Catalytic intramolecular hydroamination of aminoallenes using titanium and tantalum complexes of sterically encumbered chiral sulfonamides

Article abstract of DOI:10.1039/d0dt02557g

Catalysis using earth abundant metals is an important goal due to the relative scarcity and expense of precious metal catalysts. It would be even more beneficial to use earth abundant catalysts for the synthesis of common pharmaceutical structural motifs such as pyrrolidine and pyridine. Thus, developing titanium catalysts for asymmetric ring closing hydroamination is a valuable goal. In this work, four sterically encumbered chiral sulfonamides derived from naturally occurring amino acids were prepared. These compounds undergo protonolysis reactions with Ti(NMe2)4 or Ta(NMe2)5 to give monomeric complexes as determined by both DOSY NMR and X-ray crystallography. The resulting complexes are active for the ring closing hydroamination hepta-4,5-dienylamine to give a mixture of tetrahydropyridine and pyrrolidine products. However, the titanium complexes convert 6-methylhepta-4,5-dienylamine exclusively to 2-(2-methylpropenyl)pyrrolidine in higher enantioselectivity than those previously reported, with enantiomeric excesses ranging from 18-24%. The corresponding tantalum complexes were more selective with enantiomeric excesses ranging from 33-39%.

Full text of DOI:10.1039/d0dt02557g

View Article Online
View Journal
Dalton
Transactions
An international journal of inorganic chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: F. Sha, E. A.
Shimizu, H. S. Slocumb, S. E. Towell, Y. Zhen, H. Z. Porter, M. Takase and A. Johnson, Dalton Trans., 2020,
DOI: 10.1039/D0DT02557G.
Volume 46
Number 10
14 March 2017
Pages 3073-3412
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Dalton
Transactions
An international journal of inorganic chemistry
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
rsc.li/dalton
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
ISSN 0306-0012
COMMUNICATION
Douglas W. Stephen et al.
shall the Royal Society of Chemistry be held responsible for any errors
N-Heterocyclic carbene stabilized parent sulfenyl, selenenyl,
and tellurenyl cations (XH+, X = S, Se, Te)
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
rsc.li/dalton

Page 1 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/D0DT02557G
Catalytic intramolecular hydroamination of aminoallenes using titanium and tantalum
complexes of sterically encumbered chiral sulfonamides
Fanrui Sha†, Emily A. Shimizu†, Hannah S. Slocumb†, Sydney E. Towell†, Yi Zhen†, Hanna Z.
Porter†, Michael K. Takase‡, and Adam R. Johnson†,*
†Harvey Mudd College, 301 Platt Blvd., Claremont, CA 91711
‡Beckman Institute, California Institute of Technology, Pasadena, CA 91125
Abstract
Catalysis using earth abundant metals is an important goal due to the relative scarcity
and expense of precious metal catalysts. It would be even more beneficial to use earth
abundant catalysts for the synthesis of common pharmaceutical structural motifs such as
pyrrolidine and pyridine. Thus, developing titanium catalysts for asymmetric ring closing
hydroamination is a valuable goal. In this work, four sterically encumbered chiral sulfonamides
derived from naturally occurring amino acids were prepared. These compounds undergo
protonolysis reactions with Ti(NMe2)4 or Ta(NMe2)5 to give monomeric complexes as
determined by both DOSY NMR and X-ray crystallography. The resulting complexes are active
for the ring closing hydroamination hepta-4,5-dienylamine to give a mixture of
tetrahydropyridine and pyrrolidine products. However, the titanium complexes convert 6-
methylhepta-4,5-dienylamine exclusively to 2-(2-methylpropenyl)pyrrolidine in higher
enantioselectivity than those previously reported, with enantiomeric excesses ranging from 18-
Page 1

Dalton Transactions
Page 2 of 39
View Article Online
DOI: 10.1039/D0DT02557G
24%. The corresponding tantalum complexes were more selective with enantiomeric excesses
ranging from 33-39%.
Introduction
Hydroamination, the direct addition of an N-H bond across an unsaturated carbon-
carbon bond, is an atom-economical transformation that can result in both piperidine and
pyrrolidine heterocycles that are commonly found in US FDA approved pharmaceuticals.1
Metals from across the periodic table have been investigated as catalysts for this
transformation,2-8 including late metals9-11 and lanthanides.12-13 The development of earth-
abundant catalysts has lately become an important goal.14-16 Many groups are developing
catalysts based on group-IV and -V metals for both hydroamination and the related
hydroaminoalkylation reactions.17-42 The reaction has been under intense study, but general
procedures for hydroamination remain to be realized.
Gold catalysts have been shown to be capable of highly enantioselective
hydroamination of aminoallenes to give pyrrolidines, but the substrates must be N-tosylated.43
Ideally, this reaction would be achieved using less expensive catalysts and without the necessity
of protecting groups. Thus, our goal has been to develop a low-cost, earth-abundant catalyst
for the transformation of the unprotected substrates. Our group has studied the intramolecular
hydroamination of both di- and trisubstituted aminoallene substrates (Scheme 1, compounds
1a and 1b) using the early transition metals titanium and tantalum.44-46 We have found that
substrate 1a can give a mixture of tetrahydropyridine 2a and pyrrolidine 3a products, while
substrate 1b gives only the pyrrolidine product 3b.
Page 2

Page 3 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/D0DT02557G
H
N
NH2
R
N
+
R
·
1a, R = H
2a
3a, R = H
1b, R = CH3
3b, R = CH3
Scheme 1 Catalytic hydroamination of di- and trisubstituted aminoallenes (1) gives achiral
tetrahydropyridine (2a) or chiral -vinylpyrrolidines (3a and 3b).
We have observed that tantalum catalysts perform better than their corresponding
titanium catalysts in this hydroamination reaction. Tantalum derived catalysts have achieved
enantioselectivities up to 80% ee,45 although the highest enantioselectivity we have observed
for a titanium derived catalyst is below 20% ee.46-47 We have obtained several crystal structures
of titanium complexes with bidentate amide-alkoxide ligands, and all of them are dimeric with
bridging oxygen atoms.48-49 The crystal structures of tantalum complexes we have obtained
have all been monomeric.45, 50 Interestingly, the highest enantioselectivity we have observed
with a titanium catalyst was with a ligand that contained an additional donor atom resulting in
a monomeric species.46 This result led us to postulate that if the dimeric nature of the titanium
complexes was maintained in solution, a possible explanation for the reduced selectivity for the
titanium complexes relative to their tantalum counterparts is their dimeric nature.
Our group has focused on using “chiral-pool”51 derived ligands based principally on the
naturally occurring amino acids phenylalanine and valine. We published a series of sulfonamide
ligands in 2011.52 The ligands in that publication had only H and Me substituents  to the
oxygen, while ligands containing Ph substituents at that position have generally been more
selective.47 Sulfonamides are known to sometimes coordinate to titanium centers through a
sulfonamide oxygen,53-60 and this additional coordination site could lead to different reactivity,
Page 3

Dalton Transactions
Page 4 of 39
View Article Online
DOI: 10.1039/D0DT02557G
especially with the larger phenyl substituent that can project the ligand chirality more
efficiently. With a sterically more encumbered sulfonamide ligand, we hoped to observe two
results: first, the isolation of monomeric titanium complexes, and second, increased
stereoselectivity for the hydroamination of aminoallene substrates.
Results and Discussion
Synthesis of ligands
The synthesis of the ligands followed procedures similar to those reported previously.52,
54-58 Treatment of the phenylalanine-derived amino alcohol (S)-2-amino-1,1,3-triphenylpropanol
with a substituted aryl sulfonyl chloride gave the desired sulfonamides in good yield.
Spectroscopic data for sulfonamide L-H2Ph1 matched the literature.61 All compounds were
purified either by recrystallization (H2Ph1 and H2Ph2) or by chromatography (H2Ph3 and
H2Ph4). The D-enantiomers of H2Ph2 and H2Ph3 were also prepared to verify enantiopurity of
the sulfonamides. Chiral contact shift NMR spectroscopy showed no evidence of racemization
during their synthesis. Plots of the contact shift NMR studies are presented in the ESI.
Page 4

