Iridium complexes containing N-donor-functionalized η1-fluorenyl ligands

Article abstract of DOI:10.1016/j.poly.2015.08.023

Iridium complexes of two different nitrogen-donor-functionalized fluorenyl ligands have been synthesized and characterized, L¹Ir(COD) and L²Ir(COD), where L^1- = [9-(2-dimethylamino)ethyl]fluorenyl anion, L^2- = [

Full text of DOI:10.1016/j.poly.2015.08.023

Polyhedron xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Iridium complexes containing N-donor-functionalized
ligands
g
1-fluorenyl
Fioralba Taullaj a, David Armstrong a, Alan J. Lough b, Ulrich Fekl a,
⇑
a Department of Chemical and Physical Sciences, University of Toronto Mississauga, 3359 Mississauga Road, Mississauga, ON L5L 1C6, Canada
b X-ray Crystallography Lab, University of Toronto St. George, 80 St. George St., Toronto, ON M5S 3H6, Canada
a r t i c l e i n f o
a b s t r a c t
Article history:
Iridium complexes of two different nitrogen-donor-functionalized fluorenyl ligands have been synthe-
sized and characterized, L1Ir(COD) and L2Ir(COD), where L1À = [9-(2-dimethylamino)ethyl]fluorenyl
anion, L2À = [9-(2-ortho-pyridyl)ethyl]fluorenyl anion, and COD = 1,5-cyclooctadiene. The structures of
both ligands (in their protonated form) and of both complexes were determined by single-crystal X-
ray crystallography. Structural consequences of the nature of the nitrogen ligand (dimethylamino sub-
stituent versus ortho-pyridyl substituent) are significantly different Ir–N distances (shorter for L2Ir
(COD)) and indirect cis-influence on the COD ligand resulting in a shorter Ir–olefin bond cis to the nitro-
gen donor in L2Ir(COD). Both complexes act as precatalysts for arene hydrogenation.
Received 2 June 2015
Accepted 24 August 2015
Available online xxxx
Keywords:
Constrained geometry
Fluorenyl
Nitrogen donor
Late metal
Ó 2015 Elsevier Ltd. All rights reserved.
Hydrogenation
1. Introduction
p-coordinating ligand that the same ligand can switch binding
modes, even with the same metal, depending on oxidation state
Constrained geometry complexes are an important class of
and ligand environment of the metal [4]. As will be detailed below,
compounds [1]. They typically involve a
p
-donor (cyclopentadienyl
in L1Ir(COD) and L2Ir(COD) both L1À and L2À act as C,N
r
-donors
or substituted cyclopentadienyl) ligand tethered with a short lin-
(g p-donating constrained geometry
1-fluorenyl) and not as
ker (bridge) to a group containing a donor atom (Fig. 1).
ligands. Preliminary hydrogenation studies with the two com-
plexes suggest that under mild H2 pressures (1 atm) and ambient
temperature, these complexes are reduced to finely dispersed irid-
ium metal which performs heterogeneous arene hydrogenation
under the reaction conditions, such that L1Ir(COD) and L2Ir(COD)
act as precatalysts for arene hydrogenation.
Most often used for olefin polymerization, these complexes
became popular in part due to their increased thermal stability
over traditional metallocene catalysts [2]. Other transformations
that can be catalyzed by constrained geometry complexes include
polar bond hydrogenation as well as hydroboration, hydroamina-
tion, and hydrosilylation of olefins [3].
The majority of constrained geometry complexes contain early
transition metals or rare-earth metals [3]. There are very few
examples of constrained geometry complexes of late transition
metals [4]. Motivated in part by the potential for hydrogenation
catalysis, we set out to synthesize constrained geometry
2. Results and discussion
2.1. Synthesis
The procedure for the synthesis of L1Ir(COD) is depicted in
Scheme 1. This procedure was applied analogously for the synthesis
of L2Ir(COD), where 2-(bromomethyl)pyridine hydrobromide was
substituted for 2-chloro-N,N-dimethylethylamine hydrochloride.
