Binding catalysis and inhibition by metal ions and protons in the enolization of phenylacetylpyrazine

Binding Catalysis and Inhibition by Metal Ions and Protons in the

Enolization of Phenylacetylpyrazine

Anne-Gaelle Collot, Michael Courtney, Dara Coyne, Stephen E. Eustace, and

Rory A. More O’Ferrall*

School of Chemistry and Chemical Biology, UniVersity College Dublin, Belﬁeld, Dublin 4, Ireland

rmof@ucd.ie

ReceiVed December 2, 2008

A study of the enolization of phenylacetylpyrazine (PzCOCH₂Ph) catalyzed by acid, base and metal

ions in aqueous solution shows, unusually, that metal ions are more effective catalysts than protons,

e.g., for zinc k_Zn/k_H) 600. Such behavior contrasts with that of the structurally related

phenacylpyridine (PyCH₂COPh) for which k_Zn/k_H) 0.0065. To interpret this difference, equilibrium

constants for the tautomerization of phenylacetylpyrazine and for binding of protons and metal ions

to its keto tautomer and enolate anion have been measured or estimated and are compared with

existing measurements for phenacylpyridine. A tautomeric constant, K_E) 1.2 × 10^-3(pK_E) 2.9),

is derived by combining forward and reverse rate constants for enolization measured, respectively,

by iodination or bromination of the keto tautomer and relaxation of the less stable enol. For the keto

tautomer, NMR measurements yield a pK_a) -0.90 for N-protonation, and spectrophotometric

measurements give pK_a) 11.90 for ionization to an enolate anion. For the enol, pK_avalues of 0.44

and -4.80 for mono- and diprotonation are obtained from the pH proﬁle for ketonization and

absorbance measurements for the transient enol reactant. Binding constants for metal ions (Cu²⁺

,

Ni²⁺, Zn²⁺, Co²⁺, and Cd²⁺) are derived from the saturation of their catalysis of the ketonization

reaction. It is found that ketonization is efﬁciently catalyzed by metal ions but inhibited by acid.

These ﬁndings, and the striking difference from phenacylpyridine, are ascribed to differences in

thermodynamic driving force arising from stronger binding of the proton to the more basic pyridine

than pyrazine nitrogen atom in both the reactant keto tautomer and in the enaminone or zwitterion

product of the rate-determining (proton transfer) step of the enolization.

Introduction

carboxylic acid esters,^16,17amides,¹⁷and anions,¹⁴undertaken

by a number of authors over the past 20 years. The present

paper describes measurements of equilibrium constants for

Studies of the ionization and tautomerism of R- and

ꢀ-heterocyclic ketones^1-9form part of a wider investigation

of keto-enol¹⁰or keto-phenol¹¹tautomerism of aldehydes,

ketones,¹²ketenes,¹³and carboxylic acids^14,15as well as

(6) Eustace, S. E.; McCann, G. M.; More O’Ferrall, R. A.; Murphy, M. G.;

Murray, B. A.; Walsh, S. M. J. Phys. Org. Chem. 1998, 11, 519–528.

(7) Bunting, J. W.; Stefanidis, D. J. Am. Chem. Soc. 1988, 110, 4008–4017.

(8) Stefanidis, D.; Bunting, J. W. J. Am. Chem. Soc. 1990, 12, 3163–3168.

(9) More O’Ferrall, R. A.; Murray, B. A. J. Chem. Soc., Perkin Trans. 2

1994, 2461–2470. Katritzky, A. R.; Ghiviriga, J.; Oniciu, D. C.; More O’Ferrall,

R. A.; Walsh, S. M. J. Chem. Soc., Perkin Trans. 2 1997, 2605–2608. Greenhill,

J. V.; Loghmani-Khouzani, H.; Maitland, D. J. J. Chem. Soc., Perkin Trans. 2

1991, 2831–2840.

(10) The Chemistry of Enols; Rappoport, Z., Ed.; Wiley: New York, 1990.

(11) Capponi, N.; Gut, I. G.; Hellrung, B.; Persy, G.; Wirz, J. Can. J. Chem.

1999, 77, 605–613.

(12) Chiang, Y.; Kresge, A. J.; Nikolau, V. A.; Popik, V. V. J. Am. Chem.

Soc. 2003, 125, 6478–6484.

(13) Kresge, A. J. Acc. Chem. Res. 1990, 23, 43–48.

* To whom correspondence should be addressed. Tel: 353-1-716-2285/

2165.

(1) Katritzky, A. R.; Kucharska, H. Z.; Rowe, J. D. J. Chem. Soc. 1965,

3093–3096. Elguero, J.; Marzin, C.; Katritzky, A. R.; Linda, P. AdVances in

Heterocyclic Chemistry, Supplement 1; Academic Press: London, 1976.

(2) Carey, A. R. E.; Eustace, S. E.; More O’Ferrall, R. A.; Murray, B. A.

J. Chem. Soc., Perkin Trans. 2 1993, 2285–2296.

(3) Carey, A. R. E.; Fukata, G.; More O’Ferrall, R. A.; Murphy, M. G.

J. Chem. Soc., Perkin Trans. 2 1985, 1171–1722.

(4) Carey, A. R. E.; More O’Ferrall, R. A.; Murphy, M. G.; Murray, B. A.

J. Chem. Soc., Perkin Trans. 2 1994, 2471–2479.

(5) McCann, G. M.; More O’Ferrall, R. A.; Walsh, S. M. J. Chem. Soc.,

Perkin Trans. 2 1997, 2761–2772.

(14) Richard, J. P.; Williams, G.; O’Donoghue, A. C.; Amyes, T. L. J. Am.

Chem. Soc. 2004, 124, 2957–2968.

3356 J. Org. Chem. 2009, 74, 3356–3369

10.1021/jo802650s CCC: $40.75 2009 American Chemical Society

Published on Web 04/03/2009

Enolization of Phenylacetylpyrazine

the ionization and tautomerization of phenylacetylpyrazine

1. It compares catalysis of enolization by protons and metal

ions and explores the concept of “binding at the transition

state” as a basis for interpreting the relative effectiveness of

the two catalysts.

A feature distinguishing heterocyclic ketones from simpler

ketones is that migration of hydrogen from carbon to nitrogen

to form an imine provides an alternative to enolization as a

pathway for tautomerization. Indeed, for the ꢀ-heterocyclic

ketones 2- and 4-phenacyl-pyridine²or quinoline³(e.g., 2) in

aqueous solution, the enaminone 3 rather than the enol 4 is the

principal minor tautomer.

This ineffective acid catalysis is complemented by pro-

nounced catalysis by metal ions. The ratio of rate constants for

catalysis by zinc ions and protons, k_Zn/k_H) 591, appears to be

the highest recorded for Lewis acid catalysis in aqueous

solution.^6,18This marked advantage for metals contrasts with

the previously reported behavior⁶of 2-phenacylpyridine 1, for

which protons are favored over zinc ions by a factor of 1600,

a difference in relative rates between the two substrates of nearly

10⁶. A principal aim of the present paper is to establish the extent

to which this difference can be accounted for in terms of binding

constants of catalysts to their respective substrates and the

thermodynamic driving force conferred by complexation of the

catalysts to these substrates

The enol is restored as the preferred second tautomer⁴if the

basicity of the heterocyclic nitrogen atom is reduced, as in

phenacylprazine 5. The same is true of R-heterocyclic ketones,

such as the phenylacetylpyridines (e.g. 6), for which the

alternative tautomer is a zwitterion 7 lacking the stabilizing

resonance interaction between positive and negative charge

Results

Equilibrium Measurements. In aqueous solution phenyl-

acetylpyrazine exists predominantly as its keto tautomer (1), as

indicated by its UV-vis spectrum (with λ_max) 270, ε ) 1.9 ×

10⁴) and proton and ¹³C NMR spectra. In dilute aqueous sodium

hydroxide it forms a long wavelength species with λ_max) 320

nm (ε ) 1.15 × 10⁴) and a shoulder at 395 nm (ε ) 6.6 × 10³)

for which the dependence of the intensity on base concentration

is consistent with ionization to an enolate anion with pK_a)

