Acid strength of tetrafluoroboric acid: The hydronium ion as a superacid and the inapplicability of water as an indicator of acid strength

Article abstract of DOI:10.1039/a700798a

The Hammett acidity function, H₀, of 50.7% to 79.2% aqueous fluoroboric acid has been determined by ¹³C NMR spectroscopy (the Δδ⁰ method) with mesityl oxide as indicator. At all concentrations, the H₀ values are more negative than those of sulfuric or trifluoromethanesulfonic acid solutions, increasingly so at higher concentrations. The acid strengths measured represent the acidities of various clusters of hydronium ions and water, ion-paired with fluoroborate anions; in such systems the acid strength is also dependent upon the anion. At 79.2% concentration (0.28 mol excess water per mol hydronium ions) the solution is as strong as, or slightly stronger than, pure sulfuric acid, meaning that mixtures with less excess of water are superacidic. The hydronium ion (as fluoroborate) being a stronger acid than 100% sulfuric acid means that water dissolved in hydronium fluoroborate is a weaker base than the hydrogen sulfate ion in sulfuric acid. At the same concentration (wt.%), 'fluoroboric acid' is stronger than perchloric acid, but the two are equally strong when concentration is measured in molar ratios water:acid, meaning that perchloric acid is also fully converted in concentrated solutions to hydronium perchlorate. In acetic acid solution, hydronium perchlorate and fluoroborate are again equally strong, as expected, but they are significantly weaker than phosphotungstic acid, even though the hydron-donating species should be H₃O⁺ · xH₂O, in all cases. The acid strength of H₃O⁺ is thus strongly dependent upon the nature of the anion, which means that the basicity of water is strongly dependent upon the acid. Water is, therefore, not useful as a probe base for estimating acid strengths.

Full text of DOI:10.1039/a700798a

View Article Online / Journal Homepage / Table of Contents for this issue
Acid strength of tetraÑuoroboric acid
The hydronium ion as a superacid and the inapplicability of water as an indicator
of acid strength
Dan Farcasiu*¤ and Dan Hancu
Department of Chemical and Petroleum Engineering, University of Pittsburgh,
1249 Benedum Hall, Pittsburgh, PA 15261, USA
The Hammett acidity function, H , of 50.7% to 79.2% aqueous Ñuoroboric acid has been determined by 13C NMR spectroscopy
0
(the *d0 method) with mesityl oxide as indicator. At all concentrations, the H values are more negative than those of sulfuric or
0
triÑuoromethanesulfonic acid solutions, increasingly so at higher concentrations. The acid strengths measured represent the
acidities of various clusters of hydronium ions and water, ion-paired with Ñuoroborate anions; in such systems the acid strength
is also dependent upon the anion. At 79.2% concentration (0.28 mol excess water per mol hydronium ions) the solution is as
strong as, or slightly stronger than, pure sulfuric acid, meaning that mixtures with less excess of water are superacidic. The
hydronium ion (as Ñuoroborate) being a stronger acid than 100% sulfuric acid means that water dissolved in hydronium Ñuoro-
borate is a weaker base than the hydrogen sulfate ion in sulfuric acid. At the same concentration (wt.%), “Ñuoroboric acidÏ is
stronger than perchloric acid, but the two are equally strong when concentration is measured in molar ratios water : acid,
meaning that perchloric acid is also fully converted in concentrated solutions to hydronium perchlorate. In acetic acid solution,
hydronium perchlorate and Ñuoroborate are again equally strong, as expected, but they are signiÐcantly weaker than phos-
photungstic acid, even though the hydron-donating species should be H O` É xH O, in all cases. The acid strength of H O` is
3
2
3
thus strongly dependent upon the nature of the anion, which means that the basicity of water is strongly dependent upon the
acid. Water is, therefore, not useful as a probe base for estimating acid strengths.
