Selectivity effects in zirconium-catalyzed heterodehydrocoupling reactions of phosphines

Article abstract of DOI:10.1080/10426507.2015.1128933

Zirconium compounds are known to dehydrocouple phosphines catalytically. An exploration of the factors that may promote selective heterodehydrocoupling was performed, revealing that steric factors are important but do not provide substantial selectivity. It was observed that ⁵-(Me₃SiNCH₂CH₂)₂N(CH₂CH₂NSiMe₂CH₂)Zr (1) may be sufficiently Lewis acidic to perform Lewis acid or frustrated Lewis pair catalysis.

Full text of DOI:10.1080/10426507.2015.1128933

Phosphorus, Sulfur, and Silicon and the Related Elements
ISSN: 1042-6507 (Print) 1563-5325 (Online) Journal homepage: http://www.tandfonline.com/loi/gpss20
Selectivity effects in zirconium-catalyzed
heterodehydrocoupling reactions of phosphines
Michael B. Ghebreab, Sierra Costanza & Rory Waterman
To cite this article: Michael B. Ghebreab, Sierra Costanza & Rory Waterman (2016)
Selectivity effects in zirconium-catalyzed heterodehydrocoupling reactions of phosphines,
Phosphorus, Sulfur, and Silicon and the Related Elements, 191:4, 668-670, DOI:
10.1080/10426507.2015.1128933
To link to this article: http://dx.doi.org/10.1080/10426507.2015.1128933
Accepted author version posted online: 14
Jan 2016.
Submit your article to this journal
Article views: 19
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=gpss20
Download by: [New York University]
Date: 09 April 2016, At: 05:22

PHOSPHORUS, SULFUR, AND SILICON
, VOL. , NO. , –
http://dx.doi.org/./..
Selectivity effects in zirconium-catalyzed heterodehydrocoupling reactions of
phosphines
Michael B. Ghebreab, Sierra Costanza, and Rory Waterman
Department of Chemistry, University of Vermont, Burlington, VT, USA
ABSTRACT
ARTICLE HISTORY
Received  December 
Accepted  December 
Zirconium compounds are known to dehydrocouple phosphines catalytically. An exploration of
the factors that may promote selective heterodehydrocoupling was performed, revealing that
steric factors are important but do not provide substantial selectivity. It was observed that κ5-
(Me3SiNCH2CH2)2N(CH2CH2NSiMe2CH2)Zr (1) may be sufficiently Lewis acidic to perform Lewis acid or
frustrated Lewis pair catalysis.
KEYWORDS
Phosphine; dehydrocoupling
catalysis; zirconium;
diphosphine
GRAPHICAL ABSTRACT
Introduction
Despite a steady effort in metal-catalyzed dehydrocoupling reac-
tions involving phosphines, a variety of challenges remain.1,2 In
particular, selectivity in dehydrocoupling catalysis is tantalizing.
There are examples of selective heterodehydrocoupling reac-
tions between silanes or germanes and phosphines using group
4 metal catalysts.3,4 In those reactions, where catalysis is likely to
involve σ-bond metathesis steps, 5–7 selectivity appears to arise
from preferential formation of phosphido compound interme-
diates and is amplified by the relative stability of a heavier main
group element in the β-position of the σ-bond metathesis tran-
sition state (Scheme 1).8
We have observed unusual selectivity in the heterodehydro-
coupling of primary phosphines using zirconium compounds
such as κ5-(Me3SiNCH2CH2)2N(CH2CH2NSiMe2CH2)Zr (1)
or (N3N)ZrMe (N3N) = N(CH2CH2NSiMe2CH2)33–).6 In par-
ticular, it was found that reaction of PhPH2 and CyPH2 (Cy =
Scheme . Heterodehydrocoupling of silanes and germanes with phosphines cat-
C6H11) gave a nonstatistical excess of PhHP–PHCy.5 What is alyzed by 1.
even more remarkable about this reaction is that the homocou-
pling of the CyPH2 substrate appears to be completely inhibited
by the PhPH2 as supported by the observation of (CyPH)2 for-
mation only after PhPH2 is completely consumed.5 During the
catalysis, only (N3N)ZrPHPh (2) was detected by NMR spec-
troscopy, leading to the tentative hypothesis that this compound
is more stable than (N3N)ZrPHCy (3, Scheme 2). However,
structural and computational data provide no support for that
notion and no explanation for the observation, where computa-
tional data, for example, predicted virtually identical Zr–P bond
dissociation energies for these two compounds.3,9
The possibility that selectivity in heterodehydrocoupling
reactions may be governed by tunable factors at the phosphine
CONTACT Rory Waterman
rory.waterman@uvm.edu,
Department of Chemistry, University of Vermont, Burlington, VT , USA.
Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/gpss.
©  Taylor & Francis Group, LLC

