Regiochemical variation in the electrophilic addition of HBr to 1-phenylprop-1-yne

Article abstract of DOI:10.1039/b300042g

The reaction of aryl alkynes with dilute methylene chloride solutions of quaternary ammonium bromide and 20% trifluoroacetic acid produces primarily the syn Markovnikov adducts of hydrogen bromide. At moderate concentrations of the bromide, the principal

Full text of DOI:10.1039/b300042g

View Article Online / Journal Homepage / Table of Contents for this issue
Regiochemical variation in the electrophilic addition of HBr to
-phenylprop-1-yne
1
Hilton M. Weiss,* Kim M. Touchette, Frantz Andersen and David Iskhakov
Department of Chemistry, Bard College, Annandale-on-Hudson, New York 12504, USA
Received 7th January 2003, Accepted 30th April 2003
First published as an Advance Article on the web 15th May 2003
The reaction of aryl alkynes with dilute methylene chloride solutions of quaternary ammonium bromide and 20%
trifluoroacetic acid produces primarily the syn Markovnikov adducts of hydrogen bromide. At moderate
concentrations of the bromide, the principal product is the Markovnikov anti adduct. At high concentrations of
bromide, the anti-Markovnikov anti addition product predominates.
Table 1 Fractions of HBr addition products at each concentration of
tetrabutylammonium bromide
Introduction
1
In a previous paper, we showed that the electrophilic addition
of HBr to unconjugated alkynes proceeds via an Ad 3 mechan-
[Bromide]
E1B
Z1B
Z2B
E
ism where the bromide ion attacks an acid–alkyne complex in
the rate determining step. Thus, the rate was bromide depend-
ent and the initial products resulted from an anti addition.
Fahey et al. had reported that 1-phenylprop-1-yne adds to
HCl (in acetic acid) by a competition between this concerted
0.050
0.794
0.795
0.777
0.735
0.631
0.301
0.067
0.020
0.000
0.000
0.000
0.000
0.000
0.206
0.205
0.223
0.265
0.366
0.618
0.668
0.619
0.527
0.465
0.418
0.321
0.153
0.000
0.000
0.000
0.001
0.003
0.081
0.265
0.361
0.473
0.535
0.582
0.679
0.847
0
.101
2
0.209
0
0
.397
.597
mechanism and an Ad 2 process involving a vinyl cation conju-
E
0.815
0.983
1.100
1.244
gated with the aryl ring. The latter cationic intermediate was
3
also postulated in the acid catalyzed hydration of aryl alkynes
ϩ
which showed a kinetic dependence on the σ value for the
1
1
1
1
.300
.399
.600
.995
substituted aryl alkynes. The p-methoxyphenyl vinyl cation has
4
also been seen by NMR at Ϫ80 ЊC.
We initiated this study to look at these reactions more fully
under the conditions of our previous work in which we were
able to vary the acidity of the solutions and the concentrations
of the relatively nucleophilic bromide ion. The unconjugated
alkynes were reacted with a large excess of 20% trifluoroacetic
acid in methylene chloride containing variable, excess amounts
of tetra-n-butylammonium bromide. Low concentrations of
bromide ion were seen to increase the acidity of these tri-
fluoroacetic acid solutions whereas higher concentrations of
At the lowest concentration of bromide (0.05 M), the princi-
pal product was the (E)-1-bromo-1-phenylprop-1-ene (E1B).
This is good evidence for the intermediacy of a free vinyl cation
which is primarily attacked by bromide on the sterically less
hindered side (Scheme 2). This four to one preference for the
syn adduct is repeated at double the bromide concentration
and is changed little at a tenfold increase of bromide ion. We
can assume that this ratio of products is a result of the cationic
mechanism operating in solutions of high acidity and low
bromide concentration. Above a 0.4 molar concentration of
bromide ion, we begin to find the anti-Markovnikov product,
5
bromide decreased the acidities as measured by their abilities
1
to protonate a Hammett base. It should be noted that part of
the acid-weakening effect of the concentrated salt solutions
derives from the dilution of the acid by large volumes of salt.
In a 2.0 molar solution of bromide, the concentration of
the “20%” trifluoroacetic acid is only 8%. Over this range, the
H -pK value was found to rise only 0.04 units for each one
(
Z)-2-bromo-1-phenylprop-1-ene (Z2B) being formed, pre-
sumably by a concerted Ad3 mechanism.† The decreasing acid-
ity slows the independent protonation of the alkyne while the
increasing bromide concentration also promotes its attack on
o
a
percent drop in the concentration of trifluoroacetic acid when
diluted by solvent. By contrast, the dilution of a 20% tri-
fluoroacetic acid solution by the quaternary ammonium salt to
5
a 12% TFA solution spans more than four H units. The simple
o
TFA dilution factor is therefore seen to be insignificant.
Results and discussion
Product studies
Scheme 2
Our initial studies looked at the addition of HBr to 1-phenyl-
prop-1-yne as this substrate could provide data on the regio-
selectivity and stereoselectivity of the reaction. Three products
were found (Scheme 1) and these results are reported in Table 1.
†
The increasingly important role of the nucleophile in these reactions
has prompted us (as well as Fahey and co-workers) to drop the “E”
from the Ad 3 descriptor for this mechanism.
E
Scheme 1
2
148
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 1 4 8 – 2 1 5 1
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3

