Flash photolytic generation of primary, secondary, and tertiary ynamines in aqueous solution and study of their carbon-protonation reactions in that medium

Article abstract of DOI:10.1021/ja9601797

A group of nine phenylynamines (PhC≡CNH₂, PhC≡CNHCH(CH₃)₂, PhC≡CNHC₆H₁₁, PhC≡CNHC₆H₅, PhC≡CNHC₆F₅, PhC≡CN(CH₂)₅, PhC≡CN(CH₂CH₂)₂O, PhC≡CN(CH₂CH₂CN)₂, and PhC≡CN(CH₃)C₆F₅) were generated in aqueous solution by flash photolytic decarbonylation of the corresponding phenylaminocyclopropenones, and the kinetics of their facile decay in that medium were studied. This decay is catalyzed by acids for all ynamines-primary, secondary, and tertiary-and also by bases for primary and secondary ynamines. Solvent isotope effects and the form of acid-base catalysis show that the acid-catalyzed path involves formation of keteniminium ions by rate-determining proton transfer to the β-carbon atoms of the ynamines. The ions generated from primary and secondary ynamines then lose nitrogen-bound protons to give ketenimines, and the ketenimines obtained from secondary ynamines are hydrated to phenylacetamides, whereas that from the primary ynamine tautomerizes to phenylacetonitrile. Keteniminium ions formed from tertiary ynamines have no nitrogen-bound protons that can be lost, and they are therefore captured by water instead, and the amide enols thus produced then ketonize to phenylacetamides. The base-catalyzed decay of primary and secondary ynamines also generates ketenimines, but protonation on the β-carbon is now preceeded by proton removal from nitrogen. Rate constants for β-carbon protonation of PhC≡CNHCH(CH₃)₂ and PhC≡CN(CH₂)₅ by a series of carboxylic acids give linear Bronsted relations with exponents α = 0.29 and 0.28, respectively, whereas inclusion of literature data for protonation of PhC≡CN-(CH₂)₅ by a group of weaker acids gives a curved Bronsted relation whose exponent varies from 0.25 to 0.97. Application of Marcus rate theory to this curved Bronsted relation produces the intrinsic barrier ΔG((+))(o) = 3.26 ± 0.19 kcal mol^-1 and the work term w(r) = 8.11 ± 0.15 kcal mol^-1.

Full text of DOI:10.1021/ja9601797

4366
J. Am. Chem. Soc. 1996, 118, 4366-4372
Flash Photolytic Generation of Primary, Secondary, and
Tertiary Ynamines in Aqueous Solution and Study of Their
Carbon-Protonation Reactions in That Medium
Y. Chiang, A. S. Grant, A. J. Kresge,* and S. W. Paine
Contribution from the Department of Chemistry, UniVersity of Toronto,
Toronto, Ontario M5S 1A1, Canada
ReceiVed January 19, 1996X
Abstract: A group of nine phenylynamines (PhCtCNH2, PhCtCNHCH(CH3)2, PhCtCNHC6H11, PhCtCNHC6H5,
PhCtCNHC6F5, PhCtCN(CH2)5, PhCtCN(CH2CH2)2O, PhCtCN(CH2CH2CN)2, and PhCtCN(CH3)C6F5) were
generated in aqueous solution by flash photolyic decarbonylation of the corresponding phenylaminocyclopropenones,
and the kinetics of their facile decay in that medium were studied. This decay is catalyzed by acids for all
ynaminessprimary, secondary, and tertiarysand also by bases for primary and secondary ynamines. Solvent isotope
effects and the form of acid-base catalysis show that the acid-catalyzed path involves formation of keteniminium
ions by rate-determining proton transfer to the â-carbon atoms of the ynamines. The ions generated from primary
and secondary ynamines then lose nitrogen-bound protons to give ketenimines, and the ketenimines obtained from
secondary ynamines are hydrated to phenylacetamides, whereas that from the primary ynamine tautomerizes to
phenylacetonitrile. Keteniminium ions formed from tertiary ynamines have no nitrogen-bound protons that can be
lost, and they are therefore captured by water instead, and the amide enols thus produced then ketonize to
phenylacetamides. The base-catalyzed decay of primary and secondary ynamines also generates ketenimines, but
protonation on the â-carbon is now preceeded by proton removal from nitrogen. Rate constants for â-carbon
protonation of PhCtCNHCH(CH3)2 and PhCtCN(CH2)5 by a series of carboxylic acids give linear Bronsted relations
with exponents R ) 0.29 and 0.28, respectively, whereas inclusion of literature data for protonation of PhCtCN-
(CH2)5 by a group of weaker acids gives a curved Bronsted relation whose exponent varies from 0.25 to 0.97.
Application of Marcus rate theory to this curved Bronsted relation produces the intrinsic barrier ∆Gqo ) 3.26 ( 0.19
kcal mol-1 and the work term wr ) 8.11 ( 0.15 kcal mol-1
.
