Heterobimetallic Complexes of 1,1-Diphosphineamide Ligands

Article abstract of DOI:10.1021/acs.organomet.0c00662

We herein discuss the use of 1,1-diphosphineamide ligands for the synthesis of mono- A nd heterobimetallic complexes of Rh(I) and Ti(IV) metal centers. A small library of monometallic species was synthesized and isolated: Ti was used to complex to the amide moiety, whereas Rh was used to bind to the diphosphine. The sequential synthesis of the heterobimetallic species started from isolation of the Ti(amidate) complex followed by addition of Rh precursors in solution. Meticulous analysis of the complexes by NMR spectroscopy allowed stereometric and spatial recognition of these challenging to synthesize and isolate heterobimetallic complexes.

Full text of DOI:10.1021/acs.organomet.0c00662

pubs.acs.org/Organometallics
Article
Heterobimetallic Complexes of 1,1-Diphosphineamide Ligands
Nathan T. Coles, Danila Gasperini, Cei B. Provis-Evans, Mary F. Mahon, and Ruth L. Webster*
Cite This: Organometallics 2021, 40, 148−155
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: We herein discuss the use of 1,1-diphosphineamide ligands for the
synthesis of mono- and heterobimetallic complexes of Rh(I) and Ti(IV) metal centers. A
small library of monometallic species was synthesized and isolated: Ti was used to
complex to the amide moiety, whereas Rh was used to bind to the diphosphine. The
sequential synthesis of the heterobimetallic species started from isolation of the
Ti(amidate) complex followed by addition of Rh precursors in solution. Meticulous
analysis of the complexes by NMR spectroscopy allowed stereometric and spatial
recognition of these challenging to synthesize and isolate heterobimetallic complexes.
enhanced reactivity of the bimetallic complex compared to their
monometallic precursors.
INTRODUCTION
■
The synthesis of heterobimetallic model complexes is of pivotal
importance to the understanding of function and role of
biological systems that contain multiple metal sites,1 for
example, binuclear copper oxidases,2 iron oxidoreductase,3,4
and zinc phosphatase,5 to name but a few. Understanding their
cooperativity gives access to the knowledge to construct new
drugs;6,7 mimicking these systems also helps to formulate
enhanced catalytic sites for chemical transformations.8−11 The
construction of an appropriate ligand that allows multiple metals
to selectively bind through more than one ligating group is not
trivial, but careful design allows one to reach the desired goal of
reactivity, stability, and selectivity of the final bimetallic
complexes. We have chosen to work with 1,1-diphosphinea-
mides because of the scarce literature on the utilization of these
moieties for heterobimetallic complexes and because of the
potential for tunability of the ligand frame. This class of ligands
presents two main functionalities: (i) the amidate group and (ii)
the 1,1-diphosphine group. The former has been studied due to
its several coordination modes in metal complexes; examples of
μ2 coordination through N, O are known,12,13 and relevant to
this work, Schafer and Arnold have extensively investigated their
κ2 coordination to group 4 metals14−17 and their further use as
efficient hydroamination catalysts. The second moiety is a
derivative of 1,1-bis(diphenylphosphine)methane (dppm);
these type of chelating ligands have a small bite angle and are
extremely valuable for ligation to late transition metals, such as
Ru, Rh, and Pt.18 Complexes of these ligands have shown activity
in homogeneous catalysis, with applications in reactions such as
hydrogenation,19 hydroacylation,20 and hydrogen borrowing
reactions.21 Moreover, we targeted an early/late transition metal
combination for the potential of synergistic and cooperative
reactivity of the two centers;9,22−24 examples of homogeneous
Ti−Rh bi- or multimetallic complexes have been reported25−31
and used in catalysis, for example, hydrosilylation,32 hydro-
acylation,33 and hydrogenation reactions,34 with an observed
Therefore, the synthesis and use of a small library of 1,1-
diphosphineamides are presented. Model monometallic com-
plexes of titanium and rhodium were synthesized and
characterized; these procedures were then combined in a
sequential synthetic approach to develop heterobimetallic
complexes containing both Ti and Rh metal centers.
RESULTS AND DISCUSSION
■
In 2018, we published a rapid and mild route to highly
functionalized 1,1-diphosphines using a catalytic amount of
inexpensive bases.35 We therefore applied the optimized
procedure to synthesize 1,1-diphosphineamides with 10 mol %
of KHMDS, with a 2:1 ratio of diphenylphosphine to amide
(Scheme 1); while 2a is obtained in 62% isolated yield after 6 h,
products 2b and 2c, with bulkier substituents on the nitrogen,
were obtained with low conversion, and significant amounts of
unreacted secondary phosphine was recovered. A reoptimized
reaction procedure involved increasing the catalyst loading to 50
Scheme 1. Synthesis of 1,1-Diphosphineamides 2a−2c (Yield
in Parentheses)
Received: October 6, 2020
Published: January 8, 2021
https://dx.doi.org/10.1021/acs.organomet.0c00662
© 2021 American Chemical Society
Organometallics 2021, 40, 148−155
148

Organometallics
pubs.acs.org/Organometallics
Article
mol % and leaving the reaction to stir for 16 h. Quenching the
solution with methanol (a few drops) and, as the 1,1-
diphosphines are known to slowly oxidize in solution in air,
filtration of the crude product through a short alumina plug
under an inert atmosphere yield 2b and 2c in 66% and 68%,
respectively. Particularly, because of hindered rotation of the
bulky aryl substituents on the amide, 2c is observed as a mixture
of rotamers (5:1 2c:2c′) at room temperature (RT) by 1H NMR
spectroscopy, while rotation is less hindered for 2b showing only
one rotamer.
Next, the synthesis of titanium amidate complexes was
developed; small scale reactions (4.6 × 10−2 M) were performed
in a J-Young NMR tube by reacting 2 equiv. of ligands 2a and 2b
with the yellow solution of Ti(NMe2)4 in toluene or benzene
(Scheme 2). The addition of ligand to the Ti(IV) solution shows
attribute a signal at 3.28 ppm to the NMe2 groups (12H) and a
triplet at 4.15 (J = 7.0 Hz, 2H) to the methine of 3b; while for
3b′ a signal at 3.22 ppm (18H) correlates with a triplet at 4.36
ppm (J = 6 Hz, 1H). Pleasingly, the synthesis can be scaled up to
6.5 times of Ti starting material, and by increasing the 2b:Ti
ratio to 3:1, 3b is isolated as an orange powder in 81% yield.
Isolation of 3b′ is complicated due to potential ligand
redistribution to 3b after work-up; nonetheless, full character-
ization in situ allowed us to calculate spectroscopic yield (77%)
and analyze the diffusion coefficient in C6D6 (D = 5.11 × 10−10
m2 s−1 for 3b′ vs D = 4.70 × 10−10 m2 s−1 for 3b with rsolu 3b =
9.07 Å). When we used the bulkier pro-ligand 2c, a mixture of
3c:3c′ was found in 8:1 ratio and confirmed by DOSY NMR
(rsolu 3c = 8.63 Å vs rsolu 3c′ = 7.41 Å). However, the mixture
shows similar solubility in common organic solvents which did
not allow separation.38 The observed trend indicates that
monoligation is favored when bulkier substituents at nitrogen
are used.
Scheme 2. Synthesis of Ti Complexes 3
Bis-ligated complexes 3a and 3b can exist in multiple
conformations in solution.39 However, only one signal for the
methine and methylene groups is observed in solution for both
complexes; thus, only one isomer is expected. The symmetric
1
signals observed by H NMR rule out any asymmetric isomer,
i.e., C1 point group. To further investigate the preferred isomer,
we analyzed 3a and 3b using NOESY (nuclear Overhauser effect
spectroscopy), and we observed intramolecular interactions
through space between the N-phenyl substituents and the NMe2
groups on Ti and weak interactions with the phosphine phenyl
substituents. Thus, two potential isomers can be depicted for 3
with the NMe2 groups trans to each other, while the N amidate
atoms can be trans (C2h point group) or cis (C2v point group) to
each other; the structure drawn in Scheme 2 (C2h) will be the
least sterically encumbered39 and potentially the least energeti-
cally demanding, hence the preferred isomer in solution.
Solution NMR studies of 3a and 3b at variable temperature
suggest that the C2h group is the most likely isomer in solution;
fluxional processes at the ligand are observed to be faster than
any at the metal center. At low temperature new peaks appear,
which correspond to the methylene and phosphine groups; this
is ascribed to a hindered rotation around the C−C bond in the
ligand backbone. Two broad peaks are observed at low
temperature by 31P{1H} NMR spectroscopy which coalesce
into a single peak at RT. This dynamic behavior could be due to
syn- and anti-isomers of 3a, as previously described in the
literature.40 Slight line broadening is observed when raising the
temperature, which further suggests slow isomer interconver-
sion. It is worth noting that mono- and bis-ligated species 3b and
3b′ do not interconvert at high temperature up to 348 K.