The synthetic procedure was designed such as to allow for facile
synthesis of both the ligands and final complexes. Various syn-
thetic procedures for both ligands have been reported previously
[5], however these procedures employ column chromatography
for purification of the final product. It is known that large, organic
cations can be cleanly precipitated from aqueous solution using
hexafluorophosphate anion [6]. We therefore chose to purify the
ligands through the use of a salt metathesis reaction. Addition of
complexes of iridium, using the two potentially
p-donating
ligands L1À, [9-(2-dimethylamino)ethyl]fluorenyl anion, and L2À
,
[9-(2-ortho-pyridyl)ethyl]fluorenyl anion. Specifically, the iridium
(I) complexes L1Ir(COD) and L2Ir(COD) were obtained (COD = 1,5-
cyclooctadiene). Whether the ligands L1À and L2À actually act as
constrained geometry
p
-donating ligands (
g
5-fluorenyl residue)
or as -donor ligands (with
r
g
1-bound fluorenyl residue) will
depend on many factors. It is known from a related potentially
⇑
Corresponding author. Tel.: +1 905 828 3806.
E-mail address: ulrich.fekl@utoronto.ca (U. Fekl).
http://dx.doi.org/10.1016/j.poly.2015.08.023
0277-5387/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: F. Taullaj et al., Polyhedron (2015), http://dx.doi.org/10.1016/j.poly.2015.08.023

2
F. Taullaj et al. / Polyhedron xxx (2015) xxx–xxx
br
-
D
br
D
BR
SiR2
(CR2)n
PR
-
π-Lig
MLn
br =
D =
br
-
D
[π
-br-D]- =
-Lig
br
NR2
PR2
O(R)
S(R)
D
stab. carbene
M = Sc, Y, Ln
Ti, Zr, Hf
(in most cases)
Fig. 1. Variability of constrained geometry complexes.
Cl
H
H
H
NMe2
Cl
Li
NMe2H
nBuLi
THF
H
2
2
, LiCl
L1-H
fluorene
a) 0.2M HCl
b) KPF6/H2O
NaH
1/2 [(COD)IrCl]2
H
NMe2H
Ir
CH3
CH3
N
THF
PF6
L1Ir(COD)
+
L1-H2
Scheme 1. Synthetic procedure for L1Ir(COD).
KPF6 to doubly protonated (at C and N) ligands, denoted as L1–H2+
and L2–H+2, soluble in the form of the chloride salts, produced the
less soluble and crystalline hexafluorophosphate salts (Scheme 1).
For transfer onto iridium, a one-pot synthesis was employed, in
which deprotonation of the respective ligand by excess NaH in
the presence of [Ir(COD)Cl]2 led to successful transmetallation onto
iridium. Yields for the ligands are fair-to-good and yields for met-
allation onto iridium are excellent: L1–H2+ PFÀ6 and L1–H2+ PF6À were
synthesized in 50% yield and 66% yield, respectively. A close to
quantitative yield of 95% was obtained for both metallations, to
yield L1Ir(COD) and L2Ir(COD).