TetraÑuoroboric acid, or rather the solutions of boron tri-
Ñuoride in hydrogen Ñuoride, have long been considered to
have superacidic strength. Thus, dissolution of monocyclic
arenes in this medium1 was said to give “aromatic carbonium
ionsÏ2 (i.e. arenium ions). Hydrogen Ñuoride saturated with
most concentrated acid possible is the hydronium salt,
H O` É BF ~ (nominally 82.6% HBF ), a solid melting at
3
4
4
52 ¡C.15 Other combinations identiÐed in the melting point
diagram correspond to H O ` É BF ~, H O ` É BF ~ and
5
2
4
7
3
4
possibly H O ` É BF ~.17 The latter stoichiometry corre-
sponds to 49.4% Ñuoroboric acid. At this concentration, and
below, the Ñuoroborate anion might undergo, however, partial
11
5
4
BF was used in the determination of basicities of alkyl-
3
benzenes3 and 1H NMR spectra of arenium ions (Pfei†erÈ
Wizinger complexes4) were Ðrst obtained in this medium.5
hydrolysis to hydroxyÑuoroborates, BF (OH) ~.18
A
x
4~x
Equilibrium constants for the formation of MF ~ ions from
similar behaviour was noticed for boron triÑuoride hydrates
n`1
the Ñuorides MF gave the ordering PF \ NbF \ BF \
of higher than 2 : 1 water to BF ratio.19
n
5
5
3
3
TaF \ AsF .6 In a classiÐcation of superacids presented in
a monograph, the HFÈBF system was indicated to have a
Based on these considerations, the report on acid strength
5
5
of HBF in TFA seems to contain some anomalies: since TFA
3
4
Hammett acidity function, H ,7 varying from [10.8 to [17
is not basic enough to be hydronated signiÐcantly, even by
0
for BF concentrations between 0 and 7 mol%.8 In total
triÑuoromethanesulfonic acid, HBF could not exist in that
3
4
medium if it was a weaker acid than MeSO H. It is conceiv-
3
contrast with these representations, a report on the acid
strength of several acids in triÑuoroacetic acid (TFA) solution
able that the Ñuoroboric acid used in TFA solution contained
lists Ñuoroboric acid in the strength sequence: perchloric [
water, in which case more water than one mol per mol of
sulfuric [ É É É [ methanesulfonic [ HBr [ HBF [ HI.9
HBF , would be required to explain the reported results.9
4
4
There is, however, no compound with the formula of tetra-
The H values reported for the HFÈBF system8 are also
0
3
Ñuoroboric acid between room temperature and the freezing
point of the lowest melting component, [131 ¡C.10 Below
that temperature, mixed crystals of the two individual com-
ponents are formed, again with no sign of formation of the
open to question. The strongest solution reported (6 mol%
BF )8 is well beyond the solubility limit of BF (0.8 mol% at
3
3
0 ¡C under 1 atm BF 3a and 1.9 mol% at 19 ¡C under ca. 6
3
atm BF 10b). Moreover, at the lowest end of concentration
3
complex acid.11 In this respect, HBF is similar to the also
listed, the hydron-donating power given (H \ [10.8) is ca.
4
0
non-existing HAlCl and HAlBr .12 Such species are formed
three powers of ten less than the value established for anhy-
4
4
only in the presence of stoichiometric amounts of bases and
drous HF.20 It is again possible that water was present in the
mixtures, but the experiments from which the data presented
were obtained could not be assessed because no primary refer-
ence to those data was given.8
Information useful for the di†erentiation of strong acids
can be obtained from the study of their acidity function in
aqueous solution. Whereas, in dilute solution, all strong acids
are levelled o† by the basic solvent water, their hydron-
donating abilities diverge increasingly as the solution becomes
more concentrated.7 Comparative studies for Ñuoroboric acid
do not appear to have been made.
cannot be assigned H values.13 Their superacidity is assessed
0
from the weakness of bases that they hydronate” (or the basic-
ity of solvents in which they can be generated).14 Aqueous
Ñuoroboric acid is a solution of hydronium Ñuoroborate in
water, with no undissociated Ñuoroboric acid present. The
¤ E-mail: dfarca=vms.cis.pitt.edu
” The name “hydronÏ has been recommended for the cation of
hydrogen in natural isotopic distribution, to be distinguished from the
isotopically pure “protonÏ, “deuteronÏ, and “tritonÏ.16
J. Chem. Soc., Faraday T rans., 1997, 93(12), 2161È2165
2161

View Article Online
As a part of our investigation of the acid strength of
materials of interest as strong acid catalysts, we developed the
in the two steps of the titration was four). Because of its
hygroscopicity, the concentration of the most concentrated
acid was determined by Ðtting its freezing point, determined in
a sealed tube, to the curve provided in the literature.17 The
freezing temperature determined, 12 ¡C, corresponded to a
concentration of 79.2 wt.%; the titration gave a value lower
by 2È3%. All other solutions were titrated without difficulty.
