Cobalt- and Iron-Catalyzed Isomerization-Hydroboration of Branched Alkenes: Terminal Hydroboration with Pinacolborane and 1,3,2-Diazaborolanes

Article abstract of DOI:10.1021/acs.organomet.6b00823

The synthesis and characterization of a series of structurally varied N-phosphinoamidinate-ligated cobalt complexes is described, along with the successful application of these and a related iron complex as precatalysts in the isomerization-hydroboration of terminal, geminal, and internal alkenes. These reactions proceed under mild conditions (23-65 °C), at relatively low base-metal loadings (1-5 mol %), typically without cosolvent, and with high terminal hydroboration selectivity across a broad spectrum of branched alkenes. With some of the alkene substrates examined, the N-phosphinoamidinate-ligated precatalysts employed herein are shown to provide alternative terminal selectivity versus other previously reported precatalyst classes for such transformations. Reports of terminal-selective metal-catalyzed alkene isomerization-hydroboration disclosed thus far in the literature employ pinacolborane (HBPin); while effective in the system herein, we also report the first examples of such transformations employing either 1,3-dimethyl-1,3-diaza-2-boracyclopentane or benzo-1,3,2-diazaborolane. The application of these 1,3,2-diazaborolanes in place of HBPin in some instances enables novel terminal selectivity in the isomerization-hydroboration of branched alkenes.

Full text of DOI:10.1021/acs.organomet.6b00823

Article
pubs.acs.org/Organometallics
Cobalt- and Iron-Catalyzed Isomerization−Hydroboration of
Branched Alkenes: Terminal Hydroboration with Pinacolborane and
1,3,2-Diazaborolanes
Takahiko Ogawa,† Adam J. Ruddy,† Orson L. Sydora,*,‡ Mark Stradiotto,*,† and Laura Turculet*,†
†Department of Chemistry, Dalhousie University, 6274 Coburg Road, P.O. 15000, Halifax, Nova Scotia B3H 4R2, Canada
‡Research and Technology, Chevron Phillips Chemical LP, 1862 Kingwood Drive, Kingwood, Texas 77339, United States
S
* Supporting Information
ABSTRACT: The synthesis and characterization of a series of
structurally varied N-phosphinoamidinate-ligated cobalt com-
plexes is described, along with the successful application of these
and a related iron complex as precatalysts in the isomerization−
hydroboration of terminal, geminal, and internal alkenes. These
reactions proceed under mild conditions (23−65 °C), at
relatively low base-metal loadings (1−5 mol %), typically
without cosolvent, and with high terminal hydroboration
selectivity across a broad spectrum of branched alkenes. With some of the alkene substrates examined, the N-
phosphinoamidinate-ligated precatalysts employed herein are shown to provide alternative terminal selectivity versus other
previously reported precatalyst classes for such transformations. Reports of terminal-selective metal-catalyzed alkene
isomerization−hydroboration disclosed thus far in the literature employ pinacolborane (HBPin); while effective in the system
herein, we also report the first examples of such transformations employing either 1,3-dimethyl-1,3-diaza-2-boracyclopentane or
benzo-1,3,2-diazaborolane. The application of these 1,3,2-diazaborolanes in place of HBPin in some instances enables novel
terminal selectivity in the isomerization−hydroboration of branched alkenes.
alternative catalysts based on first-row transition metals for such
challenging transformations. A breakthrough in this regard was
published in 2013, when Obligacion and Chirik8 reported on
the ability of (bis(imino)pyridine)CoMe precatalysts to
promote such transformations under mild conditions (neat,
23 °C) using HBPin. Mechanistic investigations point to the
intermediacy of Co−H species in this chemistry, which in the
case of internal alkenes effect alkene isomerization via a series
of alkene insertion/β-hydride elimination steps. Selective
reaction of HBPin with the thus-obtained terminally situated
Co-alkyl species, apparently in the absence of competing
reactivity of internal Co-alkyl intermediates, affords the 1-
alkylboronate product and regenerates the putative Co−H.8 In
a subsequent publication the application of alternative
electronically responsive ligands was examined; within this
report an α-diimine-substituted cobalt allyl complex was shown
to be highly active for the isomerization−hydroboration of even
hindered polysubstituted alkenes with HBPin.9
1. INTRODUCTION
The synthetic utility of organoboron compounds,1 including as
reactants in Suzuki−Miyaura cross-couplings,2 provides moti-
vation for the development of efficient routes to such species.
Alkene hydroboration en route to alkylboronates is attractive in
this regard, given the inherently atom-economical nature of this
class of transformations. Whereas the uncatalyzed addition of
B−H reagents to alkenes has a long history in synthetic
chemistry,1b,3 more recently developed metal-catalyzed hydro-
boration protocols are useful in that they can provide
complementary selectivity versus uncatalyzed transformations
and can enable the use of otherwise unreactive B−H reagents.4
The application of metal catalysis can, in principle, also enable
more complex transformations such as sequential alkene
isomerization−hydroboration, whereby borylation occurs at a
position distal to the initial location of the CC double bond.
Despite the synthetic potential of such reactions,5 selective
transformations of this type have proven difficult to achieve,
owing to the requirements of facile catalytic alkene isomer-
ization6 and high selectivity in terms of reactivity with the
borane (i.e., internal versus terminal borylation).
