Dynamic Cross-Exchange in Halophosphonium Species: Direct Observation of Stereochemical Inversion in the Course of an SN2 Process

Article abstract of DOI:10.1002/anie.201708649

The complex fluxional interconversions between otherwise very similar phosphonium bromides and chlorides R₃PX⁺X^? (R=Alk, Ar, X=Cl or Br) were studied by NMR techniques. Their energy barriers are typically ca. 11 kcal mol^?1, but rise rapidly as bulky groups are attached to phosphorus, revealing the importance of steric factors. In contrast, electronic effects, as measured by Hammett analysis, are modest (ρ 1.46) but still clearly indicate negative charge flow towards phosphorus in the transition state. Most significantly, detailed analysis of the exchange pathways unequivocally, and for the first time in any such process, shows that nucleophilic attack of the nucleophilic anion on the tetrahedral centre results in inversion of configuration.

Full text of DOI:10.1002/anie.201708649

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Dynamic Cross-Exchange in Halophosphonium Species: First
Direct Observation of Stereochemical Inversion in the Course of
an SN2 Process
Authors: Kirill Nikitin, Elizabeth V. Jennings, Sulaiman Al Sulaimi,
Yannick Ortin, and Declan G Gilheany
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201708649
Angew. Chem. 10.1002/ange.201708649
Link to VoR: http://dx.doi.org/10.1002/anie.201708649
http://dx.doi.org/10.1002/ange.201708649

10.1002/anie.201708649
Angewandte Chemie International Edition
Dynamic Stereochemistry
DOI: 10.1002/anie.201708649
Dynamic Cross-Exchange in Halophosphonium Species: First
Direct Observation of Stereochemical Inversion in the Course of
an SN2 Process.**
Kirill Nikitin*, Elizabeth V. Jennings, Sulaiman Al Sulaimi, Yannick Ortin, Declan G. Gilheany*
Abstract: The complex fluxional interconversions between, otherwise
very similar, phosphonium bromides and chlorides R3PX+X- (R = Alk,
Ar, X = Cl or Br) was studied by NMR techniques. Their energy
barriers are typically ca. 11 kcal/mol but rise rapidly as bulky groups
degeneracy by simply mixing the chloro- and bromo-species
(Figure 1c). As we show below, this nearly degenerate cross-
exchange system enabled us to create a unified scale of reactivity
based on sterics10 and electronics. Most significantly, we were
are attached to phosphorus, revealing the importance of steric factors. able to observe inversion of tetrahedral configuration directly for
In contrast, electronic effects, as measured by Hammett analysis, are
modest (ρ 1.46) but still clearly indicate negative charge flow towards
phosphorus in the transition state. Most significantly, detailed analysis
of the exchange pathways unequivocally, and for the first time in any
such process, shows that nucleophilic attack of the nucleophilic anion
on the tetrahedral centre results in inversion of configuration.
the first time.11 This also validated our (and others) previous
computational predictions9, 12 of Walden-type inversion in the
course of nucleophilic attack at tetrahedral phosphorus.
The study of reactivity at tetrahedral centres is ubiquitous in
chemistry, not only because this geometry allows for strong
bonds and efficient use of space but also because of its potential
for asymmetry. Thus modern asymmetric synthesis makes
extensive use of stereospecific reactions, especially at carbon,
boron, silicon and phosphorus.1,2 A crucial feature of most of
these reactions is the act of inversion at the stereogenic centre.
However, in all the time since the discovery of Walden inversion,
this central event has not been directly observed and, typically,
inversion has been inferred from the final product configuration.
Phosphorus chemistry provides an ideal platform to study
the issue of inversion because, just as carbon chemistry relies on
3
Walden inversion, syntheses of P-stereogenic compounds
depend heavily on stereospecific inversions at phosphorus after
an initial stereoselective reaction.4,5,6 However, unlike carbon, the
phosphorus intermediates may be easier to study. This was
certainly the case in our P-stereogenic synthesis7 utilising the
8
dynamic kinetic resolution
of rapidly interconverting P-
Figure. 1. 31P NMR observations of halide exchange: a) single peak due to full
degeneracy; b) two close signals in the presence of a chiral s-Bu group9; c) in
this work degeneracy has been lifted as both P-Cl and P-Br species are present.
halophosphonium halides (Figure 1a). Since the fully degenerate
process was difficult to study, we used 9 the corresponding
diastereomeric pseudo-degenerate exchange (Figure 1b) of
model species to quantify (by NMR) the rates and energy barriers.