11.90. (Errors for ionization or rate constants are given in the

Experimental Section or Supporting Information.)

centers which leads to charge neutralization in the enaminones.⁵

There is no change in wavelength or intensity of the principal

absorption maximum upon addition of acid to a neutral solution

of phenylacetylpyrazine, and in this respect the behavior is

similar to that of 2-phenylacetylpyridine, for which practically

no change in absorption spectrum accompanies protonation of

the ring nitrogen atom.⁵Qualitative evidence for the protonation

was obtained from the acid dependence of the ﬂuorescence

spectrum, but a quantitative value for the pK_awas obtained by

NMR from the dependence of chemical shifts of hydrogen atoms

of the pyrazine ring in DCl solutions upon acid concentration,

based on the measurements listed in Table S1 of the Supporting

Information. These measurements gave pK_a) -0.42. Similar

measurements for 2-acetylpyrazine (Table S2, Supporting

Information) gave pK_a) -0.55. The values are based on best

A characteristic of heterocyclic ketones is that even when

the zwitterion or enaminone is present at lower concentrations

than the enol, these species are implicated as intermediates in

the acid-catalyzed interconversion of keto and enol tauto-

mers. This is illustrated in eq 1, in which acid-catalyzed reaction

of the ketone proceeds not with initial protonation of the keto

oxygen atom but protonation of the nitrogen atom of the

heterocyclic ring. This protonation is followed by proton

removal from carbon to yield the enaminone (or zwitterion) as

intial product, which then tautomerizes rapidly to the more stable

enol. The reaction may be monitored by trapping the enol or

enaminone with iodine or bromine or may be observed directly

in the thermodynamically favored reverse direction following

quenching of the enolate anion in excess acid or acidic buffers.^4,5

ﬁts of calculated to observed chemical shifts plotted against pH

19,20

- X_o

for phenylacetyl-pyrazine in Figure 1 and for

acetylpyrazine in Figure S1 (Supporting Information). The

measured values refer to D₂O, and for phenylacetylpyrazine the

pK_awas corrected to -0.90 in H₂O by assuming that the isotope

(15) Andraos, J.; Chiang, Y.; Kresge, A. J.; Popik, V. V. J. Am. Chem. Soc.

1997, 119, 8417–8424.

(16) Richard, J. P.; Amyes, T. L. J. Am. Chem. Soc. 1996, 118, 3129–3141.

Chiang, Y.; Jones, J. Jr.; Kresge, A. J. J. Am. Chem. Soc. 1994, 116, 8358–

8359.

(17) Chiang, Y.; Guo, H.-X.; Kresge, A. J.; Richard, J. P.; Toth, K. J. Am.

Chem. Soc. 2003, 125, 187–194.

(18) Fontana, A.; De Maria, P.; Pierini, M.; Siani, G.; Cerritelli, S.; Macaluso,

G. J. Phys. Org. Chem. 2002, 15, 247–257.

(19) Cox, R. A. AdV. Phys. Org. Chem. 2000, 35, 1–66. Bagno, A.; Scorrano,

G.; More O’Ferrall, R. A. ReV. Chem. Intermed. 1987, 7, 313–352.

(20) Cox, R. A.; Lam, S.-O.; McClelland, R. A.; Tidwell, T. T. J. Chem.

Soc., Perkin Trans. 2 1979, 272–275.

(21) Kresge, A. J.; More O’Ferrall R. A.; Powell, M., Isotopes in Organic

Chemistry Vol 7; Secondary and SolVent Isotope Effects; Buncel, E., Lee, C. C.,

Eds.; Elsevier: Amsterdam, 1987; pp 177-273.

For phenylacetylpyrazine 1 the enol 8 is again the more stable

minor tautomer. Indeed, in this case the zwitterion is so unstable

that it is doubtful if it is formed as an intermediate in the acid-

catalyzed enolization. Moreover, the catalytic effectiveness of

protonation on nitrogen is so attenuated that in the ketonization

direction protonation inhibits rather than catalyzes the reaction.

J. Org. Chem. Vol. 74, No. 9, 2009 3357

Collot et al.

base component of the buffer; it also showed an uncatalyzed

reaction. In buffer solutions and more basic media rate constants

were measured using a Durrum stopped-ﬂow spectrometer, but

in solutions of hydrochloric or perchloric acid, below pH 3,

the reaction was slow enough for measurements using a

conventional spectrophotometer. In the pH range 1-9 the

reaction was monitored from disappearance of a spectrum with

λ_max) 335 nm (ε ) 2.14 × 10⁴) which was attributed to the

enol. Above 0.1 M H⁺, the initial spectrum from quenching

the enolate anion changed to one with λ_max) 390 nm (ε ) 1.9

× 10⁴). The dependence of the intensity of this chromophore

upon acid concentration, and the observation of an inﬂection

in the pH dependence of rate constants in the same acid

concentration range, suggested that it corresponded to the

protonated enol. At very high acid concentrations (>10 M

HClO₄) a further change to a spectrum with λ_max) 510 nm

was attributed to formation of the diprotonated enol. The high

value of λ_maxfor this ion is consistent with values observed by

Bunting for N-methylarylacetylpyridimium ions in basic solu-

tion.⁷

FIGURE 1. Plot of chemical shift against pH - X_ofor the protonation

of phenylacetylpyrazine in DCl-D₂O at 25 °C.

First-order rate constants (k_obs) for ketonization were measured

in acetate, lutidine and borate buffers. The values are shown in

Tables S3-S5 of the Supporting Information. Plots of k_obs

against concentration of buffer base (B) gave straight lines of

slope k_GB^Kand intercept k_oconsistent with eq 2. There was no

dependence on buffer acid concentration. This is consistent with

the operation of general base but not general acid catalysis.

Values of k_oand k_GBare listed in Table 1.

k_obs) k_o+ k_GB^K[B]

(2)

Rate constants for ketonization in perchloric acid are shown

in Table S6 (Supporting Information). Rate constants for

“ketonization” correspond to a sum of rate constants for

conversion of the enol (or enolate anion) to ketone and the

reverse enolization. However, only at pH values greater than

the pK_aof the enol (9.0) does the “reverse” reaction contribute

appreciably to the measured rate constant.

pH Proﬁle for Ketonization. The rate constants in Table

S6 (Supporting Information) may be combined with the buffer-

independent values (k_o) in Table 1 and rate constants for

conversion of phenylacetylpyrazine to its enolate anion in dilute

sodium hydroxide, listed in the Experimental Section, to provide

a plot of log k against pH shown as a pH proﬁle in Figure 2.

The proﬁle corresponds to rate constants for relaxation of the

enol tautomer EH (and its conjugate acids EH₂⁺and EH₃²⁺) to

the keto tautomer (KH) and, in basic solutions, to ionization of

the keto tautomer to its enolate anion (E^-). At high acid

concentrations the pH is replaced by pH - X_o, which corre-

sponds to the Hammett acidity function H_o.^19,22

FIGURE 2. Plot of logk_obsagainst pH - X_ofor the ketonization of

phenylacetylpyrazine enol in HClO₄, NaOH, and aqueous buffers at

25 °C.