The solutions of Ñuoroboric acid in glacial acetic acid were
prepared in a glove bag Ðlled with dry nitrogen. The molar
method for determination of the acidity function H , by the
0
13C NMR chemical shift di†erence at inÐnite dilution (*d0),
with unsaturated ketones as indicators.21 Examples of the
application of this method to various strong acids have been
described,21,22 as was its extension to evaluation of the hydro-
gen bond donor ability of solvents and solid surfaces.23 A
review has also been published.13 We expanded our studies to
the hydronium Ñuoroborate solutions (aqueous Ñuoroboric
acid), using mesityl oxide as an indicator, and we report our
results here.
composition of the ternary mixture HBF Èacetic acidÈwater
4
was determined by weighing the amount of the glacial acetic
acid added to a solution of HBF of known concentration.
4
The water contained in the acetic acid used was taken into
account in the calculation of the acid : water ratio in the
mixture.
Experimental
Materials
Acidity measurements
Gaseous boron triÑuoride (99.5% purity), mesityl oxide
(minimum purity: 98%), hydroÑuoric acid solution (sold as
48È50%, titrated as 49.35%), potassium hydrogen Ñuoride
(minimum purity 99%) and glacial acetic acid (98.5%, deter-
mined by titration) were used as purchased.
The determination of the 13C NMR chemical shifts for Cb
and Ca of mesityl oxide (measured at 75.468 MHz) and the
calculation of the *d and *d0 parameters has been described
previously.13,21,22
Preparation of HF monohydrate
Results and Discussion
KHF (25 g, mp 237 ¡C) was introduced into a stainless-steel
reaction vessel heated with an electric tape and was decom-
Acidity function of hydronium Ñuoroborate in water
2
The 13C NMR spectra of mesityl oxide in aqueous Ñuoroboric
acid solutions of varying concentrations gave the values
shown in Table 1 for the *d and *d0 parameters. From the
posed at 250 ¡C for 2 h. The resulting HF was condensed into
the commecial hydroÑuoric acid solution (49.35%, 25 g) in a
50 ml Kel-F receiver vessel cooled with a mixture of ice and
salt. A bu†er vessel, also made of Kel-F, was introduced
between the receiver and the decomposition reactor to avoid
contamination of the solution in the receiver with small par-
ticles of alkali-metal Ñuoride carried over by the acid vapours
during the decomposition of the salt. The connection between
the reaction vessel and the bu†er vessel was made of stainless-
steel tubing, whereas the connection between the bu†er and
the absorption vessels was made of TeÑon tubing. Another
piece of TeÑon tubing went from the absorption vessel to the
back of the hood. The hydroÑuoric acid product was titrated
as 52.85% HF, which corresponds to the concentration of the
HF monohydrate.
latter, the corresponding values for the H acidity function,
0
also given in Table 1, were obtained.21 Because of the hydro-
lysis of BF ~ in dilute solutions,18 the lowest concentration
4
investigated was 50 wt.%.
Plots of H as a function of concentration (in wt.%)
0
for sulfuric,21 perchloric,21b,25 triÑuoromethanesulfonic
(TFMSA),26 and Ñuoroboric acids are presented in Fig. 1. The
last three examples represent acids which form solid mono-
hydrates (hydronium salts). It is seen that Ñuoroboric acid is
the strongest of this group, but not much stronger than per-
chloric acid. It can also be observed that the Ñuoroboric acid
concentrations of 54.9%, 61.9% and 70.9%, o†er information
about the acid strength of the stable clusters27 H O `,
9
4
4
H O ` and H O `, respectively, each paired with the BF ~
Preparation of HBF solutions
7
3
5
2
4
anion. These values might change signiÐcantly, however, when
the counterion is changed.
The HBF monohydrate was prepared by condensing gaseous
4
The results in Table 1 also show that the 79.2% Ñuoroboric
acid is about as strong a hydron donor as (actually slightly
stronger than) 100% sulfuric acid. The reactions of the two
boron triÑuoride into the solution of the HF monohydrate. In
the installation described above, the decomposition reactor
was replaced with a BF lecture bottle and the receiver vessel
3
was cooled at [25 ¡C in an acetoneÈdry ice bath. After 3 h of
bubbling, the end of the absorption was indicated by the
appearance of white fumes at the end of the installation. In
agreement with the literature,15 a crystal mass with some
liquid was obtained upon warming to room temperature. The
Kel-F receiver containing this slurry was covered with paraf-
Ðn Ðlm and stored in a desiccator over P O . The liquid part
2
5
was the most concentrated acid amenable to investigation and
more dilute solutions were obtained by addition of water, with
cooling, in a glove bag Ðlled with dry nitrogen.