However, the more general dynamics of e.g. common P-
The halophosphonium salts in this study (XPS 1-17, Scheme 1, X
= Cl or Br) were generated from the corresponding phosphine
oxides by treatment with oxalyl dihalides.13 We have previously
shown that, in the solid state, such salts have the free anion
positioned to perform a nucleophilic attack on the phosphonium
cation.13, 14 Individual XPS were fully characterised by NMR
spectroscopy and in all cases the 31P signals of the bromides
were observed 20-30 ppm upfield from the chlorides (see ESI).
Variable temperature NMR spectra were recorded using
equimolar amounts of both halides in dichloromethane (DCM) or
chloroform. Figure 2 shows the 31P NMR of a 1:1 mixture of
halides [1-Cl]+Br- and [1-Br]+Cl- where the interconversion of the
species is slow on the NMR timescale below ca. -20 °C.15 The
coalescence (Tc = 0 °C) in this case corresponds to an average
activation barrier of 11.1 kcal/mol.16 Scheme 1 shows the results
of similar experiments for XPS 2-17.
stereogenic DKR intermediates (Figure 1a: R1/R2/R3
=
Ph/Me/oTol) or simple symmetrical analogues (R1-R3 = Ph) could
not be studied in this way, which prompted us to seek a different
methodology for the systematic study of the electronic and steric
effects.
We had previously observed9 that, separately, the
interconverting dichloro- and dibromo-diastereomers (Figure 1b)
had very similar NMR behaviour. Therefore we sought to lift
[∗]
Dr. K. Nikitin, Dr. E. V. Jennings, Dr. S. Al-Sulaimi , Dr. Y. Ortin,
Prof. Dr. D. G. Gilheany, School of Chemistry, University College
Dublin, Belfield, Dublin, Ireland.
E-mail: kirill.nikitin@ucd.ie; declan.gilheany@ucd.ie
[∗∗]
This work is supported by Science Foundation Ireland.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the authors.
1
This article is protected by copyright. All rights reserved.

10.1002/anie.201708649
Angewandte Chemie International Edition
cross-exchange rate constant would be about one half of its true
value for the individual mixed halides. However the fact (vide
supra) that the cross-exchange results are comparable to the
self-exchange values shows that the “hidden” homo-halogenated
species are not dominant and the observed cross-exchange
barriers can simply be treated as very reasonable estimates.
X
P
X
P
X
P
X
P
Et
Et
Et
Et
Et
Et
4
11.0
2
10.9
3
11.1
1
11.1
X
P
X
P
X
P
X
P
OCH3
Ph
Et
Ph
Me
Ph
C8H17
C8H17
Me
C8H17
5
11.3
8
11.8
6a
11.3
7
11.8
X
P
X
P
X
P
X
P
Ph
Me
Ph
Ph
Figure 2. Coalescence of 31P NMR signals of triphenylphosphonium halides [1-
Cl]+Br- and [1-Br]+Cl- due to pseudo-degenerate exchange (DCM, 202 MHz).
i-Pr
9
11.1
11
12.4
12
13.6
10
11.4
Generally, XPS with primary aliphatic or aromatic groups, have
average barriers of 10-12 kcal/mol in excellent agreement with
our diastereomeric self-exchange systems in the absence of
branched alkyl groups.9 The fact that the reactivity of cyclic cation
10 is not significantly different from its close acyclic analogues 6a
and 7 is consistent with our earlier report of the similarity in the
stereoselectivity in their respective DKR processes.17 However
the introduction of the conformationally more strained
dibenzophosphole moiety imposes a pronounced (~1 kcal/mol)
rise in the activation energy in 11. The unusually high activation
barrier for the tri-o-tolyl cation 14 is in stark contrast to the
relatively minor effect of an individual o-tolyl group (compare 2
and 7). This cooperative effect could be attributed to the limited
access to the P-centre on the side opposite to the leaving group
X blocked by one of the o-tolyl groups (conformation of ion-pairs
14, see ESI). This indicates, but does not prove, predominant
back-side attack with inversion of the P-configuration.
