2/CH3CN). The solvent was
removed from the product-containing fractions to give a sticky yellow
solid. This was dissolved in CH2Cl2 (5 mL), and pentane was added
until a precipitate formed. The solvent was removed by oil pump
vacuum. More pentane (5 mL) was added and removed by oil
−
precipitate, which was dried by oil pump vacuum to give 2+BF4 ·(H2O)2
as a yellow powder (0.151 g, 0.306 mmol, 77%). Dec pt: 241 °C
(capillary). Anal. Calcd (%) for C14H14BF4N5ORu·(H2O)2 (493.05): C
34.16, H 3.69, N 14.23. Found: C 33.85, H 3.14, N 12.32.52 NMR (δ):
1H (DMSO-d6, 300 MHz) 11.57 (s, 2H, NH), 7.41−7.38 (m, 1H,
−
pump vacuum (2 × ) to give 2+PF6 as a yellow powder (0.039 g,
NCCH(CH)2CHCN), 7.22−7.14 (m, 3H, NCCH(CH)2CHCN), 6.54
0.076 mmol, 81%).
J
dx.doi.org/10.1021/om500704u | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
[(η5-C5H5)Ru(CO)(GBI)]+BArf− (2+BArf−). Route A. A Schlenk flask
was charged with 1+BArf−·(H2O)2 (0.257 g, 0.171 mmol) and CH2Cl2
(5 mL). The sample was saturated with CO, fitted with a balloon filled
with CO, and stirred. After 24 h, the mixture was filtered through a plug
of Celite (1 × 2.5 cm), which was rinsed with CH2Cl2
(2 × 10 mL).50 The filtrate was concentrated by rotary evaporation
(ca. 5 mL). Hexanes (25 mL) was added, and the CH2Cl2 was removed
by rotary evaporation. The solvent was decanted from the precipitate,
which was dissolved in CH2Cl2 (5 mL). The solution was added drop-
wise to stirred hexanes (25 mL), and the CH2Cl2 was removed by rotary
evaporation. The solvent was decanted from the precipitate, which was
dried by oil pump vacuum to give 2+BArf− as a yellow powder (0.194 g,
0.157 mmol, 92%).19 Route B. A Schlenk flask was charged with 2+Cl−
(0.154 g, 0.381 mmol), Na+BArf− (0.354 g, 0.401 mmol),24 and CH2Cl2
(5 mL). The mixture was stirred for 12 h and filtered through a plug of
Celite (1 × 2.5 cm), which was rinsed with CH2Cl2 (2 × 25 mL).50 The
filtrate was concentrated by rotary evaporation (ca. 5 mL). Hexanes
(25 mL) was added, and the CH2Cl2 was removed by rotary
evaporation. The solvent was decanted from the precipitate, which
was dissolved in CH2Cl2 (5 mL). The solution was added dropwise to
stirred hexanes (25 mL), and the CH2Cl2 was removed by rotary
evaporation. The solvent was decanted from the precipitate, which was
dried by oil pump vacuum to give 2+BArf−·(H2O)1.5 as a yellow powder
(0.420 g, 0.333 mmol, 88%). Dec pt: 187 °C (capillary). Anal. Calcd (%)
for C46H26BF24N5ORu·(H2O)1.5 (1260.11): C 43.86 H 2.32, N 5.56.
Found: C 44.06, H 2.77, N 4.94.52 NMR (δ): 1H (DMSO-d6, 400 MHz)
12.02 (br s, 2H, NH), 7.78 (s 8H, o-B(C6H3(CF3)2)4), 7.71 (s, 4H,
(DMSO-d6, 75 MHz) 205.4 (s, CO), 153.6 (s, NHCNH2), 145.7 (s,
NC(NH)2), 142.5 (s, NCCHCHCHCHCN), P(O2C6Cl4)3 at 141.2
(d, 2JCP = 6.5 Hz), 122.1 (s), 113.2 (d, 3JCP = 19.6 Hz); 131.8 (s, NCCH-
CHCHCHCN), 122.7 (s, NCCHCHCHCHCN), 122.4 (s, NCCH-
CHCHCHCN), 117.0 (s, NCCHCHCHCHCN), 111.2 (s, NCCHC-
HCHCHCN), 82.5 (s, C5H5); 31P{1H} (DMSO-d6, 161 MHz) −79.7
(s). IR (cm–1, powder film): 3381 (w), 3354 (m), 3308 (m), 3204 (w), 3184
(w), 3161 (w), 3119 (w), 3038 (w), 2930 (w), 1945 (s, νCO), 1683 (s), 1637
(w), 1586 (m), 1563 (s), 1494 (s), 1463 (m), 1436 (m), 1409 (w), 1239
(m), 992 (s), 741 (m), 694 (s). MS:45 371 (59) [2]+, 341 (100) [2 – CO]+.
[(η5-C5Me5)Ru(CO)(GBI)]+Cl− (3+Cl−). A Schlenk flask was charged
with (η5-C5Me5)Ru(PPh3)2(Cl) (see above; 2.303 g, 2.893 mmol), GBI
(0.760 g, 4.34 mmol), and benzene (15 mL) and fitted with a condenser.
The mixture was refluxed with stirring. After 16 h, the yellow-brown pre-
cipitate was collected by filtration and washed with toluene (4 × 5 mL)
and hexanes (2 × 15 mL). A Schlenk flask was charged with the yellow-
brown powder and CHCl3 (25 mL). The suspension was saturated with
CO, fitted with a balloon filled with CO, and stirred. After 72 h, the
mixture was concentrated by rotary evaporation (ca. 5 mL) and filtered
through a plug of Celite (1 × 1 cm), which was rinsed with CHCl3
(40 mL). The filtrate was concentrated by rotary evaporation (ca.
10 mL) and added dropwise to stirred n-pentane (35 mL). The solvent
was decanted from the precipitate, which was dissolved in CHCl3
(15 mL).51 The solution was added dropwise to stirred hexanes
(100 mL). The solvent was decanted from the precipitate, which was
dried by oil pump vacuum to give 3+Cl−·(CHCl3)0.17 as an off-white
powder (1.128 g, 2.281 mmol, 77%). Dec pt: 221 °C (capillary). Anal.