counted as a ligand. Both complexes exhibit g
1-binding of the
anionic fluorenyl moiety. Several interesting structural features
are observed in the two complexes. First, the Ir–N bond lengths
show a significant difference between the two complexes. The Ir–
N distance in L1Ir(COD) is 2.180 Å whereas the corresponding Ir–N
distance in L2Ir(COD) is 2.099 Å. The effect observed is in accord
with literature findings: a search of the CSD reveals that the aver-
age Ir–N distance for iridium coordinated to sp3 nitrogen donors
(788 examples) is 2.146 Å, while the average Ir–N distance for
sp2 nitrogen donors (8293 examples) is shorter, 2.089 Å (see Sup-
porting information). This is in part, of course, an electronic effect,
since the donor orbital for a pyridyl donor (ca. sp2) has more s-
character and is thus more compact than the donor orbital for an
amine donor (ca. sp3). A pronounced and unexpected difference
between the two complexes concerns the iridium–olefin bond
lengths, and this appears to be due to a steric effect. The irid-
ium–olefin bond length trans to the nitrogen donor is almost the
same for both complexes, at 2.026 Å for L1Ir(COD) and 2.028 Å
for L2Ir(COD). In contrast the iridium–olefin distance trans to the
fluorenyl carbanion, that is cis to the nitrogen donor, shows a dras-
tic difference with a distance of 2.128 Å for L1Ir(COD) and 2.016 Å
for L2Ir(COD). Since the ligand trans is the same in both cases
(alkyl-substituted fluorenyl), this is almost certainly the result of
a cis influence. The increased Ir–olefin bond length for L1Ir(COD)
2.2. X-ray crystal structures
Crystals suitable for X-ray diffraction were obtained by recrys-
tallization from dichloromethane for both L1–H2+ PF6À and L1–H+2
PFÀ6 , and by recrystallization from concentrated toluene solutions
for L1Ir(COD) and L2Ir(COD). The structures were solved and
refined with standard methods (details in Table 1). Anisotropic dis-
placement plots for the structures of L1–H2+, (nitrogen-protonated
L1–H), L2–H+2, L1Ir(COD) and L2Ir(COD) are given in Fig. 2.
The structures of L1Ir(COD) and L2Ir(COD) are closely related.
Both L1Ir(COD) and L2Ir(COD) show pseudo-square planar geome-
try about the iridium center if the center of each COD
p-bond is
Please cite this article in press as: F. Taullaj et al., Polyhedron (2015), http://dx.doi.org/10.1016/j.poly.2015.08.023

F. Taullaj et al. / Polyhedron xxx (2015) xxx–xxx
3
Table 1
Crystal and Structure refinement for L1–H2+ PF6- and L1–H2+ PFÀ6 , L1Ir(COD), L2Ir(COD).
À
À
L1–H+2 PF6
L2–H+2 PF6
L1Ir(COD)
L2Ir(COD)
Empirical formula
Formula weight
T (K)
C
17H20F6NP
C19H16F6NP
403.30
147(2)
C25H30IrN
536.70
147(2)
C27H26IrN
556.69
147(2)
383.31
147(2)
k (Å)
1.54178
orthorhombic
Pca21
4.54179
monoclinic
P21/c
0.71073
monoclinic
P21/n
0.71073
monoclinic
P21/n
Crystal system
Space group
Unit cell dimensions
a (Å)
b (Å)
c (Å)
12.8020(4)
10.4478(3)
12.9417(4)
90
9.0234(3)
30.2757(11)
6.4577(2)
90
9.1676(4)
24.0869(12)
9.4239(5)
90
9.2735(6)
16.4837(10)
13.5643(8)
90
a
(°)
b (°)
90
90
95.024(3)
90
1757.4(1)
4
1.524
1.994
100.976(1)
90
2042.91(17)
4
1.745
6.545
104.371(2)
90
2008.6(2)
4
1.841
6.661
c
(°)
V (Å3)
1730.99(9)
4
1.471
1.983
792
0.370 Â 0.180 Â 0.040