Encouraged by our previous observations regarding the
catalytic utility of N-phosphinoamidine/amidinate complexes of
the first-row metals, including the chromium-catalyzed selective
tri/tetramerization of ethylene10 and the iron-catalyzed hydro-
silylation of aldehydes, ketones, and esters to alcohols,11 in
2013 we turned our attention to addressing challenges in alkene
Until recently, alkene isomerization−hydroboration chem-
istry with terminal selectivity was limited to a small collection of
reports employing rhodium or iridium catalysts in combination
with pinacolborane (HBPin).7 Notwithstanding the importance
of such contributions, the cost and rarity of the platinum group
metals continue to provide motivation for the development of
Received: October 28, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.6b00823
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
hydroboration chemistry. We were pleased to observe that the
(N-phosphinoamidinate)cobalt(amido) precatalyst C1 was
particularly useful in promoting the selective transformation
of octenes (1a−c) into 1-octylboronate products using HBPin
(Scheme 1);12 other terminal or cyclic (nonbranched) alkenes
and ketones were also successfully hydroborated by use of C1
with HBPin under analogous conditions.13
Scheme 2. Synthesis of New (N-Phosphinoamidinate)cobalt
Complexes
Scheme 1. Isomerization−Hydroboration of Octenes with
Pinacolborane (HBPin), Catalyzed by C1
In light of the fact that published examples of terminal-
selective cobalt-catalyzed alkene isomerization−hydroboration
chemistry are limited only to our work employing N-
phosphinoamidinate-ligated C113 and that of Chirik and co-
workers using cobalt complexes featuring bis(imino)pyridine,
terpyridine, or α-diimine ligation,8,9,14 it is evident that much
remains to be learned regarding the influence of ancillary ligand
design on such transformations. Furthermore, the observation
that HBPin is the only borane that has proven compatible in all
reported examples of terminal-selective metal-catalyzed alkene
isomerization−hydroboration provides motivation to identify
alternative suitably reactive boranes for such applications.15 In
this context we report herein on our efforts to evaluate (i) the
impact of modifications to the ancillary P,N ligand, as well as
the reactive Co−X fragment, on the catalytic performance of
C1; (ii) the ability of the derived precatalysts (C1−C7) to
promote the alkene isomerization−hydroboration of branched
alkenes; and (iii) the feasibility of employing 1,3,2-diazabor-
olanes as alternatives to HBPin in this chemistry. Selected
reactivity comparisons involving the iron variant of C1 (i.e.,
C8) are also presented.
respectively (Scheme 2). Elemental analysis data confirmed
the absence of additional coligands in each of C2−C7, in
keeping with single-crystal X-ray structures of these complexes
(Figure 1).16
2. RESULTS AND DISCUSSION
2.1. Precatalyst Synthesis and Characterization. Our
modular synthesis of N-phosphinoamidines10a was exploited in
the construction of analogues of C1 in which the R groups on
phosphorus, and the backbone NC(R′)N substituent, were
varied (Scheme 2); we strategically opted to retain the bulky
(2,6-iPr2C6H3N)C(R′)NHPR2 group within the newly
prepared N-phosphinoamidine variants employed herein, on
the basis of reactivity trends observed in our previous studies
focused on iron-catalyzed carbonyl reduction chemistry.11,16
Structural variants of C1 in which the R groups on
phosphorus were changed from tBu to 1-adamantyl (C2) or
o-tolyl (C3), or whereby the backbone R′ group was changed
from Ph to 3,5-F2C6H3 (C4) or N-piperidyl (C5), were
prepared using methods analogous to those employed in the
synthesis of C1.13a Variation of the reactive Co−X fragment in
C1 from N(SiMe3)2 to NH(2,6-iPr2C6H3) (C6) or η3-benzyl
(C7) was achieved by use of a similar synthetic protocol
employing only one equivalent of LiN(SiMe3)2 followed by
treatment with either LiNH(2,6-iPr2C6H3) or BnMgBr,
Figure 1. Single-crystal X-ray structures of C2−C7, shown with 50%
thermal ellipsoids and with selected hydrogen atoms omitted for
clarity.
Crystallographic characterization of the (PN)CoN(SiMe3)2
complexes C2−C5 (Figure 1 and Table S116) in each case
revealed a three-coordinate, distorted trigonal-planar structure
analogous to that observed previously for C1.11 Despite altering
both the steric and electronic character of the R and R′
substituents in these complexes relative to C1 (as defined in
Scheme 2), only modest variation of the Co−P (2.3504(4)−
2.3873(7) Å), Co−N1(aryl) (1.9626(13)−1.9881(18) Å),
Co−N(SiMe3)2 (1.8806(11)−1.9031(13) Å), P−N2
(1.6631(13)−1.6815(14) Å), N1−C1N2 (1.350(3)−
1.3705(17) Å), and N2−C1N1 (1.316(3)−1.3242(17) Å)
distances was observed across the C1−C5 series. Conversely,
B
DOI: 10.1021/acs.organomet.6b00823
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
complexes C6 and C7 are best described as featuring a ligand
bond number of four, arising from either an agostic interaction
(C6, Co···H60 ca. 1.87 Å17) or η3-benzyl coordination (C7).