Chemical analysis of acids
The concentration of HF was determined by titration with
NaOH solution 1 M in the presence of Methyl Orange. After
reaching the equivalence point, CaCl É 2H O (1 g) was added
2
2
to the acid sample (0.5 g) to convert the NaHF to HCl, and
2
titration was continued to the equivalence point. The solu-
tions of HBF were analysed by a two-step titration, as
4
described previously.22a,24 The concentrations determined in
Fig. 1 Variation of H with the concentration of acids in aqueous
0
the Ðrst step (for HF) and second step (for boric acid) were in
agreement (the ratio between the volumes of NaOH measured
solutions: (>) HBF , this work; (=) H SO , ref. 21(b); (È)
4
2
4
CF SO H, ref. 26(b); (…) HClO , ref. 25; (L) HClO , ref. 21(b)
3
3
4
4
2162
J. Chem. Soc., Faraday T rans., 1997, V ol. 93

View Article Online
Table 1 Chemical shift di†erences, *d, for mesityl oxide (MSO) and
Ñuoroboric acid solution is a weaker base than the hydrogen
sulfate anion in sulfuric acid. Such a conclusion goes against
Hammett acidity functions, H , for hydronium Ñuoroborate in water
0
solution (aqueous HBF )
the usual representations of strong acid chemistry. The pK
value of the hydronium ion was taken, by deÐnition, as
4
a
concentration
MSO conc.
[1.74([log C
).28 Later, a value of [2.38 was determined
water
HBF
HBF : H O
(molar)
/mol MSO
(kg acid)~1
*da
in glacial acetic acid, obtained on the assumption that per-
chloric acid is fully ionized in that solution.29 A study of acidÈ
base equilibria in formic acid provided a value of [3.43,
together with a criticism of the value of [1.74 obtained “by
4
4
2
(wt.%)
(ppm)
sb
H
0
0.760
0.404
0.297
0.000
52.57
55.01
55.82
57.87c
51.7
1 : 4.56
[6.98
[4.15
[4.52
[4.92
[5.43
[5.67
[6.23
[7.07
[7.67
[8.37
deÐnitionÏ earlier.30 For sulfuric acid, a pK value of [3 was
a
listed,31 but this was later indicated to be only a predicted
value.32,33 In the same source, the pK of HCl was given as
0.661
0.362
0.110
0.000
57.18
59.00
60.62
61.29c
a
[6.1.31 Whatever we can make of these values, it is clear, and
53.5
55.1
57.2
58.8
61.1
65.4
67.5
69.4
1 : 4.24
1 : 3.97
1 : 3.65
1 : 3.42
1 : 3.10
1 : 2.58
1 : 2.35
1 : 2.15
[6.24
[6.0
is also well known, that the behaviour of acids in dilute
aqueous solutions is not an appropriate predictor of their
acidity in very concentrated solutions. At the same time, it is
an open question which of the two, water or the hydrogen
sulfate anion, would be the stronger base in a competition in
yet another medium.
The relative positions of the curves for sulfuric acid and
TFMSA below 90% concentration are due to the di†erence in
their molecular masses. It was observed earlier that, for acid
strength comparisons, the concentrations should be expressed
as molar ratios of acid to water, rather than as wt.%.22d When
the Hammett acidity function for the four acids shown in Fig. 1
is re-plotted as in Fig. 2, the points for perchloric and Ñuoro-
boric acids fall on exactly the same line, which thus seems
to measure the strength of hydronium ions in water solution,
in the absence of an undissociated acid [eqn. (3)]. In concen-
trated solution, the hydronium ion as Ñuoroborate and as per-
chlorate is a superacid.
0.631
0.351
0.194
0.000
62.13
63.81
64.76
65.92c
0.659
0.366
0.184
0.000
63.83
65.91
67.11
68.41c
[6.93
[5.70
[5.68
[3.01
[3.71
[3.94
0.555
0.328
0.220
0.000
66.82
68.11
68.73
69.98c
0.647
0.372
0.168
0.000
69.18
70.63
71.73
72.62c
0.670
0.410
0.139
0.000
73.44
74.34
75.04
75.49c
A~ ] H O` ] In H A~ ] InH` ] H O
(3)
3
2
The high acid strength of hydronium ions paired with
complex anions is in perfect agreement with the long-known
observations that water in signiÐcant amounts activates alu-
minum chloride and aluminum bromide for the catalytic
isomerization of saturated hydrocarbons34 and isomerization
of the side chain of alkylbenzenes.35 A liquid phase, or
“sludgeÏ,36 at the surface of the catalyst particles, responsible
for the catalytic activity, should be a mixture of MX É OH
0.698
0.352
0.122
0.000
74.78
76.07
76.91
77.37c
0.623
0.359
0.192
0.000
75.87
77.00
77.56
78.35c
n
2
and MX OH~ É H O` and possess superacidic strength. The
n
3
corresponding mixture in which BF is the Lewis acid was
shown, by acidity measurements, to be superacidic at molar
0.540
0.374
0.130
0.000
79.32
79.68
3
ratios water/BF lower than ca 1.4.22a
73.8
79.2
1 : 1.73
1 : 1.28
80.186 [2.11
[10.08
[12.00
3
80.46c
It was surprising to see, however, that TFMSA is weaker
than perchloric and Ñuoroboric acids (Fig. 2). One could
rationalize this observation by incomplete ionization of the
former acid in water,26b but this is a tenuous proposition,
because TFMSA also forms a solid monohydrate, melting at
0.546
0.471
0.353
0.153
0.098
0.000
80.79
80.90
81.0
81.20
81.30
81.39c
[1.08
a *d \ d(Cb) [ d(Ca). b The slope of correlation *dÈmolal concentra-
tion of MSO. c *d0, value of *d extrapolated to inÐnite dilution of
base.