TABLE 1. Rate Constants for Buffer-Independent and Buffer

(Base)-Catalyzed Ketonization of Phenylacetylpyrazine Enol in

Aqueous Buffers (Ionic Strength 0.1) at 25 °C

buffer base

acetate

buffer ratio ([B]/[BH⁺])

pH

k_o

k_GB

0.25

1.0

4.0

0.25

1.0

4.0

4.05

4.65

5.25

6.20

6.80

7.40

9.12

10.07

0.012

0.019

0.011

1.90

1.85

1.99

lutidine

borate

22.6

26.9

3.5^a

1.4^a

1.0

9.0

0.36

0.33

Figure 2 presents a staircase-like proﬁle, with three pH-

independent segments separating three, or possibly four, base-

catalyzed or acid-inhibited reactions. Despite its apparent

complexity an interpretation is straightforward. Between pH 9

and pH 13 the observed reactions are ionization of phenyl-

acetylpyrazine to its enolate anion and the reverse protonation

^aSubject to large uncertainties: 3.5 ( 2.9 or 1.4 ( 0.8.

effect on the equilibrium was the similar to that measured for

N-protonation of the enol tautomer (3.8) and the normal value

of K_a^H2O/K_a^D2O) 3.0 for dissociation of weak acids.²¹

Acid-Base Catalysis of Ketonization. Rate constants for

relaxation of the enol of phenylacetylpyrazine to its more stable

keto tautomer were measured in aqueous solution at 25 °C by

quenching a solution of the enolate anion in 0.1 M sodium

hydroxide in excess acid or acidic buffer. The reaction was

susceptible to catalysis by hydroxide ion, hydrogen ion and the

of the anion by water with rate constants k_OH(40.5 M^-1s^-1

)

and k_H2O(0.33 s^-1). Below pH 9 the enol exists predominantly

in its un-ionized form and the reaction again becomes base-

dependent reﬂecting ionization of the enol prior to its reaction

(22) More O’Ferrall, R. A.; O’Brien, D. M. Can. J. Chem. 2000, 78, 1594–

1612.

3358 J. Org. Chem. Vol. 74, No. 9, 2009

Enolization of Phenylacetylpyrazine

with water. These reactions are summarized in eqs 3 and 4, in

which K_a^EHis the ionization constant of the enol and KH, EH

and E^-denote the ketone, enol and enolate anion, respectively.

diprotonated enol, K_a^EH2+is the acid dissociation constant of

the monoprotonated enol and k_AH^H3O+refers to the rate constant

for protonation of the zwitterion (at carbon) by H₃O⁺to form

the N-protonated ketone 11 (or of the enol to form the

O-protonated ketone). The derived equation is simpliﬁed by the

fact that this mechanism is observed only under conditions that

the enol (EH) is fully protonated.

K_a^EH/[H ]

EH y

k_H2O

+

z E^-y

z KH

(3)

(4)

k_OH[HO^-]

k_H2O

(1 + [H⁺]/K_a^EH

EH3++

+

/[H ]

EH2+

+

/[H ]

H3O+

+

+ k_OH[HO^-]

+

K_a

k_AH

EH₃⁺⁺y

z EH₂⁺y

z ZH⁽

^{[H ]}8 KH₂

k_obs

)

Below pH 7, reaction of the enolate anion with H₃O⁺(with

rate constant k_H) becomes more important than with H₂O, and

this accounts for the pH-independent reaction (k ) 1.19 × 10^-2

s^-1) between pH 7 and pH 1. Between pH 1 and 0 a further

change to a pH-dependent reaction (k ) 1.51 × 10^-3s^-1) is

interpreted as a change in reactant from enol to protonated enol

(EH₂⁺). This interpretation is supported by a complementary

change in spectrum and λ_maxof the reactant (from 335 to 395

nm) and a dependence of the relative intensities of the two

chromophores upon acid concentration consistent with an

ionization equilibrium with pK_a) 0.45 ( 0.1 (Table S7 of the

Supporting Information). These reactions are summarized in eqs

5 and 6, in which the acid dissociation constant of the protonated

(8)

(9)

k_AH^H3O+K_a^EH2+

(1 + [H⁺]/K_a^EH3++

k_obs

)

In practice, this kinetic expression needs to be modiﬁed to

take account of the more complex dependence of rate and

equilibrium constants upon acid concentration characteristic of

strong acid solutions. As discussed in more detail below, this

is done by replacing pH by pH - X_o(and [H⁺] by 10^-(pH-Xo)).

The parameter X_ois a measure of the capacity of the medium

to protonate a series of reference aniline bases and corresponds

to log (K_a/K_a^H2O) where K_ais the measured acid dissociation

constant of a protonated aniline at a particular acid concentration

and K_a^H2Ois the dissociation constant in water.¹⁹In so far as X_o

approaches zero in dilute acid solutions, pH - X_oreduces to

pH at values greater than 1.

The line drawn through the experimental points in Figure 2

represents a best ﬁt to the measured rate constants based on

assignment of rate and equilibrium constants to the sum of the

right-hand sides of eqs 4, 6, and 9. As discussed in the

Experimental Section, the derived equilibrium constants are

consistently somewhat larger than those determined spectro-

photometrically. At least in the case of monoprotonation of the

enol the difference (pK_a) -0.58 compared with 0.45) appears

to beyond the range of experimental error. The explanation for

this discrepancy is considered below but remains uncertain.

Solvent Isotope Effects and Catalysis in HCl and DCl. In

addition to kinetic measurements of ketonization in HClO₄

measurements were undertaken in aqueous HCl. This was to

assess solvent isotope effects on rate and equilibrium constants

by comparison with DCl (in D₂O). To our surprise, instead of

ﬁrst-order rate constants remaining constant or decreasing as

the acid concentration increased, above a value of 5 M they

showed a sharp increase. The same behavior was observed in

DCl-D₂O. Figure 3 shows plots of log k_obsagainst pH - X_o

(pH proﬁles) in the strong acid region for HCl and DCl.

Measured rate constants for these acids are shown in Table S9

(Supporting Information).

enol (9) is denoted K_a^EH2+and the rate constant for reaction of

H3O+

H₃O⁺with the enolate anion is k_BH

(the notation for this

rate constant is explained below).

EH2+

+

H3O+

k_BH

+

K_a

/[H ]

K_a^EH/[H ]

EH₂⁺y

z EH y

z E^-

^{[H ]}8 KH (5)

k_BH^H3O+[H⁺]

(1 + [H⁺]/K_a^EH2+)[H⁺]/K_a^EH

k_obs

)

(6)

In more concentrated solutions of perchloric acid, above 4

M, a further acid-independent reaction is seen. This is interpreted

as a shift in mechanism from reaction of H₃O⁺with enolate

anion as in eq 5 to reaction with the unprotonated enol or,

alternatively,⁵as shown in eq 8, the zwitterionic species 10.

Which pathway is preferred is considered below.

Finally, at high acidities (>10 M HClO₄) there is evidence

of a further change in chromophore of the reactant (derived from

quenching the enolate anion from base) from λ_max) 395 nm to

λ_max) 510 nm. This is interpreted as reﬂecting an additional

protonation of the enol, at the second nitrogen of the pyrazine

ring, to yield the diprotonated species EH₃⁺². From the

dependence on acidity of the absorbance at 510nm, as measured

by Cox’s and Yates’s acidity parameter X_o,^19,22a pK_a) -4.80

( 0.15 may be inferred based on the data in Table S8

(Supporting Information) and a plot of absorbance at 510 nm

against pH - X_oin Figure S2 (Supporting Information).