X
X
P
X
P
X
P
X
P
P
Ph
Et
Ph
Ph
Ph
Me
Ph
16
19.6
13
16.0
14
16.5
15
17.6
17
25.5
Scheme 1. P-halophosphonium salts (XPS, X = Cl, Br) 1-17 arranged in
approximate ascending order of their free energy barriers of activation for
pseudo-degenerate exchange (in blue, average of the forward and reverse
barriers, 0.5 kcal/mol).
Turning now to electronic effects, we have shown previously by
calculation that the attack of halide anion follows a pronounced
associative pathway (strongly curved-in on the More-O-Ferrall-
Jencks diagram).9 To explore this experimentally, we employed a
series of P-stereogenic XPS 6a-h (Figure 3). In our earlier studies
this series had shown very little effect of the substituent on the
stereoselectivity of our DKR process.7a,e
The consistency of the earlier self-exchange and the
present cross-exchange results also extends to branched
substrates. Thus, the reactivity of diaryl-isopropyl system 12
(cross-exchange barrier 13.6 kcal/mol) is close to the arithmetic
average of Cl/Cl and Br/Br self-exchange in comparable sec-
butylphenyltolylphosphonium (13 kcal/mol) and sec-butyl-iso-
propylphenylphosphonium (13.7 kcal/mol) salts.9 This indicates
that the cross-exchange pentacoordinate transition state (TS) can
be energetically interpolated between the corresponding self-
exchange processes and stresses the parallel with carbon SN2
reactions - a very similar interpolation in carbon chemistry was
documented for self- and cross-Finkelstein reactions by Ingold
and co-workers.10c,d Therefore, the relatively slow exchange in
iso-Pr compound 12 and, especially, t-Bu derivatives 13, 15 and
16 can readily be attributed to the steric effect of bulky groups
mimicking other SN2 processes. Accordingly, introduction of one
t-Bu (13) adds 5 kcal/mol, while the second t-Bu (17) adds
another ca. 10 kcal/mol to the barrier.
Br
P
Cl
P
CH3
CH3
kCl
Ph
Me
Me
Ph
Y
Y
6a-6h
+
Y = H. NMe2, OMe, Me, Cl, F, CF3, NHMe2
2
1.5
1
+
-NHMe2
-CF3
-F
0.5
0
-Cl
-H
-Me
-2
-1.5
-1
-0.5
0
0.5
1
-0.5
-1
σ+
-OMe
-1.5
-2
There is, nevertheless, a caveat to be associated with these
cross-exchange measurements: the degree of speciation
between homo- and hetero-halogenated ion pairs is not known.18
For example, in system 1X2 (Figure 2) equilibration of [1-Cl]+Br-
and [1-Br]+Cl- is observable by DNMR but the competing
degenerate exchange in both [1-Cl]+Cl¯ and [1-Br]+Br¯ is hidden.
Purely statistically, with equal speciation, the observed nominal
-NMe2
Figure 3. Experimental Hammett plot of the exchange rate constants (kCl) in
0.05 log units; R2 0.989) corresponding to ρ = 1.46,
series 6 (error bar
determined by 2D 31P EXSY NMR.
When series 6 was studied at -50 ˚C by 31P two-dimensional
EXSY spectroscopy using the non-equal exchange approximation
2
This article is protected by copyright. All rights reserved.

