Mechanistic Investigations of Reactions of the Frustrated Lewis Pairs (Triarylphosphines/B(C6F5)3) with Michael Acceptors

Article abstract of DOI:10.1021/acs.orglett.6b03868

Frustrated Lewis pair (FLP)-catalyzed reduction of Michael acceptors is a challenging reaction that proceeds with specific FLP structures. Kinetics and equilibrium of the reactions of two phosphines (Ar₃P), namely tri(1-naphthyl)phosphine and tri(o-tolyl)phosphine, are reported with reference electrophiles. The reason for the failure of the FLPs (Ar₃P/B(C₆F₅)₃) to reduce activated alkenes under H₂ pressure is shown to be a hydrophosphination process that inhibits the reduction reaction. Kinetic and thermodynamic factors controlling both pathways are discussed in light of Mayr's free linear energy relationships.

Full text of DOI:10.1021/acs.orglett.6b03868

Letter
pubs.acs.org/OrgLett
Mechanistic Investigations of Reactions of the Frustrated Lewis Pairs
(Triarylphosphines/B(C6F5)3) with Michael Acceptors
Jonathan Dupre, Annie-Claude Gaumont,* and Sami Lakhdar*
́
Normandie Universite,
Normandie, 6, Boulevard du Marec
́
Laboratoire de Chimie Molec
́ ́
ulaire et Thio-organique, CNRS-UMR 6507, ENSICAEN, Universite de Caen
́
hal Juin, Caen 14000, France
S
* Supporting Information
ABSTRACT: Frustrated Lewis pair (FLP)-catalyzed reduction of Michael
acceptors is a challenging reaction that proceeds with specific FLP
structures. Kinetics and equilibrium of the reactions of two phosphines
(Ar3P), namely tri(1-naphthyl)phosphine and tri(o-tolyl)phosphine, are
reported with reference electrophiles. The reason for the failure of the FLPs
(Ar3P/B(C6F5)3) to reduce activated alkenes under H2 pressure is shown to
be a hydrophosphination process that inhibits the reduction reaction. Kinetic and thermodynamic factors controlling both
pathways are discussed in light of Mayr’s free linear energy relationships.
eterolytic bond splitting of inert molecules is undoubt-
edly one of the most challenging processes that continues
boron substituents.6 On the other hand, Paradies et al. have
isolated and characterized various FLPs, for instance,
fluoroarylphosphines/B(C6F5)3.7 Importantly, it was shown
that the acidity of the phosphonium ions plays a pivotal role in
the hydrogenation of polarized CC double bonds. Indeed,
acidic phosphonium ions activate alkenes to the corresponding
carbocations that can then be reduced by the borohydrides. For
example, due to the high Brønsted acidity of its phosphonium
salt, tri(1-naphthyl)phosphine ((C10H7)3P) 1a (pKa (C2H4Cl2)
= 6.7)7 has been shown to be a good Lewis base partner for
B(C6F5)3 to reduce a large variety of electron-rich olefins under
H2 pressure.8
On the basis of these studies and especially on those of
Berionni et al., who demonstrated that phosphonium and
ammonium borohydrides react smoothly with quinone
methides, α,β-unsaturated iminium ions, and benzylidene
malononitriles under stoichiometric conditions, it is reasonable
to assume that common Michael acceptors can be reduced
under catalytic FLP conditions.6 However, FLP reduction of
Michael acceptors bearing oxygen atoms turned out to be
problematic as only specific FLPs are effective in this reaction.