Page 5 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/D0DT02557G
Ph
Ph
Ph
Ph
O
S
O
NH
OH
F3C
S
NH
OH
O
O
a) NEt3, cat. DMAP
RAr
O
S
L-H2Ph1
L-H2Ph2
Ph
Ph
0 °C, CH2Cl2,
+
Cl
H2N
OH
b) HCl/NaHCO3
O
Ph
Ph
Ph
Ph
F3C
F3C
O
O
S
NH
OH
S
NH
OH
O
O
L-H2Ph3
L-H2Ph4
Scheme 2 Synthesis of sulfonamides used in this study.
Attempts were made to prepare the even bulkier 2,4,6-tri-isopropyl sulfonamide. Two
products, both with similar NMR chemical shifts were observed. Although the two products
could be separated by column chromatography, the yield of the desired sulfonamide was low.
IR spectroscopy suggested that the major product was the corresponding sulfonate ester as it
contained S-O stretches but no S-N stretch near 950 cm-1.62 Additional attempts at increasing
bulk at the 2,6-positions of the aromatic ring were thus abandoned. IR stretches consistent with
the sulfonamide group were observed in all four compounds.63
XRD of Ph1
A single crystal of L-H2Ph1 was grown by slow evaporation from ethanol. The molecular
structure is shown in Figure 1, and crystallographic parameters are listed in the ESI (Table S1).
The molecule crystallizes with two molecules in the asymmetric unit that are essentially
isostructural; both molecules are shown in the ESI (Figure S1). The crystal structures of amino
alcohol ligands we have obtained previously often contain intramolecular hydrogen bonds,64
even with diphenyl substitution on the tertiary alcohol carbon.65 However, this molecule has no
Page 5

Dalton Transactions
Page 6 of 39
View Article Online
DOI: 10.1039/D0DT02557G
intramolecular hydrogen bonds but has several intermolecular ones. There are two short
contacts linking N(1)H(1A)···O(5), a sulfonyl oxygen, at 2.088 Å and O(4)H(4)···O(3) at 2.021 Å.
There are two longer contacts, N(2)H(2A)···O(1) at 2.280 Å and O(1)H(1)···O(4) at 2.281 Å. These
contacts form a zig-zag structure in the ab plane. There are no other short contacts observed. A
view of the hydrogen bonding network is shown in the ESI (Figure S2).
Figure 1. Thermal ellipsoid drawing of L-H2Ph1 (all hydrogen atoms except H(1A) and H(1)
omitted for clarity; ellipsoids shown at 50% probability).
Synthesis of complexes
Titanium complexes of the L-enantiomers of Ph1, Ph2, and Ph3 were prepared by adding
a chilled ether solution of the ligand to a chilled ether solution of Ti(NMe2)4 (Scheme 3). The
titanium complexes appeared red in solution but were isolated as yellow solids that
precipitated from the reaction solution in moderate yields due to their high solubility in ether.
The complexes could be isolated essentially quantitatively in impure form by removing the
solvent and subsequently purified by recrystallization from toluene. All three complexes were
isolated as the dimethylamine adducts.
Page 6

Page 7 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/D0DT02557G
Scheme 3 Synthesis of titanium sulfonamide complexes; TiPh1, RAr = 4-CH3, TiPh2, RAr = 4-CF3,
TiPh3, RAr = 3,5-di-CF3.
Since H2Ph1 is relatively insoluble in ether, the synthesis of complex TiPh1 was also
attempted in dichloromethane, but this resulted in several unknown side products. Complex
TiPh2 shows coordinated dimethylamine in both the 1H and 13C NMR spectra, but the elemental
analysis results are consistent with loss of the coordinated amine. Attempts to heat solutions of
TiPh2 and evacuate the headspace to drive off the amine were not successful as determined by
1H NMR spectroscopy. Complex TiPh3 tenaciously retains toluene after recrystallization, and
the 1H, 13C NMR spectra, elemental analysis results, and X-ray crystal structure (vide infra) are
all consistent with a toluene of solvation. Attempts to remove the toluene under vacuum were
unsuccessful.
The titanium complex of Ph4 could not be prepared by simply stirring overnight in ether.
A series of NMR tube scale reactions showed that it took approximately three days at 65 °C for
complete conversion to the desired product. It is possible that the steric bulk of the 2,6-
dimethyl substituents prevents ready complexation. For preparative scale synthesis, the
reaction was carried out in toluene in a Teflon sealed glass reaction vessel. Complex TiPh4 was
isolated as an orange-brown foam in 95% yield and was characterized by 1H and 13C NMR
Page 7

Dalton Transactions
Page 8 of 39
View Article Online
DOI: 10.1039/D0DT02557G
spectroscopy, but attempts to crystallize the material have thus far been unsuccessful (Scheme
4). Complex TiPh4 did not retain dimethylamine.
Scheme 4 Synthesis of TiPh4.
All four titanium complexes exhibit two inequivalent NMe2 resonances at around 2.8
and 3.2 ppm, and TiPh1, TiPh2, and TiPh3 also have an additional resonance for the
coordinated HNMe2 ligand as a doublet at around 1.8 ppm in their 1H NMR spectra. TiPh2 and
TiPh3 exhibit sharp singlets in their 19F NMR spectra for the CF3 groups. The spectroscopic data
are consistent with the formulation of the complexes as Ti(Ph1-3)(NMe2)2(HNMe2) and
Ti(Ph4)(NMe2)2.
The synthesis of the tantalum complexes was straightforward and followed a similar
procedure to that of their titanium analogs. Addition of a chilled ether solution of the ligand to
a chilled ether solution of Ta(NMe2)5 resulted in essentially quantitative yield of the desired
complexes as white to off-white solids after an overnight reaction (Scheme 5). All four
complexes could be purified by recrystallization from ether in low yields.
Page 8

Page 9 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/D0DT02557G
Scheme 5 Synthesis of tantalum sulfonamide complexes; TaPh1, RAr = 4-CH3, TaPh2, RAr = 4-CF3,
TaPh3, RAr = 3,5-di-CF3, TaPh4, RAr = 2,4,6-tri-CH3.
The tantalum complexes all exhibit sharp singlets for the NMe2 groups at around 3.2
ppm in their 1H NMR spectra. TaPh2 and TaPh3 exhibit sharp singlets in their 19F NMR spectra.
All spectroscopic data are consistent with their formulation as Ta(Ph1-4)(NMe2)3. The elemental
analysis results for TaPh3 suggest that two dimethylamide ligands were replaced by an oxo
group. The elemental analysis results for TaPh4 are consistent with one dimethylamide ligand
being replace by a hydroxide. These results are presumably due to sample handling as the NMR
spectra are consistent with the expected formulation.
X-Ray structures of complexes
X-ray quality crystals of TiPh3, TaPh1, TaPh2 and TaPh3 were obtained from a chilled
ether solution of the complex. Suitable crystals were mounted at 100 K and their structures
were determined. Crystallographic parameters are listed in the ESI (Table S1). The geometry
around the titanium center is best described as a distorted trigonal bipyramid with Rc(x)
parameter of 9.77 for trigonal bipyramidal and 21.44 for square pyramidal.66 The  parameter is
0.53,67 or 53% of the way from perfectly tetragonal to perfectly trigonal. The TBP axis is
between the ligand nitrogen N(1) and the NHMe2 nitrogen N(2). There is one short contact
(3.085 Å) from the hydrogen atom on the dimethylamine to O(2) on adjacent molecule. TiPh3
Page 9