Calcd (%) for C19H24ClN5ORu·(CHCl3)0.17 (494.72): C 46.23, H 4.89,
N 14.04. Found: C 46.48, H 4.89, N 14.41. NMR48 (δ): 1H (DMSO-d6,
400 MHz) 11.53 (br s, 2H, NH), 8.32 (s, trace CHCl3), 7.42−7.40 (m,
p-B(C6H3(CF3)2)4), 7.52−7.49 (m, 1H, NCCH(CH)2CHCN), 7.32−
7.27 (m, 3H, NCCH(CH)2CHCN), 6.63 (s, 2H, NH2), 6.45 (s, 1H,
NH), 5.30 (s, 5H, C5H5), 3.31 (s, H2O); 13C{1H} (CD2Cl2, 75 MHz)
203.3 (s, CO), 163.1 (q, 1JCB = 49.6 Hz, i-C6H3(CF3)2), 152.4 (s, NH
CNH2), 144.1 (s, NC(NH)2), 142.6 (s, NCCHCHCHCHCN),
135.2 (s, o-C6H3(CF3)2), 130.8 (s, NCCHCHCHCHCN), 129.5
1H, NCCH(CH)2CHCN), 7.27−7.17 (m, 2H, NCCH(CH)2CHCN),
7.12−7.10 (m, 1H, NCCH(CH)2CHCN), 6.83 (s, 2H, NH2), 5.89 (s,
1H, NH), 1.58 (s, 15H, C5(CH3)5); 13C{1H} (DMSO-d6/MeOH-d4,
100 MHz)48 208.5 (s, CO), 154.6 (s, NHCNH2), 146.1 (s, N
2
1
(q, JCF = 31.2 Hz, m-C6H3(CF3)2), 126.7 (q, JCF = 270.7 Hz,
C6H3(CF3)2), 124.9 (s, NCCHCHCHCHCN), 124.5 (s, NCCHCHC-
HCHCN), 118.4 (s, NCCHCHCHCHCN), 117.9 (s, p-C6H3(CF3)2),
C(NH)2), 141.5 (s, NCCHCHCHCHCN), 133.1 (s, NCCHCHC-
HCHCN), 123.3 (s, NCCHCHCHCHCN), 122.8 (s, NCCHCHC-
HCHCN), 117.5 (s, NCCHCHCHCHCN), 111.9 (s, NCCHCHCH-
111.4 (s, NCCHCHCHCHCN), 81.9 (s, C5H5); 19F{1H} (CD2Cl2, 282
MHz) −63.2 (s); 11B{1H} (DMSO-d6, 128 MHz) −6.63 (s). IR (cm–1,
powder film): 3713 (w), 3652 (w), 2362 (w), 2343 (w), 1961 (s, νCO),
1718 (m), 1687 (m), 1629 (w), 1575 (m), 1355 (s), 1278 (s), 1116 (s),
1061 (m), 934 (w), 887 (w), 837 (w), 745 (m), 710 (m), 671 (m).
MS:45 603 (51) [2]+, 341 (100) [2 – PPh3]+. UV–visible (nm, 0.0010 M
in DMSO (ε, M–1 cm–1)): 289 (3810), 293 (3130), 305 (2770),
416 (210).
CHCN), 93.4 (s, C5(CH3)5), 79.4 (s, trace CHCl3), 9.7 (s, C5(CH3)5).
IR (cm–1, powder film): 3374 (w), 3289 (m), 3227 (w), 3184 (w), 3146
(w), 3100 (w), 3030 (w), 2972 (w), 2918 (w), 1915 (s, νCO), 1683 (m),
1637 (w), 1586 (m), 1556 (s), 1490 (w), 1463 (s), 1382 (m), 1251 (m),
810 (m), 741 (s), 690 (m). MS:45 440 (49) [3]+, 412 (100) [3 – CO]+.
UV–visible (nm, 0.0010 M in DMSO (ε, M–1 cm–1)): 295 (3020), 300
(4850), 302 (4980), 311 (3520), 315 (3140), 405 (253).
[(η5-C5H5)Ru(CO)(GBI)]+(Δ)-TRISPHAT− (2+(Δ)-TRISPHAT−). A
Schlenk flask was charged with 2+Cl− (0.273 g, 0.427 mmol),
(n-Bu)3NH+(Δ)-TRISPHAT− (0.445 g, 0.469 mmol),26 and CH2Cl2
(5 mL) with stirring. After 12 h, the mixture was filtered through a plug
of Celite (1 × 2.5 cm), which was rinsed with CH2Cl2 (25 mL).50 The
filtrate was concentrated by rotary evaporation (ca. 5 mL). Then
CH2Cl2/water (15 mL, 1:2 v/v) was added, and the mixture shaken for
5 min. The organic phase was separated, washed with water (5 × 10 mL),
dried (Na2SO4), and filtered (sintered glass). Hexanes (25 mL) was
added, and the CH2Cl2 was removed by rotary evaporation. The solvent
was decantedfromthe precipitate, which was dissolved inCH2Cl2 (5mL).
The solution was added dropwise to stirred hexanes (25 mL), and the
CH2Cl2 was removed by rotary evaporation. The solvent was decanted
from the precipitate, which was dried by oil pump vacuum to give 2+(Δ)-
TRISPHAT− as a white powder (0.545 g, 0.371 mmol, 81%) of ca. 95%
purity by 1H NMR. Dec pt: 187 °C (capillary). NMR (δ): 1H (DMSO-d6,
−
[(η5-C5Me5)Ru(CO)(GBI)]+BF4 (3+BF4−). A Schlenk flask was
−
charged with 3+ Cl−·(CHCl3)0.17 (0.106 g, 0.214 mmol), Na+BF4
(0.117 g, 1.06 mmol), and CH2Cl2/water (10 mL, 1:1 v/v) with stirring.