Whereas the P−N and N−C interatomic distances within C6
and C7 are similar to those found in C1−C5, the Co−P (C6,
2.1341(3) Å; C7, 2.1544(4) Å) distances, and to a lesser extent
the Co−N1(aryl) (C6, 1.9021(9) Å; C7, 1.9496(11) Å)
distances, are shorter than those in C1−C5. The unsymmetrical
nature of the coordinated η3-benzyl unit in C7 (Co−C51,
1.9913(16) Å; Co−C52, 2.0856(13) Å; Co−C53, 2.2390(16)
Å) is in keeping with P(OMe)3Co(η3-benzyl)18 and other
crystallographically characterized late transition metal η3-benzyl
complexes.19 As was observed for C1, the paramagnetic three-
coordinate complexes C2−C5 were found to exhibit S = 3/2
magnetic ground states as measured in solution at room
temperature by use of Evan’s method (μeff = 4.17−5.36 μB);20
differing magnetic behavior was observed for the paramagnetic
complexes C6 (μeff = 2.31 μB) and C7 (μeff = 2.57 μB), in
keeping with an S = 1/2 magnetic ground state.21
examined subsequently (Scheme 4). In using 3 mol % Co
with 1c, each of C1, C6, and C7 afforded high conversion to
Scheme 4. Isomerization−Hydroboration of the Linear and
a
Branched trans-Alkenes 1c and 1f with HBPin
2.2. Substrate Scope Employing HBPin. Having
successfully prepared and structurally characterized C1−C7,
we turned our attention to their application as precatalysts for
the isomerization−hydroboration of substituted alkenes
employing HBPin. In an initial set of experiments, we examined
the influence of ancillary P,N ligand modifications on catalytic
performance in transformations involving the methylpentene
isomers 1d and 1e (Scheme 3). With 1d, each of C1−C5
a
1
Estimated conversion to product on the basis of H and 13C{1H}
NMR data. Catalyst loading, reaction times, and isolated yield given in
parentheses; 1 h reaction times are unoptimized.
the terminal hydroboration product 2 after only 1 h, thereby
establishing the viability of employing alternative (PN)CoX
precatalysts (X = NH(2,6-iPr2C6H3), C6; X = η3-benzyl, C7) in
this chemistry. Similarly high conversion was achieved by use of
C5, although higher loadings and longer reaction times were
required (5 mol % Co, 22 h). In using C3, a maximum of 80%
conversion to 2 was achieved (5 mol % Co, 16 h), in keeping
with the inferior performance of C3 versus C1 in the
isomerization−hydroboration of 1e (Scheme 3).
a
Scheme 3. Hydroboration of Methylpentenes with HBPin
In examining the isomerization−hydroboration of 1f with
HBPin to give 4, we opted to employ a 1 h reaction time using
3 mol % C1−C5 at 23 °C, given that Chirik’s bis(imino)-
pyridine-supported cobalt precatalyst required 72 h to go to
completion when using 1 mol % Co at 23 °C.8 Once again,
higher conversion to the target product was achieved by use of
C1 relative to C2−C5 (Scheme 4).
The isomerization−hydroboration of what can be viewed as
more challenging cyclic trisubstituted (1g and 1h) and
tetrasubstituted (1i) alkenes with HBPin was examined
subsequently (Scheme 5). The superiority of precatalysts
featuring P(tBu)2 donor groups (i.e., C1, C4, and C5), relative
to the P(o-tolyl)2-containing variant C3, was apparent in
transformations involving 1g; however, relatively poor
selectivity for terminal hydroboration (i.e., for 5 versus 5′)
was achieved. Precatalysts C1 and C4 also proved optimal
within our reactivity survey involving the isomerization−
hydroboration of 1h, cleanly affording the terminal hydro-
boration product 6. In contrast, related transformations
involving the tetrasubstituted alkene 1i proved challenging for
our N-phosphinoamidinate-supported cobalt catalysts (e.g.,
70% conversion to 7, 3 mol % C1, 18 h, 23 °C). By
comparison, an α-diimine-substituted cobalt precatalyst re-
ported by Chirik and co-workers is able to efficiently promote
the isomerization−hydroboration of this substrate.9
a
1
Estimated conversion to 3 on the basis of H and 13C{1H} NMR
data. Catalyst loading, reaction times, and isolated yield given in
parentheses; 1 h reaction times are unoptimized.
proved competent in affording the anticipated terminal
hydroboration product 3, although longer reaction times were
required when using the P(o-tolyl)2-containing precatalyst C3.
Gratifyingly, the same product 3 was also selectively formed
when using the internal alkene 1e, with C1 and C4 offering
optimal catalytic performance among the precatalysts tested;
the apparent superiority of precatalysts supported by N-
phosphinoamidinate ligands featuring P(tBu)2 and where the
backbone R′ group is aryl (Scheme 2) in the isomerization−
hydroboration of 1e is reflected in related transformations
involving other branched alkenes (vide infra). Our observed
terminal selectivity in the transformation of 1e to 3 contrasts
the isomeric terminal selectivity achieved by Obligacion and
Chirik when using a bis(imino)pyridine-supported cobalt
precatalyst under similar conditions (1 mol % Co, 23 °C).8
The isomerization−hydroboration of trans-4-octene (1c) and
(E)-2,5-dimethyl-3-hexene (1f) by use of C1−C7 was
2.3. Substrate Scope Employing 1,3,2-Diazaboro-
lanes. Collectively, the preceding alkene isomerization−
hydroboration chemistry employing HBPin (Section 2.2)
establishes the P(tBu)2-containing precatalyst C1 (or alter-
C
DOI: 10.1021/acs.organomet.6b00823
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 5. Isomerization−Hydroboration Cyclic
Scheme 6. Alkene Isomerization−Hydroboration Employing
a
Trisubstituted (1g and 1h) and Tetrasubstituted (1i)
1,3-Dimethyl-1,3-diaza-2-boracyclopentane (8)
a
Alkenes with HBPin
a
Isolated yields given; 18 h reaction times are unoptimized.
a
1
Estimated conversion to product on the basis of H and 13C{1H}
examples employing HBPin.23 The higher conversion to
product achieved by use of C8 versus C1 in reactions of 1a
or 1c with 8 was also reflected in related transformations
employing the isomeric terminal methylpentenes 1d and 1j
(Scheme 6). Notably, when using C8, 10a is observed as the
major product derived from 1d, whereas 10b is selectively
formed from 1j. By comparison, transformations involving the
internal alkene 1e proved challenging for both C1 and C8, with
poor (<25%) yields of the product alcohol(s) obtained in each
case.