Dalton Transactions
Page 10 of 39
View Article Online
DOI: 10.1039/D0DT02557G
crystallizes with one molecule of toluene in the asymmetric unit, consistent with the
spectroscopic and analysis data. The CF3 groups were modeled as two component disorders. A
complete ORTEP with disordered CF3 groups and the toluene of crystallization is shown in the
ESI (Figure S3).
Figure 2. Thermal ellipsoid drawing of complex TiPh3 (toluene molecule and all hydrogen atoms
except H(2N) are omitted for clarity; ellipsoids shown at 50% probability). Selected bond
distances (Å) and angles (°): Ti(1)-N(4) 1.877(2), Ti(1)-O(1) 1.8866(18), Ti(1)-N(3) 1.908(2), Ti(1)-
N(1) 2.115(2), Ti(1)-N(2) 2.224(2), N(4)-Ti(1)-O(1) 120.39(10), N(4)-Ti(1)-N(3) 115.90(11), O(1)-
Ti(1)-N(3) 123.36(9), N(4)-Ti(1)-N(1) 101.47(9), O(1)-Ti(1)-N(1) 75.91(8), N(3)-Ti(1)-N(1) 99.22(9),
N(4)-Ti(1)-N(2) 94.95(10), O(1)-Ti(1)-N(2) 79.82(8), N(3)-Ti(1)-N(2) 89.95(10), N(1)-Ti(1)-N(2)
155.24(9).
All three tantalum complexes are essentially isostructural with only minor variations in
bond angles around the tantalum and orientation of the dimethylamide groups. The structures
of TaPh1 and TaPh2 are reported in the ESI (Figures S4 and S5). Like TiPh3, the geometry
around the tantalum center in TaPh3 is also best described as a distorted trigonal bipyramid. It
has an Rc(x) parameter of 11.04 for trigonal bipyramidal and 17.88 for square pyramidal,66 with
Page 10

Page 11 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/D0DT02557G
a  parameter of 0.35.67 The TBP axis is between the ligand nitrogen N(1) and the NMe2
nitrogen N(2) (Figure 3).
Figure 3. Thermal ellipsoid drawing of complex TaPh3 (hydrogen atoms are omitted for clarity;
ellipsoids shown at 50% probability). Selected bond distances (Å) and angles (°): Ta(1)-O(1)
1.948(3), Ta(1)-N(3) 1.958(3), Ta(1)-N(4) 1.962(3), Ta(1)-N(2) 1.993(3), Ta(1)-N(1) 2.202(3), O(1)-
Ta(1)-N(3) 115.97(12), O(1)-Ta(1)-N(4) 135.16(15), N(3)-Ta(1)-N(4) 108.18(16), O(1)-Ta(1)-N(2)
87.39(13), N(3)-Ta(1)-N(2) 98.68(16), N(4)-Ta(1)-N(2) 93.60(13), O(1)-Ta(1)-N(1) 72.76(11), N(3)-
Ta(1)-N(1) 102.27(13), N(4)-Ta(1)-N(1) 91.15(13), N(2)-Ta(1)-N(1) 155.86(14).
Selected metrical parameters for the four complexes are given in Table 1. Aside from
the slightly shorter bond lengths to the amide and alkoxide ligands to Ti by about 0.05 Å relative
to Ta, all four complexes exhibit a similar coordination geometry. The Ti-NHMe2 bond length of
2.224(2) Å is longer than corresponding Ta-NMe2 bond lengths of about 1.99 Å. Ta-NHMe2
bonds have been reported at 2.410(5) Å.50 The bond lengths between the metal centers and the
dimethylamide ligands are comparable to related complexes at 1.88-1.91 Å for Ti48, 58-59, 68 and
Page 11

Dalton Transactions
Page 12 of 39
View Article Online
DOI: 10.1039/D0DT02557G
1.95-1.99 Å for Ta.45, 50, 69-70 The bond between the metal and the sulfonamide nitrogen is
longer than that to the dimethylamide nitrogen by approximately 0.2 Å for all four complexes,
which is consistent with previously reported sulfonamide complexes on titanium.58-59, 68, 71 The
metal oxygen bond lengths are shortened relative to similar molecules by about 0.05 Å for Ta
and 0.10 Å for Ti,45, 48, 50, 53, 58 though not outside the range of metal oxygen bonds observed.72-73
Table 1. Selected bond lengths and angles for TiPh3, TaPh1, TaPh2 and TaPh3.
Complex
M-N(1)a
M-O(1)
TiPh3
2.115(2)
1.8866(18) 1.9649(15)
2.224(2)
1.908(2)
TaPh1
2.1952(15)
TaPh2
2.1720(19)
1.9705(17)
1.993(2)
1.960(2)
1.948(2)
156.12(8)
73.85(7)
84.67(8)
116.03(8)
137.55(8)
TaPh3
2.202(3)
1.948(3)
1.993(3)
1.958(3)
1.962(3)
155.86(14)
72.76(11)
87.39(13)
135.16(15)
115.97(12)
M-N(2)a
1.9859(17)
1.9452(18)
1.9558(17)
158.68(7)
73.47(6)
87.69(7)
130.78(7)
116.32(7)
b
M-NF
b
M-NB
1.877(2)
N(1)-M-N(2)
O(1)-M-N(1)
O(1)-M-N(2)
O(1)-M-NF
155.24(9)
75.91(8)
79.82(8)
123.36(9)
120.39(10)
O(1)-M-NB
aN(1), the ligand nitrogen, and N(2) (NHMe2 for TiPh3 and NMe2 for the tantalum complexes)
are the apical atoms for an idealized trigonal bipyramidal geometry. bNF and NB are the “front”
and “back” dimethylamide nitrogen atoms, N(3) or N(4), in the pseudo trigonal plane.
DOSY of complexes
The solid state structures of TiPh3, TaPh1, TaPh2 and TaPh3 are all monomeric, but
dimerization through a bridging oxygen atom in solution is reasonable, especially for titanium,
as seen previously with our bidentate amide-alkoxide ligands,48-49 or other sulfonamide-
alkoxide complexes.53, 58 However, unlike our previously studied titanium complexes with
bidentate amide-alkoxide ligands, three of the titanium complexes in this study have a
coordinated dimethylamine, and all have bulky diphenyl substitution on the alkoxy carbon
Page 12

Page 13 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/D0DT02557G
which may prevent dimerization in solution. To address the solution structures of these
complexes, we undertook DOSY NMR measurements.
Solutions of all eight Ti and Ta complexes were prepared in C6D6 and their diffusion
constants were determined as an average value calculated from fits to the signal intensity for
two regions of the spectra: an aryl signal and an NMe2 signal. Additional details of the DOSY
experiment are presented in the ESI. Intensity was found to decay as a function of delay time
and the resulting diffusion constants are reported units of m2 sec-1 using the Bruker TopSpin 3.0
software package. Hydrodynamic radii (rh) were calculated from the diffusion constants using
the Stokes-Einstein equation and are reported in Table 2.
The rh values were not calibrated with internal standards, as can be done for more
precise molecular weight determination by DOSY.74-75 Therefore, we chose to benchmark them
by determining radii for TiPh3, TaPh1, TaPh2 and TaPh3 by several other methods. The radius
was calculated from the unit cell volume (ruc), by measuring three orthogonal radii graphically
within CrystalMaker (rx), by determining the volume by summing the Van der Waals radii using
the “molinfo” command within Olex2 (ro), or by carrying out volume calculations using Gaussian
(rG). Each radius was calculated by assuming a spherical volume or averaging the three
orthogonal radii. When carrying out the Gaussian calculations, electron density was integrated
to 0.001 e-/Å3 using the Volume keyword using the B3LYP functional and the lanl2dz basis set.
The volume and therefore the radius is defined differently by these different measures, but the
average radii for the complexes are comparable (Table 2).
Page 13