After 16 h, the organic phase was separated, and the aqueous phase was
extracted with CH2Cl2 (2 × 5 mL).50 The combined organic phases
were filtered through a plug of Na2SO4 (1 × 1 cm), which was rinsed
with CH2Cl2 (3 × 10 mL). The filtrate was concentrated by rotary
evaporation (ca. 5 mL). Hexanes (25 mL) was added, and the CH2Cl2
was removed by rotary evaporation. The solvent was decanted from the
precipitate, which was dissolved in CH2Cl2 (5 mL). The solution was
added dropwise to stirred hexanes (25 mL), and the CH2Cl2 was
removed by rotary evaporation. The solvent was decanted from the
−
precipitate, which was dried by oil pump vacuum to give 3+BF4 ·
(H2O)1.5 as a yellow powder (0.072 g, 0.130 mmol, 61%). Dec pt: 219
°C (capillary). Anal. Calcd (%) for C19H24BF4N5ORu·(H2O)1.5
(554.13): C 41.17, H 5.09, N 12.63. Found: C 41.46, H 4.89, N
12.21. NMR (δ): 1H (DMSO-d6, 300 MHz) 11.58 (s, 2H, NH), 7.42−
300 MHz) 11.55 (br s, 2H, NH), 7.39−7.37 (m, 1H, NCCH(CH)2-
CHCN), 7.18−7.14 (m, 3H, NCCH(CH)2CHCN), 6.44 (s, 1H, NH),
6.30 (s, 2H, NH2), 5.18 (s, 5H, C5H5); 1H (C6D6, 300 MHz) 10.98 (br s,
7.40 (m, 1H, NCCH(CH)2CHCN), 7.26−7.19 (m, 2H, NCC-
2H, NH), 7.31−7.12 (m, 4H, NCCH(CH)2CHCN), 6.11 (br s, 1H,
NH), 5.98 (br s, 2H, NH2), 4.61 and 4.60 (2 s, 5H, C5H5); 13C{1H}
H(CH)2CHCN), 7.12−7.10 (m, 1H, NCCH(CH)2CHCN), 6.64 (s,
2H, NH2), 5.83 (s, 1H, NH), 3.40 (s, H2O), 1.58 (s, 15H, C5(CH3)5);
K
dx.doi.org/10.1021/om500704u | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
13C{1H} (CDCl3, 75 MHz) 206.5 (s, CO), 153.1 (s, NHCNH2),
144.9 (s, NC(NH)2), 140.7 (s, NCCHCHCHCHCN), 131.5 (s,
NCCHCHCHCHCN), 123.5 (s, NCCHCHCHCHCN), 122.9 (s,
NCCHCHCHCHCN), 117.5 (s, NCCHCHCHCHCN), 111.3 (s,
(s, NCCHCHCHCHCN), 129.5 (q, 2JCF = 31.2 Hz, m-C6H3(CF3)2),
1
126.7 (q, JCF = 270.7 Hz, C6H3(CF3)2), 125.0 (s, NCCHCH-
CHCHCN), 124.5 (s, NCCHCHCHCHCN), 118.7 (s, NCCHCH-
CHCHCN), 113.1 (s, NCCHCHCHCHCN), 111.5 (s,
p-C6H3(CF3)2), 93.1 (s, C5(CH3)5), 9.9 (s, C5(CH3)5); 19F{1H}
(CD2Cl2, 282 MHz) −63.2 (s); 11B{1H} (DMSO-d6, 128 MHz) −6.64
(s). IR (cm–1, powder film): 3692 (w), 3460 (w), 3391 (w), 3290 (w),
3213 (w), 3097 (w), 2966 (w), 2927 (w), 1931 (s, νCO), 1676 (m), 1607
(m), 1560 (m), 1460 (m), 1352 (m), 1274 (s), 1120 (s), 888 (m), 834
(m), 741 (m), 672 (m). MS:45 441 (51) [3]+, 411 (100) [3 – CO]+.
UV–visible (nm, 0.0011 M in DMSO (ε, M–1 cm–1)): 290 (3510), 302
(4980), 309 (4000), 317 (2140), 401 (264).
NCCHCHCHCHCN), 92.6 (s, C5(CH3)5), 9.7 (s, C5(CH3)5);
19F{1H} (CDCl3, 282 MHz) −147.9 (s); 11B{1H} (DMSO-d6, 128
MHz) −1.03 (s). IR (cm–1, powder film): 3375 (w), 3290 (m), 3190
(w), 3144 (w), 3090 (w), 3035 (w), 2966 (w), 2920 (w), 2819 (m),
1915 (s, νCO), 1684 (m), 1637 (w), 1552 (m), 1468 (m), 1383 (w), 1328
(w), 1259 (m), 1081 (s), 1020 (s), 803 (s), 741 (m), 687 (m). MS:45
440 (33) [3]+, 411 (100) [3 – CO]+. UV–visible (nm, 0.0010 M in
DMSO (ε, M–1 cm–1)): 289 (2250), 295 (2140), 308 (3110),
402 (253).
