Alkali metal and ammonium cation-arene interactions with tetraphenylborate anion

Article abstract of DOI:10.1080/10610270903045318

Sodium, potassium, rubidium, caesium, ammonium and tetramethylammonium tetraphenylborates were studied by both positive and negative ion electrospray mass spectrometry. An affinity order of Cs⁺ > Rb⁺ > K⁺ Na⁺ was obtained. The results obtained were compared with both calculations and solid-state structures, where available. The previously reported high affinity of caesium for tetraphenylborate concluded from NMR experiments was confirmed for the gas phase. The affinity does not appear to result from steric effects and a cation-pi interaction seems likely. In the positive ion mode, a unique acetonitrile complex of NaBPh₄ was observed.

Full text of DOI:10.1080/10610270903045318

Supramolecular Chemistry
Vol. 22, No. 2, February 2010, 73–80
Alkali metal and ammonium cation–arene interactions with tetraphenylborate anion
a
Ruiqiong Li , Rudolph E.K. Winter , Joseph Kramer and George W. Gokel *
b
b
bc
a
b
Department of Chemistry, Washington University, 1 Brookings Drive, Saint Louis, MO 63130, USA; Departments of Chemistry &
Biochemistry and Biology, Center for Nanoscience, University of Missouri – Saint Louis, One University Boulevard, Saint Louis,
MO 63121, USA; Washington University School of Medicine, Saint Louis, MO, 63110, USA
c
(Received 15 March 2009; final version received 8 May 2009)
Sodium, potassium, rubidium, caesium, ammonium and tetramethylammonium tetraphenylborates were studied by both
þ
þ
þ
þ
positive and negative ion electrospray mass spectrometry. An affinity order of Cs . Rb . K , Na was obtained.
The results obtained were compared with both calculations and solid-state structures, where available. The previously
reported high affinity of caesium for tetraphenylborate concluded from NMR experiments was confirmed for the gas phase.
The affinity does not appear to result from steric effects and a cation–pi interaction seems likely. In the positive ion mode,
a unique acetonitrile complex of NaBPh was observed.
4
Keywords: alkali metal; ammonium ion; cation–pi interaction; mass spectrometry; tetraphenylborate
1
.
Introduction
(17) and many others have solidified the concept as a
significant, if weak, molecular recognition force (18).
Our own efforts to understand the sodium–arene
interaction began unsuccessfully with an X-ray structural
The alkali metal–arene (1) and ammonium–arene (2)
cation–pi interactions were explored by mass spec-
trometry in the early 1980s. During the 1990s, cation–pi
interactions were postulated by Kumpf and Dougherty (3)
as a mechanism to account for the selectivity of potassium-
conducting protein channels. The first solid-state channel
structure (4) and other evidence (5) showed that the
selectivity filter involved carbonyl groups rather than
arenes, but the suggestion had already stimulated great
interest in cation–pi contacts as a weak force interaction
0
examination of N,N -dibenzyl-4,13-diaza-18-crown-6.
The systems were designed in the expectation that a
þ
ring-bound Na would receive secondary, apical stabilis-
ation from the sidearm arenes. In fact, the sidearms were
turned away from the ring-bound cation, which interacted
with the compound’s iodide counterion (19). Remarkably,
the simple expedient of extending the sidearms from benzyl
toarylethyl permitted solid-statestructures to beobtained for
sodium or potassium pi-complexes involving phenyl (20),
hydroxyphenyl (21) and indolyl (22). Solid-state evidence
was also obtained for alkali metal–arene interactions with
double (23) and triple bonds (24). Ultimately, a broad range
of solid-state examples was obtained (25).
(
6–8).
Kebarle and co-workers showed by mass spectrometric
analysis that potassium cations formed a stable complex
with benzene and estimated its enthalpic stabilisation at
1
9 kcal/mol (1). Meot-Ner and Deakyne also used mass
spectrometry to demonstrate arene–ammonium ion
interactions (2). Burley and Petsko recognised the
possibility of ammonium–pi interactions by 1986 when
a search of the Protein Data Bank (PDB) revealed close
contacts in a number of protein structures (9). A similar,
but more recent, survey concerned arene–arginine
interactions (10). Sodium–arene interactions were studied
by Castleman et al., and found to be stronger than the
Among the simplest alkali metal–arene interactions
possible are those observed in the known solid-state
structures of the tetraphenylborate salts. The tetraphenyl-
borate anion is tetrahedral and offers a cation four
individual arenes or two v-shaped pockets in which a
cation could be accommodated. Electrostatic consider-
ations suggest that the cation should approach the
negatively charged boron as closely as possible. To the
extent sterically possible, the cation should insinuate itself
between two arenes to maximise the cation–pi interaction.