Chart 1
1-Aminoacetylenes, commonly called ynamines, are very
reactive substances. Primary and secondary ynamines, for
example, undergo facile tautomerization to nitriles and keten-
imines, and they therefore have been observed previously only
in the gas phase or in low-temperature matrices.1 Tertiary
ynamines cannot isomerize in this way, and they have conse-
quently been prepared and characterized,2 but they are neverthe-
less quite reactive, and only one previous investigation of their
chemistry in aqueous solution has been carried out.3 We have
found that all three classes of ynamines can be generated by
photodecarbonylation of phenylaminocyclopropenones, eq 1 (R1,
R2 ) H, alkyl, aryl), and that their reactions in aqueous solutions
R1
R2
2a
H
H
H
H
H
H
2b
2c
2d
2e
2f
2g
2h
2i
CH(CH3)2
c-C6H11
C6H5
C6F5
(CH2)5
(CH2CH2)2O
CH2CH2CN
CH3
CH2CH2CN
C6F5
here the results of a detailed examination of the representative
series of phenylynamines listed in Chart 1.4
Experimental Section
Materials. Phenylaminocyclopropenones were prepared by treating
the corresponding amines with phenylchlorocyclopropenone, as shown
in eq 2, which was itself obtained either by partial hydrolysis of
may be studied using flash photolytic techniques. We present
X Abstract published in AdVance ACS Abstracts, April 15, 1996.
(1) Wentrup, C.; Winter, H.-W. Angew. Chem., Int. Ed. Engl. 1980, 19,
720-721. Buschek, J. M.; Holmes, J. L. Org. Mass Spectrom. 1986, 21,
729-731. van Baar, B. L. M.; Koch, W.; Lebrilla, C.; Terlouw, J. K.;
Weiske, T.; Schwarz, H. Angew. Chem., Int. Ed. Engl. 1986, 25, 827-828.
Terlouw, J. K.; Bergers, P. C.; van Baar, B. L. M.; Weiske, T.; Schwarz,
H. Chimia 1986, 40, 357-359. Wentrup, C.; Briehl, H.; Lorencak, P.;
Vogelbacher, U. J.; Winter, H.-W.; Maquestiau, A.; Flammang, R. J. Am.
Chem. Soc. 1988, 110, 1337-1343. August, J.; Klemm, K.; Kroto, H. W.;
Walton, D. R. M. J. Chem. Soc., Perkin Trans. 2 1989, 1841-1844.
(2) See, for example: Viehe, H. G. In Chemistry of Acetylenes; Viehe,
H. G., Ed.; Marcel Dekker: New York, 1969; Chapter 12. Gladych, J. M.
Z.; Hartley, D. In ComprehensiVe Organic Chemistry; Barton, D., Ollis,
W. D., Eds.; Pergamon: New York, 1979; pp 75-79.
phenyltrichlorocyclopropene,5 eq 3, or by treatment of phenylhydroxy-
(3) Verhelst, W. F.; Drenth, W. J. Am. Chem. Soc. 1974, 96, 6692-
6697.
(4) A preliminary account of some of this work has appeared in: Chiang,
Y.; Grant, A. S.; Kresge, A. J.; Pruszynski, P.; Schepp, N. P.; Wirz, J.
Angew Chem., Int. Ed. Engl. 1991, 30, 1356-1358.
(5) Chickos, J. S.; Patton, E.; West, R. J. Org. Chem. 1974, 39, 1647-
1650.
S0002-7863(96)00179-5 CCC: $12.00 © 1996 American Chemical Society

Flash Photolytic Generation of Ynamines
J. Am. Chem. Soc., Vol. 118, No. 18, 1996 4367
cyclopropenone5 with thionyl chloride, eq 4. The partial hydrolysis
were also thermostated at 25.0 ( 0.05 °C. Observed rate constants
were calculated by nonlinear least-squares fitting of exponential
functions.
Rates of reaction of authentic 1-(N-methyl-N-(pentafluorophenyl)-
amino)-2-phenylacetylene were measured using a Hi-Tech SF-S1
stopped-flow spectrometer operating at 25.0 ( 0.05 °C.
Results
Identification of Transients. Flash photolysis of the primary
aminocyclopropenone, 1a, or any of its secondary analogs, 1b-
e, in aqueous solution produced an immediate bleaching of the
strong absorbance of these substances at λ = 310 nm, followed
by a slower but still quite rapid decay of absorbance at this
wavelength. This slower decay was accompanied by a simul-
taneous rise in absorbance at λ = 270 nm that occurred at the
same rate, and this new absorbance then decayed away
somewhat more slowly. These spectral changes suggest that
two transient species are formed in the flash photolysis of these
aminocyclopropenones: a substance absorbing at λ ) 310 nm
and another, somewhat longer-lived, absorbing at λ ) 270 nm.
Two transient species are also formed in the flash photolysis
of arylhydroxycyclopropenones, 3, eq 5, and the transients there
was carried out by stirring an acetone (25 mL) solution of the trichloride
(1.0 g) with ice (10 g) at 0 °C for 2 h. The acetone was then removed
under vacuum at room temperature, and the residual white oil was used
directly without further purification.
The other synthesis of phenylchlorocyclopropenone was performed
by stirring a mixture of phenylhydroxycyclopropenone (0.15 g) and
thionyl chloride (1.5 mL), plus a catalytic amount of dimethylformamide
(6 µL), in a dry flask under a nitrogen atmosphere at 0 °C for 10 min.