Dalton Transactions
Page 14 of 39
View Article Online
DOI: 10.1039/D0DT02557G
Table 2. Measured diffusion coefficients for TiPh1, TiPh2, TiPh3, TiPh4, TaPh1, TaPh2, TaPh3,
and TaPh4 at 25 °C in C6D6. Radii calculated from diffusion coefficients (rh), from unit cell, (ruc),
from X-ray structure measurements (rx), from Olex2 (ro), and from volume Gaussian calculations
(rG).
Complex D (m2/s)
rh (Å)
ruc (Å)
rx (Å)
ro (Å)
rG (Å)
TiPh1
TiPh2
TiPh3
TiPh4
TaPh1
TaPh2
TaPh3
TaPh4
5.08 x 10-10
6.8
6.8
6.7
6.6
6.3
6.2
6.5
6.6
-
-
-
-
-
-
-
-
5.06 x 10-10
5.12 x 10-10
5.21 x 10-10
5.42 x 10-10
5.55 x 10-10
5.25 x 10-10
5.19 x 10-10
6.3*
-
5.8
-
5.8
5.6
6.3
-
5.2
-
5.1
5.1
5.2
-
5.6
-
5.4
5.5
5.9
-
5.8
5.9
6.0
-
*unit cell contains a molecule of toluene so radius is inflated relative to others.
The average hydrodynamic radius for all eight complexes is 6.6 ± 0.2 Å by our NMR
techniques. Given that the rh values for all eight complexes are similar, and the good agreement
between those radii and those calculated by other methods, there is little evidence of
dimerization or increased average molecularity of any of the complexes in solution. There is
also no evidence of differing molecularity between the Ti and Ta complexes in solution. If the
complexes were dimeric, we would anticipate an increase, though not necessarily a doubling, of
the hydrodynamic radius. Other researchers have observed differences in dimer/monomer
hydrodynamic radii ranging from 1.21:1 to >2:1 for a range of metal ligand combinations.74-78
The evidence supports that the molecularity of the Ti and Ta complexes in this study are the
same both in the solid state and in solution. It is reasonable that they would therefore behave
similarly when coordinated to the primary amine of the substrate or the imido of the
postulated intermediate during catalysis.
Page 14

Page 15 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/D0DT02557G
Intramolecular hydroamination of aminoallenes
Hydroamination was carried out using our previously described in situ procedures with
either hepta-4,5-dienylamine (1a)44, 46 or 6-methyl-hepta-4,5-dienylamine (1b).44-45, 52 We have
not observed differences in catalytic activity or enantioselectivity when using isolated and
purified complexes relative to in situ catalysis. Catalyst precursor solutions were prepared by
mixing stock solutions of the desired ligand and either Ti(NMe2)4 or Ta(NMe2)5 in benzene-d6 in
a J. Young NMR tube. Catalysis began by the addition of a stock solution of the aminoallene
substrate in benzene-d6 followed by heating to 110 °C (substrate 1a) or 135 °C (substrate 1b).
All runs were carried out at 5% catalyst loading with the exception of TiPh3 with substrate 1b,
which was also run at 10% catalyst loading due to very low conversion. The reactions were
monitored by 1H NMR spectroscopy and were quenched either when they were complete or
showed no further reaction progress after 24 hours. Enantioselectivity was determined by
chiral GC-MS of the corresponding benzyl derivatives. Run-to-run repeatability and repeated
injections of the same sample gave errors in the calculated enantioselectivity of ±2% ee. All
catalytic reactions were run in duplicate.
The catalytic hydroamination of hepta-4,5-dienylamine (1a) gave all three possible
products (Table 3). The titanium-derived catalysts favored the tetrahydropyridine (2a) by a
factor of 2.5:1 to 5.5:1, while the tantalum-derived catalysts gave approximately equal amounts
of 2a and the 3a pyrrolidine products. For TiPh4 (entry 4) and TaPh4 (entry 8), a small impurity
peak interfered with the integration of the third peak to elute, Z-3a. Including the peak in the
integration would increase the %ee by about 2%. The peak was not integrated for the purposes
of reporting our %ees in Table 3, but representative GC traces are shown in the ESI.
Page 15

Dalton Transactions
Page 16 of 39
View Article Online
DOI: 10.1039/D0DT02557G
Table 3. Hydroamination of hepta-4,5-dienylamine (1a) at 110 ℃ with in situ catalysts (5 mol%
catalyst) to give tetrahydropyridine 2a, and E- or Z--vinylpyrrolidines E- or Z-3a.
H
N
H
N
NH2
N
·
1a
2a
E-3a
Z-3a
Yielda (%)
Entry Complex t/h % conv. 2a E-3ab (ee) Z-3ab (ee)
1
2
3
4
5
6
7
8
TiPh1
TiPh2
TiPh3
TiPh4
TaPh1
TaPh2
TaPh3
TaPh4
64 84
40 95
23 98
30 83
30 96
30 97
23 98
23 95
60 6 (17%)
73 4 (46%)
82 5 (55%)
59 4 (19%)
49 10 (40%) 37 (17%)
50 10 (25%) 36 (16%)
18 (5%)c
18 (20%)c
10 (45%)c
20 (8%)c
48 18 (8%)
49 10 (40%) 36 (15%)
32 (21%)
aRelative amount of tetrahydropyridine:pyrrolidine determined by 1H NMR, ±2%. bOf the benzyl
derivative, determined by GC, ±2%. bDetermined by comparison to literature values.45, 79-81
cEnantiomer with longer retention time was favored. Data reported as the average of two
individual runs.
The relative yield of Z-3a to E-3a ranged from 2:1 to 4.5:1, though no obvious trends are
apparent for the regioselectivity. Our prior work with this substrate and bidentate amino-
alcohol ligands on titanium had similar Z- to E-selectivity, although with a much lower yield of
the achiral tetrahydropyridine product at 15-30%.44 Our more recent work with tridentate
imine diol ligands had a much lower Z- to E-selectivity of approximately 1:1, but again generally
gave a lower yield of product 2a.46 The enantioselectivity of the resulting pyrrolidine products
were substantially higher than previously observed by titanium-derived catalysts, up to 40-50%
for E-3a.44 However, the yields of the pyrrolidine products are too low (4-6%) to be synthetically
useful. We have typically observed higher enantioselectivity of the E-substrates, suggesting that
the regiochemistry of the substrate is at least somewhat responsible for the selectivity.
Attempts to model the transition states of the reaction have been inconclusive.82 The
Page 16

Page 17 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/D0DT02557G
enantioselectivity of the products using TaPh1 and TaPh4 were more than double the
corresponding titanium catalysts, while for TiPh2 and TiPh3, the enantioselectivity is 2-4 times
higher than the corresponding tantalum catalysts for both regioisomers.
In the hydroamination of 6-methyl-hepta-4,5-dienylamine (1b), the titanium reactions
tended to stall at about 70-90% conversion. The tantalum systems gave complete conversion
with slightly higher enantioselectivity than the titanium systems. Catalysis using TiPh3 only
reached 18% conversion at 5 mol% loading (entry 3), though the enantioselectivity for
pyrrolidine product was 41 %ee. Increasing the catalyst loading to 10% increased the
conversion to 100%, with a lowered enantioselectivity of only 21 %ee. We observed low
conversion at 5% catalyst loading in our prior report on sulfonamide ligands on titanium,52 and
in other systems,83 but increasing the catalyst loading to 10% usually resulted in increased
conversion with similar enantioselectivity rather than reducing the selectivity. With the
exception of entry 3, no clear difference in enantioselectivity amongst the four ligands was
observed in the titanium systems, which gave enantioselectivity of 23±4%. The 27 %ee for
TiPh4 is the highest observed enantioselectivity for an asymmetric hydroamination of an
aminoallene using a titanium catalyst. The corresponding tantalum catalysts give enantiomeric
excesses of 36±3%. Within error, there is no apparent influence of the sulfonamide substituents
on the selectivity of the reaction, and catalysts derived from both metals favored the formation
of the S-(-) enantiomer of product 3b.
Page 17