The compatibility of benzo-1,3,2-diazaborolane (11) in
alkene isomerization−hydroboration chemistry employing C1
or C8 as precatalysts was subsequently evaluated (Scheme 7);
in contrast to the neat reaction mixtures employed in our work
described thus far, the physical properties of 11 required the
use of a cosolvent (C6D6). In transformations of 1-octene (1a)
NMR data. Catalyst loading, reaction times, 5:5′ ratio (5′ = product(s)
of hydroboration at secondary position(s); ND = ratio not
determined), and isolated yields (5:5′ mixtures, 6, or 7) are given in
parentheses; 1 h reaction times are unoptimized; otherwise for 1g the
time to maximum observed conversion is given.
natively C6 or C7) as being superior among the N-
phosphinoamidinate-supported cobalt catalysts featured in our
reactivity survey, in some cases providing alternative terminal
selectivity versus other reported precatalysts.8,9 Encouraged by
the favorable reactivity profile of C1, we sought to establish
hitherto unknown alkene isomerization−hydroboration chem-
istry involving boranes other than HBPin, as a complement to
catalyst design in the long-term quest to establish expanded
reaction scope and/or product selectivity. Our efforts to apply
catecholborane (HBCat) or 9-borabicyclo[3.3.1]nonane (9-
BBN) in place of HBPin in several alkene isomerization−
hydroboration test reactions employing C1 were unsuccessful,
affording complex product mixtures; in some cases, significant
uncatalyzed background reactivity was also observed. We then
turned our attention to the use of 1,3,2-diazaborolanes as
alternatives to HBPin, as these are known to be less reactive
than HBCat yet are useful in metal-catalyzed alkene hydro-
boration chemistry.22 We initially focused on transformations
involving 1,3-dimethyl-1,3-diaza-2-boracyclopentane (8) em-
ploying as precatalysts either C1 or the iron analogue C811,13 at
65 °C (Scheme 6). Notably, we found the alkene isomer-
ization−hydroboration products derived from 8 to be some-
what susceptible to hydrolysis; as a result, we opted to convert
each product to the corresponding alcohol (i.e., 9 and 10a,b)
prior to isolation and NMR analysis.
Scheme 7. Alkene Isomerization−Hydroboration Employing
a
Benzo-1,3,2-diazaborolane (11)
We were pleased to observe that in transformations of 1-
octene (1a) or the more challenging trans-4-octene (1c) with 8,
each of C1 and C8 afforded good to excellent conversion to the
terminal hydroboration product selectively, as evidenced by the
isolation of 1-octanol 9. The efficacy of C8 in this
transformation is noteworthy, in that the terminally selective
isomerization−hydroboration of internal alkenes employing
alternative iron precatalysts is limited to a small number of
a
1
Estimated conversion to product on the basis of H and 13C{1H}
NMR data. Reaction temperature or catalyst loading and isolated
yields given in parentheses; reaction times are unoptimized.
D
DOI: 10.1021/acs.organomet.6b00823
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
with 11, excellent conversion to the expected terminal
hydroboration product 12 was achieved at room temperature
with C1 (90%), whereas lower conversion (50%) was realized
by use of C8; in both cases, excellent conversion to 12 (>95%)
was observed at 65 °C. As expected, reactions involving trans-4-
octene (1c) proved to be more challenging; in contrast to
transformations involving 8 with 1c in which both C1 and C8
(3 mol %, 65 °C) performed reasonably well (Scheme 6), when
using 11 with 1c, only the use of C1 at 65 °C afforded high
conversion to the terminal alkene isomerization−hydroboration
product 12. Higher conversion and product selectivity at lower
catalyst loading were also achieved by use of C1 relative to C8
in related transformations employing 1d, 1e, and 1j in
combination with 11. Whereas the use of C8 (5 mol %)
afforded mixtures of the terminal hydroboration products 13a,b
(from 1d and 1e) with incomplete conversion, both high
conversion and selectivity for the less-hindered terminal
hydroboration product 13b resulted when using C1 (3 mol
%) in combination with each of 1d, 1e, and 1j. Notably, the site
of terminal hydroboration in 13b is divergent from that
achieved with C1 by use of HBPin (affording 3, Scheme 3) in
combination with 1d or 1e.
The divergent catalytic performance of C1 and C8 and the
product selectivity changes arising from the use of HPBin or
the 1,3,2-diazaborolane 11 highlight the complex interplay
between catalyst composition and borane structure in terms of
achieving high conversion and terminal selectivity in alkene
isomerization−hydroboration chemistry. To probe such
selectivity concepts further, the reaction of 2-ethyl-1-butene
(1k) with each of HBPin and benzo-1,3,2-diazaborolane (11)
in the presence of C1 (3 mol %) was examined (Scheme 8).