acids with the indicator are shown in eqn. (1) and (2):
BF ~ ] H O` ] In H BF ~ ] InH` ] H O
(1)
(2)
4
3
4
2
H SO ] In H HSO ~ ] InH`
2
4
4
One can note that, at 79.2% concentration, the ratio Ñuoro-
boric acid : water is 1 : 1.28, which means that mixtures with
less excess of water over H O` are deÐnitely stronger than
3
100% sulfuric acid and, therefore, superacidic. Based on eqn.
(1) and (2), this observation means that the hydronium ion in
hydronium Ñuoroborate containing small amounts of water is
a stronger acid than pure sulfuric acid; reading the equations
from right to left, it means that water in the concentrated
Fig. 2 Variation of H with the molar ratio acid/H O in aqueous
0
2
solution (>) HBF , this work; (=) H SO , ref. 21(b); (È) CF SO H,
4
2
4
3
3
ref. 26(b); (…) HClO , ref. 25; (L) HClO , ref. 21(b)
4
4
J. Chem. Soc., Faraday T rans., 1997, V ol. 93
2163

View Article Online
34 ¡C37 (hydronium salt).37b It is hard to argue that TFMSA
could be only ca. 60% ionized (a B 0.6)26b in a composition
proven as hydronium triÑuoromethanesulfonate.37b Unlike
Ñuoroboric acid, for which an attempt to extrapolate the
correlation line beyond H O` É BF ~ would be meaningless.
that of dichloroacetic acid (H \ [0.75) and diÑuorochloro-
0
acetic acid (H \ [2.50),21b whereas the phosphotungstic acid
0
solution has an H value close to [4. There is no doubt,
0
however, that the hydronating species is, in each case, a sol-
vated hydronium ion, H O` É 1.9H O. Combining this
3
4
3
2
TFMSA can be studied on both sides of H O` É CF SO ~
observation with the result of the comparison between hydro-
3
3
3
and interpolation gives a value of ca. [9.5 for H in the vicin-
nium Ñuoroborate and H O` É CF SO ~, we Ðnd that the
0
3
3
3
ity of that point,26 indicating that hydronium ion paired with
the TFMSA anion is at least a three orders of magnitude
weaker acid than when paired with perchlorate or tetra-
Ñuoroborate (almost four orders of magnitude for a ratio
acid strength of hydronium ion changes very much with the
nature of the counterion. Reading eqn. (3) from right to left,
we Ðnd that the basicity of water changes very much with the
nature of the acid system. ConÑicting results for basicity of
hydroxylic compounds, including water, have been cited
before.38
AH/H O of 0.78, i.e. 1/1.28).
2
Acidity comparisons in acetic acid solution
Usefulness of water as an indicator of acid strength
A study of the acidity of acetic acid solutions of hydronium
Ñuoroborate hydrates also gave signiÐcant results. The *d and
*d0 values for mesityl oxide dissolved in mixtures of composi-
tions 1HBF : xH O : 14.3AcOH are listed in Table 2. The
The use of water as a probe base for evaluation of acid
strength, especially of solids, has been repeatedly reported in
the literature. Various methods of investigation did not agree
as to whether water was hydronated or not in some typical
solid acids. An extensive discussion among leading investiga-
tors of solid acidity could not reach a common conclusion.39
Thus, high-resolution magic-angle spinning 1H NMR spec-
troscopy at room temperature40 and deconvolution of broad-
band 1H NMR spectra of water at 4 K41 indicated that water
is partially hydronated in H-Y or H-ZSM-5 zeolites. In a
comparison of several materials by the latter method, it was
concluded that H-Y gives a higher proportion of hydronium
ions than H-ZSM-5 for a ratio of water : acid sites of 1 : 1 and,
therefore, H-Y is the stronger acid.41b Note, however, that the
examination of the *d parameter for mesityl oxide led to the
opposite conclusion.42
4
2
*d0 values for Ñuoroboric, perchloric, and phosphotungstic
acid 1(H PW
O
), the latter two taken from our previous
3
3
12 40
study,22d are compared in Fig. 3. It is seen again that the
points for Ñuoroboric and perchloric acids fall on the same
line. The heteropolyacid, however, is signiÐcantly stronger.