[(η5-C9H7)Ru(PPh3)(GBI)]+Cl− (4+Cl−). A Schlenk flask was charged
with (η5-C9H7)Ru(PPh3)2(Cl) (0.417 g, 0.537 mmol),23 GBI (0.117 g,
0.671 mmol), and toluene (15 mL). The mixture was refluxed with
stirring. After 24 h, the precipitate was collected by filtration, washed
with toluene (3 × 15 mL) and hexanes (2 × 15 mL), and dried by oil
pump vacuum at 120 °C to give 4+Cl−·(H2O)0.5 as an orange powder
(0.263 g, 0.377 mmol, 72%). Dec pt: 222 °C. Anal. Calcd (%) for
C35H31ClN5PRu·(H2O)0.5 (699.11): C 60.21, H 4.62, N 10.03. Found:
C 59.92, H 5.04, N 10.20. NMR (δ, DMSO-d6): 1H (400 MHz)48 11.97
−
[(η5-C5Me5)Ru(CO)(GBI)]+PF6 (3+PF6−). A Schlenk flask was
−
charged with 3+Cl−·(CHCl3)0.17 (0.182 g, 0.368 mmol), Na+PF6
(0.309 g, 1.84 mmol), and CH2Cl2 (5 mL). The mixture was stirred
for 16 h and filtered through a plug of Celite (1 × 1 cm), which was
rinsed with CH2Cl2 (3 × 5 mL).50 The filtrate was concentrated by
rotary evaporation (ca. 5 mL). Hexanes (25 mL) was added, and the
CH2Cl2 was removed by rotary evaporation. The solvent was decanted
from the precipitate, which was dissolved in CH2Cl2 (5 mL). The
solution was added dropwise to stirred hexanes (25 mL), and the
CH2Cl2 was removed by rotary evaporation. The solvent was decanted
from the precipitate, which was dried by oil pump vacuum to give
3+PF6− as a yellow powder (0.182 g, 0.310 mmol, 84%). Dec pt: 211 °C
(capillary). Anal. Calcd (%) for C19H24F6N5OPRu (585.07): C 39.05, H
4.14, N 11.98. Found: C 39.05, H 4.21, N 11.94. NMR (δ): 1H (DMSO-
3
(s, 1H, NH), 9.96 (s, 1H, NH), 7.69 (d, JHH = 7.8 Hz, 1H,
3
CCHCHCHCHC), 7.47 (d, JHH = 8.1 Hz, 1H, CCHCHCHCHC),
7.36−7.28 (m, 9H, P(C6H5)3), 7.14−7.08 (m, 8H, P(C6H5)3 and
NCCH(CH)2CHCN), 7.04−6.98 (m, 2H, NCCH(CH)2CHCN),
6.77−6.74 (m, 2H, CCHCHCHCHC), 6.53 (s, 1H, NH), 6.13 (s,
2H, NH2), 5.08 (br s, 1H, CCHCHCHC),55 4.86 (br s, 1H,
CCHCHCHC),55 4.30 (dd, 1H, 3JHH = 2.4 Hz, 3JHH = 2.0 Hz, CCHC-
HCHC),55 3.33 (s, H2O); 13C{1H} (100 MHz)48 151.4 (s, NH
CNH2), 143.9 (s, NC(NH)2), 141.7 (s, NCCHCHCHCHCN),
d6, 300 MHz) 11.70 (s, 2H, NH),54 7.42−7.40 (m, 1H, NCCH-
(CH)2CHCN), 7.28−7.20 (m, 2H, NCCH(CH)2CHCN), 7.13−7.11
(m, 1H, NCCH(CH)2CHCN), 6.65 (s, 2H, NH2), 5.83 (s, 1H, NH),
1.58 (s, 15H, C5(CH3)5); 13C{1H} (CD2Cl2, 75 MHz) 206.9 (s, CO),
153.4 (s, NHCNH2), 145.1 (s, NC(NH)2), 141.1 (s, NCCHCH-
CHCHCN), 132.2 (s, NCCHCHCHCHCN), 123.9 (s, NCCHCH-
CHCHCN), 123.5 (s, NCCHCHCHCHCN), 118.1 (s, NCCHCH-
1
2
135.3 (d, JCP = 42.3 Hz, i-C6H5), 132.5 (d, JCP = 10.5 Hz, o-C6H5),
3
131.6 (s, NCCHCHCHCHCN), 129.3 (s, p-C6H5), 127.7 (d, JCP
=
9.4 Hz, m-C6H5), C9H7 at 126.2 (s), 125.6 (s), 121.6 (s), 120.7 (s),
117.1 (s), 106.0 (s), 83.2 (s), 59.2 (s), 55.4 (s); 124.6 (s, NCC-
HCHCHCHCN), 121.2 (s, NCCHCHCHCHCN), 110.5 (s,
CHCHCN), 111.6 (s, NCCHCHCHCHCN), 93.2 (s, C5(CH3)5), 9.9
(s, C5(CH3)5); 31P{1H} (CD2Cl2, 121 MHz) −142.7 (sep, 1JPF = 710.3
Hz); 19F{1H} (CD2Cl2, 282 MHz) −70.6 (d, 1JFP = 712.1 Hz). IR (cm–1,
powder film): 3499 (w), 3406 (m), 2966 (w), 2920 (w), 1922 (s, νCO),
1684 (m), 1637 (w), 1560 (m), 1460 (m), 1383 (w), 1313 (w), 1259
(w), 1097 (m), 1027 (m), 934 (w), 842 (s), 814 (w), 756 (m). MS:45
442 (61) [3]+, 412 (100) [3 – CO]+. UV–visible (nm, 0.0010 M in
DMSO (ε, M–1 cm–1)): 292 (2890), 308 (3260), 403 (258).
NCCHCHCHCHCN), 110.3 (s, NCCHCHCHCHCN); 31P{1H}
(161 MHz) 70.8 (s). IR (cm–1, powder film): 3428 (m), 3266 (w),
3235 (w), 3192 (w), 3157 (w), 3076 (w), 3034 (m), 2968 (w), 2918
(w), 2826 (w), 2351 (m), 1942 (w), 1675 (s), 1633 (m), 1610 (m),
1559 (s), 1478 (m), 1436 (m), 1413 (m), 1320 (m), 1262 (m), 1185
(m), 1154 (m), 1092 (m), 1027 (w), 972 (w), 926 (w), 856 (w), 806
(w), 737 (m), 694 (m). MS:45 653 (82) [4]+, 391 (100) [4 – PPh3]+.
UV–visible (nm, 0.0010 M in DMSO (ε, M–1 cm–1)): 289 (3890), 296
(4570), 299 (4360), 311 (5950), 315 (5840), 325 (5260), 329 (5160),
335 (3680), 424 (2440), 445 (3050), 455 (3370).
[(η5-C5Me5)Ru(CO)(GBI)]+BArf− (3+BArf−). A Schlenk flask was
charged with 3+Cl−·(CHCl3)0.17 (0.187 g, 0.379 mmol), Na+BArf−
(0.352 g, 0.398 mmol),24 and CH2Cl2 (5 mL). The mixture was stirred
for 16 h and filtered through a plug of Celite (1 × 2.5 cm), which was
rinsed with CH2Cl2 (2 × 15 mL).50 The filtrate was concentrated by
rotary evaporation (ca. 5 mL). Hexanes (25 mL) was added, and the
CH2Cl2 was removed by rotary evaporation. The solvent was decanted
from the precipitate, which was dissolved in CH2Cl2 (5 mL). The
solution was added dropwise to stirred hexanes (25 mL), and the
CH2Cl2 was removed by rotary evaporation. The solvent was decanted
from the precipitate, which was dried by oil pump vacuum to give
3+BArf−·H2O as a yellow powder (0.400 g, 0.303 mmol, 80%). Dec pt:
184 °C (capillary). Anal. Calcd (%) for C51H36BF24N5ORu·H2O
(1321.17): C 46.38, H 2.90, N 5.30. Found: C 46.51, H 2.91, N 5.35.