β-methylstyrene 17 to give products of type 18 or 19 were
unsuccessful (Scheme 9). In the transformations of 16, high
Scheme 9. Attempted Alkene Isomerization−Hydroboration
a
of the Silyl Ether 16 and β-Methylstyrene 17
a
1
Estimated conversions on the basis of H and 13C{1H} NMR data.
conversion of the starting materials (>75%) was observed in all
cases; however low yields of the putative products 18 were
observed throughout, along with small amounts of 2 (using
HBPin) and 12 (using 11) arising from apparent C−O bond
cleavage. No conversion of 17 was achieved by use of C1 or C8
with HPBin at 23 °C. Whereas analogous reactions conducted
at 65 °C using HBPin (with C1) or 11 (with C1 or C8)
resulted in substantial conversion of 17 (>50%), complex
product mixtures resulted in all cases, including the formation
of 1-phenylpropane (ca. 20−40%).
3. CONCLUSIONS
In summary, a series of structurally varied N-phosphinoamidi-
nate-ligated cobalt complexes have been prepared, character-
ized, and applied successfully, along with a related iron
complex, as precatalysts in the isomerization−hydroboration
of terminal, geminal, and internal alkenes. These trans-
formations proceed under mild conditions (23−65 °C; 1−5
mol % base metal loading), typically without cosolvent, and
with high selectivity across a broad spectrum of hindered
alkenes. Whereas the P(tBu)2-containing C1 in general
displayed optimal catalytic performance in our study, useful
reactivity profiles were also observed for some of the other
cobalt precatalysts disclosed herein, as well as the iron analogue
of C1 (i.e., C8). In this regard, variation of the N-
phosphinoamidinate ligand architecture appears to be well-
tolerated and can, along with metal selection, be exploited in
addressing alternative substrate classes and reaction types.
It is worthy of mention that all previous reports of terminal-
selective metal-catalyzed alkene isomerization−hydroboration
have employed pinacolborane. In addition to successful
transformations employing HBPin with C1, we also report
herein the first examples of alkene isomerization−hydro-
boration employing either 1,3-dimethyl-1,3-diaza-2-boracyclo-
pentane (8) or benzo-1,3,2-diazaborolane (11). We demon-
strate in selected cases that the successful use of 1,3,2-
diazaborolanes in this chemistry can provide a means of
obtaining previously unobserved terminal selectivity in the
isomerization−hydroboration of branched alkenes, in a manner
that is complementary to that achieved by use of HBPin.
Having established the utility of N-phosphinoamidinate
ligation in the base-metal-catalyzed isomerization−hydrobora-
tion of hindered alkenes and the utility of 1,3,2-diazaborolanes
in this chemistry, we are currently exploring the application of
such transformations involving increasingly complex substrate
molecules. We will report on our progress in this regard in due
course.
Scheme 8. Divergent Terminal Selectivity in Alkene
Isomerization−Hydroboration Employing HBPin and
a
Benzo-1,3,2-diazaborolane (11)
a
1
Estimated conversion to product on the basis of H and 13C{1H}
NMR data. Isolated yields given in parentheses; reaction times are
unoptimized.
Interestingly, while under the conditions employed high
conversion but poor selectivity was achieved when using
HPBin (leading to 14a,b), clean conversion to the less-
hindered terminal isomerization−hydroboration product 15
was achieved by use of 11, despite the potential for direct
hydroboration of the terminal alkene 1k in the absence of
alkene isomerization chemistry.
It is worthy of mention that some important scope
limitations were encountered in this isomerization−hydro-
boration chemistry. For example, our efforts to employ C1 or
C8 in combination with either HPBin or 11 in the terminal-
selective isomerization−hydroboration of the silyl ether 16 or
E
DOI: 10.1021/acs.organomet.6b00823
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
C39H58CoN3P: C, 71.10; H, 8.87; N, 6.38. Found: C, 70.92; H, 9.13;
N, 6.36.