For x \ 2.9 (2.9 molecules of water per mole of “acidic
hydrogensÏ) one could evaluate that the solution of hydro-
nium Ñuoroborate (or perchlorate) has a strength between
Table 2 Chemical shift di†erences, *d, for MSO and Hammett
acidity, functions, H , for mixtures 1HBF : xH O : 14.3 AcOH
0
4
2
MSO
conc.
/mol
Studies by IR spectroscopy, on the other hand, have con-
cluded that no hydronation of water, up to one molecule per
acid site, takes place in zeolites. The only interaction evi-
denced was hydrogen bonding, and hydron transfer was said
to occur only if at least two water molecules were present at
the acid site, forming H O ` or higher aggregates.43 Theo-
HBF : H O
(molar)
*da
4
2
(kg. acid)~1
(ppm)
sb
H
0
0.694
0.338
0.247
0.000
44.00
45.94
46.40
47.75c
1 : 2.13
[5.39
[3.14
5
2
retical calculations (ab initio and using the density functional
theory technique) led to the same conclusion.44 In contrast,
neutron di†raction on samples of water in zeolites has indi-
cated the existence of single molecules of water hydrogen-
bonded to some sites and single hydronium ions ion-paired at
other sites.45 Finally, pumpÈprobe measurements, by pico-
second IR laser pulses, on water present in concentrations
lower than 1 : 1 to acid sites in zeolites, showed both
hydrogen-bonded water and ion-paired hydronium ions as
energy minima, the latter about 7 kcal mol~1 higher in energy
than the former.46 It was estimated that this energy di†erence
should ensure a rate of hydron transfer in the endothermic
direction of about 103.4 s~1.13
0.691
0.472
0.145
0.000
38.48
38.74
39.14
39.31c
1 : 3.84
1 : 5.92
[1.21
[0.52
0.644
0.389
0.644
0.000
36.76
36.88
37.01
37.09c
a *d \ d(Cb) [ d(Ca). b The slope of correlation *dÈmolal concentra-
tion for MSO. c *d0, values of *d extrapolated to inÐnite dilution of
base.
In the light of our results, all these studies are relevant for
the chemistry of water molecules or clusters, but do not give
information about the relative strength of the acids investi-
gated toward any other base. It has already been noted that
prediction of acidity of a catalytic material toward a substrate
has to be based on measurements with a probe base similar in
structure to the substrate.14,21 Water is certainly not useful as
an acidity indicator for any application.
This work was suported by a grant (CTS-9528412) from the
US National Science Foundation and by a grant, also from
NSF (CTS-9413698), for the purchase of a DMX300 NMR
instrument. Professor J. H. Ridd is thanked for helpful
correspondence.
References
Fig. 3 Variation of *d0 with the molar ratio H O/(strong acid) in
1
(a) D. A. McCaulay, B. H. Shoemaker and A. P. Lien, Ind. Eng.
Chem., 1950, 43, 2103; (b) M. Kilpatrick and F. E. Luborsky, J.
Am. Chem. Soc., 1953, 75, 577.
2
acetic acid (systems 1AH : xH O : 14.33AcOH): (|) HClO , ref. 22(d);
2
4
(>) HBF , this work; (+) 1/3(H PW
O
), ref. 22(d)
4
3
12 40
2164
J. Chem. Soc., Faraday T rans., 1997, V ol. 93

View Article Online
2
3
C. Reid, J. Am. Chem. Soc., 1954, 76, 3264.
Meeting on Mass Spectrometry, ASTM, Atlantic City, 1960; (d)
M. Eigen, Angew. Chem., 1963, 75, 489.
(a) D. A. McCaulay and A. P. Lien, J. Am. Chem. Soc., 1951, 73,
2013; (b) E. L. Mackor, A. Hofstra and J. H. van der Waals,
T rans. Faraday Soc., 1958, 54, 186.