In this context, Paradies stated “FLP-catalyzed hydrogenation
of functionalized, especially oxygen containing, groups is highly
challenging because of the strong Lewis basic character
combined with insufficient steric shielding. Consequently,
catalyst inhibition is one of the most problematic issues in
FLP-mediated reductions of enones, malonates, or nitro-
olefins”.1g
H
to stimulate a large spectrum of research in modern organic
chemistry.1 While this type of activation has long been
restricted to transition metals,2 it has recently been
demonstrated that pure organic molecules are effective in
catalyzing such a reaction.2 For instance, Stephan, Erker, and
Paradies, to name a few, have shown that the combination of
sterically hindered Lewis acids (typically B(C6F5)3) with Lewis
bases (phosphines, carbenes and amines) results in the
formation of the so-called frustrated Lewis pairs (FLPs).3
These complexes have proven to be efficient in cleaving H2
heterolytically to form borohydrides that are nucleophilic
enough to reduce a large variety of alkenes.4
From a simplified mechanistic point of view, the catalytic
reduction of alkenes depends on the efficiency of FLPs to
cleave H2 (step 1, Scheme 1) and also on the ability of the
Scheme 1. General Mechanistic Scheme of FLP Catalytic
a
Reductions of Alkenes
a
LA and LB denote Lewis acid and Lewis base, respectively.
formed borohydride salts to reduce alkenes (step 2). While the
heterolytic splitting process of H2 (step 1) has been extensively
investigated, both computationally and experimentally,5 mech-
anistic understanding of the reduction step remained scantily
explored. To the best of our knowledge, only two reports
dealing with this aspect have been reported. Berionni et al.
quantified the hydricity of phosphonium and ammonium
borohydrides by using the well-known Mayr’s free energy
relationship (vide infra). They demonstrated that the hydride-
donating ability of these salts depends on the nature of the
In order to gain more mechanistic information about the
reasons for the failure of Ar3P/B(C6F5)3 FLPs in the reduction
of Michael acceptors, we decided to study kinetics and
equilibrium of their reactions with phosphines 1a,b. Because
1a,b are known to be effective in cleaving H2 in the presence of
Received: December 27, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b03868
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
B(C6F5)3 and the resulting phosphonium borohydrides 2a,b−
HB(C6F5)3 salts have been isolated and characterized, we
envisioned investigating their reactions with Michael acceptors
(Scheme 2).8,9
ions 6a−c (see Figure 2). When a large excess of phosphines
1a,b is combined with benzhydrylium ions 6a−c, mono-
exponential decays of the absorbance of 6a−c are observed
(Figure 1). In line with previous studies by Mayr et al. on the
Scheme 2. Structures of Phosphines 1a,b and the
Corresponding Phosphonium Salts 2a,b−HB(C6F5)3
To begin, the catalytic reduction of the methylvinylketone 3
with 1a,b−B(C6F5)3 (10 mol %) has been selected as a
benchmark reaction. As shown in Scheme 3, both reactions did
Figure 1. Plot of the absorbance at 586 nm versus time for the
reactions of 6a (1.35 × 10−4 mol/L) with 1a at different initial
concentrations.
Scheme 3. Catalytic Reduction of Methylvinyl Ketone 3 with
1a,b/B(C6F5)3 (10 mol %) under H2 Atmosphere (Top).
Reaction of Phosphonium Ions 2a,b−HB(C6F5)3 with
Methylvinyl Ketone 3 in CH2Cl2 at Room Temperature
(Bottom)
reactivity of tertiary phosphines, reversible processes have also
been observed when phosphines 1a,b were combined with 6a−
c.11
Plotting rate constants kobs, derived from monoexponential
decays, against concentrations of nucleophiles 1a,b gives linear
correlations where the slopes yielded second-order rate
constants for the reaction of 1a,b with 6a−c. Figure 2 shows
not proceed under atmospheric pressure of H2 at room
temperature, and only traces of the reduction adducts 4 can be
detected within 12 h. However, 31P NMR analysis of the crude
mixtures showed the appearance of new peaks, which have been
assigned to the phosphination adducts 5a,b. The reaction
outcomes have been found to be counterion independent as
almost similar yields have been obtained when reactions have
been performed when triflate phosphonium salts have been
used instead of the trihydroborate salt(see the Supporting
Information). In order to confirm the structures of 5a,b,
phosphonium borohydrides 2a,b−HB(C6F5)3 have been
synthesized according to protocols previously described in
the literature.8,9 Interestingly, when 2a,b−HB(C6F5)3 were
combined with 1 equiv of 3, the expected reduced adducts 4
were observed along with the hydrophosphination adduct 5a,b
in 61:39 and 47:53 ratios, respectively (Scheme 3).