Compared to Ti(IV) amidate complexes reported in the
literature, our system is likely dominated by the flexible yet
bulky diphosphine substituents. Density functional theory
(DFT) was used to give insight into the preferred geometry of
3a (Figure 1): as anticipated, the C2h conformation is lowest
energy, with the C2 geometries lying at higher energies (+5.6
kcal/mol Oamidate trans, NMe2 cis; +9.6 kcal/mol Namidate trans,
NMe2 cis). Interestingly the C2v geometry could not be
optimized: convergence to C2 (Namidate trans) is observed.18
To further confirm our hypothesis, we tested the mono- and
bis-ligated complexes as precatalysts for intramolecular hydro-
amination reactions (Table 1). When subjecting 2,2-diphenyl-
hex-5-en-1-amine to 20 mol % of freshly isolated precatalyst 3b,
no conversion to the piperidine derivative is observed. However,
after mixing an equimolar amount (or 2:1 ratio) of 2b with
a
b
Spectroscopic yield. With 2b:Ti of 3:1.
an immediate color change to orange. When 2a is used and the
reaction analyzed by 31P{1H} NMR spectroscopy, a small but
distinctive shift of the 1,1-diphosphine signal from −6.6 to −4.5
ppm is found; by use of 1H NMR spectroscopy, the
characteristic methine signal in 2a shifts downfield to a major
product at 4.41 ppm and a minor one at 4.53 ppm in a 7:1 ratio
(cf. 4.16 ppm in 2a). We attribute the major signals to a bis-
ligated amido species, 3a, with a κ2-coordination of two ligands
to the Ti center (76% spectroscopic yield). The minor species is
assigned as monoligated species 3a′ with κ2-binding of one
ligand to the Ti center.36,37 When reacting an equimolar amount
of 2a and Ti(NMe2)4, a clearer orange solution forms that
shows, by 1H NMR analysis, a 3a:3a′ ratio of 1:3. 3a′ exhibits a
downfield shift with a peak at −3.9 ppm in the 31P{1H} NMR
spectrum (cf. −4.5 ppm for 3a). Similar solubilities made
isolation and separation of 3a′ from 3a extremely difficult.
However, diffusion ordered spectroscopy (DOSY) NMR
analysis allowed us to distinguish the two complexes: D = 4.66
× 10−10 m2 s−1 for 3a (rsolu = 8.97 Å) vs D = 6.46 × 10−10 m2 s−1
for 3a′. When attempting scaling up the reaction and increasing
the ratio of ligand to Ti of 3:1, or vice versa, several byproducts
form and are visible in the 31P{1H} NMR spectrum. There is
concomitant darkening of the reaction from orange to dark
brown; this is possibly due to the steric bulk of pro-ligand 2a,
which is relatively low when compared to the known alkyl or aryl
amidate ligands used in the literature,17 where groups such as
2,6-dimethylphenyl, 2,6-diisopropylphenyl, or tert-butyl groups
are used. Thus, we set out to use pro-ligand 2b which contains a
N-2,6-dimethylphenyl substituent. When performing a small-
scale reaction (4.6 × 10−2 M), we observe a 1:1 mixture of
3b:3b′ and two peaks at −3.5 and −2.9 ppm in the 31P{1H}
NMR spectrum corresponding to 3b and 3b′, respectively.
1
When analyzing the H NMR spectrum of the mixture, we
149
https://dx.doi.org/10.1021/acs.organomet.0c00662
Organometallics 2021, 40, 148−155

Organometallics
pubs.acs.org/Organometallics
Article
Scheme 3. Synthesis of Rh Complexes 5a and 6a
a
Isolated yield using [{Rh(COE)Cl}2] (7).
is not observed, probably due to the ease of COE vs COD
displacement.
Orange single crystals of complex 5a crystallized in the
monoclinic P21/c space group as a square-planar Rh(I) cation
from a concentrated C6D6 solution (Figure 2).43 This complex
Figure 1. Graphical representation of the modeled geometric isomers
of 3a.
Table 1. Intramolecular Hydroamination of Aminoalkene
a
Ti (20 mol %)
L (mol %)
conv (%)
3b
0
>99
>99
Ti(NMe2)4
Ti(NMe2)4
2b (20)
2b (40)
a
1
Conversions determined by H NMR spectroscopy.
Figure 2. ORTEP representation of 5a. Thermal ellipsoid at 30%
probability, with hydrogens, solvate and anions omitted for clarity.
Ti(NMe2)4 and confirming the formation of complexes 3b and
3b′, full conversion into the product is observed. This indicates
that the active species in the reaction is the monoligated
pentacoordinate complex 3b′, which is in line with active
pentacoordinate titanium complexes found in the literature.36,41
The failed activity of complex 3b in the reaction further supports
our hypothesis of a C2h geometry where a trans arrangement of
the NMe2 ligands prevents the formation of the imido species
necessary for productive cycloaddition.42
Moving on to investigate the 1,1-diphosphine moiety, initial
studies on Rh(I) complexes were performed by using [{Rh-
(COD)Cl}2] (4, COD = 1,5-cyclooctadiene) as the metal
precursor. However, when adding ligand 2a to 4 in C6D6 at RT,
several species were observed by NMR spectroscopy. The
addition of 1 equiv. of sodium tetrakis[3,5-bis(trifluoromethyl)-
phenyl]borate (NaBArF) to the mixture of 2a and 4 allowed the
reaction to proceed, and two major products were visible in the
31P{1H} NMR spectrum after 1 h (Scheme 3). Two doublets are
can be directly compared to [Rh(dppm)2]+ reported by
Leitner;44 Rh−P bond lengths of 5a (2.2969(9)−2.3029(9)
Å) are slightly longer than literature reports (2.2832(8)−
2.2908(6) Å) which is to be expected due to the increase in steric
bulk in the backbone of the ligand. This additional bulk also
leads to a slight increase in the C−P distances with bond lengths
ranging from 1.866(3) to 1.875(3) Å for 5a (dppm: 1.843(3)−
1.847(3) Å). Steric bulk is also reflected by the increase in the
P−C−P and P−Rh−P bond angles in 5a (93.98(14)° and
72.82(3)° respectively compared to dppm: 91.80(10)° and
70.80(3)°).
Further attempts at heteroleptic Rh complexes were trialed;
[Rh(dppe)Cl]2 (dppe = diphenylphosphinoethane) (8) was
F
reacted in CH2Cl2 in the presence of NaBAr and 2a or 2b
yielding complexes 9a and 9b (Scheme 4). 9a displays two
distinct distinctive signal patterns in the 31P{1H} NMR
visible at −7.9 (d, 1JRh−P = 116 Hz) and at −15.2 ppm (1JRh-P
=
131 Hz). The ligated and free COD ligand is observed in the 1H
NMR spectrum, thus suggesting partial ligand displacement.
The signal at −7.9 ppm is assigned to the homoleptic
mononuclear complex 5a with two diphosphine ligands bound
to the Rh(I) center, while the signal at −15.2 ppm is tentatively
assigned to the mononuclear heteroleptic complex 6a
containing 2a and a COD ligand. The structure of 5a was
confirmed by reacting [{Rh(COE)Cl}2] (7, COE = cyclo-
Scheme 4. Synthesis 9a and 9b (Isolated Yield in
Parentheses)
F
octene) with 2 equiv of 2a and NaBAr generating 5a as an
orange powder in 60% yield. In this case the COE analogue of 6a
150
https://dx.doi.org/10.1021/acs.organomet.0c00662
Organometallics 2021, 40, 148−155