Dalton Transactions
Page 18 of 39
View Article Online
DOI: 10.1039/D0DT02557G
Table 4. Hydroamination of 6-methyl-hepta-4,5-dienylamine (1b) at 135 ℃ with in situ catalysts
(5 mol% catalyst)
H
NH2
N
·
1b
3b
Entry Complex t/h % conv. % eea Config.b
1
2
3
4c
5
6
7
8
9
TiPh1
TiPh2
TiPh3
TiPh3
TiPh4
TaPh1
TaPh2
TaPh3
TaPh4
18 82
32 95
106 18
20 100
75 71
57 100
49 100
23 100
47 100
18
24
41
21
27
37
33
35
39
S-(–)
S-(–)
S-(–)
S-(–)
S-(–)
S-(–)
S-(–)
S-(–)
S-(–)
aOf the benzyl derivative, determined by GC, ±2%. bdetermined by comparison to literature
values.45, 79-81 c10 mol% catalyst loading. Data reported as the average of two individual runs.
Overall, an increase in enantioselectivity was observed in both of these new titanium
and tantalum systems comparing to the previously reported analogs with either hydrogen or
methyl substituents α to the oxygen.52 These results indicate that the most important
consideration for selectivity of the hydroamination reaction using these sulfonamide ligands is
the substituent  to the oxygen atom and not substituents on the sulfonamide group.
One intriguing result is the substantially higher enantioselectivity of 41 %ee at 18%
conversion for TiPh3 (Table 4, entry 3). The cause of this higher selectivity at lower conversion
is not currently known. Preliminary DFT calculations on both titanium and tantalum systems are
consistent with [2+2] cycloaddition and cycloreversion being relatively low barrier processes
with a higher barrier step to remove the product by proton transfer reactions.82 Others have
reported similar energetic profiles for early metal systems both experimentally84 and
Page 18

Page 19 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/D0DT02557G
computationally.85 However, the relative rates of cyclization and proton transfer are similar and
temperature dependent, and other work shows that the proton transfer barrier is lower.86-88
Schafer has recently reported that hemilabile amidate ligands are important as hydrogen bond
acceptors to help mediate the protonolysis step, though in her system, the [2+2] reaction is
turnover limiting.42
Conclusions
A series of four sterically and electronically differentiated sulfonamides were prepared.
These underwent protonolysis reactions with either Ti(NMe2)4 or Ta(NMe2)5 to give their
corresponding complexes. Three of the titanium complexes retain a dimethylamine ligand. The
complexes were characterized by DOSY NMR and X-ray diffraction and were found to be
monomeric both in solution and in the solid state, as opposed to previously characterized
titanium complexes which were dimeric in the solid state. These complexes gave pyrrolidine
products with higher enantioselectivity during the asymmetric hydroamination of aminoallenes
compared to previously reported sulfonamide derived catalysts, but the largest contributor to
the increased selectivity is the substituent  to the oxygen of the ligand, as the enantiomeric
excesses are the same within error for each set of metal catalysts. While these catalysts are not
yet competitive with current state of the art precious metal hydroamination catalysts, these
show a promising improvement in selectivity relative to previously studied titanium catalysts.
Conflicts of Interest
There are no conflicts of interest to declare.
Page 19

Dalton Transactions
Page 20 of 39
View Article Online
DOI: 10.1039/D0DT02557G
Acknowledgements
Financial support for this project was received from the National Science Foundation,
NSF-MRI-1725142, the John Stauffer Fund for Summer Research in Chemistry, and Harvey
Mudd College. We gratefully acknowledge NSF support for quantum mechanics calculations
through the award CHE-1565743 (Robert J. Cave). The authors thank the Beckman Institute for
their support of the X-Ray Crystallography Facility at Caltech and the Dow Next Generation
Instrumentation Grant.
Experimental
General
All reagents were obtained from commercial suppliers and were purified by standard
methods89 or used as received. The ligand precursor (S)-2-amino-1,1,3-triphenylpropanol,90-91
was prepared by literature procedures and recrystallized from ethanol before use. L-H2Ph1 was
prepared by slight modification of literature procedures.61 The purity of compounds was
established by NMR spectroscopy and elemental analysis. Solvents were purified by vacuum
transfer from sodium/benzophenone (C6D6) or by passage through a column of activated
alumina (Innovative Technology PS-400-5-MD) and stored under nitrogen (diethyl ether,
tetrahydrofuran, dichloromethane, hexane, and toluene). Column chromatography was carried
out using a CombiFlash NextGen 300+ system (Teledyne ISCO). Solutions of ligands (ca. 0.05 M
in C6D6) and substrates (ca. 1.5 M in C6D6) for catalysis were dried over molecular sieves
overnight and stored at -35 °C. All air and/or moisture sensitive compounds were manipulated
under an atmosphere of nitrogen using standard Schlenk techniques, or in a glovebox (MBraun
Unilab). 1H and 13C NMR spectra were recorded at ambient temperature on a Brüker Avance
Page 20

Page 21 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/D0DT02557G
NEO 400 spectrometer and referenced to internal tetramethylsilane or residual solvent peaks.
Carbon assignments were made using DEPT experiments. J values are given in Hz. Polarimetry
was carried out using a JASCO P1010 instrument. IR spectroscopy was carried out using a
Thermo-Nicolet iS5 FTIR using a diamond anvil ATR accessory. GC-MS analysis was carried out
using a Hewlett Packard 5890 Series II Gas Chromatograph. Mass spectra were obtained using
an Advion expressionL APCI Mass Spectrometer with quadrupole mass analyzer. Specific
rotation values []D, are given in 10-1 deg cm2 g-1. Elemental analyses were performed by
Midwest Microlab, Indianapolis, IN.
(2S)-2-(p-toluenesulfonylamino)-(1,1,3-triphenyl-1-propanol) (L-H2Ph1)
IR characterization data not previously reported61 is given here for comparison. IR (ATR,
diamond): SO, asym = 1288 cm-1, SO, sym = 1151 cm-1, SN = 957 cm-1.
(2S)-2-(p-trifluoromethylbenzenesulfonylamino)-(1,1,3-triphenyl-1-propanol) (L-H2Ph2)
Under an atmosphere of N2, (S)-2-amino-1,1,3-triphenylpropanol (8.00 mmol, 2.4280 g,
1 eq.) was dissolved in 80 mL anhydrous dichloromethane together with triethylamine (NEt3)
(16.00 mmol, 2.3 mL, 2 eq.) and 4-dimethylaminopyridine (DMAP) (0.4 mmol, 0.486 g, 0.05 eq.).
After cooling to 0 °C, a solution of 4-(trifluoromethyl)benzenesulfonylchloride (8.00 mmol, 1.96
g, 1eq.) in 16 mL anhydrous dichloromethane was added slowly, and the reaction was left
stirring overnight at room temperature. The reaction was washed with 1 M HCl (20 mL),
followed by 5% NaHCO3 (10 mL). The aqueous phase was extracted with 20 mL
dichloromethane twice, and the organic layer was washed with saturated NaCl solution and
dried over MgSO4. Crude product was obtained by removing solvent under vacuum. The ligand
was then recrystallized from hot methanol (2.987 mmol, 1.528 g, 37%). Mp: 107-110 °C. [α]D=
Page 21