28 J. N. BrÔnsted and E. A. Guggenheim, J. Am. Chem. Soc., 1927,
49, 2554.
4
5
P. Pfei†er and R. Wizinger, L iebigs Ann. Chem., 1928, 461, 132.
C. MacLean and E. L. Mackor, Discuss. Faraday Soc., 1962, 34,
165.
(a) R. Gut and K. Gautschi, J. Inorg. Nucl. Chem., (Hyman Mem.
V ol.), 1976, 95; (b) J. Devynck, A. Ben Hadid, P-L. Fabre and B.
Tremillon, Anal. Chim. Acta, 1978, 100, 343.
(a) L. P. Hammett and A. J. Deyrup, J. Am. Chem. Soc., 1932, 54,
2721; (b) L. P. Hammett, Physical Organic Chemistry, McGraw-
Hill, New York, 2nd edn., 1970; (c) C. H. Rochester, Acidity
Functions, Organic Chemistry Monographs, ed. Blomquist, Aca-
demic Press, New York, 1970, vol. 17.
29 H. Lemaire and H. J. Lucas, J. Am. Chem. Soc., 1951, 73, 5198.
30 L. P. Hammett and M. A. Paul, J. Am. Chem. Soc., 1934, 56, 827.
31 A. Albert and E. P. Serjeant, T he Determination of Ionization
Constants, Chapman and Hall, London, 1984, p. 162.
32 J. E. Macintyre, Dictionary of Inorganic Compounds, Chapman
and Hall, London, 1992, vol. 2, p. 3364.
6
7
33 A value of [3.25 was deduced in: P. A. H. Wyatt, Discuss.
Faraday Soc., 1957, 24, 162.
34 (a) C. D. Nenitzescu and I. P. Cantuniari, Ber. Dtsch. Chem. Ges.,
1933, 66, 1097; (b) C. D. Nenitzescu, M. Avram and E. Sliam,
Bull. Soc. Chim. France, 1955, 1266.
8
9
G. A. Olah, G. K. Surya Prakash and J. Sommer, Superacids,
WileyÈInterscience, New York, 1985, p. 9.
J. Bessiere, Anal. Chim. Acta, 1970, 52, 55.
35 (a) R. M. Roberts, J. Am. Chem. Soc., 1957, 79, 5484; (b) C. D.
Nenitzescu, I. Necsoiu, A. Glatz and M. Zalman, Chem. Ber.,
1959, 92, 10.
36 A. Schneider, R. W. Warren and E. J. Janoski, J. Org. Chem.,
1966, 31, 1617.
37 (a) T. Gramstad and R. N. Haszeldine, J. Chem. Soc., 1957, 4069;
(b) J. B. Spencer and J. O. Lundgren, Acta Crystallogr., Sect. 3,
1973, 29, 1923; (c) R. D. Howells and J. D. McCown, Chem. Rev.,
1977, 77, 69.
38 E. M. Arnett and C. Y. Wu, J. Am. Chem. Soc., 1962, 84, 1680.
39 Discussions in Catalytic Chemistry. I. Catalysis by Strong Solid
Acids (Post-congress workshop after the 11th International
Congress on Catalysis), Baltimore, MD, USA, 6È7 July 1996.
40 M. Hunger, D. Freude and H. Pfeifer, J. Chem. Soc., Faraday
T rans., 1991, 87, 657.
10 (a) E. C. Hughes and S. M. Darling, Ind. Eng. Chem., 1951, 43,
746; (b) M. Kilpatrick and F. E. Luborsky, J. Am. Chem. Soc.,
1954, 76, 5865.
11 S. Pawlenko, Z. Anorg. Allg. Chem., 1965, 334, 292.
12 H. C. Brown and H. Pearsall, J. Am. Chem. Soc., 1951, 73, 4081.
13 D. Farcasiu and A. Ghenciu, Prog. NMR Spectrosc., 1996, 29,
129.
14 (a) D. Farcasiu, S. L. Fisk, M. T. Melchior and K. D. Rose, J.
Org. Chem., 1982, 47, 453; (b) D. Farcasiu, Acc. Chem. Res., 1982,
15, 46; (c) D. Farcasiu, G. Marino, G. Miller and R. V. Kastrup,
J. Am. Chem. Soc., 1989, 111, 7210.
15 S. Pawlenko, Z. Anorg. Allg. Chem., 1962, 315, 291.
16 J. F. Bunnett and R. A. Y. Jones, Pure Appl. Chem., 1989, 60,
1115.
41 (a) P. Batamack, C. Doremieux-Morin, J. Fraissard and D.