4. EXPERIMENTAL SECTION
General Considerations. All manipulations were carried out
under an atmosphere of purified nitrogen gas within a glovebox or by
use of standard Schlenk techniques, unless otherwise noted. Dry,
deoxygenated solvents were used unless otherwise indicated. Diethyl
ether was dried over sodium/benzophenone and distilled under
nitrogen. All alkenes, pyridine, and C6D6 were degassed via three
repeated freeze−pump−thaw cycles and stored over activated 4 Å
molecular sieves prior to use. Pinacolborane was degassed via three
repeated freeze−pump−thaw cycles prior to use. All other solvents
and reagents were used as received. C1,13a C8,11 1,3-dimethyl-1,3-
diaza-2-boracyclopentane (8),22b and benzo-1,3,2-diazaborolane
(11)22a were prepared according to literature procedures. Precatalysts
C2−C5 were prepared as outlined in Scheme 2 as analytically pure red
crystalline solids from the appropriate N-phosphinoamidine10a,11,16 by
using synthetic protocols directly analogous to those described for the
synthesis of C1 without further optimization.13a Isolated yields thus-
obtained are C2 (58%), C3 (90%), C4 (11%), and C5 (17%). NMR
spectra were measured on AV-300 or AV-500 spectrometers operating
at 300 K. Infrared spectra were recorded as thin films between KBr
plates using an FT-IR spectrometer. Single-crystal X-ray data
collection was carried out by Dr. Robert MacDonald and Dr. Michael
J. Ferguson at the University of Alberta X-ray Crystallography
Laboratory (Edmonton, Alberta, Canada). Magnetic moments
(Evan’s method in C6D6) were determined at 300 K according to
literature procedures.20
Synthesis and Characterization of C7. LiN(SiMe3)2 (78.8 mg,
0.471 mmol) and CoCl2 (61.6 mg, 0.474 mmol) were added to a glass
vial containing a magnetic stir bar and a solution of the appropriate N-
phosphinoamidine (R = tBu, R′ = Ph,11 as in Scheme 2; 200 mg, 0.471
mmol) in Et2O (5 mL). After 16 h, pyridine (37.8 μL, 0.469 mmol)
and benzylmagnesium chloride (1.0 M, 0.470 mL, 0.470 mmol) were
added to the reaction mixture, and the resulting mixture was stirred
magnetically for 1 h. The reaction mixture was then filtered through a
Celite pad to remove the white precipitate that had formed, and the
red eluent solution was concentrated to ca. 1 mL in vacuo. Storage of
this concentrated solution at −35 °C afforded C7 as a red, crystalline
1
solid (118 mg, 44%). H NMR (300 MHz, C6D6): δ −95.1 (br, 2H),
−23.0 (7H), −12.7 (7H), −4.14 (23H), 8.03 (1H), 9.11 (2H), 15.5
(3H), 23.9 (br, 3H), 68.0 (br, 1H). μeff = 2.6 μB. FT-IR (cm−1): 2958,
2866, 1581, 1496, 1435, 1047, 858, 696. Anal. Calcd for
C34H47CoN2P: C, 71.19; H, 8.26; N, 4.88. Found: C, 71.11; H,
8.60; N, 4.62.
General Procedure for Catalytic Alkene Hydroboration. In an
inert-atmosphere glovebox, an oven-dried glass vial was charged with
the alkene substrate (0.4 mmol), borane (0.4 mmol), and C6D6 (200
μL, only when using 11). To this mixture was added precatalyst C1−
C8 (1−5 mol %), and the vial was capped. Following the specified
reaction time, the vial was removed from the glovebox and opened to
air. Ferrocene (0.0108 mmol; 100 μL of a stock solution in CDCl3)
was then added to the reaction mixture for use as a standard for NMR
quantification, followed by the addition of more CDCl3 (900 μL). The
contents of the vial were filtered through a small silica pad and
analyzed by use of 1H, 11B, and 13C{1H} NMR spectroscopic methods.
Isolated yields were obtained in a similar manner, whereby upon
exposure of the reaction mixture to air, diethyl ether (1 mL) was
added and the contents of the vial were filtered through a small silica
pad; the eluent was collected and concentrated in vacuo to afford the
isolated product. In this regard, the isolated yields are not optimized.
In the case of products 9 and 10a,b derived from 8, a modified
isolation protocol was employed, whereby the crude reaction mixture
was exposed to air, diluted with diethyl ether (2 mL), and treated with
NaOH (1.0 M in H2O, 1 mL) and H2O2 (30% in H2O, 1 mL). The
organic layer was extracted with diethyl ether (3 × 2 mL), dried over
MgSO4, and concentrated under reduced pressure. The residual
alcohol product was isolated and analyzed by use of 1H and 13C NMR
methods.
Crystallographic Solution and Refinement Details. Crystallo-
graphic data were obtained between 173(2) and 193(2) K on a Bruker
D8/APEX II CCD diffractometer equipped with a CCD area detector
using graphite-monochromated Mo Kα (α = 0.710 73 Å) radiation
(except for C2, where Cu Kα, α = 1.541 84 Å, radiation was used)
employing samples that were mounted in inert oil and transferred to a
cold gas stream on the diffractometer. Unit cell parameters were
determined and refined on all reflections. Data reduction, correction
for Lorentz polarization, and absorption correction were each
performed. Structure solution by direct methods and least-squares
refinement on F2 were used throughout. All non-hydrogen atoms were
refined with anisotropic displacement parameters. During the structure
solution process for C7, two hydrogen atoms (H51A and H51B)
attached to the C(51)H2−Ph group were each located in the difference
map and refined isotropically. Otherwise, all hydrogen atoms were
added at calculated positions and refined by use of a riding model
employing isotropic displacement parameters based on the isotropic
displacement parameter of the attached atom. For C7, the disordered
diethyl ether solvate was modeled in a satisfactory manner as two half-
occupied diethyl ether molecules, whereby all atoms (C71−C74 and
O1) were refined isotropically with an occupancy factor of 0.5.
1
Compound C2. H NMR (300 MHz, C6D6): δ −111.9 (4H),
−55.9 to −55.7 (overlapping signals, 7H), −43.9 (6H), −30.6 (6H),
−20.4 (6H), −17.9 (6H), 6.36 (overlapping signals, 19H), 15.0 (8H),
17.5 (3H), 36.3 (1H), 47.7 (2H), 112.7 (2H). μeff = 5.3 μB. FT-IR
(thin film, cm−1): 2906, 2850, 1464, 1427, 867. Anal. Calcd for
C45H70CoN3PSi2: C, 67.63; H, 8.83; N, 5.26. Found: C, 67.49; H,
8.75; N, 5.34.