Organometallics
pubs.acs.org/Organometallics
Article
spectrum that confirm the heteroleptic nature at the metal
center: one at 57.8 ppm assigned to [Rh]-dppe and one at −7.4
ppm assigned to [Rh]-2a. Magnetic inequivalence of the two
phosphine environments around Rh(I)40 can be observed by the
highly ordered second-order spin coupling. Simulation of the
AA′BB′X pattern allowed calculation of the coupling constants
2JP−P trans = 284 Hz, 1JRh‑dppe = 133 Hz, and 1JRh−P = 116 Hz and
heteroleptic complex 9b (Scheme 4)50 or monosubstituted Rh/
Ti complex formed by loss of a [Rh(dppe)]+ fragment, whose
signal was detected by HRMS at 1725.5083 m/z. Decom-
position of the Ti amidate bond might be related to the extreme
sensitivity of these complexes, which can react with adventitious
traces of water in solution, releasing TiO2.51,52 Reacting 3b with
8 in fluorobenzene at RT for 1 h gives conversion of 3b to
[Rh(dppe)(C6H5F)][BArF] (doublet at 73.5 ppm, JRh-P = 204
1
27 Hz,45 JP−P cis2a
=
69 Hz, and
2
2
also the JP−P cis dppe
=
Hz)53 while no traces of 10b were found. Heating the reaction
mixture to 80 °C for 10 min results in a subtle color change to
burnt orange and full conversion to heterobimetallic 10b. To
test stability, a solution of 10b was periodically monitored over 1
week by NMR spectroscopy. However, the 1H NMR spectrum
shows multiple signals, in particular, the NMe2 and methyl
signals, which indicate that multiple isomers of 10b exist in
solution, or that steric bulk of the ligand and the complex results
in highly inequivalent environment at the NMR time scale and at
298 K.
2JP−P cis dppe‑2a = −28 Hz.46 A similar coupling pattern is observed
for complex 9b. Several unsuccessful attempts at obtaining single
crystals were made, but it is worth noting that just a handful of
cationic Rh complexes with small bite angles have been reported
to date.47
With the model studies on the monometallic complexes
giving promising results, we moved on to synthesize
heterobimetallic complexes (Figure 3a). Using 8 as the Rh(I)
Starting from the 3a:3a′ mixture and reacting it with 8 in
fluorobenzene for 10 min at 80 °C (Figure 3a), two sharp signals
at 58.0 and −4.6 ppm are observed, and the second-order spin
couplings are comparable to the signals observed for 10b; we
assigned the set of signals as 10a (Figure 3b). Performing a
1H{31P} NMR simplifies overlapping signals, and we can clearly
observe the methylene of the 1,1-diphosphineamidate ligand at
2.22 ppm, while dppe signals appear at 2.15 and 1.95 ppm. The
NMe2 signal for 10a is observed at 3.09 ppm, the CH(PPh2)2
signal at 5.61 ppm, and the ortho proton of the amide phenyl at
6.28 ppm. As suggested for 10b, less steric bulk around the
ligand decreases the complexity and allows full spectroscopic
characterization of 10a.54 The complexes were isolated as oils,
thus preventing crystallization and isolation as pure samples.
Analysis of 10b by 1H DOSY allowed us to confirm that higher
aggregates formed in solution with a diffusion coefficient of 3.41
× 10−10 m2 s−1 (rsolu = 11.5 Å); moreover, mass spectrometry
analysis confirmed the double charge and isotopic distribution
of the early/late trimetallic complex.
CONCLUSIONS
■
In summary, 1,1-diphosphineamides have been developed. The
ligands have been complexed with both Ti and Rh precursors to
synthesize and fully characterize a small library of complexes.
Interestingly, mono- and bis-ligated Ti species were identified
and used in the catalytic intramolecular hydroamination of
aminoalkenes. These model complexes were then used in a
sequential synthetic route to heterobimetallic complexes, with
strong evidence obtained indicating the existence of these
complexes in solution. Further studies into the structural
parameters and crystallization of the heterobimetallic Ti−Rh
complexes of 1,1-diphosphineamides, other than application in
catalysis, are ongoing.
Figure 3. (a) Synthesis of Ti−Rh complexes 10a and 10b. (b) 1H and
31P{1H} NMR spectra details of 10a and 10b.
source, we synthesized the Ti(IV) complex prior to addition of
8.48 Thus, complex 3b was reacted with 1 equiv of 8 in C6H6 in
the presence of NaBAr (Figure 3a),49 which results in the
F
formation of a dark oil. After removal of the solvent, the 31P{1H}
NMR spectrum shows a AA′BB′X splitting pattern with two sets
at 58.2 ppm which is attributed to Rh−P(dppe). The
coordinated 1,1-diphosphine shows a broad signal at −3.2
ppm, potentially due to steric congestion around the
diphosphine (Figure 3b). Once again, simulation of the
second-order spin coupling allowed calculation of the coupling
EXPERIMENTAL SECTION
■
General Considerations. Where appropriate, reagents were
handled under an inert atmosphere by using either a Schlenk line
(nitrogen or argon) or a glovebox (argon). Glassware was oven-dried
(100 °C) and flamed under vacuum where applicable. 1H, 13C, and 31
P
2
1
1
constants JP−P trans = 282, JRh‑dppe = 133 Hz, JRh−P = 113 Hz,
NMR spectra were recorded on 400/500 Bruker or 500 Agilent
2JP−P cis dppe
= =
29, 2JP−P cis2a 68, and 2JP−P cis dppe‑2a = −28 Hz.
spectrometers at 298 K unless stated otherwise. H and 13C NMR
1
We attribute these signals to complex 10b.28 However, after 1
week in solution, a second species was observed. The 31P{1H}
NMR spectrum in the region of the coordinated dppe shows a
set of signals slightly shifted to 57.9 ppm, which is attributed to
spectra in deuterated solvents were referenced to residual solvent peaks.
31P was referenced to an external standard of PPh3 in C6D6 (−5 ppm).
When fluorobenzene was used as a solvent, the samples were locked to a
capillary of C6D6 and referenced to the BArF− counterion (8.31 ppm). In
151
https://dx.doi.org/10.1021/acs.organomet.0c00662
Organometallics 2021, 40, 148−155

Organometallics
pubs.acs.org/Organometallics
Article
CDCl3, 1H and 13C{1H} NMR chemical shifts are reported relative to
CHCl3 at 7.26 and 77.16 ppm, respectively; in CD2Cl2, 1H and 13C{1H}
NMR chemical shifts are reported relative to CH2Cl2 at 5.32 and 54.0
(s, 1H, NH), 4.13 (t, J = 5.4, 1H, CH(PPh2)2), 2.47 (sept, J = 6.9, 2H,
CH, obscured by H of 2c), 2.03 (td, J = 10.8, J = 5.4, 2H,
CH2CH(PPh2)2), 0.83 (d, J = 6.9, 12H, CH3), 0.72 (d, J = 6.9, 12H,
CH3). 13C{1H) NMR (126 MHz, CDCl3): 2c, δ 170.5 (t, J = 3.3, CO),
146.3 (s, o-Ar) 136.5 (m, PPh2), 134.5 (app t, J = 11.0, PPh2H), 134.0
(app t, J = 10.5, PPh2H), 131.0 (i-Ar), 129.1 (PPh2H), 128.9 (PPh2H),
128.4 (m, p-ArH/PPh2H), 123.5 (s, m-ArH), 35.9 (t, J = 9.8,
CH2CH(PPh2)2), 28.5 (s, CH(CH3)2), 25.8 (t, J = 24.3, CH2CH-
(PPh2)2), 23.7 (s, CH3); 2c′, 146.9 (s, o-Ar), 134.5 (app t, J = 11.1,
PPh2H), 134.4 (app t, J = 10.7, PPh2H), 131.0 (i-Ar), 129.0 (PPh2H),
128.7 (PPh2H), 128.2 (m, p-ArH/PPh2H), 123.8 (s, m-ArH), 28.3 (s,
CH(CH3)2), 25.0 (s, CH3), 14.3 (s, CH3). 31P{1H} NMR (202 MHz,
CDCl3): δ −3.2 (2c′), −3.4 (2c); mp (°C) 134−137. FTIR-ATR: ν
(cm−1): 3285 (NH), 1683 (CO). HRMS (ESI−): calcd for
C40H42NO3P2− [M − HCOO−] 646.2640; found: 646.2659.
[Ti(NMe2)2(O,N-2a)2] 3a. A J-Young NMR tube was charged with
2a (25 mg, 0.046 mmol), and then C6D6 (0.5 mL) was added without
stirring. Ti(NMe2)4 (5.3 μL, 0.023 mmol) was added to the solution,
which turned orange, and they were mixed at RT and left for 1 h.
Compound 3a was formed together with 3a′ in a 7:1 ratio. The mixture
was dried in vacuo, and n-hexane was added (2 mL) at 0 °C followed by
cannula filtration to obtain 3a as an oil (76% spectroscopic yield) which
was analyzed in situ. 1H NMR (500 MHz, C6D6): δ 7.74−7.67 (m, 8H,
PPh2), 7.49−7.43 (m, 8H, PPh2), 7.11−6.93 (m, 28H, PPh2/m-ArH),
6.87 (t, J = 7.5, 2H, p-ArH) 6.57 (d, J = 7.4, 4H, o-ArH), 4.41 (t, J = 6.3,
2H, CH2CH(PPh2)2), 3.41 (s, 12H, NMe2), 2.72 (td, J = 11.1, J = 6.3,
4H, CH2CH(PPh2)2). 13C{1H} NMR (126 MHz, C6D6): δ 180.5 (m,
CO), 145.6 (s, i-Ar), 137.4 (app t, J = 5.7, PPh2), 136.9 (app t, J = 6.2,
PPh2), 134.2 (app t, J = 10.9, PPh2), 134.1 (app t, J = 10.7, PPh2), 128.5
(PPh2), 128.3 (app t, J = 3.6, PPh2), 128.2 (app t, J = 4.6, PPh2), 128.2
(s, m-ArH). 128.1 (app t, J = 3.5, PPh2), 124.5 (o-ArH), 123.9 (p-ArH),
46.5 (NMe2), 33.0 (t, J = 11.6, CH2CH(PPh2)2), 26.2 (t, J = 29.0,
CH2CH(PPh2)2). 31P{1H} NMR (202 MHz, C6D6): δ −4.5; D = 4.66
× 10−10 m2 s−1 (3.26 × 10−11 m2 s−1).