Figure 2. Plot of log k2 vs E for reactions of phosphines 1a (triangles)
and 1b (circles) with benzhydrylium ions 6a,c.
linear correlations between the logarithms of the second order
rate constants (log k2) and the E parameters of the
benzhydrylium ions 6a−c, as required by the eq 1. The slopes
of the correlation lines give the nucleophile-specific sensitivity
parameters sN, and the intercepts on the abscissa yield the
nucleophilicity parameters N, which are gathered in Table 1.
Table 1 shows that 1a is about 18 times more reactive than
1b. This can presumably be due to the steric hindrance of the
tolyl group due to its high Tolman angle (θ[1b] = 194°).12
As the reactions of phosphines 1a,b with benzhydrylium ions
proceeded incompletely, we also studied the corresponding
equilibrium constants (K) by UV−vis spectroscopy. The Lewis
basicity of 1a,b has been quantified by using an approach
recently introduced by Mayr, who found that the K values for
In order to gain mechanistic information into these insightful
observations, we decided to quantify nucleophilicity and carbon
Lewis basicity of phosphines 1a,b. Indeed, Mayr et al.
demonstrated, during the last two decades, the capability of
the free linear energy relationship (eq 1) to describe kinetics of
numerous nucleophile−electrophile combinations.10
log k(20°C) = sN(E + N)
(1)
In this equation, while electrophiles are characterized by a
single parameter E, nucleophiles are described by two
parameters: nucleophilicity, N, and a nucleophile-specific
susceptibility parameter, sN. Based on this approach, we studied
kinetics of the reactions of phosphines 1a,b with benzydrylium
B
DOI: 10.1021/acs.orglett.6b03868
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 1. Second-Order Rate Constants, Nucleophilicity, and
Lewis Basicity for Reactions of 1a,b with Benzhydrylium
Ions 6a−c in MeCN or CH2Cl2 at 20 °C
Scheme 4. Proposed Mechanisms for Phosphine−
(B(C6F5)3) Hydrogenation of Michael Acceptors
phosphines
electrophiles
k2 (M−1 s−1
)
N, sN
LB
a
1a
6a
6b
6c
6a
6b
1.11 × 104
5.23 × 103
2.05 × 103
1.31 × 103
2.87 × 102
9.48 × 101
13.17, 0.44
8.62
b
1b
8.40, 0.68
7.77
6c
a
b
In acetonitrile. In dichloromethane.
reaction of benzhydrylium ions with various Lewis bases can be
expressed as the sum of a Lewis acidity parameter LA and a
Lewis basicity parameter LB (eq 2).13
acceptor to give the carbanion 8, which is subsequently
protonated to yield 9. In this case, the nucleophilicity of the
borohydrides is crucial for the feasibility of the reduction. This
explanation holds also for ammonium borohydrides,14 which
have low Brønsted acidities and thus can efficiently reduce
Michael acceptors.1a (b) If now the generated phosphonium
borohydride is acidic, for instance, triarylphosphonium salts,
then two nucleophiles are present in the media: the free
phosphine and the borohydride.
log K(20°C) = LA + LB
(2)
By using the previously determined Lewis acidity (LA) of
benzhydrylium ions and the measured equilibrium constants
(K), the Lewis basicity of 1a was determined to be 10 times
higher than that of 1b (Table 1).
Having the rate constants of the reactions of 1a,b with
benzhydrylium ions and the related equilibrium constants in
hand, we then employed the Marcus equation (eq 3, working
terms neglected) to calculate barriers ΔG⧧0 for these reactions.