Some study of tetraphenylborate salts has been
reported. Interactions between the CZH group of
þ
corresponding K –benzene contact (11). Lisy and co-
workers (12) explored selectivity in the gas phase between
þ
þ
Na and K . Many additional studies have been
undertaken in order to define the scope and occurrence
of the cation–pi interaction, both theoretically and in a
biological context (13). Contributions from the labora-
tories of Dunbar (14), Wouters (15), Waters (16), Hunter
2
imidazole and an arene in BPh4 were studied in both
solid and solution phases (26). Interactions were also
*Corresponding author. Email: gokelg@umsl.edu
ISSN 1061-0278 print/ISSN 1029-0478 online
q 2010 Taylor & Francis
DOI: 10.1080/10610270903045318
http://www.informaworld.com

7
4
R. Li et al.
þ
aromatic hydrocarbons favoured Cs over the other ions
seemed surprising. Direct gas-phase competition exper-
iments should provide evidence one way or the other.
The results of a mass spectrometric study involving both
positive and negative ionisation methods are reported here.
2
.
Results and discussion
þ
þ
þ
þ
Compounds 1–4 are Na , K , Rb and Cs , tetra-
2
þ
phenylborate salts, respectively. Compound 5, NH BPh ,
4
4
is not an alkali metal salt, but it is of obvious interest in this
context. Lithium tetraphenylborate is not included in these
studies in part because it has a significant covalent
behaviour (32). The only structure available in the
Cambridge Structural Database (CSD) is for the hydrated
salt (CSD: YIRRIA) (33). X-ray structures are available
for 1–5 in the CSD. The solid-state structure of
tetramethylammonium tetraphenylborate (6) is also
reported (34) and comparisons with it are made below.
4
Figure 1. Solid-state structures of NaBPh (1, yellow cation),
þ þ þ þ
4 4
KBPh ·K (2·K , magenta cation), RbBPh ·Rb (3·Rb , grey
þ
þ
cation) and CsBPh ·Cs (4·Cs , cream cation).
4
2
examined between BPh4 and cations ligated by sigma-
donors (27, 28). Ion-selective electrodes have been
2
developed that use the pi-coordinating ability of BPh4
for function (29). The effort most closely related to our
own interest involved a solid-state NMR study of sodium
2
and potassium BPh salts by Wu and co-workers (30). Zhu
and co-workers showed by deuterium NMR studies that
polycyclic aromatic hydrocarbons interacted more
4
þ
þ
þ
þ
strongly with Cs cation than with Na , K or Rb in
aqueous methanol solutions (31).
2.1 Solid-state structures of tetraphenylborate salts
The study reported here was prompted by two
questions. First, do the solid-state structures of MBPh4
reflect only the packing matrix for these salts?
The symmetrical arrangement of the anions, which
surround each cation with four arenes, is obviously
efficient but does not indicate whether the stronger Lewis
The solid-state structures for NaBPh4 (1, CSD:
ZZZUPI01), KBPh (2, CSD: KTPHEB02), RbBPh (3,
4
4
CSD: RBPBOR), CsBPh4 (4, CSD: ZZZURS01),
NH BPh (5, CSD:AMPHEB02) and NMe BPh (6,
4
4
4
4
CSD: MATPHB) have been reported. Structures generated
from the coordinates ((35); www.Pov-Ray.org) are shown
in Figure 1. The details of the angles and distances are
summarised in Table 1.
þ
acid (i.e. Na ) would be favoured over a larger cation such
þ
as Cs . Second, the finding of Zhu and co-workers that
Table 1. Summary of tetraphenylborate crystal structure data.
˚
Data in A
Data in degrees
a
B $ Mþ dist. A˚
b
Calc’d B $ Mþc
þ
d
1
1
C –B–C angle
e
f
Salt
Descriptor
ZZZUPI10
M –centroid
Centroid–B–centroid
NaBPh
4
4.002
3.904
4.026
4.168
4.002
5.580
3.470
3.906
4.125
4.379
–
3.028
2.960
3.041
3.158
3.027
4.402
103.68, 103.68
103.41, 103.41
103.76, 103.82
104.06, 104.06
103.65, 103.65
102.04, 104.90
97.00, 97.00
96.87, 96.87
96.90, 96.92
97.76, 97.78
97.00, 97.00
95.73, 103.63
KBPh
RbBPh4
CsBPh
NH BPh
NBPh
4
KTPHEB02
RBPBOR
ZZZURS01
AMPHEB02
MATPHB
4
4
4
Me
4
4
–
a
Cambridge Structural Database.
b
c
d
e
f
Distance from a cation or N to boron measured with Mercury software.
Calculated using Gaussian 03W, see Section 4.
Distance from a cation or N to either centroid.
Angle between C of each benzene and boron.
Angle between each centroid and boron.
1

Supramolecular Chemistry
75
þ
þ
þ
þ
þ
All of the cations [Na , K , Rb , Cs , NH and
þ
aqueous methanol studies with benzene and polycyclic
aromatic hydrocarbons, reported an order of decreasing
4
þ
þ
NðCH3Þ [note that NH and NðCH3Þ are not shown in
2
4
4
4
þ
þ
þ
þ
Figure 1]] are situated approximately symmetrically
among the four arenes of two adjacent Ph–B –Ph clefts.
complexation strength of Cs . K . Rb . Na .
The order for the latter three cations correlates well with
that calculated using a simple electrostatic model.
The Gaussian software reports Mullikan charges, which
confirm the order reported (37). The experimental result
This arrangement places each cation proximate to four
electron-rich benzene rings and as close as possible to the
negative charges on boron.