4.231–67.096
À14 6 h 6 15,
À12 6 k 6 12,
À15 6 l 6 15
39364
Z
Dcalc (Mg/m3)
Absorption coefficient (mmÀ1
F(0000)
)
824
1056
1088
Crystal size (mm3)
h (°)
0.180 Â 0.140 Â 0.030
2.919–66.920
À10 6 h 6 10,
À35 6 k 6 35,
À7 6 l 6 7
22960
0.190 Â 0.150 Â 0.100
1.691–27.585
À11 6 h 6 10,
À31 6 k 6 31,
À12 6 l 6 12
26871
0.180 Â 0.140 Â 0.050
1.982–27.521
À12 6 h 6 12,
À21 6 k 6 21,
À13 6 l 6 17
33932
Index ranges
Reflections collected
Independent reflections (Rint
Completeness to h = 25.00°
Absorption correction
)
3067 (0.0429)
98.4%
Semi-empirical from
equivalents
0.7529 and 0.5962
3089 (0.0662)
98.8%
Semi-empirical from
equivalents
0.7529 and 0.5962
4720 (0.0244)
100.0%
Semi-empirical from
equivalents
0.7456 and 0.5760
4613 (0.0522)
100.0%
Semi-empirical from
equivalents
0.7456 and 0.5112
Maximum and minimum
transmission
Refinement method
Full-matrix least-squares on
F2
Full-matrix least-squares on
F2
Full-matrix least-squares on
F2
Full-matrix least-squares on
F2
Data/restraints/parameters
3067/1/232
1.088
3089/0/303
1.035
4720/0/246
1.227
4613/0/262
1.083
Goodness-of-fit (GOF) on F2
Final R indices [I > 2
R indices (all data)
r
(I)]
R1 = 0.0293,
wR2 = 0.0768
R1 = 0.0310,
wR2 = 0.0781
0.341 and À0.218
R1 = 0.0410,
wR2 = 0.0872
R1 = 0.0554,
wR2 = 0.0943
0.182 and À0.269
R1 = 0.0221,
wR2 = 0.0383
R1 = 0.0291,
wR2 = 0.0400
1.260 and À0.802
R1 = 0.0241,
wR2 = 0.0441
R1 = 0.0448,
wR2 = 0.0513
3.027 and À1.914
Largest difference peak and hole
(e ÅÀ3
)
Selected distances (Å) and angles (°); L1–H2+ PF6À: N1-C1, 1.515(3), C1–C2, 1.511(4), C2–C3, 1.545(4), N1–C1–C2, 112.61(2), C1–C2–C3, 113.18(3). L2–H2+ PF6À: C16–C1, 1.377
(3), C1–N1, 1.352(3), C1–C2, 1.490(3), C2–C3, 1.545(3), N1–C1–C2, 118.41(2), C1–C2–C3, 113.32(19). L1Ir(COD): (C3/C4 centroid)–Ir1, 2.026(3), (C7/C8 centroid)–Ir1, 2.128
(3), C9–Ir1, 2.174(3), N1–Ir1, 2.180(3), (C3/C4 centroid)–Ir1–(C7–C8 centroid), 84.73(12), (C3/C4 centroid)–Ir1–C9), 95.01(12), (C7/C8 centroid)–Ir1–N1, 98.10(11), C9–Ir1–
N1, 82.83(11), Ir1–C9–C22, 106.57(2), Ir1–N1–C23, 107.06(19). L2Ir(COD): (C3/C4 centroid)–Ir1, 2.028(4), (C7/C8 centroid)–Ir1, 2.016(4), C9–Ir1, 2.180(4), N1–Ir1, 2.099(3),
(C3/C4 centroid)–Ir1–(C7–C8 centroid), 86.52(14), (C3/C4 centroid)–Ir1–C9), 98.99(14), (C7/C8 centroid)–Ir1–N1, 95.04(14), C9–Ir1–N1, 79.45(13), Ir1–C9–C22, 105.88(2),
Ir1–N1–C23, 116.20(3).
is most likely caused by steric clash between N-methyls of L1 and
the COD diolefin ligand, as highlighted in the space-filling part of
Fig. 2.
icantly slower rates of conversion. As can be seen from Fig. 3, this
appears in part to be due to a longer induction period for L1Ir(COD)
and L2Ir(COD) when compared to [Ir(COD)Cl]2.
2.3. Reaction with hydrogen
3. Summary and conclusion
Addition of 1 atm of hydrogen at room temperature to both
complexes in benzene-d6 resulted in formation of dark, iridium-
containing powder, protonated ligand L1–H/L2–H, cyclohexane-d6,
and cyclooctane (as determined by 1H NMR). This indicates that
both complexes are effective hydrogenation precatalysts. The
decomposition of both L1Ir(COD) and L2Ir(COD) appears to be due
to hydrogenation of COD and metal reduction. Iridium nanoparti-
cles are known to catalytically hydrogenate aromatic species [7].