Dalton Transactions
Page 22 of 39
View Article Online
DOI: 10.1039/D0DT02557G
16.4° (c= 0.0042 g / mL, EtOAc). Anal. Calcd for C28H24F3NO3S: C, 65.74; H, 4.73; N, 2.74. Found:
C, 65.93; H, 4.88; N, 3.01. 1H NMR (400 MHz, CDCl3):  7.52-6.97 (m, 19H, PhH), 5.11 (d, 1H, J =
8.1, NH), 4.69 (br s, 1H, CHCH2Ph), 3.24 (dd, 1H, J = 3.4, 14.2, CHCHaHbPh), 2.87 (dd, 1H, J = 6.8,
14.2, CHCHaHbPh), 2.46 (brs, 1H, OH). 13C NMR (100 MHz, CDCl3):  144.35 (4°), 143.82 (4°),
137.09 (4°), 133.34 (4°, q, J = 33, CCF3), 129.89 (CH), 128.98 (CH), 128.82 (CH), 128.33 (CH),
127.80 (CH), 127.32 (CH), 126.87 (CH), 126.08 (CH), 125.96 (q, J = 4, CH ortho to CF3), 125.42
(CH), 123.48 (q, J = 271, CF3), 81.06 (CPh2OH), 62.26 (CHCH2Ph), 38.29 (CH2Ph), One aromatic
CH and one aromatic 4° were not observed. 19F NMR (376 MHz, CDCl3):  = 63.19. MS (APCI):
m/z 494.4 ([M-OH+H]+, 60%). IR (ATR, diamond): SO, asym = 1321 cm-1, SO, sym = 1155 cm-1, SN
971 cm-1.
=
(2R)-2-(p-trifluoromethylbenzenesulfonylamino)-(1,1,3-triphenyl-1-propanol) (D-H2Ph2).
The D-enantiomer was prepared according to a similar procedure starting from (R)-2-
amino-1,1,3-triphenylpropanol (1.65 mmol, 0.4980 g, 1eq.), NEt3 (3.30 mmol, 0.46 mL), DMAP
(0.08 mmol, 0.0117g), and 4-(trifluoromethyl)benzenesulfonylchloride (1.65 mmol, 0.4059 g,
1eq.) in anhydrous CH2Cl2 (20 mL) The product was purified by chromatography (EtOAc/hexane)
(0.2277 mmol, 0.1165 g, 14%). [α]D= -20.1° (c= 0.00398 g / mL, EtOAc). The 1H NMR spectrum
matched that of the L-enantiomer.
(2S)-2-(3,5-bis(trifluoromethyl)benzenesulfonylamino)-(1,1,3-triphenyl-1-propanol) (L-H2Ph3).
L-H2Ph3 was prepared according to a similar procedure starting from (S)-2-amino-1,1,3-
triphenylpropanol (5.284 mmol, 1.6032 g, 1 eq.), NEt3 (10.568 mmol, 1.473 mL, 2 eq.), DMAP
(0.2642 mmol, 0.0323 g, 0.05 eq.), and 3,5-bis(trifluoromethyl)benzenesulfonylchloride (5.283
mmol, 1.6518 g, 1 eq.) in anhydrous CH2Cl2 (120 mL). The product was purified by
Page 22

Page 23 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/D0DT02557G
chromatography (EtOAc/hexane) (1.29 g, 2.23 mmol, 42%). Mp: 68.5-71 °C. [α]D= 29.6° (c=
0.00452 g / mL, EtOAc). Anal. Calcd for C29H23F6NO3S: C, 60.10; H, 4.00; N, 2.42; S, 5.53. Found:
C, 59.80; H, 4.14; N, 2.33; S, 5.76. 1H NMR (400 MHz, CDCl3):  7.41 (s, 3H), 7.52 – 6.91 (m, 15H,
ArH), 5.21 (br s, NH), 4.72 (dd, 1H, J = 3.3, 7.3, CHCH2Ph), 3.21 (dd, 1H, J = 3.2, 14.2,
CHCHaHbPh), 2.88 (dd, 1H, J = 7.4, 14.2, CHCHaHbPh), 2.39 (br s, 1H, OH). 13C NMR (100 MHz,
CDCl3):  143.86 (4°), 143.58 (4°), 143.54 (4°), 136.82 (4°), 132.40 (4°, q, J = 34, CCF3,), 129.78
(CH), 128.98 (CH), 128.94 (CH), 128.29 (CH), 127.95 (CH), 127.42 (CH), 126.67 (q, J = 3, ortho-
(F3C)2ArCH), 125.97 (CH), 125.58 (sept, J = 3, para-(F3C)2ArCH), 125.24 (CH), 122.59 (4°, q, J =
271, F3CAr), 80.99 (CPh2OR), 62.97 (NCHCO), 38.43 (CCH2). One aromatic CH was not observed.
19F NMR (376 MHz, CDCl3):  62.88. MS (APCI): m/z 304.3 ([M - (CF3)2C6H3SO2 +H]+, 100%). IR
(ATR, diamond): SO, sym = 1153 cm-1, SN = 959 cm-1.
(2R)-2-(3,5-bis(trifluoromethyl)benzenesulfonylamino)-(1,1,3-triphenyl-1-propanol) (D-
H2Ph3).
The D-enantiomer was prepared according to a similar procedure starting from (R)-2-
amino-1,1,3-triphenylpropanol (1.65 mmol, 0.4999 g, 1 eq.), NEt3 (3.30 mmol, 0.46 mL, 2 eq.),
DMAP (0.08 mmol, 0.0103 g, 0.05 eq.), and 3,5-bis(trifluoromethyl)benzenesulfonylchloride
(1.65 mmol, 0.5171 g, 1 eq.) in anhydrous CH2Cl2 (20 mL). The product was purified by
chromatography (EtOAc/hexane) (0.5558g, 0.959 mmol, 58%). [α]D= -31.1° (c= 0.0035 g / mL,
EtOAc). The 1H NMR spectrum matched that of the L-enantiomer.
(2S)-2-(2,4,6-trimethylbenzenesulfonylamino)-(1,1,3-triphenyl-1-propanol) (L-H2Ph4).
L-H2Ph4 was prepared according to a similar procedure starting from (S)-2-amino-1,1-
diphenyl-3-phenylpropanol (0.9105 g, 3.001 mmol, 1 eq.) NEt3 (0.835 mL, 5.991 mmol, 2 eq.)
Page 23

Dalton Transactions
Page 24 of 39
View Article Online
DOI: 10.1039/D0DT02557G
DMAP (0.0369 g, 0.3020 mmol, 0.10 eq.), and 2,4,6-trimethylbenzenesulfonyl chloride (0.6562
g, 3.000 mmol, 1 eq.) in anhydrous CH2Cl2 (36 mL). The product was purified by
chromatography (EtOAc/hexane) (1.0195 g, 2.099 mmol, 70%). Mp: 75.8-78.3 °C. []D = -7.95 °
(c = 0.00352 g/mL, EtOAc). Anal. Calcd for C30H31NO3S: C, 74.20; H, 6.43; N, 2.88. Found: C,
73.90; H, 6.40; N, 3.11. 1H NMR (400 MHz, CDCl3):  7.27 – 6.96 (m, 15 H, ArH), 6.72 (s, 2H,
meta-ArH), 5.01 (d, J = 8, NH), 4.72 (dd, J = 4, 6, CHCH2Ph), 3.33 (dd, J = 4, 14, CHCHaHbPh), 2.88
(dd, J = 6, 14, CHCHaHbPh), 2.60 (s, 1H, OH), 2.38 (s, 6H, Ar(CH3)2), 2.28 (s, 3H, Ar(CH3). 13C NMR
(100 MHz, CDCl3):  144.14 (4°), 144.05 (4°), 141.32 (4°), 138.16 (4°), 137.18 (4°), 134.87 (4°),
131.86 (CH), 129.57 (CH), 128.66 (CH), 128.57 (CH), 128.10 (CH), 127.42 (CH), 126.81 (CH),
126.73 (CH), 125.88 (CH), 124.89 (CH), 81.16 (CPH2), 61.14 (CH), 37.90 (CH2), 22.98 (CH3), 20.81
(CH3). IR (cm-1): SO, asym 1322.73 cm-1, SO, sym = 1147.70 cm-1, SN = 966 cm-1.
Chiral contact shift experiments
Chiral shift studies were performed with H2Ph2 and H2Ph3 using the following
procedure. L-H2Ph2 (20.5 mg, 0.040 mmol) was dissolved in CDCl3 and examined by 1H NMR
spectroscopy. Shift reagent (S)-(+)-1-(9-anthryl)-2,2,2-trifluoroethanol (11.0 mg, 0.040 mmol, 1
eq.) was added to the solution, which was reexamined by 1H NMR spectroscopy. An additional
equivalent of shift reagent was then added to the solution for a third 1H NMR spectrum. The
same procedure was repeated with the D-H2Ph2 ligand. The peaks shifted without splitting
throughout the process. Finally, the L- and D-H2Ph2 solutions with 2 eq. shift reagent were
combined for a final spectrum. L-H2Ph2 exhibited splitting of the peaks at 2.8 and 3.2 ppm while
L-H2Ph3 exhibited broadening of the corresponding peaks and loss of resolution but no contact
shift splitting.
Page 24