þ
þ
þ
þ
þ
The ionic radii of Na , K , Rb and Cs increase
suggesting that Cs interacts especially strongly with
arenes seems counter-intuitive but can be examined for the
gas phase using mass spectrometry as described below.
2
þ
monotonically, but the B M distances, obtained from the
solid-state structures, show that the M $ B distance in
NaBPh is longer than in KBPh and nearly as long as in
4
4
the Rb complex. This seemed surprising considering that
þ
the counteranions are identical and Na is the smallest of
the three cations. The data are recorded in Table 1.
2
.2 Mass spectrometric assay
Mass spectrometry was used early (1, 2) and has been
applied frequently in the study of cation–pi interactions
Table 1 also shows data for bond angles and various
2
distances. Each BPh ion has four angles: those shown are
for the arenes on the sides facing the cation. These angles
are all smaller than the expected tetrahedral angle of
4
(see above). Aggregates can be detected in both negative
and positive ion modes. The salts themselves, MBPh4,
are neutral and not observed. It is assumed that these
are present, if not predominant, in all experiments.
1
09.58. The external angles are, as expected, correspond-
ingly larger and are not discussed here. It is interesting to
note that the arenes bend towards the cations, presumably
to shorten the cation–pi contact distance. The average
deformation of the angle between the arenes for the
metallic tetraphenylborates is 13.2 ^ 0.68.
þ
In the positive ion mode, we expect to observe ½ðM Þ ꢀ
2
2
ðBPh Þꢁ which has a positive charge. Higher aggregates
4
þ
[M (BPh )
n
]
would also be observed as a negatively
4 n21
þ
2
4
charged species. In the negative ion mode, ½M ðBPh Þ ꢁ
2
would be observed as would such higher aggregates
as [M (BPh )
The boron–cation distance is consistent throughout the
solid-state matrix, measured along any axis. Figure 2
compares the observed and calculated (Gaussian 03W)
2
] . The spectra show relative abundance
n
4 n þ 1
on the ordinate axis, which is the observed ion
current normalised to the most abundant (parent) peak
set to 100%.
þ
2
M –B distances. The observed distances were obtained
from the X-ray structural data identified above. When each
trend was modelled as linear, the calculated line gave
2
r ¼ 0.99 but the attempted linear fit for the observed
2
2
.2.1 Negative ion mass spectrometry
distances was an obviously unacceptable (r ¼ ) 0.29.
Ion size and the charge-to-size ratio are not thought to
control cation–pi interactions. Kumpf and Dougherty (3)
Mass spectrometric analyses were conducted using a
JEOL MStation (JMS-700) mass spectrometer equipped
with an electrospray ionisation source, operating in the
negative ion mode and scanning from m/z 100 to 1600.
Negative ion experiments for individual salts were run at a
þ
þ
þ
predicted the order K . Rb q Na for binding
between benzene and these cations in aqueous solution
based on an electrostatic model. Caldwell and Kollman
showed that polarisability was an important additional
consideration for molecular mechanical computations of
cation–pi interactions (36). Zhu et al. (31), in their
2
4
concentration of 2 £ 10 M in CH CN solution. Only
3
three peaks were observed in the mass spectrum for
NaBPh . They were m/z 319 (100%), 661 (60%) and 1003
4
2
4
2
(,1%). These peaks correspond to BPh , [Na(BPh ) ]
and [Na (BPh ) ] , respectively. As noted above, the
4 2
2
2
4 3
neutral salt NaBPh cannot be observed by an ion detector.
4
The relative intensities of the cluster peaks (e.g.
[
2
Na (BPh ) ] ) were, as expected, greater at higher salt
2 4 3
concentrations (data not shown). The mass spectrum is
shown in Figure 3.
Similar, but not identical, data were obtained for KBPh4,
RbBPh , CsBPh , NH BPh and NMe BPh . In all cases, the
4
4
4
4
4
4
2
4
BPh ion (m/z 319) is the base peak but clusters involving a
2
cation and two anions [MX ] as well as two cations and
2
2
three anions [M X ] are also typically observed. The latter
peak is weak in all cases but nearly undetectable in the
2
3
2
NaBPh and Me NBPh cases. The [M(BPh ) ] ion has a
relative abundance of 58 ^ 3% for sodium, potassium
þ
2
Figure 2. Calculated and observed M · · ·B bond distances for
left to right) NaBPh , KBPh , RbBPh and CsBPh .
4
4
4
4 2
(
4
4
4
4

7
6
R. Li et al.
the abundance for caesium was nearly twice that for
sodium, potassium or ammonium and three times that of
tetramethylammonium. We interpret this to mean that
caesium does, indeed, have an unusually high affinity for
arenes, as asserted by Zhu et al., based on NMR
experiments (31). This seems not to be simply a size-
based phenomenon as tetramethylammonium cation is
larger still, singly charged and less capable of organising
two tetraphenylborate anions.
2.2.2 Positive ion mass spectrometry
The experiments were conducted as noted above but the
instrument was operated in the positive ion mode while
scanning from m/z 100 to 1600. These experiments were run
2
4
Figure 3. Negative ion ESI spectrum of NaBPh , 2 £ 10
M
4
24
in CH
3
CN.
at a salt concentration of 2 £ 10 M in CH CN solution.