In principle, the reactions and the kinetic expression for the

dependence of k_obsupon acid concentration over the high acidity

range are described by eqs 8 and 9. In eq 8, ZH⁽refers to the

zwitterionic species 10 and KH₂⁺to the monoprotonated ketone

11. In eq 9 K_a^EH3++is the acid dissociation constant of the

The discrepancy between rate constants in HCl and HClO₄

at the same X_ovalue suggests catalysis by chloride ions. When

the reaction was carried out in HBr a similar increase in rate

was observed. From a comparison of UV-vis spectra the

products in HClO₄and HCl appear to be the same, i.e., the

N-protonated ketone. It remains difﬁcult to envisage the

involvement of chloride ion as a base or through an unusual

salt effect.

At lower acid concentrations, rate constants in HCl and HClO₄

at the same pH - X_ovalue are similar. The measurements in

DCl lead to a pH-independent rate constant 1.53 × 10^-3s^-1

and corresponding solvent isotope effect k_H2O/k_D2O) 7.0, which

is consistent with rate-determining proton transfer from H₃O⁺

J. Org. Chem. Vol. 74, No. 9, 2009 3359

Collot et al.

SCHEME 1

rate constant 1.37 × 10^-5s^-1, close to the average of those

determined from iodination (1.27 × 10^-5).

The independence of measured rate constants of the nature

of the halogen (bromine or iodine) and their dependence upon

the concentrations of halide ions, substrate, and acetate buffer,

as well as upon pH, are consistent with their expected cor-

respondence to the enolization of phenylacetylpyrazine. This

interpretation is conﬁrmed insofar as the ratios of rate constants

for halogenation and ketonization measured under similar

conditions gave consistent values of K_T) 1.20 × 10^-3and 1.10

× 10^-3for the acetate-catalyzed and uncatalyzed reactions

respectively which may be identiﬁed as the keto-enol tautom-

erisation constant for the reaction. As described below a further

value of K_T) 1.50 × 10^-3was obtained from the ratio of rate

constants for iodination and ketonization catalyzed by zinc ions.

FIGURE 3. Plot of logk_obsagainst pH - X_ofor the ketonization of

phenylacetylpyrazine enol at 25 °C: upper curve H₂O-HCl; lower

curve, D₂O-DCl.

(or D₃O⁺) following ionization of the enol to its enolate anion.²¹

An isotope effect can also be evaluated for the dissociation

constant corresponding to the increase in initial absorbance at

λ_max) 395 nm in the concentration range 0.01-1 M DCl. In

HCl as in HClO₄this was ascribed to conversion of the

protonated to unprotonated phenylacetylpyrazine enol (cf. eq

8). Values of pK_awere determined from intital absorbances

recorded in Table S10 (Supporting Information) as 0.42 ( 0.05

and 1.00 ( 0.1 in H₂O and D₂O respectively, which give a

The average of these values is 1.27 × 10^-3

.

The observation of acid catalysis of bromination in dilute

perchloric acid solutions was surprising because the measured

rate constant (3.8 × 10^-4s^-1) was too large to be consistent

with an acid-catalyzed enolization. Because a faster acid-

catalyzed bromination was found for pyrazine itself (i.e. 1

lacking the PhCH₂CO group) it seems likely that this represents

bromination of the ring. However, no attempt was made to

isolate a product, and investigation of the reaction was not

pursued. Bromination of pyrazine in the absence of acid was

not observed.

solvent isotope effect on the protonation equilibrium of K_a^H2O

/

K_a^D2O) 3.8. As described above, this solvent isotope effect

was taken into account when correcting NMR measurements

of the pK_afor the protonated keto tautomer from D₂O to H₂O.

It is less easy to evaluate accurately an isotope effect for the

“chloride ion-catalyzed” reaction, but the calculated pH proﬁles

yield an approximate value of k_H2O/k_D2O) 4. Both pH and pD

proﬁles also imply a second acid-independent reaction at high

acidities prior to the onset of the chloride ion-promoted reaction

with approximate rate constants k_H2O) 3.3 × 10^-3s^-1and k_D2O

) 6.4 × 10^-4s^-1and hence a further large solvent isotope effect

k_H2O/k_D2O) 5, which is again consistent with a pre-equilibrium

deprotonation, this time of the protonated enol, followed by rate-

determining protonation (of the enol) by H₃O⁺. However, a

Microscopic Rate Constants. The rate constants measured

for enolization and ketonization reactions catalyzed by hydrogen

ions, hydroxide ions, and buffer bases (eq 2) can be dissected

-

into microscopic values k_BHand k_Bfor protonation of the eno-

late anion by an acid (BH) or proton abstraction from the keto

form of phenylacetylpyrazine by a base (B^-). These rate

constants are shown in Scheme 1. They are related to the rate

K

constant k_GBfor (general) buffer base-catalyzed ketonization

by equations (also included in Scheme 1) in which K_a^BHand

K_a^EHare ionization constants for the buffer acid and enol of

phenylacetylpyrazine respectively.³

discrepancy between the rate constant in H₂O/HCl and k_H2O

)

1.51 × 10^-3s^-1from aqueous HClO₄suggests that this value

may not be accurately estimated from the pH proﬁle for HCl

(Figure 3) and that this isotope effect is overestimated.

The equations need a little adaptation when BH is H₃O⁺or

H₂O. Thus, for H₃O⁺, k_BH) k_H2O^KK_a^EHand k_B- ) k_H2O^KK_T,

where k_H2O^Kis the rate constant for the uncatalyzed ketonization

of the enol. For H₂O, k_B- ) k_OH^K, the directly measured rate

constant for ionization of phenylacetylpyrazine to its enolate

Enolization Reaction. In addition to the ketonization reac-

tion, rate constants were measured for the reactions of iodine

and bromine with the keto tautomer of phenylacetylpyrazine.

In the presence of a large excess of the ketone and at not too

high acid concentrations (for the iodination) the halogenation

reactions were zero order in halogen and ﬁrst order in ketone.

The iodine and bromine were complexed with excess iodide or

bromide ions and the reactions were monitored spectrophoto-

anion, and k_BH) k_OH^KK_w/K_a^EH

.

Values of k_BHand k_B- for H₃O⁺, acetic acid, lutidinium ion,

and water are listed in Table 2. Although measurements were

also made in borate buffers, the catalysis was quite weak and

the values of k_GB^K(eq 2) from Table 2 were not deemed accurate

enough for calculation of microscopic rate constants.

-

metrically from depletion of I₃and Br₃^-, respectively, as the

halogenations progressed. The reaction was subject to general

base catalysis in acetic acid buffers, and measured ﬁrst-order

rate constants for these buffers are contained in Table S11

(Supporting Information). Table S12 (Supporting Information)

contains rate constants for bromination in dilute perchloric acid

solutions Bromination in dilute solutions of perchloric acid

showed mild acid catalysis and a pH-independent reaction with

In principle, microscopic rate constants may be derived also

from enolization measurements. Thus, the rate constant k_GB

E

for acetate-catalyzed enolization of phenylacetyl pyrazine cor-

responds to k_B- (Scheme 1), and the same is true of the

uncatalyzed reaction (with water as the base). In practice, the

magnitude of these values is taken into account in deriving

average values for the relevant k_B- and k_BHrate constants in

3360 J. Org. Chem. Vol. 74, No. 9, 2009

Enolization of Phenylacetylpyrazine

TABLE 2. Microscopic Rate Constants for the Ionization of

TABLE 3. pK_aValues for 2- and 3-Protonation of Substituted

Pyrazines^aand for Protonation of 2- and 3-Substituted Pyridines in

Aqueous Solution at 25 °C

Phenylacetylpyrazine to its Enolate Anion (k_B-) and the Reverse

Protonation of the Enolate Anion (k_BH

)

b

H₃O⁺

AcOH

lutidine

H₂O

substituent

pK_app

pK_a(pyridine) ∆pK_a^cpK_a(pyrazine)

pK_a

-1.76

4.78

6.69

15.76

0.33

40.5

H

2-CH₃

3-CH₃

0.8

1.45

0.40

-0.90

0.50

0.75

0.48

3.14

5.17

5.97

5.68

5.03

4.87

2.30

3.43

2.30

2.59

4.0

0

0.29

0.16

0.5

1.27

0.98

0.17

0.01

-2.00

-0.87

-1.37

-1.08

0.15

0.25

-0.86

0.74

-0.41

0.42

2.96

a

k_BH

1.19 × 10⁷

3.20 × 10⁴

4.90 × 10³

k_B-^b

1.51 × 10^-5

2.46 × 10^-3

0.030

2-PhCOCH₂

3-PhCOCH₂

2-CH₃CO

3-CH₃CO

2-PhCH₂CO

3-PhCH₂CO

2-PhCH)C(OH)