NMR (δ): 1H (DMSO-d6, 300 MHz) 11.35 (s, 2H, NH),54 7.65 (s, 8H,
o-B(C6H3(CF3)2)4), 7.60 (s, 4H, p-B(C6H3(CF3)2)4), 7.42−7.40 (m,
Friedel−Crafts Alkylations (Table 5).56 An NMR tube was
charged with catalyst (0.010 mmol), trans-β-nitrostyrene (6, 0.015 g,
0.10 mmol), an indole (5a,b, 0.20 mmol), an internal standard
(mesitylene for 5a; tridecane for 5b), and CD2Cl2 (0.5 mL). The tube
was sealed, and 1H NMR spectra were periodically recorded. The CH
CH signals of the trans-β-nitrostyrene and the product CH2NO2 signals
at ca. 5 ppm were integrated versus those of the standards.
1H NMR data for 7a (δ, CDCl3, 300 MHz): 8.08 (br s, 1H,
C8H5NH), 7.55−6.96 (m, 10H, C8H5NH and C6H5), 5.19 (t, 1H, 3JHH
=
2
3
8.2 Hz, CHCH2NO2), 5.07 (dd, 1H, JHH = 12.4 Hz, JHH = 7.2 Hz,
CHH′NO2), 4.95 (dd, 1H, 2JHH = 12.4 Hz, 3JHH = 8.2 Hz, CHH′NO2).
Literature chemical shift values (CDCl3)56a agree within 0.01 ppm, and
data in CD2Cl2 are supplied elsewhere.22
1H, NCCH(CH)2CHCN), 7.26−7.17 (m, 2H, NCCH(CH)2CHCN),
7.12−7.10 (m, 1H, NCCH(CH)2CHCN), 6.61 (s, 2H, NH2), 5.85 (s,
[2-Guanidinium-1-methyl-3-hydrobenzimidazole] Tetrakis-
(3,5-bis(trifluoromethyl)phenyl)borate ([1-methylGBI-
H]+BArf−). A round-bottom flask was charged with [1-methylGBI-
H]+Cl− (0.022 g, 0.100 mmol),12b Na+BArf− (0.089 g, 0.100 mmol),
CH2Cl2 (2.0 mL), and H2O (1 mL) with stirring. After 2 h, the organic
1H, NH), 3.34 (s, H2O), 1.56 (s, 15H, C5(CH3)5); 13C{1H} (CD2Cl2,
1
75 MHz) 206.2 (s, CO), 161.8 (q, JCB = 49.6 Hz, i-C6H3(CF3)2),
152.5 (s, NHCNH2), 143.6 (s, NC(NH)2), 141.0
(s, NCCHCHCHCHCN), 135.2 (s, o-C6H3(CF3)2), 131.8
L
dx.doi.org/10.1021/om500704u | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
layer was separated and washed with H2O (3 × 1.0 mL). The solvent was
removed by rotary evaporation. The residue was chromatographed on a
silica gel column (5 × 1 cm; 98:2 v/v CH2Cl2/MeOH). The solvent was
removed from the product-containing fractions to give [1-methylGBI-
H]+BArf− as a pale pink powder (0.060 g, 0.058 mmol, 58%). Mp: 110−
113 °C (capillary). Anal. Calcd (%) for C41H24BF24N5 (1053.45): C
46.75, H 2.30, N 6.65. Found: C 47.28, H 2.41, N 6.66. NMR (δ,
CD2Cl2): 1H (500 MHz) 7.71 (s, 8H, o-B(C6H3(CF3)2)4), 7.55 (s, 4H,
C.; Crabtree, R. H.; Dannenberg, J. J.; Hobza, P.; Kjaergaard, H. G.;
Legon, A. C.; Mennucci, B.; Nesbitt, D. J. Pure Appl. Chem. 2011, 83,
1637−1641.
(2) (a) Sinden, R. R. DNA Structure and Function; Academic Press: San
Diego, CA, 1994. (b) Whitesides, G. M.; Simanek, E. E.; Mathias, J. P.;
Seto, C. T.; Chin, D. N.; Mammen, M.; Gordon, D. M. Acc. Chem. Res.
1995, 28, 37−44. (c) Perrin, C. L.; Nielson, J. B. Annu. Rev. Phys. Chem.
1997, 48, 511−544. (d) Sherrington, D. C.; Taskinen, K. A. Chem. Soc.
Rev. 2001, 30, 83−93. (e) Amendola, V.; Fabbrizzi, L.; Mosca, L. Chem.
Soc. Rev. 2010, 39, 3889−3915.
(3) (a) Reviews: Doyle, A. G.; Jacobsen, E. N. Chem. Rev. 2007, 107,
5713−5743. (b) Yu, X.; Wang, W. Chem.Asian J. 2008, 3, 516−532.
(c) Zhang, Z.; Schreiner, P. R. Chem. Soc. Rev. 2009, 38, 1187−1198.
(4) Representative reports from 2014: (a) Kim, H. J.; Ajitha, M. J.; Lee,
Y.; Ryu, J.; Kim, J.; Lee, Y.; Jung, Y.; Chang, S. J. Am. Chem. Soc. 2014,
136, 1132−1140. (b) Wang, W.; Kumar, M.; Hammond, G. B.; Xu, B.
Org. Lett. 2014, 16, 636−639.
(5) (a) White, A. J.; Wharton, C. W. Biochem. J. 1990, 270, 627−637.
(b) Gerlt, J. A.; Kreevoy, M. M.; Cleland, W. W.; Frey, P. A. Chem. Biol.
1997, 4, 259−267. (c) Cleland, W. W.; Frey, P. A.; Gerlt, J. A. J. Biol.
Chem. 1998, 273, 25529−25532. (d) Guo, H.; Salahub, D. R. Angew.
Chem., Int. Ed. 1998, 37, 2985−2990; Angew. Chem. 1998, 110, 3155−
3160. (e) Tuchscherer, G.; Lehmann, C.; Mathieu, M. Angew. Chem., Int.
Ed. 1998, 37, 2990−2993; Angew. Chem. 1998, 110, 3160−3164.