[Ti(NMe2)3(O,N-2a)] 3a′. A Schlenk flask was charged with 2a
(11.9 mg, 0.023 mmol), and then C6D6 (0.5 mL) was added without
stirring. Ti(NMe2)4 (5.3 μL, 0.023 mmol) was added to the solution,
which turned light orange, and they were stirred slowly at RT and left
for 1 h then analyzed in situ by NMR spectroscopy. The ratio between
3a:3a′ was of 1:6.5, and attempts at separation were unsuccessful. 1H
NMR (300 MHz, C6D6): δ 7.72 (m, 6H, PPh2), 7.48 (m, 4H), 7.01 (m,
22H), 6.91−6.82 (m, 1H), 6.61−6.51 (m, 3a), 4.53 (t, J = 6.0, 1H,
CH2CH(PPh2)2), 4.41 (t, J = 6.3, 3a), 3.40 (s, 2H, 3a), 3.26 (s, 18H,
NMe2), 3.11 (s, 10H), 2.71 (td, J = 10.5, 6.1, 2H, CH2CH(PPh2)2),
2.20 (d, J = 6.4, 12H, HNMe2). 13C{1H} NMR (75 MHz, C6D6): δ
182.1, 145.7, 137.6, 137.2, 137.1, 134.9, 134.7, 134.6, 134.5, 134.4,
134.3, 134.2, 128.9, 128.8, 128.5, 124.6, 124.1, 46.8, 45.7, 45.0, 44.2,
44.1, 39.0, 32.7 (t, J = 10.9, CH2CH(PPh2)2), 26.61 (t, J = 27.5,
CH2CH(PPh2)2). 31P{1H} NMR (122 MHz, C6D6): δ −3.9, −39.8
(traces HPPh2 from the pro-ligand synthesis). D = 6.46 × 10−10 m2 s−1
(1.677 × 10−10 m2 s−1).
[Ti(NMe2)2(O,N-2b)2] 3b. A Schlenk flask was charged with 2b
(245.5 mg, 0.45 mmol), and then C6D6 (0.5 mL) was added without
stirring. Ti(NMe2)4 (34.6 μL, 0.15 mmol) was added to the solution,
which turned orange, and they were stirred slowly at RT for 1 h. After
this time residual benzene and HPPh2 were removed from the crude
product by dissolving the solid in hot hexane (5 mL), cooling the
solution to −78 °C, and then filtering through a cannula. The resulting
solid was washed with a further 2 mL of cold hexane. The compound
was dried in vacuo and used without further purification. The
compound was isolated as an orange powder in 81% yield (142 mg,
0.12 mmol), with 1H NMR showing 0.5 equiv of n-hexane.
Crystallization attempts to obtain single crystals were unsuccessful.
1H{31P} NMR (500 MHz, C6D6): δ 7.66 (br s, 8H, PPh2), 7.21 (br s,
8H, PPh2), 7.11−6.93 (m, 30H, PPh2/Ar/H), 4.15 (t, J = 7.0, 2H,
CH2CH(PPh2)2), 3.48 (s, 12H, NMe2), 2.49 (d, J = 7.0, 4H,
CH2CH(PPh2)2), 2.20 (s, 12H, ArMe). 13C{1H} NMR (126 MHz,
C6D6): δ 181.7 (CO), 143.7 (o-ArMe), 137.6−137.5 (m, PPh2),
137.4−137.3 (m, PPh2), 134.8−134.6 (m, PPh2), 134.2−134.0 (m,
PPh2), 133.3 (m-ArH), 129.0 (PPh2), 128.8 (app t, J = 3.6, PPh2), 128.5
1
ppm, respectively For the assignment of the H and 13C{1H} NMR
spectra 2D NMR (COSY, HSQC, HMBC) experiments were also
performed. Coupling constants (J) are reported in hertz (Hz).
Multiplicities are indicated by br s (broad singlet), s (singlet), d
(doublet), t (triplet), q (quartet), and m (multiplet). Simulation NMR
spectra were performed by using gNMR 5.0. Air sensitive mass
spectrometry was performed by using a Bruker microTOF-Q (ESI)
spectrometer. Crystal structures were obtained from either a Rigaku
Oxford Diffraction Xcalibur (Mo Kα (λ = 0.71073 Å)) or Supernova
(Cu Kα (λ = 1.54184 Å)) diffractometer. FTIR-ATR analyses were
performed on a PerkinElmer Spectrum100. Solution analyses of air
sensitive samples were performed by using a Mettler Toledo ReactIR
and iC IR 7.0 with a Diamond DiComp probe. Elemental analyses were
performed by Elemental Microanalysis Ltd.
General Procedure for the Synthesis of 1,1-Diphosphinea-
mide Ligands (2). A Schlenk flask was charged with KHMDS (300
mg, 1.5 mmol) before adding MeCN (3 mL). To this suspension
HPPh2 (1.04 mL, 6 mmol) was added with stirring. A solution of the
propynamide (3 mmol) in MeCN (4 mL) was added dropwise over 10
min, and the solution was stirred for 16 h. The reaction was quenched
by addition of MeOH. Solvent was removed in vacuo, yielding the crude
product. The crude product was mounted on alumina and purified by
filtration under an inert atmosphere (5% CH2Cl2/hexane (removes
most HPPh2), 50% CH2Cl2/hexane).
Synthesis of 2a. Ligand 2a was synthesized by using the general
procedure with N-phenyl-2-propynamide (72.6 mg, 0.5 mmol). The
product was isolated as a white solid in 62% yield (161 mg, 0.31 mmol).
Data correspond to previous literature. 1H NMR (500 MHz, CDCl3): δ
7.64−7.63 (m, 4H, P(PhH)2), 7.52−7.50 (m, 4H, P(PhH)2), 7.32−
7.27 (m, 7H, P(PhH)2), 7.25−7.18 (m, 7H, P(PhH)2/NPhH), 7.10 (d,
J = 8.0, 2H, N-o-PhH), 7.01 (t, J = 7.4, 1H, N-p-PhH), 6.15 (br s, 1H,
NH), 4.16 (t, J = 5.7, 1H, CH(PPh2)2), 2.37 (td, J = 9.2, J = 5.7, 2H,
CH2CH(PPh2)2). 13C{1H} NMR (126 MHz, CDCl3): δ 169.4 (CO),
137.6 (N−Ph), 136.2 (app t, J = 4.3, PPh), 136.0 (app t, J = 3.2, PPh),
134.8 (app t, J = 11.0, PPhH), 133.7 (app t, J = 10.3, PPhH), 129.4
(PPhH) 128.9 (NPhH), 128.7 (PPhH), 128.5 (app t, J = 3.6, PPhH),
128.3 (app t, J = 3.9, PPhH), 124.1 (NPhH), 119.7 (NPhH), 37.8 (t, J =
8.6, CH2CH(PPh2)2), 27.2 (t, J = 22.9, CH(PPh2)2). 31P{1H} NMR
(202 MHz, CDCl3): δ −6.6. FTIR-ATR: ν (cm−1): 3319 (N−H), 1678
(CO). HRMS (API): calcd for C33H30NOP2+ [M + H]+: 518.1802;
found: 518.1805.
Synthesis of 2b. Ligand 2b was synthesized by using the general
procedure with N-dimethylphenylpropynamide (520 mg, 3 mmol).
The product was isolated as a white powder in 66% yield (1.08 g, 1.98
1
mmol). H NMR (500 MHz, C6D6): δ 7.84−7.78 (m, 4H, PPh2),
7.73−7.67 (m, 4H, PPh2), 7.08−6.95 (m, 12H, PPh2), 6.94 (app dd, J =
8.5, J = 6.3, 1H, p-ArH), 6.89 (d, J = 7.4, 2H, m-ArH), 5.08 (s, 1H, NH),
4.63 (t, J = 5.0, 1H, CH(PPh2)2), 2.49 (td, J = 9.9, J = 5.0, 2H,
CH2CH(PPh2)2), 1.83 (s, 6H, ArCH3). 13C{1H) NMR (126 MHz,
C6D6): δ 168.9 (t, J = 2.9, CO), 137.6 (app q, J = 5.6, PPh2), 135.8 (s, o-
ArMe), 135.0 (app t, J = 11.2, PPh2H), 134.9 (i-Ar), 134.3 (app t, J =
10.7, PPh2H), 129.1 (PPh2H), 129.0 (PPh2H), 128.7 (app t, J = 3.6,
PPh2H), 128.5 (app t, J = 3.7, PPh2H), 128.1 (s, m-ArH), 127.1 (s, p-
ArH), 36.0 (t, J = 10.1, CH2CH(PPh2)2), 26.1 (t, J = 24.5,
CH2CH(PPh2)2), 18.7 (s, ArMe). 31P{1H} NMR (202 MHz, C6D6):
δ −3.6; mp (°C) 143−146. HRMS (ESI−): calcd for C35H32NOP2− [M
− H]− 544.2037; found: 544.1972.