2-PhCH)C(OH

2-CH₃O

3-CH₃O

2-CH₃S

3-CH₃S

2-NH₂

^aRate constant for protonation of the enolate anion by indicated acid.

^bRate constant for proton abstraction from phenylacetylpyrazine by

conjugate base of indicated acid.

0.16

-1.13

-0.29

-0.10

-1.60

-0.83

0.68

Table 2. The values in Table 2 are also consistent with averaged

measurements of the tautomeric constant K_T.

4.1

One further microscopic rate constant may be identiﬁed in

connection with the H⁺-catalyzed ketonization of the enol

implied in eq 8. The measured rate constant is 1.51 × 10^-3s^-1

and refers to the pH-independent reaction in the perchloric acid

concentration range 4-9 M. This is interpreted as a reaction of

the monoprotonated enol in which the rate-determining step

involves protonation by H₃O⁺of either the neutral enol or, as

is observed for 2-phenylacetylpyridine, its zwitterionic tautomer

10. The rate constant for reaction of an acid with the neutral

enol (or zwitterion) is denoted k_AHby analogy with k_BHfor

reaction with its conjugate base. If the enol is the reactive

intermediate, k_AH) 1.51 × 10^-3/K_a^EH2+) 4.16 × 10^-3based

3.28

4.88

3.59

4.42

6.71

6.03

3-NH₂

0.68

2.38

^aDerived as described in the text from values of pK_app

.

^bValues of

pK_appfor protonation of substituted pyrazines; data from refs 4 and 23

or this study. ^cThe difference in pK_avalues of 2- and 3-substituted

pyridines are assumed to be the same for 2- and 3-protonation of the

correspondingly substituted pyrazine; data from refs 2, 5, and 23.

If we assume that the difference in pK_avalues at the two

positions, ∆pK_a, is 0.9 times the difference in pK_avalues of the

corresponding 2- and 3-substituted pyridines (13 and 14), then

we can infer individual pK_avalues for the pyrazine nitrogen

atoms from the measured value of pK_a^appby combining eq 12

with eq 13 (for 2- and 3-substituted pyridines) to obtain eqs 14

and 15. The multiplication by 0.9 is based on the slope of the

plot of pK_avalues of pyrazines against pK_avalues of pyridines

obtained below.

on the (averaged) value of 0.44 for pK_a^EH2+

.

-

It is now possible to evaluate a rate constant k_Afor the

reverse reaction of protonated ketone with water based on eqs

10 and 11. Taking the ionization constant of the N-protonated

ketone pK_a^KH2+as -0.90 and K_Tas 1.27 × 10^-3gives k_A- )

4.20 × 10^-5. The magnitude of this rate constant does not

depend on whether it is the enol or zwitterion which undergoes

protonation in the ketonization direction. However, if it is the

zwitterion which is subject to protonation k_AHwill be modiﬁed

by the equilibrium constant between the zwitterion and the enol.

An attempt to estimate this equilibrium constant is described

below.

KH2+

K_a

k_AH

EH + H₃O⁺y z KH₂⁺+ H₂O y

z KH + H₃O⁺(10)

pK_a^-2- pK_a^-3) ∆pK_a

pK_a^-4) pK_a^app-log(1 + 10^0.9∆pK

pK_a^-1) pK_a^-4+ 0.9∆pK_a

(13)

(14)

(15)

-

k_A

a

k_A) k_AHK_a^KH2+K_T

(11)

)

-

Correlation of pK_aValues for Substituted Pyridines

and Pyrazines. An equilibrium constant for interconversion of

the enol 8 and zwitterion 10 tautomers of phenylacetylpyrazine

could be derived if a pK_afor N-protonation of the enolate anion

to form the zwitterion was known. This pK_acannot be measured

directly but it may be estimated from corresponding values for

2- and 3-phenylacetylpyridines and a correlation between pK_a

values of substituted pyridines and pyrazines.

This correlation is complicated by the presence of alternative

sites for protonation of the pyrazine. For a substituted pyrazine,

the substituent is situated in a 2-position with respect to the

1-nitrogen atom and a 3-position with respect to the 4-nitrogen

atom (12). The experimental pK_ais then an apparent value

(pK_a^app) related to the pK_avalues for the individual positions

by eq 12 in which the superscripts -1 and -4 for K_aindicate

protonation at the 1- and 4-nitrogens, respectively.

Values of pK_a^-1and pK_a^-4derived in this way are listed in

Table 3 together with pK_avalues for the corresponding 2- and

3-substituted pyridines.²³For phenylacetylpyrazine itself, for

example, pK_app) -0.9 was combined with pK_avalues 2.30

and 2.59, respectively, for the 2- and 3-phenylacetylpyridines

(∆pK_a) -0.29) to derive pK_a^-1) -1.35 and pK_a^-4) - 1.09

for the 1- and 4-nitrogen atoms. The correlation between pK_a

values of pyrazines and pyridines is shown in Figure 4, with

values of pK_a^-1plotted against the pK_avalues of 2-substituted

pyridines and values of pK_a^-4plotted against pK_avalues of

3-substituted pyridines. The correlation shows considerable

scatter, but the best straight line through the points yields the

following relationship.

(23) Jencks, W. P.; Regenstein, J. Handbook for Biochemistry, 2nd ed.; Sober,

H. A., Harte, H. J., Eds.; Chemical Rubber Co.: Cleveland, OH, 1970; pp J187-

226.

1/K_app) 1/K_a^-1+ 1/K_a^-4

(12)

J. Org. Chem. Vol. 74, No. 9, 2009 3361

Collot et al.

pK_a^pyradine) (0.90 ( 0.08)pK_a^pyradine-3.70 ( 0.37

(16)

Zwitterion Tautomers of Phenylacetylpyrazine. As noted

above, the required pK_avalues for loss of a proton from the 1-

and 4-zwitterionic tautomers of phenylacetylpyrazine to yield

enolate anions may be inferred from the correlation of Figure

4 and knowledge of the pK_avalues of the zwitterionic isomers

of phenylacetylpyridine. The pK_avalues for the latter ions may

be derived from measurements of pK_avalues for the deproto-

nation of the corresponding N-methylphenylacetylpyridinium

ions corrected for replacement of the N-hydrogen atom by a

methyl group.⁷The dissociation eqilibrium is shown for the

2-substituted cation below.

FIGURE 5. Plots of k_obsfor the ketonization of phenylacetylpyrazine

enol in 0.1 M HClO₄against concentrations of metal ions in aqueous

solution at 25 °C.