The excess thionyl chloride was then removed under vacuum at room
temperature to leave a yellowish solid, mp 35-38 °C; this also was
used directly without further purification.
The reactions of phenylchloropropenone with amines occurred
readily and were effected simply by adding 2-4 equiv of the amine,
either neat or dissolved in dichloromethane, to a dichloromethane
solution of the chloride. With the more basic amines this usually
produced a precipitate of the amine hydrochloride, which was removed
by filtration. Evaporation of the solvent from the filtrate then left a
crystalline solid, which was purified either by chromatography or by
recrystallization. The phenylaminocyclopropenones so produced were
characterized by their infrared, mass, and NMR spectra; details are given
in Table S1.6
have been identified9 as the ynol, 4, produced by photodecar-
bonylation of 3, and the ketene, 5, formed by isomerization of
4. In aqueous solution the ketene becomes hydrated, giving
the arylacetic acid final product, 6. A similar process can be
formulated for the aminocyclopropenone systems studied here,
as shown in eq 6: photodecarbonylation of the aminocyclopro-
N-Methylpentafluoroaniline was prepared by methylating the N-(p-
toluenesulfonyl) derivative of pentafluoroaniline and then removing the
sulfonyl group. The methylation was performed using methyl sulfate
and the sulfonamide sodium salt, and sulfonyl group removal was
effected with sulfuric acid in acetic acid solution. The final product
was characterized by its mass and NMR spectra; details are given in
Table S1.6
1-(N-Methyl-N-(pentafluorophenyl)amino)-2-phenylacetylene was
synthesized by the reaction of phenylchloroacetylene7 with the lithium
salt of N-methylpentafluoroaniline, generated by treating the aniline
with butyllithium. The ynamine was characterized by its mass and
NMR spectra; details are given in Table S1.6
N-(Phenylacetyl)piperidine and N-(phenylacetyl)morpholine were
prepared by treating phenylacetyl chloride with the corresponding
amines; their spectral properties are listed in Table S1.6
peneone, 1, gives the ynamine, 2, whose isomerization then leads
to the ketenimine, 7, and hydration of that produces the
acetamide final product, 8. Evidence that this sequence of
reactions does in fact take place in the presently studied systems
comes from a previous investigation of ketenimine hydration.10
The phenylketenimines investigated there had strong absorption
bands at λ = 270 nm, and their hydration to the corresponding
phenylacetamides occurred with rate constants numerically
identical with the rate constants for decay of the 270 nm
absorbance determined here. This establishes the identity of
the longer lived of the two transient species observed here as
ketenimines.
This comparison with previous work can be made for three
of our ketenimines, those whose nitrogen substituents are
isopropyl (7b), cyclohexyl (7c), and phenyl (7d). As did the
previous investigators, we found the hydration of these sub-
stances to be acid-catalyzed and also to have a significiant
uncatalyzed component. Rates of the acid-catalyzed reactions
of 7b and 7d were measured in dilute perchloric acid solutions
over the concentration range 0.02-0.10 M, and rates of the
uncatalyzed reactions of 7b and 7c were measured in 0.010 M
All other materials were the best available commercial grades.
Kinetics. Flash photolysis rate measurements were made using both
a conventional flash lamp system and an excimer laser system that
have already been described in detail.8 Excitation in the conventional
system8a was provided by a pair of xenon lamps that produced light
over the entire spectral region from the ultraviolet through the visible;
the laser system8b operated at λ ) 248 nm. In both cases the
temperature of the reacting solutions was controlled at 25.0 ( 0.05
°C.
Some reactions were too slow to be monitored accurately by flash
photolysis, and these were therefore followed using Cary 118 and 2200
spectrometers. The reactions were first initiated by a single flash from
the conventional flash system, and the reacting solutions were then
quickly transferred to the Cary instruments, whose cell compartments
(6) Supporting information; see paragraph at the end of this paper
regarding availability.
(7) Organic Synthesis; Wiley: New York, 1973; Collect. Vol. 5, pp 921-
(9) Chiang, Y.; Kresge, A. J.; Hochstrasser, R.; Wirz, J. J. Am. Chem.
Soc. 1989, 111, 2355-2357. Chiang, Y.; Kresge, A. J. Popic, V. V. J. Am.
Chem. Soc. 1995, 117, 9165-9171. Wagner, B. D.; Zgierski, M. Z.; Lusztyk,
J. J. Am. Chem. Soc. 1994, 116, 6433-6434.
(10) McCarthy, D. G.; Hegarty, A. F. J. Chem. Soc., Perkin Trans. 2
1980, 579-591.
923.
(8) (a) Chiang, Y.; Hojatti, M.; Keeffe, J. R.; Kresge, A. J.; Schepp, N.
P.; Wirz, J. J. Am. Chem. Soc. 1987, 109, 4000-4009. (b) Andraos, J.;
Chiang, Y.; Huang, G.-C.; Kresge, A. J.; Scaiano, J. C. J. Am. Chem. Soc.
1993, 115, 10605-10610.

4368 J. Am. Chem. Soc., Vol. 118, No. 18, 1996
Chiang et al.