First, we calculated the Gibbs energy of activation when the
effect of the thermodynamic driving force is eliminated (ΔG0 =
0).12
On one hand, if the nucleophilicity of the borohydride is
significantly greater than that of the phosphine, the hydro-
genation pathway is predominant and only the saturated
product 9 is formed. On the other hand, if the phosphine is
more nucleophilic than the borohydride, such as in the case of
tri(1-naphthyl)phosphine 1a (N = 13.17 vs N = 10.01), the
phosphorus addition at the terminal methylene group is
kinetically favorable and the hydrophosphination adduct 10
can be formed. This may rationalize the failure of those
phosphines in reducing activated alkenes in the presence of
B(C6F5)3 and under H2 atmosphere. However, if the phosphine
has a lower nucleophilicity than the borohydride, as in the case
of 1b (N = 10.01 vs 8.40), the hydride transfer step can take
place faster than the addition of the phosphine, leading to the
formation of 9. The kinetic and thermodynamic data also
rationalize the reaction outcomes for the addition of 2a,b−
HB(C6F5)3 to 3. The predominance of the hydrophsphination
over the reduction adduct in the case of the reaction of 2a−
HB(C6F5)3 with 3 (Scheme 3) is a consequence of the low
Lewis basicity of 1b and the high intrinsic reactivity ΔG⧧0 . To
confirm our proposed mechanism, the reactions of 2a,b−
HB(C6F5)3 with four Michael acceptors 3b−e have been
conducted in dichloromethane at room temperature. As
depicted in Figure 3, while only reduction adducts 4b−e have
been observed when 2a,b−HB(C6F5)3 were combined with
3d,e, a mixture of the hydrophosphination and reduction
adducts has been detected for the reactions 2a,b−HB(C6F5)3
with the enones 3b,c. Because the Mayr’s rule of thumb
Thus, by substituting ΔG⧧ and ΔG0 into the Marcus eq 3,
⧧
the intrinsic barriers ΔG0 for the reactions of 1a,b with 6a−c
have been calculated (Table 2).
ΔG⧧ = ΔG0 + 0.5ΔG0 + ((ΔG0)2/16ΔG0 )
⧧
⧧
(3)
Table 2. Activation Energies ΔG⧧, Reaction Free Energies
ΔG0, and Intrinsic Barriers ΔG⧧0 (in kJ/mol) for the
Reactions of Benzhydrylium Ions 6a−c with Phosphines 1a,b
phosphines
Ar2CH+
ΔG⧧
ΔG0
ΔG⧧0
1a
6a
6b
6c
6a
6b
6c
49.1
50.9
53.2
54.3
58.0
60.7
−18.4
−15.5
−10.7
−12.8
−10.9
−6.66
57.9
58.4
58.4
60.5
63.3
63.9
1b
As shown in Table 2, the intrinsic barriers of the reactions of
1b with 6a−c are ∼3−5 kJ/mol greater than those for 1a. This
indicates that more reorganizing energy is needed for the
reaction of 1b than for the reactions with 1a, most probably
due to the steric hindrance.
Based on the hydricity of phosphonium and ammonium
borohydride determined by Berionni in one hand, and the
nucleophilicity, carbon Lewis basicity, and intrinsic reactivity of
phosphines, especially 1a,b, on the other hand, the following
reaction mechanism can be proposed (Scheme 4).
When the phosphonium borohydride 2, formed upon
heterolytic cleavage of H2 by the FLPs, reacts with Michael
acceptors 7, two pathways are possible: (a) If the phosphonium
ion is not acidic enough (pKa ≥ 6), typically in the case of
trialkylphosphines, the borohydride is the only nucleophilic
species present in the media and can react with the Michael
Figure 3. Product distributions of the reactions of 2a,b−HB(C6F5)3
with Michael acceptors 3b−e.
C
DOI: 10.1021/acs.orglett.6b03868
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
predicts that reactions of 3b (E = −23.54)15 and 3c (E =
−18.92)15 with phosphines 1a,b and borohydrides are not
kinetically possible at room temperature, the observed reaction
outcome given in Figure 3 is presumably the result of a proton
induction at the carbonyl oxygen that enhances the electro-
philicities of the enones 3b,c. In the case of the reactions of
2a,b−HB(C6F5)3 with the Michael acceptors 3d,e, phosphine
and hydride transfer events are possible because of the high
reactivity of both electrophiles. However, as proton activation is
less favorable, the phosphine attacks are highly reversible and
reduction pathway predominates.