Production of micro/nano-powder was observed to occur within
5 min for L1Ir(COD) and within 1 h for L2Ir(COD). We then tested
for catalytic hydrogenation of neat toluene under 1 atm of H2 at
room temperature using a 0.1 molar solution of catalyst (1% cata-
lyst loading) of L1Ir(COD) and L2Ir(COD). The reaction was moni-
tored by 1H NMR, and we observed slow hydrogenation of
toluene (Fig. 3). The two complexes were compared to the activity
of [Ir(COD)Cl]2 under identical conditions, and both showed signif-
Two new iridium complexes were obtained and structurally
characterized, rare late-transition metal complexes of an N-
donor-functionalized g1 fluorenyl ligand. The complex containing
a dimethylamino group, in ligand L1, contains a longer Ir–N bond
length compared to the complex containing a pyridyl group, in
ligand L2. Somewhat surprisingly, and likely due to steric clash,
the COD also appears more weakly bonded in L1Ir(COD). Consistent
with the idea that ligands are more weakly bonded in L1Ir(COD),
this complex decomposes faster than L2Ir(COD) under identical
hydrogenation conditions (1 atm of H2, toluene, RT). It may be con-
cluded that even more strongly bonded spectator ligands would be
needed to keep such iridium complexes stable under hydrogena-
tion conditions, for homogeneous hydrogenation catalysis. The
two complexes do, however, serve as pre-catalysts for heteroge-
neous catalysis of arene hydrogenation.
Please cite this article in press as: F. Taullaj et al., Polyhedron (2015), http://dx.doi.org/10.1016/j.poly.2015.08.023

4
F. Taullaj et al. / Polyhedron xxx (2015) xxx–xxx
Fig. 2. Molecular structures of the cationic, protonated ligands L1–H2+ and L2–H2+, left (PF6À counterions not shown) and the metal complexes L1Ir(COD) and L2Ir(COD), middle
(using 30% probability anisotropic displacement ellipsoids) and right (space-filling). The steric clash between N-methyls and COD for L1Ir(COD) is highlighted in the space-
filling picture. Selected distances and angles are reported in the legend of Table 1.
Fig. 3. Hydrogenation of toluene to methylcyclohexane as catalyzed by each of [Ir(COD)Cl]2, L1Ir(COD) and L2Ir(COD) (0.1 M solution; 1% catalyst loading) at 1 atm hydrogen
and room temperature. [Ir(COD)Cl]2 achieved 95% conversion after 72 h, L1Ir(COD) achieved 90% conversion after 168 h and L2Ir(COD) achieved 14% conversion after 144 h.
Please cite this article in press as: F. Taullaj et al., Polyhedron (2015), http://dx.doi.org/10.1016/j.poly.2015.08.023

F. Taullaj et al. / Polyhedron xxx (2015) xxx–xxx
5
amount of THF present. Elemental analysis of C19H16N+ PF6À
(404.10 g/mol): Calc. C, 56.58; H, 4.00; N, 3.47; found: C, 58.10;
H, 3.99; N, 3.55%.
4. Experimental
4.1. General information, reagents and precursors
L1Ir(COD): 100 mg (0.26 mmol) of L1–H2+ PFÀ6 were suspended
with 24 mg (1 mmol) NaH in THF, with 1% potassium tert-butoxide
and allowed to react with stirring overnight at room temperature.
87 mg (0.13 mmol) of [Ir(COD)Cl]2 was added to the NaL1 solution
and allowed to react with stirring for 1 h. The resulting (COD)Ir(L1)
solution was filtered through basic alumina. The solvent was
removed in vacuo to yield 120 mg (95%) of L1Ir(COD) X-ray quality
crystals were obtained from a saturated toluene solution at room
temperature.