Table 1. Comparison of Rates of Hydration of Ketenimines
Studied Here with Literature Valuesa
kH+/103 M-1 s-1
ko/10-3 s-1
substrate
present study lit.b present study lit.b
PhCHdCdNCH(CH3)2
PhCHdCdN-c-C6H11
PhCHdCdNC6H5
2.19
2.20
1.00
1.15
2.17
1.35
2.20
0.986
a Aqueous solution at 25 °C. b Reference 10.
sodium hydroxide. Some rate measurements for the N-phenyl
substrate were also made in more concentrated, 0.1-1.0 M,
perchloric acid solutions. The data are summarized in Tables
S2 and S3.6
Observed rate constants determined in dilute acid solutions
proved to be directly proportional to acid concentration, and
hydronium-ion catalytic coefficients were determined by linear
least-squares analysis. Observed rate constants measured in the
more concentrated acid solutions, on the other hand, increased
more rapidly than in direct proportion to acid concentration,
and the Cox-Yates method11 using the Xo acidity function12
was used to correlate the data. The two sets of measurements
Figure 1. Comparison of rates of reaction of authentic 1-(N-methyl-
N-(pentafluorophenyl)amino)-2-phenylacetylene, O, with rates of decay
of transient formed by flash photolysis of phenyl(N-methyl-N-(pen-
tafluorophenyl)amino)cyclopropenone, 4, in aqueous perchloric acid
solutions at 25 °C.
+
Table 2. Rate Constants for the Carbon Protonation of Ynamines
by Hydronium Ion in Aqueous Solution at 25 °Ca
gave results that agreed well with each other: kH ) 97.5 (
+
0.3 M-1 s-1 from dilute solutions and kH ) 99.7 ( 0.8 M-1
+
+ b
ynamine
PhCtCNH2
PhCtCNHCH(CH3)2
PhCtCNH-c-C6H11
PhCtCNHC6H5
PhCtCNHC6F5
PHCtCN(CH2)5
PHCtCN(CH2CH2)2O
PhCtCN(CH2CH2CN)2
PhCtCN(CH3)C6F5
kH+/M-1 s-1
kH+/kD
pKBH
s-1 from concentrated solutions. Best values of rate constants
for the uncatalyzed reactions were obtained by taking averages
of the 4-6 replicate determinations made in sodium hydroxide
solution.
6.70 × 105
6.42 × 106
7.01 × 106
2.43 × 104
8.83 × 102
7.73 × 106
8.12 × 105
1.76 × 105
4.16 × 103
2.95
1.98
9.24
10.67
10.64
4.60
-0.28
11.12
8.49
2.42
3.25
2.60
These results are summarized in Table 1. Comparison with
the previously determined values, also listed in Table 1, shows
that there is close correspondence between the two sets of
measurements.
Tertiary ynamines cannot tautomerize to ketenimines, but
evidence that they are formed by photodecarbonylation of our
tertiary aminocyclopropenones comes from yet another previous
investigation.3 It was found in this previous work that authentic
tertiary ynamines prepared by established methods protonate
rapidly on carbon, eq 7, giving keteniminium cations, 9, which
2.63
2.71
5.2
0.68
a Ionic strength ) 0.10 M. b pKa of the conjugate acid of the simple
amine formed by substituting H for the PhCtC group of the ynamine.
and our study in wholly aqueous solution. Although the
difference is small and in the direction expected for a change
in solvent polarity for such an ion-forming reaction, we
nevertheless supplied a more exact comparison by preparing
an authentic tertiary ynamine and measuring its rate of reaction
in wholly aqueous solution. Figure 1 shows that the results we
obtained for 1-(N-methyl-N-(pentafluorophenyl)amino)-2-phen-
ylacetylene agree well with rates of decay of the transient species
formed by flash photolysis of the corresponding aminocyclo-
propenone. This establishes the identity of this transient and
shows that all three classes of ynamines are formed by flash
photolysis of corresponding aminocyclopropenones.
Ynamine Rate Measurements. Rates of reaction of the
ynamines studied here were determined in perchloric acid,
sodium hydroxide, and buffer solutions. The measurements in
acid solutions were done in H2O and in D2O over a range of
acid concentrations, usually spanning a 5-10-fold variation, and
replicate determinations were made at each concentration. The
rate data so obtained are sumarized in Table S5.6
In each case, observed first-order rate constants were found
to be accurately proportional to acid concentration (see Figure
1), and bimolecular catalytic coefficients were determined by
linear least-squares analysis. The results so obtained are listed
in Table 2.
The measurements in sodium hydroxide solutions were also
done in H2O and D2O over a range of base concentrations, and
replicate determinations were made at each concentration. The
data are summarized in Table S4.6
upon hydration produce amide enols, 10; the enols then ketonize
to give amides, 8, as the ultimate reaction products. Reaction
rates were measured in this previous work by monitoring the
decay of ynamine absorbance at λ = 290 nm. We found that
flash photolysis of tertiary phenylaminocyclopropenones pro-
duces transient species that absorb light in this same region and
decay at rates consistent with those determined in the previous
study. Rate comparisons can be made for two sets of sub-
strates: the previous investigation found k ) 1.0 s-1 for the
reaction of 1-(piperidino)-2-phenylacetylene (2f) and k ) 0.015
s-1 for that of 1-(morpholino)-2-phenylacetylene (2g) in basic
solution where water is the proton donor, and we found k )
1.8 s-1 and k ) 0.026 s-1 for decay of the transients formed
from the corresponding phenylaminocyclopropenones (1f and
1g) under the same conditions.