10.1002/anie.201708649
Angewandte Chemie International Edition
(see ESI for details), 19 electron-donating groups (EDG) were
found to slow the exchange while electron-withdrawing groups
(EWG) accelerated it. Interestingly it was the σ+ parameter20 that
a blue cross-peak as shown in Figure 4b. Likewise, Br ion attack
on SP-chlorophosphonium with retention of configuration leads to
SP-bromophosphonium (red peak). Conversely, an attack of
bromide anion on SP-chlorophosphonium cation with inversion of
configuration leads to RP-bromophosphonium cation giving rise to
a cross-peak as shown in Figure 4c in grey. Should, however,
both retention and inversion of the P-configuration happen at an
appreciable rate, all coloured and grey cross-peaks, would be
observed (Figure 4d).
demonstrated best linearity with
a small positive reaction
parameter (Figure 3, ρ = 1.46 0.15). This somewhat uncommon
combination reflects the diminution of positive charge on
phosphorus in the TS during the substitution process.21
The above observations show that the XPS cross-exchange
has all the standard characteristics of an SN2 mechanism.
However, in common with all other studies of this type, the
inversion of the reaction centre is not directly seen. Thus in
Figure 3, we have intentionally indicated stereospecific halide-ion
attack with inversion of the P-configuration but this is not yet
proven – the rapid interconversion of the phosphonium species
could be accompanied by either inversion or retention at the P-
centre, or both. Just as in our recently reported19 phenomenon of
two independent and geometrically orthogonal stereomutations at
a single asymmetric phosphonium centre, one cannot rule out a
concomitant retention process via equatorial nucleophilic attack
or via a pseudorotation-type stereomutation. We realized that
cross-exchange techniques provide the only way to directly probe
the stereochemistry of halide-ion attack on the P-centre by
combining the two degeneracy-lifting methods together (Figure 1b
with 1c).
We tested diastereomeric systems 18 and 19 (Scheme 2a).
System 18 proved unsatisfactory22 but the introduction (Scheme
2b) of the α–alkoxyalkyl group in 19 enabled better 31P NMR
peak separation (ca. 4 ppm) and slightly lower average barrier
(11.4 0.2 kcal/mol). Both 19Cl2 and 19Br2 are unstable at r.t.
and were handled at -20 ˚C.
a
X
P
X
P
X
P
X
P
Ph
O
Ph
OMe
Ph
Me
MeO
Ph
i-Pr
18
-X
19
-X
20
-X2
b
iii
iv
TFA
OEt
19Cl2
19Br2
Me
O
P
OEt
Me
i
ii
P
P
Me
Ph
Me
Ph
Ph
Ph
21
Ph
Me
19
P-Cl
P-Br
22
=O
b
a
SP RP
SP RP
Scheme 2. a) diastereomeric P-stereogenic systems studied; b) synthesis of
19X2: i) ethylvinyl ether/TFA; ii) alkaline hydrolysis; iii) oxalyl chloride or iv)
oxalyl bromide, -20 ˚C, DCM.
When a mixture of 19Br2 and 19Cl2 was examined by 31P NMR at
-60 ˚C four well-resolved resonances of individual chloro- and
bromo-species were observed at 87, 82, 72, 68 ppm
corresponding to SP-19-Cl, RP-19-Cl, SP-19-Br, RP-19-Br. The 2D
EXSY spectrum recorded with short mixing time (Tm = 20 ms)
displayed four main diagonal peaks and a set of cross-peaks
(Figure 5). The peaks at 68/72 and 72/68 ppm clearly correspond
to bromide self-exchange SP-19-Br → RP-19-Br (Figure 4a,
double “↔” arrows).
retention
P-Cl
P-Br
P-Cl
P-Br
c
SP RP
SP RP
d
SP RP
SP RP
19-Br
SP
19-Cl
RP
SP
RP
ppm
inversion
Figure 4. Halide exchange pathways in a generic diastereomeric mixed-
halogen XPS: a) tetrahedron representation: single arrows - Walden-type
inversion of the P-configuration with change of the halogen; double arrows
-
inversion without change of halogen; curled arrows – putative retention pathway
with change of the halogen; b) expected 2D 31P EXSY signature of exclusive
retention; c) of exclusive inversion; d) of the mixed case.