PHOSPHORUS, SULFUR, AND SILICON
669
However, much of the deuterium is found on the trimethylsilyl
substituents of the N3N ligand. This observation suggests equi-
librium formation of 1 followed by reversible formation of 3-
d2. Equilibrium access to less stable (N3N)ZrX compounds from
more stable derivatives via cyclometalation and formation of 1
has been demonstrated.10
Despite the lack of insight these experiments as well as prior
structure and computational study seem to provide, the appar-
ent selectivity for the cyclohexyl-substituted substrate in the
β-position of the transition state (Scheme 2) provided a good
lead for further study. It was hypothesized that secondary phos-
phines would amplify steric factors and potentially afford greater
selectivity. Methyl substitution (e.g., PhPH2 to PhMePH, etc.) is
appropriate in this system because β-hydrogen elimination from
(N3N)ZrX compounds is not favorable.11
The heterodehydrocoupling of phosphines using 1 as the
catalyst was tested with one-to-one mixtures of phosphines
(RR’PH) (R = Cy or Ph; R’ = H, Me, Cy, Ph), in an attempt
to selectively form heterodehydrocoupled products (Equa-
tion (1), Table 1). An equimolar mixture of CyMePH and
PhMePH was reacted with catalytic 1, and the homocoupled
product (PhMeP)2 was obtained while the heterodehydrocou-
pling product PhMeP–PMeCy was not observed (entry 1). The
attempted heterodehydrocoupling reaction of PhMePH and
Ph2PH resulted first in the formation of (PhMeP)2 followed by
dehydrocoupling of Ph2PH to diphosphine, (Ph2P)2, only after
PhMePH is consumed (entry 2). The anticipated heterodehy-
drocoupling product was not observed.
Scheme . Heterodehydrocoupling of PhPH and CyPH catalyzed by 1 that illus-
trates the proposed origin of selectivity.
substrates was intriguing, and a rational analysis of the effects
were undertaken. If this system is consistent with others that
engage in σ-bond metathesis, then electronic effects should not
be important.8 However, an electronic effect may indicate a
mechanism other than σ-bond metathesis or afford new insight
into that process.
Results and discussion
The observation of exclusively compound 2 in the catalytic
heterodehydrocoupling of PhPH2 and CyPH2 was perplexing.
Indeed, treatment of one-half equivalent of 1 with a 1:1 mix-
ture of PhPH2 and CyPH2 resulted in the sole formation of 2
in benzene-d6 solution as monitored by 31P{1H} NMR spec-
(1)
The pattern observed from the attempted heterodehydrocou-
troscopy. The selectivity is not kinetic. Reaction of isolated sam- pling reactions of primary and secondary phosphine mixtures
ples of compound 3 with PhPH2 in benzene-d6 solution afford was that reactions of PhMePH with PhPH2 or CyPH2 were more
compound 2. Indeed, compound 3 can be completely converted facile than those of CyMePH with PhPH2 or CyPH2 (entries 4–
to 2 by one equivalent of PhPH2. Compound 3 remains, how- 7). In both sets of reactions, it was found that the homocoupling
ever, kinetically accessible. If isolated samples of 2 are treated products of the least sterically hindered substrate prevailed.
with CyPD2, then the deuterium is gradually exchanged from The most facile heterodehydrocoupling reaction, PhMePH and
phosphorus. It is known that compound 2 engages in degenera- PhPH2–, gave PhMeP–PHPh, based on a distinctive AB splitting
tive ligand and H/D exchange rapidly at ambient temperature.5 pattern in the 31P NMR spectrum at δ = −49.8 (d, JPP = 215 Hz),
Table . Summary of attempted dehydrocoupling catalysis.a
Entry
Substrates
Major product
Minor products
PhMeP–PMeCy
Unobservedb