Compound C3. 1H NMR (300 MHz, C6D6): δ −55.1 (2H), −42.4
(5H), −24.0 (2H), 0.86 (7H), 2.07 (2H), 13.8−12.7 (19H), 16.1
(6H), 19.2 (4H), 25.5 (2H), 35.1 (2H), 77.8 (3H). μeff = 4.2 μB. FT-
IR (cm−1): 3058, 2958, 1456, 1419, 831. Anal. Calcd for
C39H54CoN3PSi2: C, 65.89; H, 7.62; N, 5.91. Found: C, 65.29; H,
7.66; N, 5.90. Although these results in %C are outside the range
viewed as establishing analytical purity, they are provided to illustrate
the best values obtained to date.
Compound C4. 1H NMR (300 MHz, C6D6): δ −55.4 (6H), −55.2
(1H), −44.1 (18H), 1.23 (2H), 8.42 (18H), 12.1 (6H), 14.8 (2H),
34.9 (1H), 106.2 (2H). μeff = 5.0 μB. FT-IR (cm−1): 2958, 2970, 1620,
1593, 1479, 1360, 866. Anal. Calcd for C33H56CoF2N3PSi2: C, 58.38;
H, 8.31; N, 6.19. Found: C, 57.95; H, 8.34; N, 5.82. Although these
results in %C are outside the range viewed as establishing analytical
purity, they are provided to illustrate the best values obtained to date.
Compound C5. 1H NMR (300 MHz, C6D6): δ −65.7 (1H), −45.9
(6H), −40.4 (15H), 6.50 (18H), 15.7 (10H), 19.7 (4H), 33.9 (2H),
44.8 (4H), 97.7 (3H). μeff = 5.4 μB. FT-IR (cm−1): 2937, 2865, 1475,
1417, 864. Anal. Calcd for C32H63CoN4PSi2: C, 59.13; H, 9.77; N,
8.62. Found: C, 59.22; H, 9.63; N, 8.71.
Synthesis and Characterization of C6. LiN(SiMe3)2 (78.8 mg,
0.471 mmol) and CoCl2 (61.6 mg, 0.474 mmol) were added to a glass
vial containing a magnetic stir bar and a solution of the appropriate N-
phosphinoamidine (R = tBu, R′ = Ph,11 as in Scheme 2; 200 mg, 0.471
mmol) in Et2O (5 mL). Magnetic stirring at room temperature was
initiated. After 16 h, lithium 2,6-diisopropylphenylamide (86.8 mg,
0.474 mmol) was added to the reaction mixture, and the resulting
mixture was stirred magnetically for 1 h. The reaction mixture was
then filtered through a Celite pad to remove the precipitate that had
formed, and the purple eluent solution was concentrated to ca. 1 mL in
vacuo. Storage of this concentrated solution at −35 °C afforded C6 as
1
a purple crystalline solid (201 mg, 0.305 mmol, 65%). H NMR (300
MHz, C6D6): δ −2.95 (8H), 0.09 (br, 26H), 2.17 (br, 4H), 5.21 (br,
12H), 6.87 (2H), 7.80 (3H), 10.5 (1H), 28.2 (br, 2H). μeff = 2.3 μB.
FT-IR (cm−1): 2961, 2868, 1589, 1496, 1436, 698. Anal. Calcd for
F
DOI: 10.1021/acs.organomet.6b00823
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(11) Ruddy, A. J.; Kelly, C. M.; Crawford, S. M.; Wheaton, C. A.;
Sydora, O. L.; Small, B. L.; Stradiotto, M.; Turculet, L. Organometallics
2013, 32, 5581−5588.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.6b00823.
■
S
(12) The aformentioned report by Obligacion and Chirik8 appeared
while our C1-catalyzed transformations were under internal legal
review. For our issued patent on this chemistry, see: Sydora, O. L.;
Stradiotto, M.; Turculet, L.; Small, B. L.; Coffin, R. C.; Bischof, S. M.
PCT Int. Appl. U.S. 009120826 B1, 2015.
Crystallographic data (CIF)
Complete experimental procedures, characterization
data, NMR spectra for new N-phosphinoamidine ligands
and catalytic reaction products, as well as tabulated
crystallographic metrical data for C1−C7 (PDF)
(13) (a) Ruddy, A. J.; Sydora, O. L.; Small, B. L.; Stradiotto, M.;
Turculet, L. Chem. - Eur. J. 2014, 20, 13918−13922. (b) Ruddy, A. J.;
Sydora, O. L.; Small, B. L.; Stradiotto, M.; Turculet, L. Chem. - Eur. J.
2016, 22, 12589.
(14) A single example of terminal-selective alkene isomerization−
hydroboration with HBPin in the presence of (Ph3P)3CoH(N2) (5
mol %) is featured in the following report: Scheuermann, M. L.;
Johnson, E. J.; Chirik, P. J. Org. Lett. 2015, 17, 2716−2719.
(15) By comparison, the commonly employed hydroboration reagent
catecholborane (HBCat) has proven incompatible with rhodium-
catalyzed alkene isomerization−hydroboration chemistry: Evans, D.
A.; Fu, G. C.; Anderson, B. A. J. Am. Chem. Soc. 1992, 114, 6679−
6685.
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: sydorol@cpchem.com.
*E-mail: mark.stradiotto@dal.ca.
*E-mail: laura.turculet@dal.ca.
ORCID
(16) Experimental details regarding the synthesis of the new N-
phosphinoamidines employed in the preparation of C2−C5, as well as
tabulated crystallographic metrical data for C1−C7 (Table S1), are
provided in the Supporting Information.
Mark Stradiotto: 0000-0002-6913-5160
Notes
The authors declare no competing financial interest.