Freude, J. Phys. Chem., 1991, 95, 3790; (b) L. Heeribout, V.
Semmer, P. Batamack, C. Doremieux-Morin, R. Vincent and J.
Fraissard, Stud. Surf. Sci. Catal., 1996, 101, 831; (c) for the
method, see: C. Doremieux-Morin, J. Magn. Reson., 1976, 21,
419, and subsequent papers.
17 S. Pawlenko, Z. Anorg. Allg. Chem., 1964, 328, 133.
18 M. Stacey, J. C. Tatlow and A. G. Sharpe, Advances in Fluorine
Chemistry, Butterworths, London, 1960, ch. 3.
19 (a) A. Ghenciu, MS Thesis, University of Pittsburgh, 1994; (b) the
hydrolysis of BF and HBF in much more dilute solutions was
42 T. Xu, E. J. Munson and J. F. Haw, J. Am. Chem. Soc., 1994, 116,
3
4
studied by C. A. Wamser, J. Am. Chem. Soc., 1948, 70, 1209;
1951, 73, 409.
1962.
43 (a) A. Zecchina, R. Buzzoni, S. Bordiga, F. Geobaldo, D. Scarano,
G. Ricchiardi and G. Spoto, Stud. Surf. Sci. Catal., 1995, 97, 213;
(b) F. Wakabayashi, J. N. Kondo, K. Domen and C. Hirose, J.
Phys. Chem., 1996, 100, 1442; (c) A. Zecchina, F. Geobaldo, G.
Spoto, S. Bordiga, G. Ricchiardi, R. Buzzoni and G. Petrini, J.
Phys. Chem., 1996, 100, 16584.
44 (a) F. Haase and J. Sauer, J. Am. Chem. Soc., 1995, 117, 3780; (b)
M. Krossner and J. Sauer, J. Phys. Chem., 1996, 100, 6199; (c) J.
Sauer, Science, 1996, 271, 774; (d) R. van Santen and G. J.
Kramer, Chem. Rev., 1995, 95, 637; (e) S. A. Zygmunt, L. A.
Curtiss, L. E. Iton and M. K. Erhardt, J. Phys. Chem., 1996, 100,
6663.
45 L. Smith, A. K. Cheetham, R. E. Morris, L. Marchese, J. M.
Thomas, P. A. Wright and J. Chen, Science, 1996, 271, 799.
46 (a) K. Domen (Tokyo Inst. of Technology), at the meeting of ref.
39; (b) T. Fujino, M. Kashitani, S. S. Kano, F. Wakabayashi, F.
Goto, M. Ishida, A. Wada, K. Domen and C. Hirose, in prep-
aration.
20 R. J. Gillespie and J. Liang, J. Am. Chem. Soc., 1988, 110, 6053.
21 (a) D. Farcasiu, A. Ghenciu and G. Miller, J. Catal., 1992, 134,
118; (b) D. Farcasiu and A. Ghenciu, J. Am. Chem. Soc., 1993,
115, 10901.
22 (a) D. Farcasiu and A. Ghenciu, J. Catal., 1992, 134, 126;
(b) D. Farcasiu and H. Cao, J. Mol. Catal., 1994, 87, 215;
(c) D. Farcasiu and J. Q. Li, J. Phys. Chem., 1994, 98, 6893;
(d) D. Farcasiu and J. Q. Li, J. Catal., 1995, 152, 198; (e) D. L.
Hughes, J. Phys. Org. Chem., 1994, 7, 625.
23 D. Farcasiu and A. Ghenciu, Catal. L ett., 1995, 31, 351.
24 Standard Methods of Chemical Analysis, ed. F. J. Welcher, Nos-
trand Reinhold, New York, 6th edn., 1973, vol. 2, part A, p. 574.
25 K. Yates and H. Wai, J. Am. Chem. Soc., 1964, 86, 5408.
26 (a) J. H. Ridd, personal communication; (b) M. Sampoli, N. C.
Marziano and C. Tortato, J. Phys. Chem., 1989, 93, 7532;
(c) N. C. Marziano, Org. Reactivity, 1996, 30, 29.
27 (a) K. N. Bascombe and R. P. Bell, Discuss. Faraday Soc., 1957,
24, 158; (b) H. D. Beckey, Z. Naturforsch. A, 1959, 14, 712; 1960,
15, 822; (c) P. F. Knewstubb and A. W. Tickner, presented at the
Paper 7/00798A; Received 3rd February, 1997
J. Chem. Soc., Faraday T rans., 1997, V ol. 93
2165