Agreement between these two sets of rate constants is not
exact because the measurements were made in different
solvents: the previous work in water containing 8.6% dioxane
(11) (a) Cox, R. A.; Yates, K. J. Am. Chem. Soc. 1978, 100, 3861-
3867. (b) Kresge, A. J.; Chen, H. J.; Capen, G. L.; Powell, M. F. Can. J.
Chem. 1983, 61, 249-256.
(12) Cox, R. A.; Yates, K. Can. J. Chem. 1981, 59, 2116-2124.
(13) Andraos, J.; Chiang, Y.; Grant, A. S.; Guo, H.-X.; Kresge, A. J. J.
Am. Chem. Soc. 1994, 116, 7411-7412.
For primary and secondary ynamines, observed first-order
rate constants proved to be accurately proportional to base

Flash Photolytic Generation of Ynamines
J. Am. Chem. Soc., Vol. 118, No. 18, 1996 4369
Figure 3. Bronsted relation for the carbon protonation of 1-(N-
isopropylamino)-2-phenylacetylene (4) and 1-(piperidino)-2-phenyl-
acetylene (O) by a series of carboxylic acids in aqueous solution at 25
°C.
Figure 2. Buffer dilution plot for the reaction of 1-(N-isopropylamino)-
2-phenylacetylene in acetic acid buffer solutions at 25 °C. Inset:
dependence of catalytic coefficients upon fraction of buffer present in
acidic form.
Table 3. Rate Constants for the Reaction of Ynamines in Sodium
Hydroxide Solution at 25 °Ca
-
ynamine
PhCtCNH2
PhCtCNHCH(CH3)2
PhCtCNH-c-C6H11
PhCtCNHC6H5
PhCtCN(CH2)5
kHO-/M-1 s-1
kHO-/kDO
ko/s-1
3.99 × 106
6.79 × 106
3.52 × 107
2.40 × 109
1.91
4.09
b
6.81×105
1.84
PhCtCN(CH2CH2)2O
2.64×10-2
a Ionic strength ) 0.10 M. b Reference 13.
Figure 4. Rate profile for the carbon protonation of 1-(N-phenyl-
amino)-2-phenylacetylene in aqueous solution at 25 °C.
concentration, and bimolecular catalytic coefficients were
determined by linear least-squares analysis. Observed first-order
rate constants for tertiary ynamines, on the other hand, did not
vary with base concentration, and best values here were therefore
obtained by taking simple averages. The results are listed in
Table 3.
Table 4. Carboxylic Acid Catalytic Coefficients for the Carbon
Protonation of Ynamines in Aqueous Solution at 25 °Ca
acid catalyst
PhCtCNH2:
CH3CO2H
PhCtCNHCH(CH3)2:
CNCH2CO2H
ClCH2CO2H
HCO2H
kHA/103 M-1 s-1
8.47, 6.24
The rate measurements in buffer solutions were made with
CNCH2CO2H, ClCH2CO2H, HCO2H, CH3CO2H, C2H5CO2H,
769, 1030
1120, 1310
575, 539
-
H2PO4-, (CH2OH)3CNH3+, NH4+, and HCO3 as the buffer
acids. Series of solutions of constant buffer ratio and constant
ionic strength (0.10 M) but varying buffer concentration were
used; this served to hold hydronium ion concentrations constant
along most of the buffer solution series. In the case of CNCH2-
CO2H and ClCH2CO2H buffers, however, buffer failure14
occurred, and this was compensated for by adjusting observed
rate constants to a common hydronium ion concentration using
the hydronium ion catalytic coefficients determined in perchloric
acid solutions; hydronium ion concentrations needed for this
purpose were obtained by calculation using literature pKa’s of
the buffer acids and activity coefficients recommended by
Bates.15 The data are summarized in Table S6.6
As Figure 2 illustrates, pronounced buffer catalysis was found
in the carboxylic acid solutions, and analysis of results obtained
at different buffer ratios showed the catalysis to be completely
of the general acid type (see inset to Figure 2). Carboxylic
acid catalytic coefficients were evaluated by least-squares
analysis of the relationship between observed rate constants and
buffer acid concentration. These catalytic coefficients, sum-
marized in Table 4, were used to construct Bronsted plots for
CH3CO2H
242, 264, 239
290, 239
CH3CH2CO2H
PhCtCNHPh:
HCO2H
0.628, 0.614
PhCtCN(CH2)5:
CNCH2CO2H
ClCH2C2H
2080, 1900
1320, 1200
775, 674
HCO2H
CH3CO2H
CH3CH2CO2H
468, 471, 436
406, 324, 388
a Ionic strength ) 0.10 M.