(f) Schowen, K. B.; Limbach, H.-H.; Denisov, G. S.; Schowen, R. L.
p-B(C6H3(CF3)2)4), 7.11−7.04 (m, 4H, NCCH(CH)2CHCNCH3),
5.49 (br s, 4H, NH),57 3.62 (s, 3H, NCCH(CH)2CHCNCH3);
13C{1H} (125 MHz) 162.0 (q, 1JCB = 49.8 Hz, i-C6H3(CF3)2), 158.4 (s,
NHCNH2), 149.7 (s, NC(NH)2), 135.1 (s, o-C6H3(CF3)2), 130.8
and 128.1 (2 s, NCCHCHCHCHCNCH3), 129.1 (q, 2JCF = 31.5 Hz, m-
C6H3(CF3)2), 124.9 (q, 1JCF = 272.3 Hz, C6H3(CF3)2), 125.6 and 125.5
(2 s, NCCHCHCHCHCNCH3), 117.9 (s, p-C6H3(CF3)2), 112.0 and
111.0 (s, NCCHCHCHCHCNCH3), 39.6 (s, NCCHCHCHCH-
CNCH3). IR (cm–1, powder film): 3520 (w), 3444 (w), 3419 (w),
1625 (m), 1585 (s), 1556 (m), 1490 (m), 1456 (w), 1413 (w), 1354 (s),
1315 (w), 1273 (s), 1109 (s), 1097 (s), 931 (w), 885 (s), 835 (s), 746
(s), 709 (s), 680 (s).
Crystallography. A. A CH2Cl2 solution of 1+PF6− was layered with
−
n-pentane. After 7 d, yellow prisms of 1+PF6 ·CH2Cl2 were collected
and data were recorded using a Nonius KappaCCD diffractometer as
outlined in Table s3. Cell parameters were obtained from 10 reflections
using a 10° scan and refined with 7191 reflections. Lorentz, polarization,
and absorption corrections were applied.58 The space group was
determined from systematic absences and subsequent least-squares
refinement. The structure was solved by direct methods. The parameters
were refined with all data by full-matrix least-squares on F2 using
SHELXL-97.59 Non-hydrogen atoms were refined with anisotropic
thermal parameters. The hydrogen atoms were fixed in idealized
positions using a riding model. Scattering factors were taken from the
literature.60 The two cations and anions in the unit cell were related by
an inversion center. B. A 1:1 v/v CH2Cl2/benzene solution of 1+BArf−
was layered with hexanes. After 21 d at −18 °C, yellow blocks of 1+BArf−−·
CH2Cl2 were collected and were analyzed as described for 1+PF6 ·
CH2Cl2 (cell parameters from 10 frames using a 10° scan; refined with
4492 reflections). The structure was solved and refined as in A. The
fluorine atoms of one CF3 group showed displacement disorder
(F12a:F12a′, F12b:F12b′, F12c:F12c′), which could be refined to a
62:38 occupancy ratio.
Biochim. Biophys. Acta 2000, 1458, 43−62. (g) Simon
́
, L.; Goodman, J.
M. J. Org. Chem. 2010, 75, 1831−1840.
(6) (a) Kirby, A. J. Angew. Chem., Int. Ed. 1996, 35, 707−724; Angew.
Chem. 1996, 108, 770−790. (b) Zhao, Y.; Rodrigo, J.; Hoveyda, A. H.;
Snapper, M. L. Nature 2006, 443, 67−70. (c) Davie, E. A. C.; Mennen, S.
M.; Xu, Y.; Miller, S. J. Chem. Rev. 2007, 107, 5759−5812. (d) Knowles,
R. R.; Jacobsen, E. N. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 20678−
20685. (e) Wennemers, H. Chem. Commun. 2011, 47, 12036−12041.
(7) (a) Etter, M. C.; Urbanczyk-Lipkowska, Z.; Zia-Ebrahimi, M.;
̃
Panunto, T. W. J. Am. Chem. Soc. 1990, 112, 8415−8426. (b) Etter, M.
C. Acc. Chem. Res. 1990, 23, 120−126.
(8) Abbreviations: en = 1,2-ethylenediamine; BArf = B(3,5-
C6H3(CF3)2)4; TRISPHAT = (tris(tetrachlorobenzenediolato)phos-
phate(V) or P(o-C6Cl4O2)3.
(9) (a) Ganzmann, C.; Gladysz, J. A. Chem.Eur. J. 2008, 14, 5397−
5400. (b) Lewis, K. G.; Ghosh, S. K.; Bhuvanesh, N.; Gladysz, J. A.
Manuscript in preparation. (c) See also: Lewis, K. G.; Ghosh. S. K.;
Gladysz, J. A. U.S. Patent WO 2014018978A2; Chem. Abstr. 2014, 160,
240833.
(10) Representative salts with y/z = 3/1 from the Cambridge
Crystallographic Data Centre (CCDC Refcodes): A07020, A12326,
OB6070, A18811, 231235, 645729.
ASSOCIATED CONTENT
* Supporting Information
■
S
Additional experimental details and procedures, NMR and UV−
visible spectra, tables of 1H and 13C NMR data for GBI
complexes, and a table and CIF files with crystallographic details.
This material is available free of charge via the Internet at http://
pubs.acs.org.
(11) King, F. E.; Acheson, R. M.; Spensley, P. C. J. Chem. Soc. 1948,
1366−1371.
(12) (a) Barba-Behrens, N.; Vaz
́
quez-Olmos, A.; Castillo-Blum, S. E.;
dez, R. M.; Rosales-Hoz, M. J.;
Vincente, R.; Escuer, A. Transition Met. Chem. 1996, 21, 31−37.
(b) Andrade-Lopez, N.; Ariza-Castolo, A.; Contreras, R.; Vazquez-
Olmos, A.; Barba-Behrens, N.; Tlahuext, H. Heteroat. Chem. 1997, 8,
397−410. (c) Andrade-Lopez, N.; Cartas-Rosado, R.; García-Baez, E.;
Contreras, R.; Tlahuext, H. Heteroat. Chem. 1998, 9, 399−409.
(d) Fialon, M.-P.; Andrade-Lopez, N.; Barba-Behrens, N.; Contreras,
R. Heteroat. Chem. 1998, 9, 637−641. (e) Ceniceros-Gomez, A. E.;
Barba-Behrens, N.; Bernes, S.; Noth, H.; Castillo-Blum, S. E. Inorg.
Hojer, G.; Meza-Hojer, S.; Hernan
́
̈
̈
́
́
AUTHOR INFORMATION
Corresponding Author
*Tel: 979-845-1399. E-mail: gladysz@mail.chem.tamu.edu.