Synthesis of 2c. Ligand 2c was synthesized by using the general
procedure with N-diisopropylphenylpropynamide (520 mg, 3 mmol).
The product was isolated as a white powder in 68% yield (1.23 g, 2.05
mmol). 1H and 31P{1H} NMR show there are rotamers in solution (5:1,
2c:2c′). 1H NMR (500 MHz, CDCl3): 2c, δ 7.54−7.50 (m, 4H, PPh2),
7.43−7.40 (m, 4H, PPh2), 7.26−7.08 (m, 13H, p-ArH/PPh2), 7.01 (d, J
= 7.8, 2H, m-ArH), 6.04 (s, 1H, NH), 4.09 (t, J = 5.1, 1H, CH(PPh2)2),
2.55 (td, J = 9.9, J = 5.1, 2H, CH2CH(PPh2)2), 2.49 (sept, J = 6.9, 2H,
CH), 0.96 (d, J = 6.9, 12H, CH3); 2c′, δ 7.34−7.31 (m, 4H, PPh2),
7.26−7.08 (m, 17H, p-ArH/PPh2), 6.95 (d, J = 7.8, 2H, m-ArH), 6.26
152
https://dx.doi.org/10.1021/acs.organomet.0c00662
Organometallics 2021, 40, 148−155

Organometallics
pubs.acs.org/Organometallics
Article
(PPh2), 128.4 (app t, J = 3.2, 237 PPh2), 124.9 (p-ArH), 47.3 (NMe2),
34.7 (t, J = 11.9, CH2CH(PPh2)2), 25.7 (t, J = 32.4, CH2CH(PPh2)2),
18.8 (s, ArMe). i-Ar carbon not found. 31P{1H} NMR (202 MHz,
C6D6): δ −4.3. IR: ν (cm−1): 2060, 1818, 1478 (CO), 1035 (C−N).
D = 4.70 × 10−10 m2 s−1 (7.35 × 10−11 m2 s−1). Satisfactory elemental
analysis could not be obtained. Anal. Calcd for C74H76N4O2P4Ti: C,
72.54; H, 6.25; N, 4.57. Found; C, 69.65; H, 6.10; N, 4.23.
7.61 (d, J = 7.2, 8H, PPh2), 7.56 (br s, 4H, BArF), 7.50 (d, J = 7.0, 8H,
PPh2), 7.44−7.01 (m, 34H, PPh2/ArH), 6.33 (s, 2H, NH), 5.93 (t, J =
7.2, 2H, CH(PPh2)2), 2.09 (d, J = 7.2, 4H, CH2CH(PPh2)2). 11B{1H}
NMR (160 MHz, CD2Cl2): δ − 6.6 (m). 13C{1H} NMR (126 MHz,
CD2Cl2): δ 167.1 (m, CO), 162.3 (q, J = 50, ArF), 137.6 (iPh), 136.3
(m, PPh2), 135.4 (br s, ArF), 132.5 (m, PPh2), 132.3 (PPh2), 131.3
(PPh2), 131.1 (PPh2), 129.7 (PPh2H), 129.5 (s, m-NPh), 129.4 (qq, J =
32, 3, ArF), 129.2 (PPh2), 128.4 (m, PPh2), 125.2 (q, J = 272, ArF),
125.2 (s, p-NPh) 120 0.2 (s, o-NPh), 118.0 (sept J = 4, ArF), 52.3 (m,
CH2CH(PPh2)2), 38.6 (s, CH2CH(PPh2)2). 19F NMR (470 MHz,
CD2Cl2): δ −62.9. 31P{1H) NMR (202 MHz, CD2Cl2): δ −7.9 (d,
1JRh−P = 115.9). HRMS (ESI): calcd for C66H58N2O2P4Rh+ [M]+:
1137.2498; found: 1137.2427. Satisfactory elemental analysis could not
be obtained. Anal. Calcd for C98H70BF24N2O2P4Rh: C, 58.82; H, 3.53;
N, 1.40. Found; C, 57.69; H, 3.36; N, 1.42.
[Ti(NMe2)3(O,N-2b)] 3b′. The reaction was performed at 0.15
mmol; unwanted side products formed when the reaction was scaled up
further than this. A Schlenk flask was charged with 2b (81.8 mg, 0.15
mmol), and then C6D6 (0.6 mL) was added without stirring.
Ti(NMe2)4 (38.1 μL, 0.165 mmol) was added to the solution, which
turned light orange, and they were stirred slowly at RT for 1 h and then
analyzed in situ by NMR spectroscopy which showed a 3b:3b′ ratio of
1:6 (77% spectroscopic yield of 3b′). Attempts at separation and
isolation of 3b′ were unsuccessful because after recrystallization in n-
hexane or pentane only compound 3b was found in 37% (40 mg, 0.06
mmol), which shows potential ligand redistribution during work-up.
Nonetheless, VT temperature experiments do not show interconver-
sion between 3b and 3b′ in C6D6 between 298 and 348 K (see below),
as also supported by DOSY NMR which shows different diffusion
coefficient for 3b and 3b′. Complex 3b′ was analyzed in situ. 1H NMR
(300 MHz, C6D6): δ 7.68 (dtd, J = 7.2, 3.7, 2.2, 4H), 7.42−7.32 (m,
5H), 7.13−7.00 (m, 7H), 6.99−6.86 (m, 9H), 4.34 (t, J = 6.0, 1H,
CH2CH(PPh2)2), 4.13 (t, J = 6.9, 3b), 3.21 (s, 18H, NMe2), 3.11 (s,
5H), 2.47 (td, J = 11.3, 6.0, 3H, CH2CH(PPh2)2), 2.20 (d, J = 6.4, 8H,
HNMe2), 1.89 (s, 6H, Me). 13C{1H} NMR (75 MHz, C6D6): δ 182.3
(CO), 143.6 (3b), 143.2 (Ar), 137.5 (t, J = 5.7, PPh2), 137.4−137.1 (m,
PPh2), 134.4 (td, J = 10.8, 5.5, PPh2), 133.2 (3b), 132.6 (Ar), 129.0−
128.5 (m, Ar), 124.9, 45.6 (NMe2), 44.1, 39.0, 33.9 (t, J = 11.8,
CH2CH(PPh2)2), 25.4 (t, J = 29.1, CH2CH(PPh2)2), 18.8 (3b), 18.4
(Me). 31P{1H} NMR (122 MHz, C6D6): δ −2.9 (3b′), −3.5 (3b),
−39.8 (traces of HPPh2 from ligand synthesis); D = 5.11 × 10−10 m2 s−1
(9.82 × 10−11 m2 s−1).
[Rh(P,P-2a)(dppe)][BArF] 9a. A Schlenk flask was charged with
[Rh(dppe)Cl]2 (42.4 mg, 0.04 mmol), 2a (42 mg, 0.08 mmol), and
NaBArF (70.8 mg, 0.08 mmol). To this CH2Cl2 (3 mL) was added, and
the mixture was stirred at RT for 4 h. The solvent was removed,
resulting in 9a as a yellow powder in 71% yield (107 mg, 0.057 mmol).
Crystallization attempts were unsuccessful. 1H{31P} NMR (500 MHz,
CD2Cl2): δ 7.73 (br s, 8 H, BArF), 7.61 (m, 2H, PPh2H), 7.56 (br s, 4 H,
BArF), 7.46−7.04 (m, 40H, PPh2H/NPhH), 7.05 (d, J = 7.8, 2H, o-
NPhH), 7.03 (m, 1H, p-NPhH, assigned from HSQC), 6.38 (s, 1H,
NH), 5.63 (t, J = 7.1, 1H, CH(PPh2)2), 2.36−2.29 (m, 2H, CH2CH2),
2.25 (d, J = 7.1, 2H, CH2CH(PPh2)2), 2.19−2.13 (m, 2H, CH2CH2).
13C{1H) NMR (126 MHz, CD2Cl2): δ 166.9 (s, CO), 162.3 (q, J = 50,
ArF), 137.6 (s, i-NPh), 136.0 (m, PPh2H), 135.4 (br s, ArF), 133.9 (m,
PPh2H), 133.0 (m, PPh2H), 132.2 (s, PPh2H), 131.4 (m, PPh2H),
129.81−29.0 (m, ArF/m-NPh/PPh2H), 127.7 (PPh2), 125.2 (q, J =
272, ArF) 125.2 (s, p-NPh), 120.2 (s, o-NPh), 118.0 (sept J = 4, ArF),
52.3 (m, CH2CH(PPh2)2, assigned form HSQC), 38.6 (s, CH2CH-
(PPh2)2, assigned form HSQC), 28.3 (m, CH2PPh2, assigned from
HSQC). 31P{1H} NMR (202 MHz, CD2Cl2): δ 57.8 (2JP−P trans = 284,
[Ti(NMe2)3(O,N-2c)] 3c′. A Schlenk flask was charged with 2c (99.3
mg, 0.15 mmol), and then C6D6 (0.6 mL) was added without stirring.