View Article Online
Ϫ1
Ϫ3
the acid–alkyne complex. At the same time, the proportions of
anti added Markovnikov product (Z1B) increases as a result of
the antiperiplanar bromide attack on the acid–alkyne complex.
If we assume a constant 4 : 1 ratio of Markovnikov products
Table 2 Rate constants (min × 10 ) for each mechanism and total
[Bromide]
k (PPy ϩ Br)
k (Ad 2)
k (Ad3)
E
0
0
0
0
.050
.101
.209
.397
481
347
385
100
481
347
377
93.0
0.000
0.000
8.25
7.47
4.74
from the Ad 2 mechanism, we can partition the fraction of
E
reaction occurring by each route (Fig. 1a and 1b). We can use
this to partition the pseudo-first order rate constants into rate
constants for each mechanism.
0.597
23.1
18.4
0
0
1
1
1
1
1
.815
.983
.100
.244
.300
.399
.600
0.924
0.139
0.091
0.041
0.044
0.028
0.009
0.350
0.012
0.002
0.000
0.000
0.000
0.000
0.574
0.127
0.089
0.041
0.044
0.028
0.009
Kinetic studies
1,5
In our previous studies, we showed that the pseudo-first order
rate constants for the addition of HBr to alkynes divided by the
concentrations of bromide for these additions were linearly
proportional to the pseudo-first order rate constants for the
addition of HBr to the corresponding alkenes under the same
conditions. This was strong kinetic evidence that alkenes react
by the Ad 2 mechanism while alkynes use the Ad3 route includ-
E
ing the bromide ion in the rate determining step. The rate con-
stants for the addition of HBr to 1-phenylprop-1-yne are shown
in Table 2 along with the rate constants for each mechanism
calculated from the partitioning of the products.
When we plot the logarithm of the rate constants for the HBr
addition to 1-phenylprop-1-yne against the logarithm of the
rate constants for the Ad 2 addition to oct-1-ene under the
E
same conditions (Fig. 2a), we see clear curvature. When we plot
the logarithm of the calculated pseudo-first order rate con-
stants for the Ad 2 portion of this reaction against the log of
E
the pseudo-first order rate constants for the HBr addition to
Fig. 1 (a) Fraction of products coming from the Ad 2 route vs.
E
bromide concentration. (b) Fraction of products coming from the Ad3
route vs. bromide concentration.
As the concentration of bromide ion in the reaction medium
increases, the acidity decreases, the Ad 2 reaction slows signifi-
E
cantly and the concerted mechanism produces an increasing
fraction of the products. The Markovnikov adduct (Z1B)
predominates at the mid-range concentrations of bromide
but as the solution becomes even less acidic (above 1.3 M
bromide), the anti-Markovnikov adduct (Z2B) becomes the
major product. We initially believed that the increasing bromide
concentration weakening the acidity of the reaction medium
caused the transition state to develop less cationic character
and, therefore, less preference for the Markovnikov product.
However, this explanation fell short of explaining the continu-
ation of the trend to the point of favoring the anti-Markov-
nikov product in the most concentrated bromide solutions.
2
Fahey et al. found a similar distribution pattern of the corre-
sponding chloride products in the addition of HCl (in acetic
acid) to 1-phenylprop-1-yne. With no added chloride, the syn
Markovnikov adduct predominated. Added tetramethyl-
ammonium chloride increased the proportion of the anti
adduct and also produced a small amount of the (Z)-2-chloro-
1
1
-phenylprop-1-ene. This anti-Markovnikov product rose to
6% of the chloride anti-addition adducts at the maximum
chloride concentration (1.0 molar) but never became pre-
dominant. As in our studies, the anti-Markovnikov (E) isomer
was never observed. Unlike our results, the increased chloride
ion concentrations increased the rates of these reactions as it
had less of an acid-weakening effect in the acetic acid solvent.
With those results in hand, Fahey was comfortable postulating
Fig. 2 (a) Logarithm of pseudo-first order rate constants for the
HBr addition to 1-phenylprop-1-yne vs. HBr addition to oct-1-ene.
(
b) Logarithm of pseudo-first order rate constants for Ad 2 (᭺) and
E
Ad3 (᭹) contributions to the HBr addition to 1-phenylprop-1-yne vs.
HBr addition to oct-1-ene (Log ktotal is included for comparison).
a competition between the Ad 2 and Ad3 mechanisms.
E
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 1 4 8 – 2 1 5 1
2149