Page 25 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/D0DT02557G
Complex TiPh1, Ti(NMe2)2(Ph1)(NHMe2)
Ti(NMe2)4 (0.073 g, 0.323 mmol) was dissolved in ether (2 mL) and L-H2Ph1 (0.323 mmol,
0.148 g) was dissolved in ether (12 mL). Both solutions were cooled to -30 °C. The ligand was
then slowly added to Ti(NMe2)4 solution. The solution changed color from yellow to red, and a
yellow solid formed. The product was obtained as a light yellow solid by vacuum filtration
(0.142 g, 0.223 mmol, 69% yield). The complex can be recrystallized from toluene. Mp: 213-215
°C. Anal. Calcd: C, 64.14; H, 6.97; N, 8.80. Found: C, 64.48, H, 7.07; N, 8.25. 1H NMR (400 MHz,
C6D6):  8.03 (d, 2H, J= 6.8 Hz, ortho-PhH), 7.19-6.92 (m, 17H, PhH), 5.13 (d, 1H, J= 7.3 Hz,
CHCH2Ph), 3.76 (d, 1H, J= 13.5 Hz, CHCHaHbPh), 3.35 (s, 6H, NMe2), 3.15 (dd, 1H, J= 8 Hz and 14
Hz, CHCHaHbPh), 2.85 (s, 6H, NMe2), 2.06 (s, 3H, PhCH3) 1.92 (br d, 6H, HN(CH3)2). 13C NMR (100
MHz, C6D6):  149.85 (4°), 147.70 (4°), 142.46 (4°), 142.02 (4°), 129.31 (CH), 129.27 (CH), 128.52
(CH), 128.48 (CH), 128.34 (CH), 127.79 (CH), 127.77 (CH), 127.57 (CH), 126.20 (CH), 126.10 (CH),
125.11 (CH), 94.52 (CPh2), 71.31 (NCHCH2), 46.15 (NMe2), 44.49 (NMe2), 41.78 (CCH2), 39.97
(HN(CH3)2), 21.21 (PhCH3). One aromatic 4° was not observed.
Complex TiPh2, Ti(NMe2)2(Ph2)(NHMe2)
Ti(NMe2)4 (0.173 g, 0.770 mmol) was dissolved in diethyl ether (2 mL) and L-H2Ph2
(0.394 g, 0.770 mmol) was dissolved in diethyl ether (12 mL). Both solutions were cooled to -30
°C. Ligand was then slowly added to Ti(NMe2)4 solution. The solution changed color from yellow
to red until yellow solid was formed, and reaction was left to stir overnight. A bright yellow
solid was obtained by vacuum filtration (0.271 g, 0.393 mmol, 43%). An additional crop was
obtained by reduction of solvent volume (0.216 g, 0.313 mmol, 41%). The combined solids were
recrystallized from hot toluene. Mp: 166.0-169.3 °C. Anal. Calcd for C32H34F3N3O3STi: C, 59.54;
Page 25

Dalton Transactions
Page 26 of 39
View Article Online
DOI: 10.1039/D0DT02557G
H, 5.31; N, 6.51. Found: C, 59.56; H, 5.61; N, 6.97. 1H NMR (400 MHz, C6D6):  7.99 (broad d, 2H,
J = 7, ArH), 7.35-6.81 (m, 17H, ArH), 5.00 (br d, J = 8, 1H, CHCH2Ph), 3.72 (br d, J = 14, 1H,
CHCHaHbPh), 3.27 (s, 6H, NMe2), 3.16 (dd, J = 8, 1, 1H, CHCHaHbPh), 2.76 (s, 6H, NMe2) 1.85 (d, J
= 6 Hz, HN(CH3)2), 1.51 (br s, 1H, NHMe2). 13C NMR (100 MHz, C6D6):  149.42 (4°), 148.42 (4°),
147.33 (4°), 142.30 (4°), 132.21 (q, J = 32, 4°), 129.26 (CH), 128.63 (CH), 128.35 (CH), 128.11
(CH), 127.77 (CH), 127.67 (CH), 127.26 (CH), 126.39, (CH), 126.37 (CH), 125.77 (q, J = 4, meta-
CH), 125.39 (CH), 127.37* (estimated  from peaks at 126.01 & 123.30, q, J = 271, CF3), 94.54
(CPh2), 71.25 (CH), 46.15 (CH3), 44.46 (CH3), 42.07 (CH2), 40.01 (CH3). 19F NMR (376 MHz, C6D6):
 62.31
Complex TiPh3, Ti(NMe2)2(Ph3)(NHMe2)·C7H8
Ti(NMe2)4 (0.200 g, 0.892 mmol) was dissolved in diethyl ether (2 mL) and L-H2Ph3
(0.517 g, 0.892 mmol) was dissolved in diethyl ether (15 mL). Both solutions were cooled to -30
°C. The ligand was then slowly added to Ti(NMe2)4 solution. The solution changed color from
yellow to red. A yellow solid formed after around 6 hours, and the reaction was left to stir
overnight. A bright yellow solid was obtained by vacuum filtration (0.200 g, 0.264 mmol, 30%
yield). A second crop was obtained by reducing solvent volume (0.119 g, 0.157 mmol, 17%). The
crude 1H NMR spectrum showed coordinated dimethylamine but no toluene. The combined
solids were recrystallized from hot toluene, and a toluene of solvation was observed. Mp:
146.8-150 °C. Anal. Calcd for C35H40F6N4O3STi: C, 59.54; H, 5.31; N, 6.51. Calcd for
C35H40F6N4O3STi·C7H8: C, 59.29; H, 5.69; N, 6.59. Found: C, 58.99; H, 5.78; N, 6.90. 1H NMR (400
MHz, C6D6): δ 8.55 (broad s, 2H, ortho-ArH), 7.72 (s, 1H, para-ArH), 7.64-6.83 (m, 20H, ArH,
C6H5Me), 5.16 (broad d, 1H, J = 8 Hz, CHCH2Ph), 3.48 (br d, 1H, J = 14, CHCHaHbPh), 3.25 (s, 6H,
Page 26