In summary, we have performed kinetic and thermodynamic
assessments of the nucleophilcity and carbon Lewis basicity of
two hindered phosphines 1a,b, commonly used in FLP
chemistry. Though this work treats the reactivity of these two
specific phosphines, it clearly demonstrates how one can use
Mayr’s kinetic and thermodynamic parameters to deeply
understand reaction mechanisms of FLP−catalyzed reduction
of Michael acceptors. Assessment of the nucleophilicity and
Lewis basicity of other hindered phosphines are currently
ongoing in our laboratories and will be reported in due course.
(4) (a) Paradies, J. Angew. Chem., Int. Ed. 2014, 53, 3552. (b) Liu, Y.;
Du, H. J. Am. Chem. Soc. 2013, 135, 6810. (c) Liu, Y.; Du, H. J. Am.
Chem. Soc. 2013, 135, 12968. (d) Wei, S.; Du, H. J. Am. Chem. Soc.
2014, 136, 12261. (e) Zhang, Z.; Du, H. Angew. Chem., Int. Ed. 2015,
54, 623. (f) Ren, X.; Du, H. J. Am. Chem. Soc. 2016, 138, 810.
(5) For selected examples, see: (a) Wu, D.; Jia, D.; Liu, A.; Liu, L.;
Guo, J. Chem. Phys. Lett. 2012, 541, 1. (b) Spies, P.; Erker, G.; Kehr,
G.; Bergander, K.; Frohlich, R.; Grimme, S.; Stephan, D. W. Chem.
̈
̈
Commun. 2007, 5072. (c) Ozgun, T.; Bergander, K.; Liu, L.; Daniliuc,
̈
C. G.; Grimme, S.; Kehr, G.; Erker, G. Chem. - Eur. J. 2016, 22, 11958.
(d) Whittemore, S. M.; Edvenson, G.; Camaioni, D. M.; Karkamkar,
A.; Neiner, D.; Parab, K.; Autrey, T. Catal. Today 2015, 251, 28.
(e) Rokob, T. A.; Hamza, A.; Pap
́
ai, I. J. Am. Chem. Soc. 2009, 131,
10701. (f) Momming, C. M.; Otten, E.; Kehr, G.; Frohlich, R.;
̈
̈
Grimme, S.; Stephan, D. W.; Erker, G. Angew. Chem., Int. Ed. 2009, 48,
6643.
(6) Morozova, V.; Mayer, P.; Berionni, G. Angew. Chem., Int. Ed.
2015, 54, 14508. According to Mayr’s rule of thumb, a nucleophile−
electrophile combination is possible when E + N > −5. Accordingly,
Michael acceptors having electrophilicity smaller or equal to −15
should react with phosphonium borohydride (N ≈ 10) at 20 °C.
(7) Greb, L.; Tussing, S.; Schirmer, B.; Ona-Burgos, P.; Kaupmees,
̃
K.; Lokov, M.; Leito, I.; Grimme, S.; Paradies, J. Chem. Sci. 2013, 4,
2788.
̃
(8) Greb, L.; Ona-Burgos, P.; Schirmer, B.; Grimme, S.; Stephan, D.
̃
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.6b03868.
■
W.; Paradies, J. Angew. Chem., Int. Ed. 2012, 51, 10164.
(9) Ullrich, M.; Lough, A. J.; Stephan, D. W. J. Am. Chem. Soc. 2009,
131, 52.
S
(10) (a) Lucius, R.; Loos, R.; Mayr, H. Angew. Chem., Int. Ed. 2002,
41, 91. (b) Mayr, H.; Kempf, B.; Ofial, A. R. Acc. Chem. Res. 2003, 36,
66. (c) Mayr, H.; Ofial, A. R. Pure Appl. Chem. 2005, 77, 1807.
(d) Mayr, H.; Lakhdar, S.; Maji, B.; Ofial, A. R. Beilstein J. Org. Chem.