(17) For a recent report in which the crystallographic character-
ization of a diamagnetic complex exhibiting a Co···CH agostic
interaction (Co···H, 1.72(2) Å) is disclosed, see: Murugesan, S.;
Stoger, B.; Pittenauer, E.; Allmaier, G.; Veiros, L. F.; Kirchner, K.
Angew. Chem., Int. Ed. 2016, 55, 3045−3048.
ACKNOWLEDGMENTS
■
We thank Chevron Phillips Chemical Co. for supporting this
research and permission to publish. M.S. and L.T. are also
grateful to the NSERC of Canada and Dalhousie University for
their support of this work. Dr. Michael Lumsden (NMR-3,
Dalhousie) is thanked for ongoing technical assistance in the
acquisition of NMR data, and Mr. Xiao Feng (Maritime Mass
Spectrometry Laboratories, Dalhousie) is thanked for technical
assistance in the acquisition of mass spectrometric data.
(18) Bleeke, J. R.; Burch, R. R.; Coulman, C. L.; Schardt, B. C. Inorg.
Chem. 1981, 20, 1316−1918.
(19) Trost, B. M.; Czabaniuk, L. C. Angew. Chem., Int. Ed. 2014, 53,
2826−2851.
(20) Bain, G. A.; Berry, J. F. J. Chem. Educ. 2008, 85, 532−536.
(21) (a) Leung, W. P.; Lee, H. K.; Weng, L. H.; Zhou, Z. Y.; Mak, T.
C. W. J. Chem. Soc., Dalton Trans. 1997, 779−783. (b) Figgis, B. N.;
Nyholm, R. S. J. Chem. Soc. 1954, 12−15.
(22) For selected reports documenting the use of 8 and 11 as
alternatives to HBPin and HBCat in metal-catalyzed alkene hydro-
boration chemistry, see: (a) Motry, D. H.; Brazil, A. G.; Smith, M. R. J.
Am. Chem. Soc. 1997, 119, 2743−2744. (b) Molander, G. A.; Pfeiffer,
D. Org. Lett. 2001, 3, 361−363. (c) Hadebe, S. W.; Robinson, R. S.
Eur. J. Org. Chem. 2006, 2006, 4898−4904.
(23) (a) Obligacion, J. V.; Chirik, P. J. Org. Lett. 2013, 15, 2680−
2683. (b) Zheng, J.; Sortais, J.-B.; Darcel, C. ChemCatChem 2014, 6,
763−766. (c) Tseng, K. N. T.; Kampf, J. W.; Szymczak, N. K. ACS
Catal. 2015, 5, 411−415.
REFERENCES
■
(1) (a) Davidson, M.; Hughes, A. K.; Marder, T. B.; Wade, K.
Contemporary Boron Chemistry; Royal Society of Chemistry: Cam-
bridge, UK, 2000. (b) Brown, H. C. Boranes in Organic Chemistry;
Cornell University Press: Ithaca, NY, 1972.
(2) (a) Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457−2483.
(b) Jana, R.; Pathak, T. P.; Sigman, M. S. Chem. Rev. 2011, 111, 1417−
1492. (c) Suzuki, A. Angew. Chem., Int. Ed. 2011, 50, 6722−6737.
(3) Brown, H. C.; Subba Rao, B. C. J. Am. Chem. Soc. 1956, 78,
5694−5695.
(4) (a) Beletskaya, I.; Pelter, A. Tetrahedron 1997, 53, 4957−5026.
(b) Crudden, C. M.; Edwards, D. Eur. J. Org. Chem. 2003, 2003,
4695−4712. (c) Vogels, C. M.; Westcott, S. A. Curr. Org. Chem. 2005,
9, 687−699. (d) Geier, S. J.; Westcott, S. A. Rev. Inorg. Chem. 2015, 35,
69−79.
(5) Vasseur, A.; Bruffaerts, J.; Marek, I. Nat. Chem. 2016, 8, 209−219.
(6) Larionov, E.; Li, H. H.; Mazet, C. Chem. Commun. 2014, 50,
9816−9826.
(7) (a) Pereira, S.; Srebnik, M. J. Am. Chem. Soc. 1996, 118, 909−
910. (b) Yamamoto, Y.; Fujikawa, R.; Umemoto, T.; Miyaura, N.
Tetrahedron 2004, 60, 10695−10700. (c) Edwards, D. R.; Crudden, C.
M.; Yam, K. Adv. Synth. Catal. 2005, 347, 50−54. (d) Hadebe, S. W.;
Robinson, R. S. Tetrahedron Lett. 2006, 47, 1299−1302. (e) Lata, C. J.;
Crudden, C. M. J. Am. Chem. Soc. 2010, 132, 131−137.
(8) Obligacion, J. V.; Chirik, P. J. J. Am. Chem. Soc. 2013, 135,
19107−19110.
(9) Palmer, W. N.; Diao, T. N.; Pappas, I.; Chirik, P. J. ACS Catal.
2015, 5, 622−626.
(10) (a) Sydora, O. L.; Jones, T. C.; Small, B. L.; Nett, A. J.; Fischer,
A. A.; Carney, M. J. ACS Catal. 2012, 2, 2452−2455. (b) Sydora, O.
L.; Carney, M.; Small, B. L.; Hutchison, S.; Gee, J. C. PCT Int. Appl.
WO 082192 A1, 2011.
G
DOI: 10.1021/acs.organomet.6b00823
Organometallics XXXX, XXX, XXX−XXX