Unless otherwise stated, all reagents were purchased from
Sigma Aldrich and used without further purification. Solvents were
degassed under vacuum and dried either on activated molecular
sieves (CH2Cl2) or sodium benzophenone ketyl (THF, Et2O, toluene,
C6D6). All reactions were performed under an atmosphere of dry
argon in either a glove box or using Schlenk techniques. NMR data
was obtained on a Bruker Avance III 400 MHz spectrometer, all
spectra were referenced to solvent residual peaks.
1H NMR (400 MHz, 298 K, C6D6 ref d 7.16): d 8.01 (d, 2H,
J = 7.6 Hz, fluorenyl); d 7.48 (t, 2H, J = 7.5 Hz, fluorenyl); d 7.44 (t,
2H, J = 7.5 Hz, fluorenyl); d 7.28 (t, 2H, J = 7.4 Hz, fluorenyl); d
3.00 (bs, 2H, COD CH); d 2.54 (t, 2H, J = 6.4 Hz, L1 CH2); d 2.09 (s,
6H, L1 CH3); d 2.02 (bs, 2H, COD CH); d 1.94 (t, 2H, J = 6.4 Hz, L1
CH2); d 1.87 (m, 2H, COD CH2); d 1.57 (m, 2H, COD CH2); d 1.18
(m, 2H, COD CH2); d 1.00 (m, 2H, COD CH2). Elemental analysis
of L1Ir(COD) (532.70 g/mol): Calc. C, 56.37; H, 4.92; N, 2.63; found:
C, 56.58; H, 4.00; N, 3.47%.
L1–H+2 PFÀ6 : 2.0 g (13.8 mmol) of 2-chloro-N,N-dimethylethy-
lamine hydrochloride was suspended in THF, and 9 mL
(14.4 mmol) of n-butyllithium (1.6 M in hexanes) were added
dropwise with stirring. The reaction was allowed to stir for 2 h at
room temperature. 2.3 g of fluorene (13.8 mmol) were dissolved
in THF and 9 mL (14.4 mmol) of n-butyllithium were added drop-
wise with stirring. The reaction was allowed to stir for ꢀ1 h at
room temperature. The fluorenyl lithium solution was added to
the 2-chloro-N,N-dimethylethylamine solution and allowed to
react with stirring overnight at 60 °C in a Pyrex bomb. The reaction
was quenched with water and extracted with ether. The ether
extract was dried with magnesium sulfate and gravity-filtered.
The solvent was removed in vacuo and the dried product was
reacted with an equivalent (13.8 mmol) of concentrated HCl in
water to yield 2-(9H-fluoren-9-yl)-N,N-dimethylethylammonium
chloride. The 2-(9H-fluoren-9-yl)-N,N-dimethylethylammonium
chloride solution was reacted with a saturated solution containing
2.5 g (13.8 mmol) of KPF6 in water. The product was allowed to set-
tle out of the aqueous solution overnight, and the water layer was
discarded, any residual water was removed in vacuo. The product
was recrystallized from CH2Cl2 and 2.6 g (50%) of 2-(9H-fluoren-
9-yl)-N,N-dimethylethylammonium hexafluorophosphate were
isolated. X-ray quality crystals were obtained during the CH2Cl2
recrystallization.
L2Ir(COD): The synthesis of L2Ir(COD) was performed similarly
to L1Ir(COD) and yielded 140 mg (95%) of L2Ir(COD). X-ray quality
crystals were obtained from a saturated toluene solution at room
temperature.
1H NMR (400 MHz, 298 K, C6D6 ref d 7.16): d 8.05 (d, 2H,
J = 7.4 Hz, fluorenyl); d 7.70 (d, 1H, J = 5.9 Hz, pyridyl); d 7.32 (t,
2H, J = 7.2 Hz, fluorenyl); d 7.26 (t, 2H, J = 7.2 Hz, fluorenyl); d
6.99 (d, 2H, J = 7.6 Hz, fluorenyl); d 6.85 (t, 1H, J = 7.9 Hz, pyridyl);
d 6.71 (d, 1H, J = 8.1 Hz, pyridyl); d 6.28 (t, 1H, J = 6.1 Hz, pyridyl); d
3.69 (s, 2H, L2 CH2); d 3.47 (bs, 2H COD CH); d 2.57 (bs, 2H COD
CH); d 1.98 (m, 2H COD CH2); d 1.86 (m, 2H COD CH2); d 1.42
(m, 4H COD CH2). Small amount of THF present. Elemental analysis
of L2Ir(COD) (552.69 g/mol): Calc. C, 58.67; H, 4.01; N, 2.53; found:
C, 58.25; H, 4.71; N, 2.52%.