Page 27 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/D0DT02557G
NMe2), 3.09 (br dd, 1H, J = 9, 14, CHCHaHbPh), 2.74 (s, 6H, NMe2), 2.10 (s, 3H, C6H5CH3), 1.80 (d,
6H, J = 6, NH(CH3)2), 1.44 (br s, 1H, NHMe2). 13C NMR (100 MHz, C6D6):  148.84 (4°), 147.90
(4°), 146.60 (4°), 141.23 (4°), 131.74 (q, J = 33, CCF3); 137.52 (4° toluene), 28.95 (toluene CH),
128.41 (CH), 128.19 (toluene CH), 127.97 (CH), 127.78 (CH), 127.53 (CH), 127.24 (CH), 126.99
(unresolved q, ortho-CH), 126.77 (CH), 126.23 (CH), 126.01 (CH), 125.32 (toluene CH), 125.08
(CH), 123.64 (unresolved sept, J = 4, para-CH), 123.29* (estimated from peaks at 124.65 and
121.93, q, J = 272, CF3), 94.22 (CPh2), 70.63 (CH), 45.71 (CH3), 43.95 (CH3), 41.40 (CH2), 39.54
(CH3), 21.04 (toluene CH3). 19F NMR (376 MHz, C6D6):  62.38.
Complex TiPh4, Ti(NMe2)2(Ph4)
Ti(NMe2)4 (0.231 g, 1.03 mmol) was dissolved in toluene (2 mL) and L-H2Ph4 (0.500 g,
1.03 mmol) was dissolved in toluene (20 mL). Both solutions were cooled to -30 °C and then
combined in a Schlenk tube with additional toluene (7 mL). The reaction mixture was heated for
3 days at 65 °C. The solvent was removed in vacuo to give an orange-brown foam (0.638 g, 9.63
mmol, 95%). It has not been possible to isolate this material as a solid. 1H NMR (400 MHz, C6D6):
 7.88 (d, 2H, J = 7, ArH), 7.65 (d, 2H, J = 8 Hz, ArH), 7.25-7.01 (m, 6 H, ArH), 6.75-6.65 (m, 5H,
ArH), 6.45 (s, 2H, meta-ArH), 5.47 (dd, 1H, J = 10, 4, CHCH2Ph), 3.39 (s, 6H, NMe2), 3.04 (s, 6H,
NMe2), 2.41 (s, 6H, ortho-ArCH3), 1.92 (s, 3H, para-ArCH3). The benzylic protons (CH2Ph) protons
are obscured by impurities. 13C NMR (100 MHz, C6D6):  148.07 (4°), 147.16 (4°), 140.75 (4°),
139.21 (4°), 138.50 (4°), 137.38 (4°), 131.63 (CH), 128.77 (CH), 127.64 (CH), 127.60 (CH), 127.48
(CH), 127.23 (CH), 126.73 (CH), 126.30 (CH), 125.31 (CH), 96.84 (CPh2), 71.63 (CH), 46.09 (NCH3),
44.17 (NCH3), 41.45 (CH2), 23.16 (ArCH3), 20.24 (ArCH3). One aromatic CH was not observed.
Page 27

Dalton Transactions
Page 28 of 39
View Article Online
DOI: 10.1039/D0DT02557G
Complex Ta1, Ta(NMe2)3(Ph1)
Ta(NMe2)5 (0.200 g, 0.498 mmol) was dissolved in diethyl ether (2 mL) and L-H2Ph1
(0.228 g, 0.498 mmol) was partially dissolved in 13 mL diethyl ether. Both solutions were cooled
at -30 °C for 15 minutes. The ligand was then slowly added to Ta(NMe2)5 solution. The reaction
was left to stir overnight. The crude product was obtained by removing solvent (0.386 g, 0.502
mmol, 100% yield) and purified in low yield by recrystallization from ether. Mp: 197.5 - 199.0
°C. Anal. Calcd for C34H43N4O3STa: C, 53.12; H, 5.64; N, 7.29. Found: C, 53.07; H, 5.52; N, 7.09. 1H
NMR (400 MHz, C6D6):  8.03 (d, 2H, J = 8, ortho-ArH), 7.14-6.81 (m, 17H, ArH), 5.21 (br d, 1H, J
= 8, CHCH2Ph), 4.10 (br d, 1H, J = 14, CHCHaHbPh), 3.34 (s, 18H, NMe2), 3.24 (dd, 1H, J = 9, 14,
CHCHaHbPh), 2.05 (s, 3H, PhCH3). 13C NMR (100 MHz, C6D6):  149.28 (4°), 146.42 (4°), 141.99
(4°), 141.57 (4°), 140.70 (4°), 129.36 (CH), 129.17 (CH), 129.02 (CH), 128.35 (CH), 128.09 (CH),
128.07 (CH), 127.66 (CH), 126.43 (CH), 126.35 (CH), 125.09 (CH), 93.58 (CPh2), 70.61 (CH), 45.50
(CH3), 40.21 (CH2), 21.21 (CH3). One aromatic CH carbon was not observed.
Complex Ta2, Ta(NMe2)3(Ph2)
Ta(NMe2)5 (0.250 g, 0.623 mmol) was dissolved in diethyl ether (2 mL) and L-H2Ph2
(0.319 g, 0.623 mmol) was dissolved in diethyl ether (15 mL). Both solutions were cooled at -30
°C for 15 minutes. The ligand was then slowly added to the Ta(NMe2)5 solution. The reaction
was left to stir overnight. The product was obtained by removing solvent (0.520 g, 0.632 mmol,
100% yield) and purified by recrystallization from ether in low yield. Mp: 207.9-209.3 °C. Anal.
Calcd for C34H40F3N4O3STa: C, 49.64; H, 4.90; N, 6.81. Found: C, 49.37; H, 4.96; N, 6.52. 1H NMR
(400 MHz, C6D6):  7.93 (d, 2H, J = 8, ArH), 7.21 (d, 2H, J = 8, ArH), 7.1-6.7 (m, 15H, ArH), 5.08 (br
d, 1H, J = 8, CHCH2Ph), 4.03 (br d, 1H, J = 14, CHCHaHbPh), 3.30 (s, 18H, NMe2), 3.23 (dd, 1H, J =
Page 28

Page 29 of 39
Dalton Transactions
View Article Online
DOI: 10.1039/D0DT02557G
10, 14, CHCHaHbPh). 13C NMR (100 MHz, C6D6):  148.71 (4°), 146.75 (4°), 145.90 (4°), 141.66
(4°), 132.75 (q, J = 32, 4°), 129.04 (CH), 128.98 (CH), 128.29 (CH), 128.17 (CH), 128.02 (CH),
127.72 (CH), 126.61 (CH), 126.50 (CH), 125.84* (part of quartet; estimate J = 3 Hz, CH), 125.80
(CH), 125.32 (CH), 123.03* (outermost peak of CF3 quartet; other peaks not observed), 93.64
(CPh2), 70.80 (CH). 45.40 (CH3), 40.42 (CH2). 19F NMR (376 MHz, C6D6): 62.39.
Complex Ta3, Ta(NMe2)3(Ph3)
Ta(NMe2)5 (0.200 g, 0.498 mmol) was dissolved in diethyl ether (2 mL) and L-H2Ph3
(0.289 g, 0.498 mmol) was dissolved in diethyl ether (10 mL). Both solutions were cooled at -30
°C for 15 minutes. The ligand was then slowly added to the Ta(NMe2)5 solution which was then
diluted to 15 mL. The reaction was left to stir overnight. The product was obtained by removing
solvent (0.440 g, 0.494 mmol, 99% yield) and purified by recrystallization from ether. Mp 155-9
°C. Anal. Calcd for C35H39F6N4O3STa: C, 47.20; H, 4.41; N, 6.29. Anal. Calcd for C31H27F6N2O4STa
[Ta(NMe2)(Ph3)(=O)]: C, 45.49; H, 3.32; N, 3.42. Found: C, 45.85; H, 3.89; N, 2.79. 1H NMR (400
MHz, C6D6, 298 K):  8.51 (s, 2H, ortho-PhH), 7.76 (s, 1H, para-PhH), 7.05-6.72 (m, 15H, PhH),
5.10 (br d, 1H, J = 9, CHCH2Ph), 3.79 (br d, 1H, J = 14, CHCHaHbPh), 3.23 (s, 18H, NMe2), 3.10 (dd,
1H, J = 10, 13, CHCHaHbPh). 13C NMR (100 MHz, C6D6, 298 K):  148.41 (4°), 146.24 (4°), 145.44
(4°), 141.03 (4°), 132.34 (q, J = 33, CCF3), 128.76 (br q, J = 3, ortho CH), 128.58 (CH), 128.22 (CH),
128.13 (CH), 127.68 (CH), 126.88 (CH), 126.52 (CH), 125.58 (CH), 125.41 (CH), 124.70 (sept, J =
4, para CH), 123.42 (q, J = 272 Hz, CF3), 93.61 (CPh2), 70.53 (CH), 45.23 (CH3), 39.98 (CH2). One
aromatic CH was not observed. 19F NMR (376 MHz, C6D6):  62.53.
Page 29