no peak
Figure 4a shows pseudo-degenerate cross-exchange in a generic
diastereomeric system. This situation allows essentially
independent self- and cross-exchange of four XPS (not counting
their enantiomers). Crucially, in 2D EXSY experiments, cross-
exchange where the P-configuration is retained (curled
pseudorotation arrows) can be distinguished from cross-
exchange with inversion (equilibrium single arrows) or self-
exchange with inversion (double “↔” arrows) as shown in the
idealised diagrams in Figures 4b-d. Specifically, attack of bromide
anion on RP-chlorophosphonium cation with retention of
configuration leads to RP-bromophosphonium cation giving rise to
inversion
ppm
Figure 5. Dynamic inversion of the P-configuration observed in the 19-Br/Cl
exchange system by EXSY (DCM, -60ºC, Tm = 20 ms), clearly showing
correspondence to Figure 4c and strongly predominant inversion on this time-
scale.
3
This article is protected by copyright. All rights reserved.

10.1002/anie.201708649
Angewandte Chemie International Edition
Significantly, four remaining peaks (86/68, 82/72, 68/86, 72/82
ppm) aligned in the same fashion as Figure 4c correspond to
cross-exchange between bromo- and chloro-species with
inversion of the P-configuration SP-19-Cl → RP-19-Br and SP-19-
Br → RP-19-Cl. Crucially, no cross-peaks were observed in the
main-diagonal fashion of Figure 4b, corresponding to retention of
P-configuration. Therefore we have unequivocally established
strongly predominant 23 inversion of the P-configuration in the
course of nucleophilic exchange processes in halophosphonium
species. Stereospecific inversion at the P-centre is in excellent
agreement with our recent findings on the fluxional dynamics of
tetrahedral phosphonium salts.9,19 To the best of our knowledge,
obtaining of such stereochemical information from an inseparable
mixture of rapidly interconverting reacting species is the first
example of a dynamic exchange experiment of this type. We note
that attribution of the lower-field 31P NMR peaks to the SP-isomers
of 19-Cl and 19-Br as shown is consistent with computed
chemical shifts.24 Should one of those assignments be incorrect,
the observed 2D EXSY pattern would mean that self-exchange
leads to inversion while cross-exchange leads to retention – an
entirely irrational situation. Work is underway on detailed analysis
of self- and cross-exchange processes in more complex double
P-stereogenic systems, for example DiPAMP halides (20-X2
Scheme 2a).
Keywords: nucleophilic substitution · NMR spectroscopy · P-
stereogenic compounds · phosphorus · configuration inversion ·
[1]
a) A. L. Thompson, D. J. Watkin, Tetrahedron: Asymmetry 2009, 712; b)
K. Lee, D. L. Silverio, S. Torker, D.W. Robbins, F. Haeffner, F.W. van der
Mei, A. H. Hoveyda, Nat. Chem. 2016, 768; c) J. Christoffers, A. Mann,
Angew. Chem. Int. Ed. 2001, 40, 4591; d) Y.-M. Cui, Y. Lin, L.W. Xu,
Coord. Chem. Rev. 2017, 37; e) O. I. Kolodiazhnyi, Russ. J. Gen. Chem.
2017, 425.
[2]
[3]
a) J. Labuta, S. Ishihara, T. Sikorsky, Z. Futera, A. Shundo, L. Hanykova,
J.V. Burda, K. Ariga, J. P. Hill, Nat. Commum. 2013, 3188; b) J.
Escorihuela, M. I. Burgette, S. L. Luis, Chem. Soc. Rev. 2013, 42, 5595;
c) B. Bieszczad, D. G. Gilheany, Angew. Chem. Int. Ed. 2017, 4272; d) P.
Walden, Chem. Ber. 1896, 133.
a) K. M. Pietrusiewicz, M. Zablocka, Chem. Rev. 1994, 94, 1375; b) M. J.
Johansson, N. C. Kann, Mini-Rev. Org. Chem 2004, 1, 233; c) D. S.
Glueck, Synlett 2007, 2627; d) A. Grabulosa, Ed.; P-Stereogenic Ligands
in Enantioselective Catalysis, RSC, Cambridge, 2011; e) O. I.