PhMePH
PhMePH
CyMePH
CyPH
CyPH
PhPH
CyMePH
PhPH
(PhMeP)
(PhMeP)
none
(PhMeP)
(CyP)
(CyMeP)
PhMeP–PPh
(CyMeP)






(PhP)
PhMePH
CyMePH
PhMePH
CyMePH
CyPH
(CyP), CyHP–PMePh
CyHP–PMeCy
PhHP–PMeCy
(PhP)
PhPH
PhPH
(PhP)
PhHP–PMePh, (CyMeP)
d

PhHP–PHCy c
(CyP)
a
All reactions were run with equal amounts of phosphine substrates and  mol % of 1 in benzene-d solutions. Reactions were monitored by P{H}NMR spectroscopy,
and the relative distributions were unchanged until complete conversion unless noted. Cy = CH.
Compounds identified here are potentially anticipated dehydrocoupling products, but no evidence was detected by H or P NMR spectroscopy for their formation.
From Ref. .
b
c
d
Observed after complete consumption of PhPH. See Ref.  for details.

670
M. B. GHEBREAB ET AL.
and δ = −54.0 (d, JPH = 215 Hz) as well as the (PhMeP)2, which
is a known dehydrocoupling product with 1 and PhMePh (entry
6).11 However, the cyclophosphine (PhP)5, associated with dehy-
drocoupling of PhPH2 by 1, was the primary product of the reac-
tion (entry 6). The heterodehydrocoupling reaction involving
bulkier phosphines, CyPH2 and CyMePH, resulted in predomi-
nate formation of (CyP)4 as the major dehydrocoupling product.
It is known that some diphosphines (PHR)2 thermally decom-
pose to stable (RP)n rings.12,13 Our current evidence suggests
that (CyP)4 and (PPh)5 are forming directly because the diphos-
phines are not observed. Thereafter, the secondary phosphines
forms (CyMeP–PHCy) at δ = −46.48 (d, JPP = 206 Hz) and δ =
−53.76 (d, JPH = 206 Hz) in the 31P{1H} NMR spectrum. Inter-
estingly, formation of (CyP(H)[CyP]nP(H)Cy) in the dehydro-
coupling of CyPH2 by Sn(IV) complexes was reported as minor
product by Wright and coworkers,14,15 but that was not observed
in this reaction.
Thus, reactions with primary phosphines and sterically
encumbered secondary phosphines yield rings. This observa-
tion is in contrast to the dehydrocoupling of primary phos-
phines with 1 under the same conditions, which yield diphos-
phine products.5,16 Compound 1 can produce rings, but only
under more forcing conditions.5 These observations are, how-
ever, consistent with frustrated Lewis acid catalyzed dehydrcou-
pling of phosphines. In a recent report by Stephan and cowork-
ers, it was found that PhPH2 was dehydrocoupled by 10 mol%
B(C6F4H)3 to P5Ph5 exclusively.17 We have accrued evidence to
support Lewis acidic reactivity at 1.18 The possibility of FLP-like
reactivity involving 1 is intriguing, though there is not enough
data to fully support such a conclusion at present.
In all reactions, resonances attributed to known
(N3N)ZrPRR’ derivatives were identified in NMR spec-
tra.4,6,8 The selective formation of phosphido complexes
suggests that there may be some preferences for steric and
electronic properties of the primary and secondary phosphines.
However, most selectivity appeared to be steric where, for
example, (N3N)ZrPPhMe was formed over (N3N)ZrPCyMe