■
́
́
Notes
́
́
The authors declare no competing financial interest.
̀
̈
Chim. Acta 2000, 304, 230−236. (f) Castillo-Blum, S. E.; Barba-Behrens,
N. Coord. Chem. Rev. 2000, 196, 3−30.
ACKNOWLEDGMENTS
We thank the Welch Foundation (Grant A-1656) for support.
■
(13) (a) Cram, D. J. Angew. Chem., Int. Ed. Engl. 1986, 25, 1039−1057;
Angew. Chem. 1986, 98, 1041−1060. (b) Desper, J. M.; Gellman, S. H. J.
Am. Chem. Soc. 1990, 112, 6732−6734. (c) Wittenberg, J. B; Isaacs, L.
Complementarity and Preorganization. In Supramolecular Chemistry:
From Molecules to Nanomaterials; Gale, P., Steed, J., Eds.; Vol. 1; Wiley &
Sons: Hoboken, NJ, 2012.
REFERENCES
■
(1) (a) Jeffrey, G. A. An Introduction to Hydrogen Bonding; Oxford
University Press: New York, 1997. (b) Steiner, T. Angew. Chem., Int. Ed.
2002, 41, 48−76; Angew. Chem. 2002, 114, 50−80. (c) Arunan, E.;
Desiraju, G. R.; Klein, R. A.; Sadlej, J.; Scheiner, S.; Alkorta, I.; Clary, D.
(14) Cram, D. J.; Lein, G. M. J. Am. Chem. Soc. 1985, 107, 3657−3668.
M
dx.doi.org/10.1021/om500704u | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(15) Mukherjee, T.; Ganzmann, C.; Bhuvanesh, N.; Gladysz, J. A.
Organometallics, doi: 10.1021/om500705s.
(35) (a) Liang-An, C.; Xu, W.; Huang, B.; Ma, J.; Wang, L.; Xi, J.;
Harms, K.; Gong, L.; Meggers, E. J. Am. Chem. Soc. 2013, 135, 10598−
10601. (b) Chen, L.-A.; Tang, X.; Xi, J.; Xu, W.; Gong, L.; Meggers, E.
Angew. Chem., Int. Ed. 2013, 52, 14021−14025; Angew. Chem. 2013, 125,
14271−14275.
(36) Thomas, C.; Gladysz, J. A. ACS Catal. 2014, 4, 1134−1138.
(37) (a) Colquhoun, H. M.; Stoddart, J. F.; Williams, D. J. Angew.
Chem., Int. Ed. Engl. 1986, 25, 487−507; Angew. Chem. 1986, 98, 483−
503. (b) Raymo, F. M.; Stoddart, J. F. Chem. Ber. 1996, 129, 981−990.
(38) Amoroso, A. J.; Arif, A. M.; Gladysz, J. A. Organometallics 1997,
16, 6032−6034.
(39) Noyori, R.; Ohkuma, T. Angew. Chem., Int. Ed. 2001, 40, 40−73;
Angew. Chem. 2001, 113, 40−75.
(40) Mukherjee, T. Doctoral Thesis, Texas A&M University, in
preparation.
(41) David, O. R. P. Planar Chiral Catalysts In Comprehensive
Enantioselective Organocatalysis: Catalyst, Reactions, and Applications, 1st
ed.; Dalko, P. I., Ed.; Wiley-VCH: Weinheim, Germany, 2013; Vol. 1,
Part II.
(42) For insightful commentary on this issue, see ref 1 in Dalko, P. I.;
Moisan, L. Angew. Chem., Int. Ed. 2004, 43, 5138−5175; Angew. Chem.
2004, 116, 5248−5286.
(43) An alternative microwave procedure is provided in the Supporting
Information.
(44) An exact yield cannot be calculated due to the indeterminate
hydration level of the RuCl3.
(45) FAB, 3-NBA, m/z (%); the peaks correspond to the most intense
signal of the isotope envelope.
(16) Bruce, M. I.; Windsor, N. J. Aust. J. Chem. 1977, 30, 1601−1604.
(17) (a) Beck, W.; Sunkel, K. Chem. Rev. 1988, 88, 1405−1421.
̈
(b) Strauss, S. H. Chem. Rev. 1993, 93, 927−942. (c) Krossing, I.; Raabe,
I. Angew. Chem., Int. Ed. 2004, 43, 2066−2090; Angew. Chem. 2004, 116,
2116−2142. (d) Díaz-Torres, R.; Alvarez, S. Dalton Trans. 2011, 40,
10742−10750.
(18) King, R. B. Inorg. Chim. Acta 1968, 2, 454−458 These page
numbers correspond to those on the PDF file downloaded from the
publisher. However, as of the submission date of this manuscript, the
publisher’s table of contents on the web indicates the page numbers
449−453.
(19) Specifically, multiple attempts were sometimes required to obtain
−
−
reasonable yields of 1+BF4 , 1+PF6 , and 1+BArf− in the initial step A1.
(20) Gill, T. P.; Mann, K. R. Organometallics 1982, 1, 485−488.
(21) Previous synthesis of this compound by a two-step procedure:
(a) Gassman, P. G.; Winter, C. H. J. Am. Chem. Soc. 1988, 110, 6130−
6135. (b) Chinn, M. S.; Heinekey, D. M. J. Am. Chem. Soc. 1990, 112,
5166−5175.
(22) Scherer, A. Doctoral Thesis, Universitat Erlangen-Nurnberg,
̈
̈
2009.
(23) Oro, L. A.; Ciriano, M. A.; Campo, M.; Foces-Foces, C.; Cano, F.
H. J. Organomet. Chem. 1985, 289, 117−131.
(24) Yakelis, N. A.; Bergman, R. G. Organometallics 2005, 24, 3579−
3581.
(25) Complex 2+Cl− is cleanly deprotonated by Na2CO3.22 NMR
evidence suggests that the CNHC proton of the chelate ring is
removed. Other salts investigated did not give clean deprotonation with
pyridine, NaOMe, or Na2CO3, at least under the conditions assayed.
(26) (a) Lacour, J.; Goujon-Ginglinger, C.; Torche-Haldimann, S.;
Jodry, J. J. Angew. Chem., Int. Ed. 2000, 39, 3695−3697; Angew. Chem.