3
The number of peaks observed in the positive mode was
similar to the number apparent in the negative ion
experiments, but the spectra were much noisier. We attribute
this to residual hydrophilic cationic species not fully purged
from the system. As in the negative mode, cluster ions are
2
4
and ammonium cations ðBPh ¼ 100%Þ. The abundance of
2
M(BPh ) ] ion is significantly higher (96%) for M ¼ Cs
þ
[
4
2
þ
and significantly lower (30%) for M ¼ NMe . The data,
4
þ
apparent. In the case of 1–4, spraying M BPh in CH CN
4
3
2
summarised in Table 2, showed good consistency over three
separate experiments. The values reported for the abun-
typically resulted in two dominant peaks. These correspond
2
2
dances of [M(BPh ) ] were in the range of ^3 to ^5%.
þ
þ
2
4
2
to ðM Þ BPh and ðM Þ ðBPh Þ .
2
4
3
4
2
Based simply on the Lewis acidity of the metal
þ
The positive ion spectrum of NaBPh is shown in
4
þ
þ
þ
cations, one would predict the trend: Na . K . Rb
þ
Figure 4. The base peak is (Na BPh ) and a small peak
2
4
þ
.
Cs . The sizes of the cations increase in the order
þ
(relative abundance ,15%) corresponds to [Na (BPh ) ] .
The positive ion spectrum of NaBPh was unique in this
3
4 2
þ
þ
þ
þ
þ
Na , K , NH , Rb , Cs , Me4N . As noted
above, computational results (3) suggested that the
4
4
study because a major (abundance ¼ 75%) peak was
apparent at m/z 406 that corresponds to [(Na) ·BPh ·CH
3-
þ
aqueous-phase binding strengths for [M·C H ] dimin-
6
6
2
4
þ
þ
þ
þ
ished in the order K . Rb q Na . Caesium was not
included in the study. In our (gas phase) study, the
CN] . Typically, the solvent is stripped in the electrospray
ionisation process. No other peak that includes the solvent
was observed in any other spectrum of 1–5, when
conducted either in the positive or negative ion mode.
The identity of the m/z 406 peak was confirmed by an
analysis of its isotope distribution (data not shown).
Sodium cation has the highest Lewis acidity of any cation in
this series and it is expected to have the highest electrostatic
affinity for acetonitrile.
2
2
MðBPh Þ ion abundance increased in the order
4
þ
þ
þ
þ
4
þ
2
þ
þ
NMe , Na , K , NH , Rb , Cs . The order
4
þ
we observe for complexation of BPh4 is Na , K ,
þ
, Rb , which is obviously different from the benzene
þ
prediction: K . Rb q Na . Still, benzene is a nearly
two-dimensional, uncharged molecule and BPh4 has four
rings and is an anion. It is interesting to note, however, that
þ
þ
2
a
Table 2. Negative ion ESI-MS analysis of tetraphenylborate salts .
b
Observed peaks and relative abundances
2
2
]
2
Compound
BPh4 (m/z 319) (%)
[M(BPh
4
)
2
(%)
m/z
[M
2
(BPh
4
)
3
]
(%)
m/z
1
2
3
4
5
6
NaBPh4
KBPh
RbBPh
CsBPh
NH BPh4
100
100
100
100
100
100
58
58
80
96
59
30
661
677
723
771
656
712
,1
8
5
4
4
1003
1035
1126
1222
992
4
4
4
4
NMe
4
BPh
4
–
–
a
See text for experimental information.
b
Peak abundance (isotope effect included) was recorded and compared with the parent peak, which is set to 100%. Each abundance value represents three
individual experiments with a variation of less than or equal to ^5%.

Supramolecular Chemistry
77
The base peaks for NaBPh , RbBPh , CsBPh and
4
4
4
þ
4
NH BPh all correspond to the general ion M BPh .
4
4
2
Potassium cation is neither the most nor the least charge-
dense, nor is it the largest or the smallest in the group. It is
interesting to note that when the anion was changed from
tetraphenylborate to tetrakis( p-chlorophenyl)borate, the
þ
[
M X ] ion still dominated (data not shown).
3
2
þ
We were surprised by the extent to which residual Na
ions incorporated into clusters when other salts were
þ
sprayed. The near absence of Na in the CsBPh spectrum
4
was noted above. By contrast, when NH BPh (5) was
4
4
sprayed in CH CN, a peak corresponding to [NaNH
3
4-
þ
þ
BPh4] was nearly as abundant as the [(NH ) BPh ]
4 2
4
þ
þ
peak. When the NH4 cation was replaced by (CH ) N ,
3 4
the sodium-containing peak was hardly detectable when
Me NBPh was sprayed in acetonitrile.
4
4
Figure 4. Positive ion ESI-MS of NaBPh4 (1) sprayed in
CH CN solution.