Ti(NMe2)4 (38.1 μL, 0.165 mmol) was added to the solution; the
mixture was stirred slowly at RT for 1 h and then analyzed in situ by
NMR spectroscopy which showed a 3c:3c′ ratio of 1:8. Attempts at
synthesizing 3c with an excess of ligand, 2c:Ti ratio of 2:1 or 3:1, were
unsuccessful, with the major species being 3c′. Separation and isolation
of 3c′ and 3c were unsuccessful because of similar solubilities in organic
solvents of the two complexes; after washing in n-hexane and cannula
filtration, the mixture of complex (72 mg) was recovered as an orange
powder. DOSY NMR of the mixture supports that two different
complexes form in C6D6 solution with 3c′ having D = 6.24 × 10−10 m2
s−1 (1.41 × 10−11 m2 s−1) and 3c with D = 4.91 × 10−10 m2 s−1 (1.08 ×
10−11 m2 s−1). The mixture was analyzed in situ. 1H NMR (500 MHz,
C6D6): δ 7.59−7.53 (m, 5H), 7.52−7.46 (m, 4H), 7.19−7.11 (m, 3H),
7.11−7.02 (m, 7H), 7.01−6.90 (m, 16H), 4.41 (td, J = 5.6, 5.1, 2.8,
1H), 4.18 (t, J = 7.2, 0.25H), 3.30 (s, 2H), 3.21 (s, 19H), 3.11 (s, 1H),
2.75 (td, J = 13.2, 6.0, 2H), 1.33−1.25 (m, 3H), 1.15 (d, J = 6.8, 8H),
0.95 (d, J = 6.8, 6H). 13C{1H} NMR (75 MHz, C6D6): δ 183.4 (CO,
3c), 182.7 (CO, 3c′), 143.9 (Ar, 3c), 142.8 (Ar, 3c′), 140.6 (Ar, 3c),
140.3 (Ar, 3c′), 137.9−137.5 (m, PPh2), 134.6 (t, J = 11, PPh2, 3c′),
134.1 (t, J = 11, PPh2, 3c′), 128.8 (Ar), 128.6 (Ar), 128.5 (Ar), 128.4
(Ar), 126.1 (Ar, 3c), 126.0 (Ar, 3c′), 124.0 (Ar, 3c), 123.8 (Ar, 3c′),
47.9 (NMe2, 3c′), 45.8 (NMe2, 3c), 44.1 (iPr), 43.9 (iPr), 39.0
(HNMe2), 35.4 (t, J = 16.1, CH2CH(PPh2)2, 3c′), 28.3 (3c′), 27.8
(3c), 25.8 (t, J = 30.7, CH2CH(PPh2)2, 3c′), 24.8 (CH2CH(PPh2)2,
3c), 24.2 (iPr, 3c), 23.8 (iPr, 3c′). 31P{1H} NMR (202 MHz, C6D6): δ
−1.6 (3c′), −2.1 (3c).
1JRh‑dppe = 133, 2JP−P cis dppe
= 284, JRh−P = 116, JP−P cis2a
=
27, 2JP−P cis dppe‑2a = −28), −7.4 (2JP−P trans
1
2
2
=
69, JP−P cis dppe‑2a = −28). HRMS
(ESI): calcd C59H53NOP4Rh+ [M]+: 1018.2127; found: 1018.2059.
Satisfactory elemental analysis could not be obtained. Anal. Calcd for
C93H70BCl3F24NOP4Rh: C, 55.37; H, 3.50; N, 0.69. Found; C, 55.28;
H, 3.15; N, 0.81.
[Rh(P,P-2b)(dppe)][BArF] 9b. A J-Young NMR tube was charged
with [Rh(dppe)Cl]2 (10.6 mg, 0.01 mmol), 2b (10.9 mg, 0.02 mmol),
and NaBArF (17.9 mg, 0.02 mmol). To this fluorobenzene (0.5 mL) was
added, and the mixture was stirred at RT for 4 h. The solvent was
removed, resulting in a yellow powder. Compound 9b has been
1
characterized in situ. Crystallization attempts were unsuccessful. H-
{31P} NMR (500 MHz, CD2Cl2): δ 7.75 (br s, 8 H, BArF), 7.57 (br s, 4
H, BArF), 7.46−7.95 (m, 43H, PPh2H/NPhH), 6.33 (s, 1H, NH), 5.72
(t, J = 6.1, 1H, CH(PPh2)2), 2.45 (d, J = 6.1, 2H, CH2CH(PPh2)2),
2.40−2.34 (m, 2H, CH2CH2), 2.19−2.11 (m, 2H, CH2CH2), 1.81 (s,
6H, ArMe). 13C{1H) NMR (126 MHz, CD2Cl2): δ 167.6 (t, J = 5.1,
CO), 162.3 (q, J = 50, ArF), 136.4 (m, PPh2H), 136.0 (i-NPh), 135.4
(br s, ArF), 133.9 (m, PPh2H), 132.8 (m, PPh2H), 132.7 (m, PPh2H),
132.3 (s, PPh2H), 129.6−129.1 (m, ArF/m-NPh/PPh2H), 127.7
(PPh2), 125.2 (q, J = 272, ArF) 118.1 (sept, J = 4, ArF), 54.3
(CH2CH(PPh2)2, assigned from HSQC, under CD2Cl2), 36.7 (s,
CH2CH(PPh2)2), 29.0 (m, CH2PPh2), 18.5 (ArMe). Not all peaks
assigned are due to impurities. 31P{1H} NMR (202 MHz, CD2Cl2): δ
58.5 (2JP−P trans = 284, 1JRh‑dppe = 133, 2JP−P cis dppe
=
27, 2JP−P cis dppe‑2b
=
1
2
−28), − 3.1 (2JP−P trans = 284, JRh−P = 116, JP−P cis2b
=
69,
2JP−P cis dppe‑2b = −28). HRMS (ESI): calcd C61H59NOP4Rh+ [M +
nH]+: 1048.2602; found: 1048.2735.
[Rh(P,P-2a)2][BArF] 5a. A Schlenk flask was charged with [Rh-
(COE)2Cl]2 (28.7 mg, 0.04 mmol), 2a (84 mg, 0.16 mmol), and
NaBArF (70.8 mg, 0.08 mmol). To this CH2Cl2 (3 mL) was added, and
the mixture was stirred at RT for 4 h. The solvent was removed, and the
solid was washed with pentane (3 × 3 mL) and dried again. Compound
5a was isolated as an orange powder in 84% yield (134 mg, 0.067
mmol). Crystals suitable for XRD were grown from a saturated benzene
solution. 1H{31P} NMR (500 MHz, CD2Cl2): δ 7.73 (br s, 8H, BArF),
{(NMe2)Ti(μ:κ2N,O,κ2P,P-2b)Rh(dppe)][2BArF] 10a. In a J-Young
NMR tube, complex 3a was synthesized in situ by using the previously
described procedure from Ti(NMe2)4 (0.023 mmol) and 1a (0.046
mmol) in C6H5F (0.5 mL). To this [Rh(dppe)Cl]2 (24.7 mg, 0.023
F
mmol) and NaBAr (40.8 mg, 0.046 mmol) were added and mixed
vigorously. Full conversion into cationic [Rh(dppe)](η6-C6H5F)]-
[BArF] was observed by 31P{1H} NMR (d 73.5 ppm, J = 204).53 The
153
https://dx.doi.org/10.1021/acs.organomet.0c00662
Organometallics 2021, 40, 148−155

Organometallics
pubs.acs.org/Organometallics
Article
reaction was heated to 80 °C for 10 min. A subtle change in color was
observed from red to dark brown. Complex 10a was analyzed in situ,
and NMR were collected in C6H5F with a C6D6 capillary to lock/shim
the sample. The NMR spectrum shows decomposition to 9a. 31P{1H}
Cei B. Provis-Evans − Department of Chemistry, University of
Bath, Bath BA2 7AY, United Kingdom; orcid.org/0000-
0003-0309-3858
Mary F. Mahon − Department of Chemistry, University of Bath,
Bath BA2 7AY, United Kingdom
1
NMR (202 MHz, C6D6): δ 57.9, (2JP−P trans = 282, JRh‑dppe = 133,
2JP−P cis dppe
= =
29, 2JP−P cis dppe‑2a = −28), −4.6 (2JP−P trans = 282, 1JRh−P
113, 2JP−P cis2a
=
68, 2JP−P cis dppe‑2a = −28). 1H{31P} NMR (500 MHz,
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.organomet.0c00662
C6D6): δ 8.31 (s, 15H, BArF), 7.60 (s, 8H, BArF), 6.16 (s, 4H, o-NPhH),
5.49 (br, 2H), 2.97 (s, 12H, NMe2), 2.13−1.68 (m, 16H, (PPh2)2CH2-
CH2(PPh2)2/CH2CH). HRMS (ESI): calcd for fragment
C59H54NOP4Rh+ [M + H]+: 1019.2211; found: 1019.2303.