Kolodiazhnyi, Tetrahedron Asym. 2012, 23, 1.
[4]
[5]
Menthol as the auxiliary: a) O. Korpiun, R. A. Lewis, J. Chickos, K.
Mislow, J. Am. Chem. Soc. 1968, 90, 4842; b) O. Berger, J-L.
Montchamp, Angew. Chem. Int. Ed. 2013, 52, 11377.
Ephedrine as the auxiliary: a) C. Darcel, J. Uziel, S. Jugé in: Phosphorus
Ligands in Asymmetric Catalysis (Ed.: A. Börner), Wiley-VCH, Weinheim,
2008, vol. 3, 1211; b) J. V. Carey, M. D. Barker, J. M. Brown, M. J. H.
Russell, J. Chem. Soc. Perkin Trans. 1 1993, 831; c) F. Chaux et al.
Compt. Rend. Chim. 2010, 13, 1213; d) T. Oshiki, T. Imamoto, J. Am.
Chem. Soc. 1992, 114, 3975; e) T. León, A. Riera, X. Verdaguer, J. Am.
Chem. Soc. 2011, 133, 5740.
To conclude, by introducing two different halogens in the
dynamic exchange of XPS species we were able to extend the
measurement of reaction activation barrier to essentially any
phosphonium structure and construct a reactivity scale for the
process. Notably, for XPS containing ordinary hydrocarbon
groups, the barriers are around 11 kcal/mol, rising rapidly with
increased steric bulk in agreement with our earlier findings.9
Secondly, we have used, to our knowledge for the first time for a
fast nucleophilic reaction, low temperature exchange experiments
on the millisecond scale to build an informative Hammett plot.
Both findings enable us to predict the reactivity of a wide variety
of XPS and are not, in principle, limited to the halogen exchange
process. Finally, by rigorously analysing the dynamic pattern of
multiple pathway exchange systems, we successfully established
the stereochemistry of the exchange process. In excellent
agreement with computations strongly favouring the di-axial
orientation of halogens in the transition state, the attack of
nucleophilic halide anion results in an inversion of the P-
configuration. This latter result is significant for Main Group
element chemistry in general as it underlines the deep similarity
between carbon and non-carbon chemistry.
[6]
[7]
Chiral aminophenol as the auxiliary: Z. S. Han et al. J. Am. Chem. Soc.
2013, 135, 2474.
a) K. V. Rajendran, L. Kennedy, D. G. Gilheany, Eur. J. Org. Chem. 2010,
5642; b) E. Bergin, C. T. O’Connor, S. B. Robinson, E. M. McGarrigle, C.
P. O’Mahony, D. G. Gilheany, J. Am. Chem. Soc. 2007, 129, 9566; c) G.
King, E. Bergin, H. Müller-Bunz. D. G. Gilheany, Acta. Cryst 2007, E63,
O3278; d) K. V. Rajendran, D. G. Gilheany, Chem. Commun. 2012, 48,
10040; e) S. Al Sulaimi, K.V. Rajendran, D.G. Gilheany Eur. J. Org.
Chem. 2015, 5959.
[8]
[9]
K. Nikitin, K.Rajendran, H. Müller-Bunz, D.G. Gilheany, Angew. Chem.
Int. Ed. 2014, 126, 1906.
E. Jennings, K. Nikitin, Y. Ortin, D.G. Gilheany, J. Am. Chem. Soc. 2014,
136, 16217.
[10] a) I. Fernandez, G. Frenking, E. Uggerud, Chem. Eur. J. 2009, 2166; b) I.
Dostrovsky, E.D. Hughes, E.D., C.K. Ingold, J. Chem. Soc. 1946, 173; c)
P.B.D. de la Mare, L. Fowden, E.D. Hughes, C.K. Ingold, D.H. Mackie, J.
Chem. Soc. 1955, 3200; d) I. Fernandez, F. M. Bickelhaupt, Chem. Soc.
Rev. 2014, 4953.
[11] a) M.S. Korobov, G.S. Borodkin, N.I. Orisenko, T.A. Ryskina, L.E.