View Article Online
oct-1-ene (Ad 2 behavior) we find a linear relationship fitting
the overall rate curve at the lower concentrations of bromide
Table 3 GC and NMR data on HBr adducts from 1-phenylprop-1-
yne.
E
(
Fig. 2b). Similarly, if we plot our calculated pseudo-first order
NMR methyl
rate constants for the Ad3 portion divided by the bromide
concentrations of the solutions in which the reactions were
performed versus the rate constants for the HBr additions
to oct-1-ene, we find a good linear relationship (with one out-
lying data point) fitting the overall rate at the higher concen-
trations of bromide. This supports Fahey’s idea that the AdE2
mechanism predominates at low bromide concentrations but
that increased bromide causes the Ad3 mechanism to be
favored.
Product
GC%
NMR vinyl protons
6.24 ppm (q)
protons
E1B
Z1B
Z2B
3.4%
61.2%
35.5%
5%
57%
38%
1.64 ppm (d)
1.93 ppm (d)
2.48 ppm (s)
6
4.5%
6.24 ppm (q)
6.68 ppm (s)
35.5%
concentrations of bromide ion, the anti-Markovnikov product
predominates. The reason for this will be addressed in the sub-
sequent paper.
Effect of acid concentration
Experimental
Recognizing that the 20% trifluoroacetic acid is diluted to an
8
% acid solution by the addition of the quaternary ammonium
The 1-phenylprop-1-yne, trifluoroacetic acid and solvents
(HPLC grade) used in these experiments were obtained
from Aldrich Chemical Co. and were used without further
purification. The 1-phenylprop-1-yne was also obtained from
GFS Chemicals. The quaternary ammonium salts were
obtained from Fluka Chemical and from Aldrich Chemical Co.
and were kept in a desiccator prior to use. The tetrabutyl-
ammonium bromide was dried under vacuum at regular
intervals.
Reactions were performed in glass-stoppered volumetric
flasks by adding 1 or‡ 2 drops of the alkyne§ to 5–100 mL¶
of the 20% trifluoroacetic acid in methylene chloride solution
containing the quaternary ammonium salt. Aliquots (approx-
imately 0.5 mL) were removed and quenched with 15 mL of
water and 10 mL of hexanes. The hexane layer was washed
with another 15 mL of water and dried over anhydrous potas-
sium carbonate prior to GC-MS analysis. Chromatographic
peaks were identified by their mass spectrum as well as by
a comparison of a mixture analyzed by NMR. Relative
detector responses for the isomeric vinyl bromides were
assumed to be equal and this was supported by the similar
quantitative results obtained from the NMR analysis (vide
supra). The vinyl bromides were shown to be stable under the
reaction conditions in spite of a slow addition of a second
molecule of HBr. Unless otherwise noted, all reactions were run
at room temperature (20 ± 2 ЊC) and showed no evidence of
exothermicity.
bromide to make it 2.0 M in salt, we sought to see whether
similar results might be found with a weak acid solution
prepared with less acid and less salt. A 1.0 M salt solution
was made with 5% trifluoroacetic acid in methylene chloride
giving a final concentration of 3.5% trifluoroacetic acid. The
1
-phenylprop-1-yne was found to react very slowly in this
medium but to produce 80% anti-Markovnikov addition;
regioselectivity quite similar to the 2.0 M salt using 20% tri-
fluoroacetic acid. The regioselectivity is therefore seen to be
primarily a result of the acidity of the solution.
Effect of solvent polarity
Since methylene chloride has a relatively high dipole moment
(
1.6 D) and dielectric constant (8.9), we substituted carbon
tetrachloride (0.0 D and 2.2) in various proportions. Even
the total replacement of CH Cl by CCl had little effect on
2
2
4
the rates or product distributions. We had hoped that the
less polar solvent might promote an increase in the Ad3 anti-