2.3 Competition experiments
3
It was of obvious interest to compare a sigma-donor anion
(
2
Cl ) with BPh . Two solutions were prepared (each
2
4
By far, the most complex spectrum was observed
for NH BPh , 5, for which numerous mixed cation
100 mM in CH CN) for study: 1:1 NaCl þ KBPh and 1:1
3
4
4
4
NaBPh þ KCl. Each mixture was sprayed and analysed in
4
þ
peaks involving Na were apparent (spectrum not shown).
In contrast to the other positive ion spectra, the spectrum
obtained for CsBPh (4) was remarkably free of competing
the negative ion mode with comparable results. Three
peaks were observed at relative abundances .5%. They
2
2
4
were BPh4 (m/z 319, base), [Na(BPh4)2]
2
and
2
[K(BPh ) ] . No significant ion that included Cl was
4 2
þ
Na ions, which is shown in Figure 5.
The data acquired for the series of compounds 1–6 are
observed. A similarly prepared mixture of KCl and
RbBPh showed (negative mode) three significant peaks.
summarised in Table 3. For all of the salts that were
þ
4
2
4
2
They were BPh (base peak), [K(BPh ) ] (m/z 677) and
analysed except KBPh , the [M (BPh ) ] ion was the
4
3
4 2
4 2
þ
þ
2
[Rb(BPh4)2] (m/z 723). Three peaks were also observed
base peak. For KBPh , the (M BPh ) and [M (BPh ) ]
4
2
4
3
4 2
peaks fall within the range of 97 ^ 3% but the intensity of
þ
when KCl and CsBPh were mixed. The base peak was
4
þ
2
2
the [K (BPh ) ] peak is surprising as K is not the
3
BPh , the second most abundant peak was [Cs(BPh ) ]
4 2
2
(m/z 771) and a minor peak was observed for [K(BPh ) ] .
4 2
4
strongest Lewis acid in this family.
It is notable that the base peak in the positive ion mode
4
2
þ
In these comparisons involving different cations versus K
þ
þ
2
and the potential for Cl or BPh4 association, Cs and
þ
2
for K is a cluster having the elements [K (BPh ) ] .
3 4 2
2
BPh4 appear to be dominant.
In the positive ion mode, a similar experiment gave
more complicated spectra. In this case, KCl and CsBPh4
solutions were prepared (500 mM, 1:1 CH CN:CH OH)
2
x 10–4 M in CH CN
3
3
3
and then sprayed. Six peaks were detected in the positive
ion mode. Only four of these had abundances .5%. They
þ
þ
were [KCsBPh ] (m/z 490, 11%), [Cs BPh ] (m/z 584,
4
2
4
þ
þ
base), [KCs (BPh ) ] (m/z 944, 6%) and [Cs (BPh ) ]
2
4 2
3
4 2
(
[
m/z 1038, 25%). Peaks attributable to the composition
þ
þ
K BPh ] and [K Cs(BPh ) ] were also observed, but
4 2
2
4
2
their abundances were only about 1%. Again, the most
þ
2
2
significant ions were attributable to Cs and BPh4
þ
complexes. The K complex was less significant and Cl
complex was not detected.
In order to directly compare cations, we mixed
equimolar solutions of NaBPh4 and MBPh4 (together
200 mM salts) in CH CN and analysed the solution using
the negative ion mode of electrospray (three runs).
3
4
Figure 5. Positive ion ESI-MS of CsBPh (4) sprayed in
CH CN solution. The insets show the isomer distributions of the
3
adjacent peaks.
2
We compared the abundance of the [M(BPh ) ] peak
4
2

7
8
R. Li et al.
Table 3. Positive ion electrospray mass spectrometry of tetraphenylborate salts.
Relative abundance
(%) [(Na·M)BPh
þ
þ
þ
þ
)
3
Compound
(M
2
BPh
4
)
(%)
[M
3
(BPh
4
)
2
]
4
]
(%)
4 4
[M (BPh ] (%)
1
2
3
4
5
6
NaBPh
KBPh4
RbBPh
4
100
94
100
100
100
100
14
100
73
35
14
,3
–
23
21
,5
92
,3
,5
,5
,5
,5
,5
–
4
CsBPh
NH BPh4
4
4
NMe BPh4
4
2
observed for the second cation with the [Na(BPh ) ]
4
2.4 The gas-phase cation–pi interaction
2
peak, arbitrarily normalised to 1.0. Compared with the
þ
The cation–pi interaction is a weak force. The
tetraphenylborate anion possesses four electron-rich
benzene rings that are positioned so that two within the
same structure may simultaneously interact with a cation.
The charge on boron is both highly diffuse and sterically
hindered. It is not expected to directly interact with a
complexed (associated) cation. The solid-state structures
of NaBPh , KBPh , RbBPh and CsBPh all suggest that
value for Na of 1.0, the following abundances were
þ
þ
þ
observed: K , 1.2; Rb , 2.5; and Cs , 5.0. In direct
þ
þ
þ
þ
competition, the order is Na , K , Rb , Cs .