1H NMR (400 MHz, 298 K, D2O ref d 4.79): d 7.85 (d, 2H,
J = 7.4 Hz, fluorenyl); d 7.62 (d, 2H, J = 7.3, fluorenyl); d 7.47 (t,
2H, J = 7.5 Hz, fluorenyl); d 7.42 (t, 2H, J = 7.4 Hz, fluorenyl); d
4.23 (bt, 1H, J = 4.3 Hz, fluorenyl); d 2.67 (s, 6H, CH3); d 2.50–2.59
(m, 4H, CH2). Elemental analysis of C17H20N+PF6À (383.31 g/mol):
Calc. C, 53.27; H, 5.26; N, 3.65; found: C, 53.94; H, 5.28; N, 3.67%.
L2–H+2 PFÀ6 : 2.5 g (15.0 mmol) of fluorene were dissolved in THF,
9.35 mL (15.0 mmol) of n-butyllithium were added to the reaction
dropwise with stirring. The reaction was allowed to stir for 2 h at
room temperature. 1.5 g (6 mmol) of 2-(bromomethyl)pyridine
hydrobromide were added to the fluorenyl lithium solution and
allowed to react overnight at room temperature. Isopropanol was
added to the reaction mixture and solvent was removed in vacuo.
The resulting 2-((9H-fluoren-9-yl)methyl)pyridine was reacted
with an equivalent (6 mmol) of concentrated HCl in water to yield
2-((9H-fluoren-9-yl)methyl)pyridine-1-ium chloride. The 2-((9H-
fluoren-9-yl)methyl)pyridine-1-ium chloride solution was reacted
with a saturated solution containing 1.1 g (6 mmol) of KPF6 in
water. The product was allowed to settle out of the aqueous solu-
tion overnight and the water layer was discarded, any residual
water was removed in vacuo. The product was recrystallized from
CH2Cl2 and isolated 1.6 g (66%) of 2-((9H-fluoren-9-yl)methyl)pyr-
idine-1-ium hexafluorophosphate. X-ray quality crystals were
obtained directly by recrystallization from CH2Cl2.
4.2. Reaction with hydrogen
The following procedure was performed using L1Ir(COD). 51 mg
(0.10 mmol) of L1Ir(COD) were dissolved in 1 mL of toluene in a
Pyrex bomb. The sample was exposed to a continuous flow of H2
at room temperature and ambient pressure while stirring. NMR
samples were obtained directly from the reaction mixture after 1,
6, 12, 24, 48, and 72 h. Turnover was determined by 1H NMR
through integration of the emerging methylcyclohexane peaks.
1H NMR (400 MHz, 298 K, C6D6 ref d 7.16) at 24 h; d 7.13 (d, 2H,
J = 7.3 Hz, toluene Ar-H); d 7.02 (m, 3H, toluene Ar-H); d 2.11 (s,
3H, toluene CH3); d 0.88 (d, 3H, J = 6.4 Hz, methylcyclohexane
CH3). The process was repeated under identical conditions and
identical catalyst loading (1 mol%) of L2Ir(COD) and [Ir(COD)Cl]2
(conversions shown in Fig. 3).
Acknowledgements
We thank the University of Toronto, as well as the Natural
Sciences and Engineering Research Council (NSERC) of Canada,
for support. A syringe grant (Hamilton Syringes) is gratefully
acknowledged.