Author Contributions
N.T.C. and D.G. contributed to this work equally.
{(NMe2)Ti(μ:κ2N,O,κ2P,P-2a)Rh(dppe)][2BArF] 10b. A J-Young
NMR tube was charged with [Rh(dppe)Cl]2 (20 mg, 0.018 mmol) and
NaBArF (32 mg, 0.036 mmol) in C6H5F (0.5 mL) and was vigorously
mixed before leaving the reaction for 15 min. Full conversion into
cationic [Rh(dppe)](η6-C6H5F)][BArF] was observed by 31P{1H}
NMR (d 73.5 ppm, J = 204).53 3b (22.8 mg, 0.018 mmol) was added
to this solution, and the reaction was heated at 80 °C for 10 min; during
this time a subtle change in color was observed. Complex 10b was
filtered through a small glass filter plug to remove NaCl, and the
complex was analyzed without further purification. NMR spectra were
collected in C6H5F with a C6D6 capillary to lock/shim the sample. The
NMR spectrum shows decomposition to 9b, potentially due to
filtration. 31P{1H} NMR (162 MHz, C6D6): δ 58.2 (2JP−P trans = 282,
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The EPSRC is thanked for funding. The authors acknowledge
Dr. John Lowe for helping with NMR simulation as well as Dr.
Shaun Reeksting, Dr. Catherine Lyall, and The University of
Bath DReaM facility for assistance with HRMS analyses.
REFERENCES
1JRh‑dppe = 133, 2JP−P cis dppe
= 282, JRh−P = 113, JP−P cis2b
=
29, 2JP−P cis dppe‑2b = −28), −3.2 (2JP−P trans
2 1
■
1
2
68, JP−P cis dppe‑2b = −28). H{31P}
(1) Blondin, G.; Girerd, J. J. Interplay of electron exchange and
electron transfer in metal polynuclear complexes in proteins or
chemical models. Chem. Rev. 1990, 90, 1359−1376.
(2) Karlin, K. D. Bioinorganic Chemistry of Copper; Springer:
Dordrecht, 1993.
=
NMR (500 MHz, C6D6): δ 8.31 (BArF), 7.59 (BArF), 5.22 (br,
CH(PPh2)2), 2.94 (NMe2), 2.86 (NMe2), 2.27−2.11 ((PPh2)2CH2-
CH2(PPh2)2/CH2CH/Ar(CH3)2). HSQC analysis on the complex
confirm several chemical environments for the NMe2 coupling with the
same carbon 13C{1H} NMR (126 MHz, C6D6): δ 46.1. 1H NMR (500
MHz, C6D6): δ 2.92, 2.84. This highlights the existence of multiple
isomers in solution at RT in the NMR time scale. D = 3.41 × 10−10 m2
s−1 (3.597 × 10−11). HRMS (ESI): calcd C126H124N4O2P8Rh2Ti+
[M]+2:1113.7624; found: 1113.7693.
(3) Borovik, A. S.; Papaefthymiou, V.; Taylor, L. F.; Anderson, O. P.;
Que, L. Models for iron-oxo proteins. Structures and properties of
FeIIFeIII, ZnIIFeIII, and FeIIGaIII complexes with (μ-phenoxo)bis(μ-
carboxylato)dimetal cores. J. Am. Chem. Soc. 1989, 111, 6183−6195.
(4) Lippard, S. J. Oxo-Bridged Polyiron Centers in Biology and
Chemistry. Angew. Chem., Int. Ed. Engl. 1988, 27, 344−361.
̈
(5) Lipscomb, W. N.; Strater, N. Recent Advances in Zinc
ASSOCIATED CONTENT
■
Enzymology. Chem. Rev. 1996, 96, 2375−2434.
sı
* Supporting Information
(6) Pelletier, F.; Comte, V.; Massard, A.; Wenzel, M.; Toulot, S.;
Richard, P.; Picquet, M.; Le Gendre, P.; Zava, O.; Edafe, F.; Casini, A.;
Dyson, P. J. Development of Bimetallic Titanocene−Ruthenium−
Arene Complexes As Anticancer Agents: Relationships between
Structural and Biological Properties. J. Med. Chem. 2010, 53, 6923−
6933.
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.organomet.0c00662.
Full characterization of the metal complexes and ligands
(PDF)
(7) Xue, F.; Xie, C.-Z.; Zhang, Y.-W.; Qiao, Z.; Qiao, X.; Xu, J.-Y.; Yan,
S.-P. Two new dicopper(II) complexes with oxamido-bridged ligand:
Synthesis, crystal structures, DNA binding/cleavage and BSA binding
activity. J. Inorg. Biochem. 2012, 115, 78−86.
Cartesian coordinates for modeled geometric isomer
(XYZ)
Accession Codes
(8) Matsunaga, S.; Shibasaki, M. Recent advances in cooperative
bimetallic asymmetric catalysis: dinuclear Schiff base complexes. Chem.
Commun. 2014, 50, 1044−1057.
CCDC 2035747−2035748 contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge via www.ccdc.cam.ac.uk/data_request/cif, or by email-
ing data_request@ccdc.cam.ac.uk, or by contacting The Cam-
bridge Crystallographic Data Centre, 12 Union Road, Cam-
bridge CB2 1EZ, UK; fax: +44 1223 336033.
(9) Stephan, D. W. Early-late heterobimetallics. Coord. Chem. Rev.
1989, 95, 41−107.
(10) Shibasaki, M.; Yamamoto, Y. Multimetallic Catalysts in Organic
Synthesis; 2004.
(11) Campos, J. Bimetallic cooperation across the periodic table. Nat.
Rev. Chem. 2020, 4, 696−702.
AUTHOR INFORMATION
■
(12) Duncan, J.; Malinski, T.; Zhu, T. P.; Hu, Z. S.; Kadish, K. M.;
Bear, J. L. Characterization of novel rhodium(II) dimers with N-
phenylacetamido bridging ligands. J. Am. Chem. Soc. 1982, 104, 5507−
5509.
(13) Huang, B.-H.; Yu, T.-L.; Huang, Y.-L.; Ko, B.-T.; Lin, C.-C.
Reactions of Amides with Organoaluminum Compounds: Factors
Affecting the Coordination Mode of Aluminum Amidates. Inorg. Chem.
2002, 41, 2987−2994.
(14) Li, C.; Thomson, R. K.; Gillon, B.; Patrick, B. O.; Schafer, L. L.
Amidate complexes of titanium and zirconium: a new class of tunable
precatalysts for the hydroamination of alkynes. Chem. Commun. 2003,
2462−2463.
Corresponding Author
Ruth L. Webster − Department of Chemistry, University of
Bath, Bath BA2 7AY, United Kingdom; orcid.org/0000-
0001-9199-7579; Email: rw498@bath.ac.uk
Authors
Nathan T. Coles − Department of Chemistry, University of
Bath, Bath BA2 7AY, United Kingdom; orcid.org/0000-
0001-6190-9080
Danila Gasperini − Department of Chemistry, University of
Bath, Bath BA2 7AY, United Kingdom
154
https://dx.doi.org/10.1021/acs.organomet.0c00662
Organometallics 2021, 40, 148−155

Organometallics
pubs.acs.org/Organometallics
Article
(15) Zhang, Z.; Schafer, L. L. Anti-Markovnikov Intermolecular
Hydroamination: A Bis(amidate) Titanium Precatalyst for the
Preparation of Reactive Aldimines. Org. Lett. 2003, 5, 4733−4736.
(16) Giesbrecht, G. R.; Shafir, A.; Arnold, J. Synthesis and Structure of
a Linked-Bis(amidate) Ligand and Some Complexes with Titanium.
Inorg. Chem. 2001, 40, 6069−6072.
(34) Ojeda, M.; Fandos, R.; Fierro, J. L. G.; Otero, A.; Pastor, C.;
Rodríguez, A.; Ruiz, M. J.; Terreros, P. Synthesis and reactivity of
heterometallic RhO−M (M = Si, Ti) complexes: Memory effect in their
catalytic performance in CO hydrogenation. J. Mol. Catal. A: Chem.
2006, 247, 44−51.
(35) Coles, N. T.; Mahon, M. F.; Webster, R. L. 1,1-Diphosphines and
divinylphosphines via base catalyzed hydrophosphination. Chem.
Commun. 2018, 54, 10443−10446.
(17) Ryken, S. A.; Schafer, L. L. N,O-Chelating Four-Membered
Metallacyclic Titanium(IV) Complexes for Atom-Economic Catalytic
Reactions. Acc. Chem. Res. 2015, 48, 2576−2586.
(36) Perry, M. R.; Gilmour, D. J.; Schafer, L. L. Mono, bis, and
tris(phosphoramidate) titanium complexes: synthesis, structure, and
reactivity investigations. Dalton Trans. 2019, 48, 9782−9790.