1-(isopropylamino)-2-phenylacetylene and 1-piperidino-2-phen-
ylacetylene, shown in Figure 3, whose slopes are R ) 0.29 (
0.03 and R ) 0.28 ( 0.02, respectively.
The intercepts of buffer dilution plots such as those shown
in Figure 2 represent reaction through solvent-related species,
and these together with the rate constants measured in perchloric
acid and sodium hydroxide solutions were used to construct
rate profiles. A representative example for a secondary ynamine
is shown in Figure 4 and another for a tertiary ynamine is shown
in Figure 5.
(14) Keeffe, J. R.; Kresge, A. J. In Techniques of Chemistry, Volume
VI, InVestigations of Rates and Mechanisms of Reactions; Bernasconi, C.
F., Ed.; Wiley: New York, 1986; Chapter XI.
(15) Bates, R. G. Determination of pH Theory and Practise; Wiley: New
York, 1973; p 49.
Product Studies. The hydration of tertiary ynamines is
known to produce carboxylic acid amides,3 as is also the
hydration of ketenimines10 formed by tautomerism of second-

4370 J. Am. Chem. Soc., Vol. 118, No. 18, 1996
Chiang et al.
ring-opening reactions of eqs 11 and 12, however, our phenyl-
acetaldehyde-producing process must be a photochemical reac-
tion, for our aminocyclopropenones are stable in the solutions
we use for flash photolysis. We are currently investigating this
additional photochemical process.
Basicity of N-Methylpentafluoroaniline. In order to cor-
relate the reactivity of our ynamines with the basic strength of
the corresponding simple amines (Vide infra), we needed to
know the pKa of the conjugate acid of N-methylpentafluoro-
aniline. We determined this by measuring the change in
absorbance at λ ) 230 nm that this substance undergoes upon
protonation in moderately concentrated aqueous sulfuric acid.
The absorbance of solutions containing a constant stoichiometric
amount of the aniline was measured at 12 different sulfuric acid
concentrations over the range 0-40 wt %, and triplicate
determinations were made at each concentrationsthe results are
summarized in Table S7.6
Figure 5. Rate profile for the carbon protonation of 1-(piperidino)-
2-phenylacetylene in aqueous solution at 25 °C.
ary ynamines. Carboxylic acid amides are thus expected to be
the ultimate products of the reactions we have studied. We
verified this by performing HPLC analyses of flashed reaction
mixtures, using spiking with authentic amide samples to identify
the products. These tests were made for phenylpiperidinocy-
clopropenone, 1f, and phenylmorpholinocyclopropenone, 1g, in
acidic, neutral, and basic solutions.
In the case of the primary aminocyclopropenone 1a, however,
the product proved to be phenylacetonitrile rather than phenyl-
acetamide. The cation formed by carbon protonation of the
ketenimine here thus loses its nitrogen-bound hydrogen more
rapidly than it is captured by water, eq 8. This is consistent
These data were analyzed using the Cox-Yates X acidity
function11a in two different ways, one by fitting the expression
shown as eq 13 and another by fitting that shown as eq 14. In
log(I/[H+]) ) pKa + mX; I ) [BH+]/[B]
(13)
AB + (ABH+[H+])10(pK +mX)
a
A )
(14)
1 + [H+]10(pK +mX)
a
fitting eq 13, the indicator ratio I () (AB - A)/(A - ABH+))
was calculated using limiting absorbances for the completely
unprotonated (AB) and completely protonated (ABH+) forms of
the substrate measured in water containing no acid and in 40%
acid, respectively. In fitting eq 14, on the other hand, AB and
with the known mechanism for hydrolysis of simple nitriles,
which occurs by rapid and reversible protonation on nitrogen
followed by reaction of the protonated species with water, eq
9:16 here again the same cation loses a nitrogen-bound proton
more rapidly than it reacts with water.
+
ABH were left as parameters to be determined by the least-
squares fit. The two methods gave nicely consistent results:
pKa ) 0.676 ( 0.012, m ) 1.22 ( 0.13 (eq 13) and pKa )
0.677 ( 0.017, m ) 1.28 ( 0.29 (eq 14).
The pKa of pentafluoroaniline, which we also need for our
correlation, had been determined before using an acidity function
based on a series of halogenated anilines.19 However, the Cox-
Yates method using the universal acidity function X provides a
better way of data treatment, and we therefore reanalyzed the
original data for pentafluoroaniline20 and obtained pKa ) -0.277
( 0.014, m ) 0.92 ( 0.04; this result is consistent with, but
slightly different from, the originally reported value pKa )
-0.36 ( 0.02.19
We found also that flash photolysis of phenylpiperidinocy-
clopropenone and phenylmorpholinocyclopropenone produced
other products in addition to the corresponding phenylaceta-
mides. One of these additional products proved to be phenyl-
acetaldehyde, which could have been formed by decarboxylation
of 2-phenylformylacetic acid, itself made by hydrolysis of the
enamine obtained by ring opening of the cyclopropenone, as
shown in eq 10. A similar ring-opening reaction has been
Discussion
Reaction Mechanism. We have found that primary, second-
ary, and tertiary phenylynamines are produced by photodecar-
bonylation of the corresponding phenylaminocyclopropenones,
according to eq 1, and that they react with acids in aqueous
solution showing general acid catalysis and appreciable hydro-
nium ion isotope effects in the normal direction, kH+/kD+ > 1.