2012, 8, 1458. (e) Mayr, H. Tetrahedron 2015, 71, 5095. (f) Lakhdar,
S. Quantitative Treatments of Nucleophilicity and Carbon Lewis
Basicity. In Lewis Base Catalysis in Organic Synthesis; Vedejs, E.,
Denmark, S., Eds.; Wiley-VCH, 2016, Chapter 4. (g) Access to all
data are provided at: Mayr, H. Database of Nucleophilicties and
Electrophilicities. http://www.cup.uni-muenchen.de/oc/mayr/
DBintro.html (accessed Jan 11, 2017).
General procedures and NMR spectra (PDF)
AUTHOR INFORMATION
Corresponding Authors
■
*E-mail: annie-claude.gaumont@ensicaen.fr.
*E-mail: sami.lakhdar@ensicaen.fr.
ORCID
(11) Kempf, B.; Mayr, H. Chem. - Eur. J. 2005, 11, 917−927.
(12) Tolman, C. A. J. Am. Chem. Soc. 1970, 92, 2956.
(13) Mayr, H.; Ammer, J.; Baidya, M.; Maji, B.; Nigst, T. A.; Ofial, A.
R.; Singer, T. J. Am. Chem. Soc. 2015, 137, 2580.
Sami Lakhdar: 0000-0002-1168-7472
Notes
The authors declare no competing financial interest.
(14) Ines
Alcarazo, M. Angew. Chem., Int. Ed. 2012, 51, 12367.
(15) Allgauer, D. Dissertation, Faculty of Chemistry and Pharmacy,
́
, B.; Palomas, D.; Holle, S.; Steinberg, S.; Nicasio, J. A.;
ACKNOWLEDGMENTS
We thank the CNRS, Normandie Universite,
(ANR-11-LABX-0029) for financial support. J.D. is grateful to
■
̈
́
and Labex Synorg
LMU Munchen, 2014.
̈
the “Minister
for a fellowship. We thank Prof. Dr. Herbert Mayr (LMU−
Munchen) for insightful discussions and Nathalie Hampel
̀ ́
e de l’enseignement superieure et de la recherche”
̈
(LMU−Munchen) for synthesizing reference electrophiles.
̈
REFERENCES
■
(1) (a) Stephan, D. W.; Erker, G. Angew. Chem., Int. Ed. 2015, 54,
6400. (b) Stephan, D. W.; Erker, G. Angew. Chem., Int. Ed. 2010, 49,
46. (c) Stephan, D. W. Org. Biomol. Chem. 2012, 10, 5740.
(d) Paradies, J. Synlett 2013, 24, 777. (e) Mahdi, T.; Stephan, D. W.
J. Am. Chem. Soc. 2014, 136, 15809. (f) Scott, D. J.; Fuchter, M. J.;
Ashley, A. E. J. Am. Chem. Soc. 2014, 136, 15813. (g) Cooperative
Catalysis: Designing Efficient Catalysts for Synthesis;Peters, R., Ed.;
Wiley-VCH: Weinheim, 2015.
(2) (a) Ero
Tarkanyi, G.; Soos
Mehdi, H.; Papai, I.; Rokob, T. A.; Kiral
Angew. Chem., Int. Ed. 2010, 49, 6559.
̋
s, G.; Nagy, K.; Mehdi, H.; Pap
, T. Chem. - Eur. J. 2012, 18, 574. (b) Ero
y, P.; Tarkanyi, G.; Soos
́
ai, I.; Nagy, P.; Kiral
s, G.;
, T.
́
y, P.;
́
́
́
̋
́
́
́
́
́
(3) (a) Stephan, D. W. Acc. Chem. Res. 2015, 48, 306. (b) Stephan, D.
W. J. Am. Chem. Soc. 2015, 137, 10018. (c) Welch, G. C.; San Juan, R.
R.; Masuda, J. D.; Stephan, D. W. Science 2006, 314, 1124.
D
DOI: 10.1021/acs.orglett.6b03868
Org. Lett. XXXX, XXX, XXX−XXX