(4). Attempts to independently prepare and isolate 4 were not
successful. Though in the attempted catalytic dehydrocoupling
of CyMePH with 1 (entry 2), a new resonance at δ = 38.2
was observed in the 31P NMR spectrum. It is hypothesized
that P–H activation of CyMePH by compound 1 is disfa-
vored because this phosphine is too sterically encumbered
and that the possible observation of 4 in equilibrium is only a
function of high concentrations of CyMePH, a phenomenon
that has been seen for reaction of 1 with bulky primary
phosphines.4,6
Concluding remarks
The steric and electronic factors that may promote more selec-
tive phosphine heterodehydrocoupling by (N3N)Zr-catalyst
were probed. Substrates that were more significantly sterically
encumbered did not promote additional selectivity because they
would fail to have sufficient reactivity. Therefore, greater selec-
tivity than the original system was not obtained. However, these
observations do imply that compound 1 may be able to act as the
Lewis acid partner in FLP chemistry based in the unusual prod-
uct formation in the dehydrocoupling of PhPH2 in the presence
of secondary phosphines.
Funding
This work was supported by the U.S. National Science Foundation (CHE-
0747612 and CHE-1265608 to R.W.). R.W. thanks the Alexander von Hum-
boldt Foundation for a Research Fellowship, and S.C. thanks the Project
SEED endowment administered by the American Chemical Society for
summer support.
References
1. Waterman, R. Curr. Org. Chem. 2008, 12, 1322.
2. Waterman, R. Curr. Org. Chem. 2012, 16, 1313.
3. Roering, A. J.; MacMillan, S. N.; Tanski, J. M.; Waterman, R. Inorg.
Chem. 2007, 46, 6855.
4. Shu, R.; Hao, L.; Harrod, J. F.; Woo, H.-G.; Samuel, E. J. Am. Chem.
Soc. 1998, 120, 12988.
5. Waterman, R. Organometallics 2007, 26, 2492.
6. Tilley, T. D. Comment. Inorg. Chem., 1990, 10, 37.
7. Tilley, T. D. Acc. Chem. Res., 1993, 26, 22.
8. Waterman, R. Organometallics 2013, 32, 7249.
9. Roering, A. J.; Maddox, A. F.; Elrod, L. T.; Chan, S. M.; Ghebreab, M. B.;
Donovan, K. L.; Davidson, J. J.; Hughes, R. P.; Shalumova, T.; MacMil-
lan, S. N.; Tanski, J. M.; Waterman, R. Organometallics 2009, 28, 573.
10. Roering, A. J.; Leshinski, S. E.; Chan, S. M.; Shalumova, T.; MacMillan,
S. N.; Tanski, J. M.; Waterman, R. Organometallics 2010, 29, 2557.
11. Ghebreab, M. B.; Newsham, D. K.; Waterman, R. Dalton Trans. 2011,
40, 7683.
12. Baudler, M.; Gruner, C.; Tschäbunin, H.; Hahn, J. Chem. Ber. 1982,
115, 1739.
13. Blaurock, S.; Hey-Hawkins, E. Zeit. Anorg. Allg. Chem. 2002, 628, 37.
14. Naseri, V.; Less, R. J.; Mulvey, R. E.; McPartlin, M.; Wright, D. S. Chem.
Commun. 2010, 46, 5000.
15. Erickson, K. A.; Dixon, L. S. H.; Wright, D. S.; Waterman, R. Inorg.
Chim. Acta 2014, 422, 141.
16. Ghebreab, M. B.; Shalumova, T.; Tanski, J. M.; Waterman, R. Polyhe-
dron 2010, 29, 42.
17. Dobrovetsky, R.; Takeuchi, K.; Stephan, D. W. Chem. Commun. 2015,
51, 2396.
18. Leshinski, S.; Waterman, R. 2015, in preparation.