Nivorozhkin, V.I. Minkin, J. Mol. Structure, 1989, 200, 61; b) M.
McGlinchey, Symmetry 2014, 6, 622; c) M. Oki, Pure Appl. Chem. 1989,
699; d) L.P. Olekhovich, Z.N. Budarina, A.V. Lesin, S.V. Kurbatov, G.S.
Borodkin, V.I. Minkin, Mend. Commun. 1994, 162.
Experimental Section
[12] a) M. A. Van Bochove, F. M. Bickelhaupt, Eur. J. Org. Chem. 2008, 649;
b) M. A. Van Bochove, M. Swart, F.M. Bickelhaupt, ChemPhysChem
2007, 2452; c) H. M. Buck, Int. J. Quant. Chem. 2010, 110, 1412–1424.
[13] K. Nikitin, H. Muller-Bunz, D. G. Gilheany, Chem Commun. 2013, 49,
1434.
Experimental procedures for dynamic measurements and characterisation
of individual compounds are given in the Electronic Supporting Information
section.
[14] Structure of [1Cl]+ salts: a) S. M. Godfrey, C. A. McAuliffe, R. G.
Pritchard, J. M. Sheffield, Chem. Commun. 1996, 2521; b) S. J. Hwang,
D. C. Powers, A. G. Maherab, D. G. Nocera, Chem. Sci. 2015, 6, 917.
[15] Ignoring possible anion-exchange in the outer sphere.
Acknowledgements
This work was supported by Science Foundation Ireland chiefly
through Principal Investigator Grant 09/IN.1/B2627 and partially
through the SFI funded Solid State Pharmaceutical Centre
12/RC/2275 Grants to DGG. We are also grateful to UCD Centre
for Synthesis and Chemical Biology for access to their extensive
analysis facilities.
[16] a) H. Shanan-Atidi, I. H. Bar-Eli, J. Phys. Chem. 1970, 74, 961; b)
barriers of forward and reverse reaction: 11.0 and 11.3 kcal/mol.
[17] D. J. Carr, J. S. Kudavalli, K. S. Dunne, H.Müller-Bunz, D. G. Gilheany, J.
Org. Chem. 2013, 10500.
[18] The concentration of pentacoordinate covalent species can safely be
ignored: the respective NMR signals are not observed in chlorinated
solvents in good agreement with DFT-level computations (see ref. 9)
indicating a 10-20 kcal/mol energy penalty.
Received:
Published online on:
4
This article is protected by copyright. All rights reserved.

10.1002/anie.201708649
Angewandte Chemie International Edition
[19] This provided somewhat improved precision ( 0.2 vs 0.5 kcal/mol) over
the VT experiments. Note that EXSY-determined barriers are known to
be slightly higher (up to 0.9 kcal/mol) than coalescence results: M. W.
Gillick-Healy, E. Jennings, K. Nikitin, Y. Ortin, D. G. Gilheany, Chem. Eur.
J. 2017, 2332.
[20] a) L.P. Hammett, Physical Organic Chemistry, 2nd Edn., New York,
USA: McGraw-Hill, 1970; b) J. E. Leffler, E. Grunwald, Rates and
Equilibria of Organic Reactions, Wiley, 1963.
[21] Strong EWGs directly on phosphorus are known to stabilise the
penatcoordinate state: S. M. Godfrey, C. A. McAuliffe, R. G. Pritchard, J.
M. Sheffield, G. M. Thompson, J. Chem. Soc., Dalton Trans. 1997, 4823.
[22] Due to unresolved closely-positioned 31P NMR peaks.
[23] No retention was observed within detection limit, 5% relative, of this
measurement. Second-generation “retention” peaks are observed in
experiments with longer mixing times (Tm = 50 ms or greater). As their
intensity is time-lagged, this behaviour is attributed to a sequential
process e.g. SP-19Br → RP-19Cl → SP-19Cl.
[24] Chemical shifts computed (SPARTAN 10, carbon S-configuration fixed )
for the SP-isomers were consistently higher than for the RP-isomers.
5
This article is protected by copyright. All rights reserved.