The data shown in Table 2 compare the abundances of
2
2
[Na(BPh ) ] and [M(BPh ) ] for the salts 1–4. If we set
4 2 4 2
the abundance of the [Na(BPh ) ] to 1.0, the Na:K:Rb:Cs
2
4
2
ratio is 1:1:1.4:1.7. In the direct comparison with NaBPh ,
4
4
4
4
4
2
the [M(BPh ) ] ions show an almost identical trend. This
4
2
orderly and largely symmetrical interactions should occur
between the cation and the anion. The calculated M–B
distances for the MBPh salt (complex) in the gas phase
þ
further confirms the greater ability of Cs to interact with
2
BPh . The crystal structures illustrated in Figure 1 show
þ
4
4
that Rb is large enough that no steric effect is likely to
þ
increase monotonically from sodium through caesium.
The observed bond distances in the solid state do not
agree: sodium cation has a longer than expected distance.
This is likely due to the steric demand of the large anion.
The K–B, Rb–B and Cs–B distances fall on a straight
line, whether the distances are calculated or observed.
There is no indication that caesium will behave differently
with tetraphenylborate than do the other cations. In fact,
caesium shows a higher affinity for tetraphenylborate than
do the other cations, in a variety of negative and positive
ion electrospray experiments.
play a role. This confirms the high affinity of Cs for
2
BPh4 and, by extension, for arenes generally.
A further experiment was undertaken in which a
þ
þ
solution (200mM, CH CN) containing equimolar Na , K ,
3
þ
þ
2
4
Rb and Cs BPh salts was sprayed and analysed in the
positive mode. Such an experiment carries with it the
inherent error of observing residual cations in addition to
those present in the solution (see above). Seven ions were
2
4
detected in which a single BPh ion appeared to accompany
two cations. The masses of five ions could be fitted to the
þ
formula [M (BPh ) ] . The data are summarised in Table 4.
3
4 2
Zhu et al. (31) have reported deuterium NMR
experiments with both deuterated benzene and polyaro-
matic hydrocarbons that suggest a special affinity of Cs for
arenes. They attribute this affinity, which is apparent in the
results presented here, to the high polarisability of this
cation. Caldwell and Kollman (36) have concluded that
polarisability is important in modelling the cation–pi
interaction and that generalisation comports with the
results presented here.
Nine of the 12 significant ions include the caesium cation.
The four ions having the highest relative abundances all
contain one or more caesium ions.
a
Table 4. Four-cation competition study .
þb
þb
[M M M (BPh ) ]
1 2 3 4 2
[
M M BPh ]
1
2
4
c
c
m/z
Amt.
M
1
M
2
m/z
849
896
944
989
1037
Amt.
1 2 3
M M M
365
381
427
474
490
537
584
23
17
15
45
30
33
54
Na2
NaK
NaRb
NaCs
KCs
10
14
17
17
9
K Cs
2
3.
Conclusions
KRbCs
KCs
2
Computations by Kumpf and Dougherty (3) predicted the
þ
RbCs
Cs3
2
following binding order with benzene (aqueous phase): K
þ
þ
þ
þ
.
.
Rb q Na . Zhu et al. (31) found the order Cs . K
Rb . Na for aqueous methanol using deuterium
RbCs
Cs
þ
þ
2
NMR methods. The order for the latter three cations agrees
generally with the aqueous-phase calculations. The order
a
Equimolar NaBPh
positive ion mode.
Observed ion.
Relative abundance.
4
, KBPh
4
, RbBPh
4
4 3
and CsBPh , in CH CN, ESI-MS
þ
b
that we observe for the tetraphenylborate anions is Cs
þ
c
þ
þ
. Rb . K , Na . Admittedly, this binding order

Supramolecular Chemistry
79
reflects contacts with electron-rich arenes, which are
expected to be more polarisable than benzene. We agree
56.66 (CH ), 123.16 (C-4), 126.96 (C-2, C-6), 137.15
3
(C-3, C-5), 163.40 (C-1). MS-FAB (negative): m/z
(Me N·BPh þ BPh , C H NB ) calculated 712.4286,
þ
with the Zhu et al. finding that Cs shows the strongest
4
4
4
52 52
2
arene–cation interaction but conclude that the polarisa-
bility dominates over electrostatics for an electron-rich
arene and a weakly Lewis acidic cation, leading to an
apparently reasonable order. The Mullikan charge
calculation showed that the atomic charge on the metals
found 712.4302.
4.3 Electrospray ionisation mass spectrometry
Mass spectra were obtained using a JEOL MStation
(JMS-700) mass spectrometer equipped with an electro-
þ
þ
þ
þ
decreased in the order Cs . K . Rb . Na . We note
that the validity of this methodology is further supported
by two experimental observations. First, ESI-MS shows
spray ionisation source, operating in the negative or positive
ion mode and scanning from m/z 100 to 1600. Slits were set
to achieve a resolution of about 2000. Stock solutions were
þ
that only Na includes acetonitrile in a complex cation.
þ
Second, only Cs shows ESI-MS spectra that are devoid of
interference by other residual cations. The latter
observation confirms the order listed above and the former
prepared in CH CN at 1 mM for each tetraphenylborate salt
3
and in CH OH at 1 mM for NaCl and KCl. The tested
3
mixing solution was injected and kept at a flow rate of
50ml/min using a Harvard syringe pump. The spray voltage
þ
comports with the high Lewis acidity of the Na cation,
showing that it is behaving normally in these experiments.
was 2.00kV, and the capillary temperature (desolvating
temperature) was 2008C. Each trial was processed using the
MSMP9020D software supplied by JEOL with a minimum
of 21 scans averaged for the final spectral presentation.