Markovnikov product.
Effect of anion nucleophilicity
We wanted to examine the behavior of the iodide ion in
solutions of the same acidity but we found it impossible to
determine the H of iodide solutions due to the unavoidable
o
oxidation of iodide. To circumvent this problem, we used a
solution which was 0.8 M in bromide and in iodide and com-
pared the relative production of isomers from each anion.
Under these conditions, the anti-Markovnikov adduct com-
prised 44% of the vinyl iodides while it accounted for 39%
of the bromide adducts. Although the more nucleophilic
iodide ion produced 90% of the hydrogen halide adducts, the
regioselectivity from both anions was not significantly differ-
ent. This result suggests that the nucleophiles are involved
in the rate and product determining steps leading to both
Markovnikov and anti-Markovnikov products from both
anions. The more nucleophilic iodide must be increasing the
rate of formation of both vinyl iodides to maintain their ratio
so close to the ratio of vinyl bromides. Therefore all products
are being formed by Ad3 mechanisms. The explanation for this
result only became clear after the extensive studies described in
our next paper.
A preparative scale reaction on 1-phenylprop-1-yne was
analyzed by GC-MS and by NMR. The NMR spectrum
showed overlapping quartets for the vinyl protons of the
Markovnikov adducts but the methyl proton signals were well
resolved. These data are collated in Table 3. The mass spectra
of the Markovnikov adducts were virtually identical while
the anti-Markovnikov adduct gave a significantly stronger
molecular ion (196, 198).
NMR spectra were recorded on a Varian Mercury 300 MHz
spectrometer utilizing a deuterium lock and TMS as internal
reference. Mass spectra and chromatographic analyses were
performed on a Hewlett Packard 5890 Chromatograph with a
1
2 m, HP-1 capillary column and a 5971A mass selective
detector.
Acknowledgements
This work was supported by the Petroleum Research Fund of
The American Chemical Society. Portions of this work were
reported at the Northeast Regional Meeting of the American
Chemical Society in 2001 at Durham, NH.
Conclusion
In strongly acidic solutions, conjugated aryl alkynes containing
low concentrations of bromide ion, undergo protonation to
form resonance stabilized vinyl cations. In less acidic solutions
containing higher concentrations of a good nucleophile, a con-
certed mechanism becomes dominant in which the nucleophile
can attack the acid–alkyne complex. The less polarized transi-
tion state for this concerted mechanism causes an increased
proportion of anti-Markovnikov product. At the highest
Ϫ4
‡
Results from 1 or 2 drops of alkyne (approximately 10 moles) are
identical within our experimental uncertainty.
In all kinetic studies, a preanalyzed mixture of alkyne and decane was
used.
¶ At least 10 moles of bromide ion was used.
§
Ϫ3
2
150
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 1 4 8 – 2 1 5 1

View Article Online
3
(a) R. W. Bott, C. Eaborn and D. R. M. Walton, J. Chem. Soc., 1965,
84; (b) A. D. Allen, Y. Chiang, J. Kresge and T. T. Tidwell, J. Org.
References
3
1
H. M. Weiss and K. M. Touchette, J. Chem. Soc., Perkin Trans. 2,
998, 1523.
(a) R. C. Fahey and Do-Jae Lee, J. Am. Chem. Soc., 1966, 88, 5555;
Chem., 1982, 47, 775.
1
4 (a) H.-U. Siehl, F.-P. Kaufman and K. Hori, J. Am. Chem. Soc., 1992,
2
114, 9343.
(
(
b) R. C. Fahey and Do-Jae Lee, J. Am. Chem. Soc., 1968, 90, 2124;
c) R. C. Fahey, M. T. Payne and Do-Jae Lee, J. Org. Chem., 1974, 39,
124.
5 H. M. Weiss and K. M. Touchette, J. Chem. Soc., Perkin Trans. 2,
1998, 1517.
1
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 1 4 8 – 2 1 5 1
2151