SCHEME 2

Substitution of pK_avalues of 5.9 and 7.4 for the 2- and

3-phenylacetylpyridine zwitterions⁷in eq 16 yields 2.9 and 1.6

as the corresponding values for the 1- and 4-N-protonated

phenylacetylpyrazine zwitterions. The tautomeric constants

(pK_T) for interconversion of enol and 1- and 4-zwitterions can

then be obtained as 6.1 and 7.4, respectively, based on the

thermodynamic cycles of Scheme 2, which include the known

ionization constant of the enol (pK_a) 9.0). In Scheme 2, the

double arrows for equilibria are replaced by single arrows to

indicate the direction of reaction to which the equlibrium

constant refers.

SCHEME 3

These equilbrium constants for the zwitterions allow con-

struction of “complete” networks of equilibrium constants for

the ionization and tautomerisation of phenylacetylpyrazine as

described in the Discussion.

Catalysis of Tautomerization by Metal Ions. The ease with

which the ketonization reaction could be monitored and the

absence of competing acid catalysis (except at high acid

concentrations) provide favorable conditions for the study of

catalysis by metal ions. First-order rate constants for reaction

of Cu²⁺, Ni²⁺, Co²⁺, Zn²⁺, and Cd²⁺ions were measured in

aqueous solutions of 0.1 M HCl. The rate constants are listed

in Table S13 (Supporting Information) and are plotted against

the concentrations of metal ions for Cu, Ni, and Zn in Figure

5. For all the ions except Cu²⁺the plots are linear or nearly

linear, and second-order rate constants may be derived from

their (limiting) slopes. These constants (k_M^K) are listed in Table

4 where they are compared with the second-order rate constant

for catalysis by hydrogen ions and ﬁrst-order rate constant for

the uncatalyzed reaction. Also shown in Table 4 are rate

constants for the reverse metal ion-catalyzed enolization reac-

tions (k_M^E) obtained by combining the rate constants for

ketonization with the equilibrium constant for tautomerisation,

K_T) 1.27 × 10^-3

.

In Figure 5 it can be seen that catalysis by Cu²⁺ions is subject

to kinetic saturation. This may be interpreted in terms of a

reaction scheme (eq 18) in which a metal bound enolate anion

(E^-M²⁺) reacts with H₃O⁺to yield the coordinated ketone.⁶

Formation of the metal-enolate complex involves exchange of

an enolic hydrogen atom for a metal ion. Saturation occurs as

the direction of this equilibrium shifts from left to right. The

observed kinetic behavior can be formulated in terms of the

equilibrium constant for the exchange reaction (K_x), the rate

constant for protonation (k_p) and the concentration of metal ion

as in eq 19. At high concentrations of metal ion the limiting

ﬁrst order rate constant k_max) k_p[H⁺]. At low concentrations

H

of metal ion the limiting second-order rate constant k_M) k_pK_x.

FIGURE 4. Plots of “dissected” pK_avalues for the protonation at 1-

and 4-positions of 2-substituted pyrazines against pK_avalues of

corresponding 2- and 3-substituted pyridines.

The metal ion concentration at half-maximal rate corresponds

to an apparent binding constant K_app) K_x/[H⁺].

3362 J. Org. Chem. Vol. 74, No. 9, 2009

Enolization of Phenylacetylpyrazine

H

K_X

k_p

are represented by single arrows to show the directions of

reactions to which the pKs refer.

EH + M²⁺y z E^-M²⁺+ H⁺8 KHM²⁺h KH + M²⁺

\

From Scheme 3 it is apparent that pK_B) pK_x- pK_a^EH(or

K_B) K_x/K_a^EH) where pK_a^EH) 9.0 refers to the ionization

constant of the enol tautomer of phenylacetylpyrazine. Values

of binding constants are, however, more conveniently expressed

as log K_Brather than pK_Bbecause this yields positive rather

than negative values and, in the case of the proton, log K_B)

pK_a. For the zinc, copper and hydrogen ions values of log K_B

are 5.66, 9.43, and 9.00, respectively. These values are also

listed in Table 4.

(18)

k^H_p[H⁺]

(1 + [H⁺]/K_X[M²⁺])

k_obs

)

(19)

A best ﬁt of eq 19 to the data for catalysis by Cu²⁺gives k_p

) 90.5 M^-1s^-1and K_x) 2.72 ( 0.23. The limiting value of

k_obsat high concentrations of metal ion, k_max, corresponds to

H

k_p[H⁺] ) 9.05 s^-1. The limiting slope of the plot of k_obsagainst

H

[M²⁺] gives k_pK_x) 246 M^-1s^-1. The latter value represents

the second order rate constant (k_M^K) for catalysis of ketonization

by Cu²⁺ions shown in Table 4.

In addition to ketonization, enolization catalyzed by zinc ions

was studied in aqueous solution in the absence of buffers.²⁴

Measured ﬁrst-order rate constants are listed in Table S15

(Supporting Information) and when plotted against metal ion

Catalysis by zinc ions was also studied in 0.02 M 1:1 acetic

acid buffers (ionic strength 1.17). At this pH, the zinc catalysis

is also subject to strong saturation. The rate constants are listed

in Table S14 (Supporting Information). They yield a value of

k_max) 1.17 s^-1and apparent binding constant for zinc ions K_app

) 18.9 ( 2.5.

concentration give a straight line with slope corresponding to a

second order rate constant k_M) 3.68 × 10^-3M^-1s^-1

.

E

Combining this rate constant with the rate constant 2.46 M^-1

s^-1for zinc-catalyzed ketonization in water yields K_T) 1.50

× 10^-3in good agreement, as we have seen, with values for

other catalysts.

For acetic acid buffers, the reaction scheme in eq 18 is

modiﬁed to that in eq 20 in which the metal-enolate complex

is now protonated by acetic acid instead of H₃O⁺. Equation 19

For enolization it was difﬁcult to detect and measure the

saturation of catalysis with increasing zinc ion concentration.

However, the extent of curvature was increased by the addition

of triﬂuoroethanol to the aqueous medium, and in a dilute

cacodylate buffer in 60% v/v TFE-H₂O a value of K_Bwas

measured as 6.9 M^-1. First order rate constants for iodination

on which the measurement is based are listed in Table S16

(Supporting Information) and plotted against [Zn²⁺] in Figure

S3 (Supporting Information). The third-order rate constant for

zinc and cacodylate-catalyzed enolization in 60%TFE is 5.4 M^-2

s^-1. In order to extrapolate the value of K_Bto water, rate

constants for the iodination of acetylpyridine catalyzed by zinc

ions in 60% v/v TFE-H₂O in the absence of buffer were

similarly measured, and a value of K_B) 65.0 was derived. Since

a value of K_B) 35.7 M^-1had previously been measured in

water,²⁴it was assumed that enolization of phenylacetylpyrazine

would be characterized by the same ratio of rate constants in

water and 60% TFE as acetylpyridine. On this basis, K_Bfor

binding of zinc ions to phenylacetylpyrazine in water is 6.9 ×

35.7/65 ) 3.8 M^-1and log K_B) 0.6. Rate constants for the

zinc-catalyzed iodination of acetylpyridine in aqueous TFE are

listed in Table S17 (Supporting Information).

is unchanged except that k_p[H⁺] in the numerator is replaced

H

by k_p^AcOH[AcOH], as shown in eq 21.