Mass values in Figures 4 and 5 were not corrected. They
varied from theory by 1 mass unit.
4
.
Experimental section
4
.1 General
NMR spectra were recorded by a Varian Unity Plus
00 MHz NMR spectrometerand a BrukerAvance300 MHz
3
NMR spectrometer. H NMR was recorded at 300 MHz in
1
4.4 Computational details
CD CN solvents and is reported in ppm (d) downfield from
3
1
internal TMS unless otherwise noted. C NMR was
3
All calculations were performed using the Gaussian 03W
suite of programs. The B3LYP as one of the density
functional theory methods was used for all geometry
optimisation and single-point energy calculations.
The method combines Becke’s three-parameter functional
1
1
recorded at 75MHz in CD CN unless otherwise stated.
B
NMR was recorded at 96MHz in CD CN unless otherwise
3
3
stated. Melting points were determined on a Thomas Hoover
apparatus in open capillaries and are uncorrected. Sodium,
potassium, rubidium, caesium and ammonium tetraphenyl-
borate are the best (non-LC) grade commercially available
and were distilled, recrystallised or used without further
purification, as appropriate. ESI-MS is the abbreviation for
electrospray ionisation mass spectrometry. Acetonitrile is of
HPLC grade, which is suitable for mass spectrometry
analysis. Fast atom bombardment (FAB) mass spectrum was
obtained using a JEOL MStation (JMS-700) mass
spectrometer (in UMSL).
(
38) with the non-local correlation provided by the
correlation functional of Lee et al. (39), which has been
shown to be suitable for vibrational frequency calculations
and geometry optimisation. We performed the restricted
B3LYP with the LanL2DZ (D95Von first row, Los Alamos
ECP plus DZ on Na–Bi; http://www.gaussian.com) basis
set. For each molecule, several reasonable conformations
were calculated to compare the final energies of stable
conformations and the one having the lowest energy was
chosen as the final structure. For the chosen structure, all
geometries were completely optimised in the gas phase.
4
(
.2 Tetramethylammonium tetraphenylborate
Me N·BPh )
4
4
Acknowledgements
It was prepared by a simplified procedure. Tetramethyl-
ammonium chloride (10.96 mg, 10 ml, 0.01 M solution in
CH CN) and sodium tetraphenylborate (34.22 mg, 10 ml,
We thank the NSF (CHE-0415586) and NIH (GM 36262) for
grants that supported this work.
3
0
.01 M solution in CH CN) were mixed and stirred at
3
room temperature for 1 h. The solvent was evaporated and
the residue was crystallised in CH CN–H O (1:9, v/v).
A white solid, 25.01 mg (64%), was obtained. Melting
References
3
2
(
1) Sunner, J.; Nishizawa, K.; Kebarle, P. J. Phys. Chem. 1981,
85, 1814–1820.
2) Meot-Ner, M.; Deakyne, C.A. J. Am. Chem. Soc. 1985, 107,
1
point: 368–3708C. H NMR (CD CN, 300 MHz) d (ppm):
3
(
3
.04 (s, 12H), 6.84 (t, 4H, J ¼ 7.20 Hz), 7.00 (t, 8H,
474–479.
(3) Kumpf, R.A.; Dougherty, D.A. Science 1993, 261,
1708–1710.
1
1
J ¼ 7.35 Hz), 7.27 (m, 8H). B NMR (CD CN, 96 MHz)
3
1
d (ppm): 66.62 (s). C NMR (CD CN, 75 MHz) d (ppm):
3
3

8
0
R. Li et al.
(
(
4) Doyle, D.A.; Cabral, J.M.; Pfuetzner, R.A.; Kuo, A.;
Gulbis, J.M.; Cohen, S.L.; Chait, B.T.; MacKinnon, R.
Science 1998, 280, 69–77.
5) Heginbotham, L.; Lu, Z.; Abramson, T.; MacKinnon, R.
Biophys. J. 1994, 66, 1061–1067.
(22) De Wall, S.L.; Meadows, E.S.; Barbour, L.J.; Gokel, G.W.
J. Am. Chem. Soc. 1999, 121, 5613–5614.
(23) Hu, J.; Barbour, L.J.; Gokel, G.W. Chem. Comm. 2001,
1858–1859.
(24) Hu, J.; Barbour, L.J.; Gokel, G.W. J. Am. Chem. Soc. 2001,
123, 9486–9487.
(
(
6) Dougherty, D. Science 1996, 271, 163–168.
7) Ma, J.C.; Dougherty, D.A. Chem. Rev. 1997, 97,
(
25) (a) Gokel, G.W.; Barbour, L.J.; Ferdani, R.; Hu, J. Acc.
Chem. Res. 2002, 35, 878–886. (b) Gokel, G.W. Chem.
Commun. 2003, 23, 2847–2852.
1
303–1324.