2000, 112, 3830−3832. (b) Favarger, F.; Goujon-Ginglinger, C.;
Monchaud, D.; Lacour, J. J. Org. Chem. 2004, 69, 8521−8524.
(46) Kohl, F. X.; Jutzi, P. J. Organomet. Chem. 1983, 243, 119−121.
(47) This complex was purchased from ACROS or prepared by a
literature method: Trost, B. M.; Older, C. M. Organometallics 2002, 21,
2544−2546. The commercial material contained up to 10% of the
−
precursor [(η5-C5H5)Ru(C6H6)]+PF6 , but gave equally good results.
(48) The 1H and 13C{1H} NMR assignments are supported by 1H,1H
COSY NMR and 1H,13C HETCOR NMR experiments and are
consistent with assignments made in previous studies of comparable
GBI derivatives; see refs 12c, d.
(27) Agbossou, F.; O’Connor, E. J.; Garner, C. M.; Men
Fernandez, J. M.; Patton, A. T.; Ramsden, J. A.; Gladysz, J. A. Inorg.
Synth. 1992, 29, 211−225.
́
dez, N. Q.;
́
(49) An alternative procedure using a static CO atmosphere is
provided in the Supporting Information.
(50) The workup was carried out under ambient laboratory conditions.
(51) In some cases, the solubility in CHCl3 was poor. When this
occurred, a CHCl3/MeOH mixture was employed.
(52) Curiously, the ruthenium cyclopentadienyl carbonyl complexes
gave consistently low microanalytical values for nitrogen.
(53) Traces of NH- and NH2-signals were detectable at δ 8.13, 6.63,
and 5.87 ppm.
(54) The very broad signal precluded an accurate integration.
(55) The 1H and 13C{1H} NMR assignments are supported by 1H,1H
COSY NMR and 1H,13C HETCOR NMR experiments and are
consistent with assignments made in previous studies of (η5-C9H7)
Ru(PPh3)2(Cl); see ref 23.
(56) (a) Herrera, R. P.; Sgarzani, V.; Bernardi, L.; Ricci, A. Angew.
Chem., Int. Ed. 2005, 44, 6576−6579; Angew. Chem. 2005, 117, 6734−
6737. (b) See also ref 30.
(57) Only a single signal was observed for the five NH protons, and it
integrated to 4H, suggestive of some H/D exchange.
(58) (a) Collect data collection software; Nonius B. V., 1998.
(b) Scalepack data processing software; Otwinowski, Z.; Minor, W. In
Methods Enzymol. 1997, 276 (Macromolecular Crystallography, Part A),
307.
(28) (a) McCrindle, R.; Ferguson, G.; McAlees, A. J.; Massod, P.;
Stephenson, D. K. J. Chem. Soc., Dalton Trans. 1982, 1291−1296.
(b) Husain, A.; Nami, A. A. A.; Siddiqi, K. S. Appl. Organomet. Chem.
2011, 25, 761−768.
(29) (a) Lungwitz, R.; Spange, S. New J. Chem. 2008, 32, 392−394.
́
(b) Claudio, A. F. M.; Swift, L.; Hallett, J. P.; Welton, T.; Coutinho, J. A.
P.; Freire, M. G. Phys. Chem. Chem. Phys. 2014, 16, 6593−6601.
(30) The following papers describe hydrogen bond donor catalysts for
the reaction of indole and trans-β-nitrostyrene. (a) Dessole, G.; Herrera,
R. P.; Ricci, A. Synlett 2004, 13, 2374−2378. (b) Zhuang, W.; Hazell, R.
G.; Jørgensen, K. A. Org. Biomol. Chem. 2005, 3, 2566−2571.
(c) Fleming, E. M.; McCabe, T.; Connon, S. J. Tetrahedron Lett.
2006, 47, 7037−7042. (d) Akalay, D.; Durner, G.; Bats, J. W.; Bolte, M.;
̈
Gobel, M. W. J. Org. Chem. 2007, 72, 5618−5624. (e) Takenaka, N.;
̈
Sarangthem, R. S.; Seerla, S. K. Org. Lett. 2007, 9, 2819−2822.
(f) Rodriguez, A. A.; Yoo, H.; Ziller, J. W.; Shea, K. J. Tetrahedron Lett.
́ ́
2009, 50, 6830−6833. (g) Marques-Lopez, E.; Alcaine, A.; Tejero, T.;
Herrera, R. P. Eur. J. Org. Chem. 2011, 3700−3705. (h) Tran, N. T.;
Wilson, S. O.; Franz, A. K. Org. Lett. 2012, 14, 186−189. (i) Shokri, A.;
Wang, X.-B.; Kass, S. R. J. Am. Chem. Soc. 2013, 135, 9525−9530.
(31) Berkessel, A.; Cleemann, F.; Mukherjee, S.; Muller, T. N.; Lex, J.
̈
Angew. Chem., Int. Ed. 2005, 44, 807−811; Angew. Chem. 2005, 117,
817−821.
(59) Sheldrick, G. M. SHELX-97, Program for Refinement of Crystal
(32) (a) Lam, S. Y.; Louis, C.; Benoit, R. L. J. Am. Chem. Soc. 1976, 98,
Structures; University of Gottingen, 1997.
̈
1156−1160. (b) Meshcheryakov, D.; Arnaud-Neu, F.; Bohmer, V.;
̈
(60) Cromer, D. T.; Waber, J. T. In International Tables for X-ray
Crystallography; Ibers, J. A.; Hamilton, W. C., Eds.; Kynoch:
Birmingham, England, 1974.
Bolte, M.; Hubscher-Bruder, V.; Jobin, E.; Thondorf, I.; Werner, S. Org.
Biomol. Chem. 2008, 6, 1004−10014. (c) Amendola, V.; Bergamaschi,
G.; Boiocchi, M.; Fabbrizzi, L.; Milani, M. Chem.Eur. J. 2010, 16,
4368−4380.
(33) Heyns, A. M. Spectrochim. Acta, Part A 1977, 33, 315−322.
́ ́
(34) Hamza, A.; Schubert, G.; Soos, T.; Papai, I. J. Am. Chem. Soc. 2006,
128, 13151−13160.
N
dx.doi.org/10.1021/om500704u | Organometallics XXXX, XXX, XXX−XXX