K_X/K^A_a^cOH

k^A_p^cOH

EH + M²⁺+ AcO^-y

z E^-M²⁺+ AcOH

8

(20)

k^A_p^cOH[AcOH]

(1 + [H⁺]/K_X[M²⁺])

k_obs

)

(21)

AcOH

Identifying k_maxwith k_p^AcOH[AcOH] gives a value of k_p

) 58.5 M^-1s^-1. As before, the apparent binding constant for

zinc ions K_appcorresponds to K_x/[H⁺], and this implies a value

of K_x) 4.64 × 10^-4(taking pK_a^AcOH) 4.61 at the prevailing

ionic strength of 1.17). The limiting second order rate constant

at low metal ion concentrations is k_p^AcOH(K_x/[H⁺])[AcOH] )

22.1 M^-1s^-1([AcOH] ) 0.02 M, pH ) 4.55).

The exchange equilibrium constants K_xfor copper and zinc

ions are conveniently converted to equilibrium constants K_Bfor

binding the metals to the enolate anion by making use of the

thermodynamic cycle shown in Scheme 3. Again the equilibria

TABLE 4. Rate Constants and Binding Constants for Catalysis of Ketonization by Protons and Metal Ions in Aqueous Solution at 25 °C

b

K

c

E

d

e,f

f,g

catalyst

acid^a

k_M

logK_B

k_p

k_-p

none

Cd²⁺

Zn²⁺

Co²⁺

Ni²⁺

Cu²⁺

H⁺

AcOH

H⁺

1.94

2.46 × 10^-3

1.51 × 10^-5

3.94 × 10^-4

2.81 × 10^-2

3.12 × 10^-3

1.10 × 10^-2

1.40 × 10^-2

0.312

0

3.2 × 10⁴

2.46 × 10^-3

1.19 × 10^-2

1.19 × 10⁷

1.51 × 10^-5

H⁺

0.31

22.1

2.46

8.67

AcOH

5.66

58.5

7.4 × 10^-3

5.6 × 10^-4

H⁺

5.30 × 10³

H⁺

11.0

H⁺

246.0

9.43

9.00

90.5

H⁺

4.16 × 10^-3

5.28 × 10^-6

4.16 × 10^-3

4.20 × 10^-5

^aThe acid protonating complexed (or for the uncatalyzed reactions uncomplexed) enolate anion in the ketonization reaction, with rate constant k_p. (In

the reverse enolization, the conjugate base (H₂O or AcO^-) abstracts a proton from the complexed (or uncomplexed) ketone, with rate constant k_-p.)

^bRate constant for catalyzed or uncatalyzed ketonization. ^cRate constant for catalyzed or uncatalyzed enolization ^dLog K_bis the log of the binding

constant of the metal ion or proton to the enolate anion. In the case of the proton it corresponds to the pK_aof the enol. ^eSee footnote a; k_pcorresponds

to k_BHfor uncatalyzed reactions and k_AHfor the H⁺-catalyzed reaction. ^fThere are no values for metal ions for which binding constants were not

determined; for Zn²⁺, K_B) 3.8 M^-1

.

^gSee footnote a; k_-pcorresponds to k_A- for the H⁺-catalyzed reaction and k_B- for the uncatalyzed reactions.

J. Org. Chem. Vol. 74, No. 9, 2009 3363

Collot et al.

SCHEME 4

SCHEME 5

-6.25²⁵reported for the protonation of pyrazine itself. The latter

measurement represents H_oat half-protonation and is not strictly

a pK_a. Nevertheless, the large difference from pK_a) -4.80 is

consistent with a signiﬁciant stabilization of the pyrazine

dication by the enolic substituent, which is also suggested by

strong conjugation between the two groups implied by the long

wavelength of the principal UV-vis absorption (λ_max) 510

nm). It is a little surprising therefore that there appears not to

be a complementary stabilization of the monoprotonated enol

(pK_a) 0.44) which is of comparable acidity to the protonated

pyrazine (pK_a) 0.5).

In aqueous buffers, the dependence of the rate constants upon

buffer concentrations and pH is consistent with a base-catalyzed

mechanism for tautomerisation, with the enolate anion as

reactive intermediate. This corresponds to the lower pathway

of Scheme 4 (if the direction of an arrow is reversed!). Below

pH 7, the reaction is independent of pH and proceeds with initial

ionization of the enol reactant followed by protonation of the

enolate anion by H₃O⁺as in eq 22.

Discussion

Equilibrium Constants for Ionization and Tautomeriza-

tion. The pK_avalues for ionization of phenylacetylpyrazine in

base and acid are readily measured as 11.90 and -0.90,

respectively. The value in acid includes a correction for a solvent

isotope effect upon the value measured as -0.40 in D₂O. That

the ionization occurs in dilute solutions of base means that the

less stable enol tautomer can be generated by quenching the

enolate anion in excess acid or acidic buffer.^2-5Rates of

relaxation of the enol to form the keto tautomer may then be

monitored spectrophotometrically. The reverse enolization reac-

tion can be monitored by iodination or bromination of the

ketone. Combining rate constants for the two reactions yields

an equilibrium constant for tautomerisation K_T) [enol]/[ketone]

) 1.27 × 10^-3and pK_T) 2.90. This constant can be combined

with the ionization constants for the ketone to yield additional

ionization constants for the enol tautomer based on the

thermodynamic cycles of Scheme 4 (in which, again, the arrows

indicate directions of reaction to which the equilibrium constants

refer).

The site of protonation of the pyrazine ring in Scheme 4 is

shown as the nitrogen atom at the 4-position. In practice, pK_a

values for 2- and 3-phenylacetylpyridine of 2.30 and 2.59,

respectively, and for 2- and 3-acetylpyridine of 2.30 and 3.43,

suggest that although protonation should occur predominantly

at this position there may be a signiﬁcant contribution from

protonation at the 1-position. As described above, the pK_a)

-0.90 may be treated as an apparent value implying pK_avalues

of -1.35 and -1.1 for the 1- and 4-nitrogen atoms respectively,

based on the difference in pK_avalues of the 2- and 3-pheny-

lacetylpyridines. The dissected pK_avalues indicate that there is

twice as much protonation at the 4- than 1-position.

EH h E^-+ H₃O⁺f KH

(22)

This reaction is retarded by acid. This is apparent from the

form of the pH proﬁle close to pH ) 0 and at higher acidities.

The downward inﬂection above pH ) 0 corresponds, as we

have seen, to protonation of the enol with pK_a) 0.44. The

decrease in rate at pHs below this pK_arepresents inhibition of

the reaction by acid. At higher acid concentrations the reaction

again becomes pH independent, and this must represent the H⁺-

catalyzed reaction of the enol, albeit with the protonated enol

as reactant.

It is unusual for keto-enol tautomerization to be inhibited

by acid. The extent of the inhibition is expressed by the ratio

of ﬁrst-order rate constants for neutral and protonated enols as

reactants in Scheme 5.

In the reverse direction (enolization), the protonated ketone

is indeed more reactive than the unprotonated ketone, but the

difference is small: 3.8 × 10^-5s^-1compared with 1.52 × 10^-5

s^-1. This reversal of reactivities occurs because K_Tfor tautom-

erisation of keto to enol tautomers is larger for the protonated

than unprotonated enol, i.e., pK_T) 1.57 compared with 2.90

(see below).

O- versus N-Protonation. As with phenacylpyrazine (5), the

following dilemma arises: does acid-catalyzed enolization

proceed with protonation of the keto tautomer at a nitrogen atom

or oxygen atom? In other words, is there a zwitterionic

intermediate on the pathway from enol to protonated ketone,

as shown in eq 8? Factors favoring conversion of the enol to a

zwitterion prior to protonation have been discussed elsewhere.⁴

As outlined in the Supporting Information, it is difﬁcult to say

whether this is true of phenylacetylpyrazine. If the reaction does

not proceed with N-protonation, phenylacetylpyrazine is the ﬁrst

heterocyclic ketone studied for which this has been found to

pH Proﬁle. Rate constants for relaxation of enol or ketone

reactants are shown as a plot of log k against pH in Figure 2.