8) Gallivan, J.P.; Dougherty, D.A. Proc. Natl Acad. Sci. USA
999, 96, 9459–9464.
9) Burley, S.K.; Petsko, G.A. FEBS Lett. 1986, 203, 139–143.
(
1
(
(
(
(
26) Dupont, J.; Suarez, P.A.; De Souza, R.F.; Burrow, R.A.;
Kintzinger, J.P. Chem. Eur. J. 2000, 6, 2377–2381.
27) Fooladi, E.; Dalhus, B.; R u` mming, C.; Tilset, M. Acta
Cryst. 2002, C58, m567–m569.
(
(
(
(
10) Flocco, M.M.; Mowbray, S.L. J. Mol. Biol. 1994, 235,
09–717.
11) Guo, B.C.; Purnell, J.W.; Castleman, A.W., Jr. Chem. Phys.
Lett. 1990, 168, 155–160.
12) (a) Cabarcos, O.M.; Weinheimer, C.J.; Lisy, J.M. J. Chem.
Phys. 1998, 108, 5151–5154. (b) Cabarcos, O.M.;
Weinheimer, C.J.; Lisy, J.M. J. Chem. Phys. 1999, 110,
429–8435.
13) Scrutton, N.S.; Raine, A.R. J. Biochem. 1996, 319 (Pt 1),
–8.
14) Dunbar, R.C. J. Phys. Chem. A 1998, 102, 8946–8952.
15) (a) Wouters, J. Protein Sci. 1998, 7, 2472–2475.
b) Dunbar, R.C. J. Phys. Chem. A 2000, 104, 8067–8074.
16) Tsou, L.K.; Tatko, C.D.; Waters, M.L. J. Am. Chem. Soc.
002, 124, 14917–14921.
17) Hunter, C.A.; Low, C.M.R.; Rotger, C.; Vinter, J.G.;
Zonta, C. Proc. Natl Acad. Sci. USA 2002, 99, 4873–4876.
18) Hunter, C.A. Angew. Chem. Int. Ed. Engl. 2004, 43,
7
28) Fukin, G.K.; Lindeman, S.V.; Kochi, J.K. J. Am. Chem. Soc.
2
002, 124, 8329–8336.
29) Bobacka, J.; Alaviuhkola, T.; Hietapelto, V.; Koskinen, H.;
Lewenstam, A.; Lamsa, M.; Pursiainen, J.; Ivaska, A.
Talanta 2002, 58, 341–349.
8
(
(
(
(
(
(
(
(
30) Wong, A.; Whitehead, R.D.; Gan, Z.; Wu, G. J. Phys.
Chem. A 2004, 108, 10551–10559.
31) Zhu, D.; Herbert, B.E.; Schlautman, M.A.; Carraway, E.R.
J. Environ. Qual. 2004, 33, 276–284.
32) Povey, A.F.; Sherwood, P.M.A. J. Chem. Soc., Faraday
Trans. 2 1974, 70, 1240–1246.
33) Bakshi, P.K.; Sereda, S.V.; Knop, O. Can. J. Chem. 1994,
(
1
(
(
(
(
(
(
(
2
7
2, 2144–2152.
34) CSD: MATPHB. Hoffmann, K.; Weiss, E. J. Organomet.
Chem. 1974, 67, 221–228.
35) Barbour, L.J. X-Seed X-Ray Crystallography Program for
Windows. http://www.x-seed.net/ 2003.
36) Caldwell, J.W.; Kollman, P.A. J. Am. Chem. Soc. 1995,
5310–5324.
19) (a) Gandour, R.D.; Fronczek, F.R.; Gatto, V.J.; Minganti,
C.; Schultz, R.A.; White, B.D.; Arnold, K.A.; Mazzocchi,
D.; Miller, S.R.; Gokel, G.W. J. Am. Chem. Soc. 1986, 108,
1
17, 4177–4178.
37) (a) Tsuzuki, M.; Uchimaru, T.; Mikami, M. J. Phys. Chem.
A 2003, 107, 10414–10418. (b) Reddy, A.S.; Sastry, N.
J. Phys. Chem. A 2005, 109, 8893–8903. (c) Tateno, M.;
Hagiwara, Y. J. Phys. Cond. Mat. 2009, 21, 064243.
4078–4088. (b) Arnold, K.A.; Viscariello, A.M.; Kim, M.;
Gandour, R.D.; Fronczek, F.R.; Gokel, G.W. Tetrahedron
Lett. 1988, 29, 3025–3028.
(
20) De Wall, S.L.; Meadows, E.S.; Barbour, L.J.; Gokel, G.W.
Proc. Natl Acad. Sci. USA 2000, 97, 6271–6276.
21) De Wall, S.L.; Barbour, L.J.; Gokel, G.W. J. Am. Chem.
Soc. 1999, 121, 8405–8406.
(38) Becke, A.D. J. Chem. Phys. 1993, 98, 5648–5652.
(39) Lee, C.; Yang, W.; Parr, R.G. Phys. Rev. B 1988, 37,
785–789.
(

Copyright of Supramolecular Chemistry is the property of Taylor & Francis Ltd and its content may not be
copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written
permission. However, users may print, download, or email articles for individual use.