1H NMR (400 MHz, 298 K, CD2Cl2 ref d 5.32): d 8.41 (d, 1H,
J = 6.8 Hz, pyridyl); d 8.04 (t, 1H, J = 7.8 Hz, pyridyl); d 7.66 (m,
3H, fluorenyl + pyridyl); d 7.56 (d, 2H, J = 6.1 Hz, fluorenyl); d
7.38 (m, 4H, fluorenyl); d 7.05 (d, 1H, J = 8.1 Hz, pyridyl); d 4.59
(t, 1H, J = 5.7 Hz, fluorenyl); d 3.87 (d, 2H, J = 5.6 Hz, CH2). Small
Appendix A. Supplementary data
CCDC 1404289, 1404290, 1404291, and 1404292 contains the
supplementary crystallographic data for L1Ir(COD), L2Ir(COD),
L1–H+2 PF6, and L1–H+2 PF6À. These data can be obtained free of charge
Please cite this article in press as: F. Taullaj et al., Polyhedron (2015), http://dx.doi.org/10.1016/j.poly.2015.08.023

6
F. Taullaj et al. / Polyhedron xxx (2015) xxx–xxx
(2005) 4760;
via http://www.ccdc.cam.ac.uk/conts/retrieving.html, or from the
Cambridge Crystallographic Data Centre, 12 Union Road, Cam-
bridge CB2 1EZ, UK; fax: (+44) 1223-336-033; or e-mail:
deposit@ccdc.cam.ac.uk.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.poly.2015.08.023.
(g) B.D. Stubbert, T.J. Marks, J. Am. Chem. Soc. 129 (2007) 4253;
(h) A. Raith, P. Altmann, M. Jokoja, W.A. Herrmann, F.E. Kühn, Coord. Chem. Rev.
254 (2010) 608;
(i) M. Delferro, T.J. Marks, Chem. Rev. 111 (2011) 2450.
[2] J. Cano, K. Kunz, J. Organomet. Chem. 692 (2007) 4411.
[3] H. Braunschweig, F.M. Breitling, Coord. Chem. Rev. 250 (2006) 2691.
[4] H. Shen, Z. Xie, Chem. Commun. (2009) 2431.
[5] (a) M. Tomoi, T. Yoneyama, H. Kakiuchi, Bull. Chem. Soc. Jpn. 48 (1975) 1285;
(b) H. Miao, S. Wang, S. Zhou, Y. Wei, Z. Zhou, H. Zhu, S. Wu, H. Wang, Inorg.
Chim. Acta 363 (2010) 1325.
References
[6] C.M. Gordon, J.D. Holbrey, A.R. Kennedy, K.R. Seddon, J. Mater. Chem. 8 (1998)
2627.
[7] (a) E. Bayram, M. Zahmakiran, S. Özkar, R.G. Finke, Langmuir 26 (2010) 12455;
(b) Y. Tonbul, M. Zahmakiran, S. Özkar, Appl. Catal. B., Environ. 148–149 (2014)
466;
[1] (a) P.J. Shapiro, E. Bunel, W.P. Schaefer, J.E. Bercaw, Organometallics 9 (1990)
867;
(b) W.E. Piers, P.J. Shapiro, E.E. Bunel, J.E. Bercaw, Synlett 2 (1990) 74;
(c) J. Okuda, Chem. Ber. 123 (1990) 1649;
(d) K.C. Hultzsch, P. Voth, K. Beckerle, T.P. Spaniol, J. Okuda, Organometallics 19
(2000) 228;
(e) L.A. Williams, T.J. Marks, Organometallics 28 (2009) 2053;
(f) C. Wang, G. Erker, G. Kehr, K. Wedeking, R. Fröhlich, Organometallics 24
(c) V. Mévellec, A. Roucoux, E. Ramirez, K. Philippot, B. Chaudret, Adv. Synth.
Catal. 346 (2004) 72.
Please cite this article in press as: F. Taullaj et al., Polyhedron (2015), http://dx.doi.org/10.1016/j.poly.2015.08.023