(37) Elkin, T.; Kulkarni, N. V.; Tumanskii, B.; Botoshansky, M.;
Shimon, L. J. W.; Eisen, M. S. Synthesis and Structure of Group 4
Symmetric Amidinate Complexes and Their Reactivity in the
Polymerization of α-Olefins. Organometallics 2013, 32, 6337−6352.
(38) See the Supporting Information for further information.
(39) Thomson, R. K.; Zahariev, F. E.; Zhang, Z.; Patrick, B. O.; Wang,
Y. A.; Schafer, L. L. Structure, Bonding, and Reactivity of Ti and Zr
Amidate Complexes: DFT and X-Ray Crystallographic Studies. Inorg.
Chem. 2005, 44, 8680−8689.
(18) Mansell, S. M. Catalytic applications of small bite-angle
diphosphorus ligands with single-atom linkers. Dalton Trans. 2017,
46, 15157−15174.
(19) Lorenzini, F.; Hindle, K. T.; Craythorne, S. J.; Crozier, A. R.;
Marchetti, F.; Martin, C. J.; Marr, P. C.; Marr, A. C.
[Rh2(COD)2(Dppm)(μ2-Cl)]BF4: Precursor for a Selective Hydro-
genation Catalyst and Its Recycling by Silica Entrapment. Organo-
metallics 2006, 25, 3912−3919.
(20) Chaplin, A. B.; Hooper, J. F.; Weller, A. S.; Willis, M. C.
Intermolecular Hydroacylation: High Activity Rhodium Catalysts
Containing Small-Bite-Angle Diphosphine Ligands. J. Am. Chem. Soc.
2012, 134, 4885−4897.
(21) Dowson, G. R. M.; Haddow, M. F.; Lee, J.; Wingad, R. L.; Wass,
D. F. Catalytic Conversion of Ethanol into an Advanced Biofuel:
Unprecedented Selectivity for n-Butanol. Angew. Chem., Int. Ed. 2013,
52, 9005−9008.
(22) Cooper, B. G.; Napoline, J. W.; Thomas, C. M. Catalytic
Applications of Early/Late Heterobimetallic Complexes. Catal. Rev.:
Sci. Eng. 2012, 54, 1−40.
(23) Gade, L. H. Highly Polar Metal−Metal Bonds in “Early−Late”
Heterodimetallic Complexes. Angew. Chem., Int. Ed. 2000, 39, 2658−
2678.
(40) Li, Z.; Yokley, T. W.; Tran, S. L.; Zong, J.; Schley, N. D.;
Brewster, T. P. Synthesis and characterization of rhodium−aluminum
heterobimetallic complexes tethered by a 1,3-bis(diphenylphosphino)-
2-propanoxy group. Dalton Trans. 2019, 48, 8782−8790.
̈
(41) Dorfler, J.; Preuß, T.; Brahms, C.; Scheuer, D.; Doye, S.
Intermolecular hydroaminoalkylation of alkenes and dienes using a
titanium mono(formamidinate) catalyst. Dalton Trans. 2015, 44,
12149−12168.
(42) Zhang, Z.; Leitch, D. C.; Lu, M.; Patrick, B. O.; Schafer, L. L. An
Easy-To-Use, Regioselective, and Robust Bis(amidate) Titanium
Hydroamination Precatalyst: Mechanistic and Synthetic Investigations
toward the Preparation of Tetrahydroisoquinolines and Benzoquino-
lizine Alkaloids. Chem. - Eur. J. 2007, 13, 2012−2022.
(43) See the Supporting Information for crystal data for complex 5a
(CCDC 2035747).
(24) Bodio, E.; Picquet, M.; P, L. G. “Early−Late” Heterobimetallic
Catalysis and Beyond. In Homo- and Heterobimetallic Complexes in
Catalysis. Topics in Organometallic Chemistry; Springer: Cham, 2015;
Vol. 59.
̈
̈
(44) Fornika, R.; Six, C.; Gorls, H.; Kessler, M.; Kruger, C.; Leitner,
W. Synthesis and structural features of complexes [(P∩P)2Rh][hfacac]
containing hexafluoroacetyl- acetonate as a noncoordinating anion.
Can. J. Chem. 2001, 79, 642−648.
́
(25) Fandos, R.; Hernandez, C.; Otero, A.; Rodríguez, A.; Ruiz, M. J.;
Terreros, P. Early−Late Heterobimetallic Alkoxides as Model Systems
for Late-Transition-Metal Catalysts Supported on Titania. Chem. - Eur.
J. 2003, 9, 671−677.
(45) The sign of the coupling did not affect the simulated spectra.
(46) Bertrand, R. D.; Ogilvie, F. B.; Verkade, J. G. Signs of
phosphorus-phosphorus coupling constants in coordination com-
pounds. J. Am. Chem. Soc. 1970, 92, 1908−1915.
(47) Newland, R. J.; Lynam, J. M.; Mansell, S. M. Small bite-angle 2-
phosphinophosphinine ligands enable rhodium-catalysed hydrobora-
tion of carbonyls. Chem. Commun. 2018, 54, 5482−5485.
(48) Several attempts were made using Wilkinson’s catalyst to isolate
both monometallic Rh(I) species and heterobimetallic Rh-Ti species;
see the Supporting Information for further information and clear NMR
analysis.
(49) Attempts at synthesising Ti(IV) in situ and then reacting it with
Rh precursors were unsuccesful; see the Supporting Information.
(50) This decomposition was confirmed for the PPh3 analogue by
NMR and mass spectrometry analysis.
(51) See the Supporting Information for crystal data collected for a Ti-
oxo cluster containing 2b (CCDC 2035748).
(52) Esarte Palomero, O.; Jones, R. A. Accessing Pentagonal
Bipyramidal Geometry with Pentadentate Pincer Amido-bis(amidate)
Ligands in Group IV and V Early Transition Metal Complexes.
Organometallics 2020, 39, 3689−3694.
(53) Dallanegra, R.; Robertson, A. P. M.; Chaplin, A. B.; Manners, I.;
Weller, A. S. Tuning the [L2Rh···H3B·NR3]+ interaction using
phosphine bite angle. Demonstration by the catalytic formation of
polyaminoboranes. Chem. Commun. 2011, 47, 3763−3765.
(54) Refer to the Supporting Information for further attempts to react
monoligated Ti complexes with Rh precursors.
(26) Kuwata, S.; Kabashima, S.-i.; Sugiyama, N.; Ishii, Y.; Hidai, M.
Synthesis of TiRu2 Heterobimetallic and TiRuM (M = Rh, Ir, Pd, Pt)
Heterotrimetallic Sulfido Clusters from a Hydrosulfido-Bridged
Titanium−Ruthenium Complex. Inorg. Chem. 2001, 40, 2034−2040.
(27) Slaughter, L. M.; Wolczanski, P. T. Ti(μ:η1,η1-OC-
Me2CH2PPh2)3Rh has a cylindrically symmetric triple bond. Chem.
Commun. 1997, 2109−2110.
́
(28) Fandos, R.; Otero, A.; Rodríguez, A.; Terreros, P.; Aullon, G.;
Alvarez, S. A New Titanium Alkoxide−Thiolate Complex as a Versatile
́
Heterofunctional Metalloligand. Eur. J. Inorg. Chem. 2009, 2009, 1079−
1085.
́
(29) Fandos, R.; Bruna, S.; Hernandez, C.; Otero, A.; Rodríguez, A.;
̃
Ruiz, M. J.; Terreros, P.; Cuadrado, I. Titanium thiosalicylate
complexes: functional metalloligands for the construction of redox-
active heterometallic architectures. Dalton Trans. 2018, 47, 15391−
15398.
́
(30) Casado, M. A.; Ciriano, M. A.; Lahoz, F. J.; Oro, L. A.; Perez-
Torrente, J. J. Reactivity of [TiM2] (M = Rh, Ir) and [TiIr3] early-late
heterobimetallic sulfido-bridged clusters. J. Organomet. Chem. 2016,
812, 123−134.
(31) Fandos, R.; Gallego, B.; Otero, A.; Rodríguez, A.; Ruiz, M. J.;
Terreros, P.; Pastor, C. A new titanium building block for early−late
heterometallic complexes; preparation of a new tetrameric metal-
lomacrocycle by self assembly. Dalton Trans. 2006, 2683−2690.
(32) Comte, V.; Le Gendre, P.; Richard, P.; Moïse, C. [TiPHOS-
(Rh)]+: A Fortuitous Coordination Mode and an Effective Hydro-
silylation Bimetallic Catalyst. Organometallics 2005, 24, 1439−1444.
(33) Morgan, J. P.; Kundu, K.; Doyle, M. P. A readily prepared neutral
heterobimetallic titanium(IV)−rhodium(I) catalyst for intramolecular
hydroacylation. Chem. Commun. 2005, 3307−3309.
155
https://dx.doi.org/10.1021/acs.organomet.0c00662
Organometallics 2021, 40, 148−155