This is classic evidence for rate-determining proton transfer to
carbon,14 and it suggests that the process under observation is
reported for the methyl ether of phenylhydroxycyclopropenone,
eq 11,17 and we have found that treatment of phenylmorpholi-
(16) Hyland, C. J.; O’Connor, C. J. J. Chem. Soc., Perkin Tranns. 2
1973, 223-227. Liler, M. Reaction Mechanisms in Sulfuric Acid; Academic
Press: New York, 1971; pp 205-208.
(17) Farnum, D. G.; Chickos, J.; Thurston, P. E. J. Am Chem. Soc. 1966,
88, 3075-3081.
(18) Grant, A. S.; Kresge, A. J. Unpublished work.
(19) Yates, K.; Shapiro, S. A. Can J. Chem. 1972, 50, 581-583.
(20) The data was fortunately still available through Dr. R. A. Cox.
nocyclopropenone with methanol, eq 12, gives the methyl ester
of the enamine-acid postulated in eq 10.18 Unlike the thermal

Flash Photolytic Generation of Ynamines
J. Am. Chem. Soc., Vol. 118, No. 18, 1996 4371
PhCtCNHR + HO- a
PhCtCNR- + H2O f PhCHdCdNR + HO- (16)
In the previous investigation of tertiary ynamines,3 reactions
were observed for 1-(piperidino)-2-phenylacetylene in acetic
acid buffers that were believed to be the carbon-protonation
process of eq 15, despite the fact that they were several orders
of magnitude slower than expected on the basis of results
obtained in more basic solutions. It is now apparent that this
assignment is incorrect, because we have also observed these
slower reactions in addition to the faster changes that we know
to be carbon protonation. We are currently investigating these
slower processes.
pKa Limits. At sufficiently high acidities, the carbon-
protonation reaction of eq 15 should be accompanied by an
equilibrium protonation on nitrogen, eq 17, which will convert
Figure 6. Correlation of hydronium ion rate constants for the carbon
protonation of phenlynamines in aqueous solution at 25 °C with the
basicity of the corresponding simple amines.
PhCdCNHR2+ a PhCtCNR2 + H+ f
simple protonation of the ynamines on their â-carbon atoms to
form keteniminium cations, as shown in eq 15. The cations
+
PhCHdCdNR2 (17)
PhCtCNR2 + HA f PhCHdCdNR2+ + A- (15)
the ynamine into a much less reactive form. This will produce
a break in the rate profile, and that break will occur at an acidity
corresponding to the acidity constant of the N-protonated
ynamine. Figures 4 and 5 show that no such break has occurred
for the substrates represented there up to an acidity of [H+] )
0.1 M. We have also examined other ynamines in more
concentrated acids and have found no breaks up to [H+] ) 4
M.22 This sets an upper limit of -log (4) ) -0.6 on the pKa’s
of the conjugate acids of these ynamines. If, as seems likely,
the protonation of ynamines follows an acidity function rather
than [H+], this upper limit will be even lower than -0.6.
This limit is much below the estimate pKa = 5 made for the
conjugate acid of 1-piperidino-2-phenylacetylene on the basis
of what we now know to be a misidentification of the reactions
of this substance in acetic acid buffers.3 The present limit also
makes ynamines remarkably weakly basic substances. Theo-
retical calculations suggest that this weak basicity is caused by
strong destabilization of ynamine conjugate acids.23
Bronsted Relations and Marcus Theory. In the previous
investigation of tertiary ynamines,3 a Bronsted relation was
constructed for the carbon protonation of 1-piperidino-2-
phenylacetylene whose exponent was R ) 0.74. This is quite
different from the Bronsted exponent for the same reaction
obtained in the present study, R ) 0.28. The previous
correlation, however, was based on a group of acids much
weaker than the presently used series of carboxylic acids. The
previous reactions were consequently considerably slower, and
they could be expected to have had later, more product-like
transition states in which proton transfer would have been further
advanced.21 Since the Bronsted exponent R is believed to
measure the extent of proton transfer at the transition state,24 it
is not unreasonable that R for the previous correlation would
be greater than that for the present one. This idea is reinforced
by the fact that the previous and present sets of data form a
self-consistent group. As Figure 7 shows, a smooth continuous
curve can be drawn through all of the data, and least-squares
then react further according to eqs 6, 7, or 8 depending upon
whether they have been generated from primary, secondary, or
tertiary ynamines.
Further support for the process of eq 15 comes from the fact
that the reactions observed are all quite fast (see Table 2), as
expected on the basis of the strong ability of amino groups to
stabilize adjacent positive charge through their electron donating
resonance effects, and that the rates of reaction respond as
expected to variations in this electron-donating ability. These
variations could be measured by the nitrogen basicity of the
ynamines, but that unfortunately is not available (Vide infra).
This basicity, however, will be proportional to the basicity of
simple amine analogs of the ynamines in which the phenyl-
ethynyl groups have been replaced by hydrogen, and Figure 6
shows that there is indeed a good correlation between this
quantity and hydronium ion rate constants. The slope of this
correlation is low, 0.33 ( 0.02, consistent with the early,
reactant-like transition state expected for a fast process such as
this.21 Primary, secondary, and tertiary ynamines, moreover,
obey the same correlation, which implies that proton loss from
nitrogen in the case of primary and secondary ynamines is not
concerted with proton addition to carbon, for such proton loss
cannot occur with tertiary ynamines.
Tertiary ynamines react by the simple process of eq 15 in
basic was well as in acidic solution. This is illutrated by the
rate profile of Figure 5 for 1-(piperidino)-2-phenylacetylene,
where acid catalysis by the hydronium ion can be seen to give
way to an “uncatalyzed” reaction in which water serves as the
proton donor. The rate-determining proton-transfer nature of
this uncatalyzed process is substantiated by the normal solvent
isotope effects and general acid catalysis by weak acids observed
in this region in the previous investigation of tertiary ynamines.3
Primary and secondary ynamimes, on the other hand, give a
base-catalyzed reaction. This is illustrated by the rate profile
of Figure 4 for 1-(N-phenylamino)-2-phenylacetylene, which
shows hydroxide-ion catalysis in addition to the hydronium-
ion catalyzed process. This base-catalyzed reaction has been
identified as a process in which hydroxide ion first removes a
nitrogen-bound hydrogen from the ynamine in a rapidly
established reversible step, and the very reactive anion thus
formed is then protonated on carbon by water, eq 16.13
fitting of a quadratic expression gives the relationship log (kHA
/
p) ) 6.69 ( 0.08 + (1.29 ( 0.26) × 10-1 log(qKa/p) - (2.62
( 0.15) × 10-2(log(qKa/p))2.
Curved Bronsted relations such as this are a direct conse-
quence of the hypothesis that the structure of transition states
(22) Paine, S. W. Can J. Chem., in press.
(23) Smith, B. J.; Radom, L. J. Am. Chem. Soc. 1992, 114, 36-41.
(24) Kresge, A. J. In Proton Transfer Reactions; Caldin, E. F., Gold,
V., Eds.; Chapman and Hall: London, 1975; Chapter 7.
(21) Leffler, J. E. Science 1953, 117, 340-341. Hammond, G. S. J. Am.
Chem. Soc. 1955, 77, 334-338.

4372 J. Am. Chem. Soc., Vol. 118, No. 18, 1996
Chiang et al.
eq 20. This shows that R will change more sharply over a given
R ) d∆Gq/d∆G° ) (1 + ∆G°/4∆Gq )/2
(19)
o
dR/d∆G° ) 1/8∆Gq
(20)
o
interval of ∆G° for intrinsically fast reactions with low values
of ∆Gq than for intrinsically slow reactions with large values
o
of ∆Gq , and intrinsically fast reactions will consequently have
o
more sharply curved Bronsted plots.
It is possible to evaluate Marcus theory parameters from the
coefficients of a quadratic Bronsted correlation.26 For the
combined data of Figure 7, this leads to wr ) 8.11 ( 0.15 kcal
mol-1 and ∆Gq ) 3.26 ( 0.19 kcal mol-1. The rather low
o
value of ∆Gqo implies that this is an intrinsically fast reaction.
Even so, however, appliction of eq 20 leads to the prediction
that R should change by only 0.12 over the range of ∆Go
provided by our series of carboxylic acid protonating agents.
The good linearity of the Bronsted plots illustrated in Figure 3
shows that this change in R is too small to be seen, which
underscores the difficulty of detecting curvature in Bronsted
plots of even intrinsically rather fast reactions.
Figure 7. Extended Bronsted relation for the carbon protonation of
1-(piperidino)-2-phenylacetylene in aqueous solution at 25 °C: O,
present work; 4, ref 3.
varies with the energetics of the reaction, being reactant-like in
exoergic processes and product-like in endoergic ones.21 The
idea is given quantitative expression by Marcus rate theory,25
which formulates reaction barriers as shown in eq 18, where
Acknowledgment. We are grateful to the Natural Sciences
and Engineering Research Council of Canada for financial
support of this work.
q
∆Gq ) wr + (1 + ∆G°/4∆Gq )2∆Go
(18)
o
Supporting Information Available: Tables S1-S7 (30
pages). This material is contained in many libraries on
microfiche, immediately follows this article in the microfilm
version of this journal, can be ordered from the ACS, and can
be downloaded from the Internet; see any current masthead page
for ordering information and Internet access instructions.
∆Gq is the observed free energy of reaction, wr is the work
required to assemble the reactants into a reaction complex, ∆G°
is the free energy of reaction within that complex, and ∆Gqo is
the intrinsic barrier. The Bronsted exponent R can be identified
with the first derivative of ∆Gq with respect to ∆G°, eq 19,
and the rate of change of R is given by the second derivative,
JA9601797
(25) Marcus, R. A. J. Phys. Chem. 1968, 72, 891-899. Kresge, A. J.
Chem. Soc. ReV. 1973, 4, 475-503.
(26) Albery, W. J.; Campbell-Crawford, A. J.; Curran, J. S. J. Chem.
Soc., Perkin Trans. 2 1972, 2206-2214.