In aqueous sodium hydroxide the reaction corresponds to

ionization of the keto tautomer to its enolate anion. At lower

pH, the enolate anion is quenched in media sufﬁciently acidic

to effect its protonation, and the observed reaction is conversion

of enol to keto tautomer. In concentrated HClO₄inﬂections in

the proﬁle may be interpreted in terms of mono and diproto-

nation of the enol. From the acid dependence of associated

absorbance changes pK_avalues of 0.45 and -4.80 can be

assigned. These values may be compared with 0.65²³and

(24) Cox, B. G. J. Am. Chem. Soc. 1974, 96, 6823–6828.

(25) Brignell, P. J.; Johnson, C. D.; Katritzky, A. R. J. Chem. Soc. B 1967,

1233–1237.

(26) Alunni, S.; Del Giacco, T.; De Maria, P.; Fiﬁ, G.; Fontana, A.; Ottavi,

L.; Tesei, I. J. Org. Chem. 2000, 69, 3276–3281.

3364 J. Org. Chem. Vol. 74, No. 9, 2009

Enolization of Phenylacetylpyrazine

SCHEME 6

SCHEME 7

SCHEME 8

TABLE 5. Ratios of Rate Constants for Catalyzed and

Uncatalyzed Enolization of Phenylacetylpyrazine for Proton and

Metal Ion Catalysts with Water as Base

H⁺

Cd²⁺

Zn²⁺

Co²⁺

Ni²⁺

Cu²⁺

k_M/k_o(l mol^-1

)

0.35

26.0

207

729

924

20,700

activated by a ꢀ-pyridyl substitutent,²⁶bromination of keto-

amides,²⁷hydrogen isotope exchange of nucleotide bases,^28,29

and a zinc-catalyzed hydride rearrangement.³⁰In this paper, we

focus on the difference between the ketonization of phenyl-

acetylpyrazine 1, which appears to be characterized by the

largest metal ion-to-proton rate ratio so far recorded, and the

apparently structurally similar phenacylpyridine 2 for which an

even larger difference in rates is observed,⁶but now with the

proton more rather than less effective than the metal ion, i.e.,

k_H/k_Zn) 1600.

be true. If it does, then the O-protonation route must be deemed

unusually slow.

Networks of Equilibrium Constants. Dissection of apparent

equilibrium constants for the acid ionization of phenylacetylpyra-

zine and inclusion of estimates of pK_avalues for zwitterionic

tautomers allows expansion of Scheme 4 into more compre-

hensive networks of equilibrium constants referring, respec-

tively, to protonation at the 1- and 4-nitrogen atoms of the ring,

as shown in Scheme 6. Note that the distinction between

positions of protonation applies only to species protonated at

nitrogen, namely KH₂⁺, EH₂⁺and ZH⁽. As already mentioned,

in Scheme 4 the pK_a) -0.90 for N-protonation of phenyl-

acetylpyrazine is an apparent value which may be separated

into pK_avalues of -1.4 and -1.1 for 1- and 4-protonation by

assuming that the ratio of the two ionization constants is the

same as for protonation of 1- and 4-phenylacetyl pyridines. Also

included in Scheme 6 are pK_avalues for mono and diprotonation

of the phenylacetylpyrazine enols.

The format of Scheme 6 follows that used for other

heterocyclic ketones.^2,4,5Acid dissociation equilibria are indi-

cated by vertical or diagonal arrows and tautomeric equilibria

by horizontal arrows. The numbers in brackets are for equilibria

involving the zwitterion tautomers for which the equilibrium

constants are based on the correlation between pK_avalues of

correspondingly substituted pyrazines and pyridines (Figure 4).

Metal Ion Catalysis. The weak catalysis of tautomerization

of phenylacetylpyrazine by acid means that catalysis by metal

ions is readily observable. Partly as a consequence, catalysis

by metal ions is considerably more effective than by protons.

This is apparent from Table 5, which compares ratios of rate

constants (k_H/k_oor k_M/k_o) for catalyzed and uncatalyzed keton-

ization reactions. The ratios represent the acceleration produced

by an (ideal) one molar concentration of catalyst. Unusually,

rate constants for the acid-catalyzed reactions are smaller than

for the uncatalyzed reactions (for water or acetate ions acting

as bases). Correspondingly, ratios of rate constants for metal

ion-catalyzed to proton-catalyzed reactions are high: e.g., k_Zn/

k_H) 591 and k_Cu/k_H) 58600.

To understand the marked contrast between these two

substrates it is helpful to represent the catalysis within the

conventional framework of Scheme 7, which shows uncatalyzed

and H⁺-catalyzed reaction pathways for enolization.³¹Enoliza-

tion indeed is the reverse of the ketonization reactions considered

above, but the magnitude of the catalysis is unchanged by the

direction of reaction. In the enolization direction, the uncatalyzed

reaction involves transfer of a proton from the methylene group

of the ketone to a base B (which in the examples just considered

is the solvent water) to form an enolate anion (E^-). In the

catalyzed reaction the proton is initially bound to the substrate

with equilibrium constant K_Band the protonated substrate then

reacts with rate constant k_catto give the enol (EH). A similar

scheme applies to catalysis by metal ions. The binding constant

corresponds to the inverse of the dissociation constant for the

substrate-catalyst complex (KH₂) and for catalysis by H⁺log

+

K_B) pK_a.

From Scheme 7 it is evident that the ratio of rate constants

for catalyzed and uncatalyzed reactions (k_H/k_o) corresponds to

K_B(k_cat/k_o). This ratio has been designated the “proﬁciency” of

a catalyst by Wolfenden.³²As noted above, it corresponds to

the acceleration of the uncatalyzed reaction produced by one

molar concentration of catalyst (although this acceleration is

realized only when the solute is less than fully complexed, i.e.,

when [catalyst] > 1/K_B).

The importance of K_B(k_cat/k_o) is emphasized by its wide

interpretation as an equilibrium constant for binding a catalyst

(27) Hynes, M. J.; Clarke, E. M. J. Chem. Soc. Perkin Trans. 2 1998, 1263–

1267.

There have been a number of investigations of the relative

effectiveness of protons and metal ions in Lewis acid catalyzed

reactions. Preliminary results for tautomerization of phenacyl-

and phenylacetyl-pyridines and also phenacylpyrazine have been

reported by ourselves,⁶and an extensive range of enolization

reactions has been investigated by Fontana and De Maria.¹⁸

Other reactions studied or reviewed recently include eliminations

(28) Jones, J. R.; Taylor, S. E. J. Chem. Soc. Perkin Trans. 2 1979, 1773–

1776.

(29) Buncel, E.; Clement, O.; Onyido, I. Acc. Chem. Res. 2000, 33, 672–

678.

(30) Crugeiras, J.; Richard, J. P. J. Am. Chem. Soc. 2004, 126, 5164–5173.

(31) Hay, R. W. In ComprehensiVe Coordination Chemistry; Wilkinson, G.,

Gillard, R. D., McCleverty, J. A., Eds.; Pergamon Press: Oxford, 1987; Vol. 6,

pp 411-485.

(32) Miller, B. G.; Wolfenden, R. Annu. ReV. Biochem. 2002, 71, 847–885.

J. Org. Chem. Vol. 74, No. 9, 2009 3365

Collot et al.

to the transition state of an uncatalyzed reaction.^6,32-38We will

return to this point but focus ﬁrst on the expression of k_H/k_oas

the product of a binding constant (K_B) and rate constant ratio

(k_cat/k_o). The ratio (k_cat/k_o) has been called by Stewart and

Srinivasan a “proton activating factor” or “paf”.³⁹By analogy,

for metal ion catalysis, k_cat/k_obecomes a metal-activating factor

“maf”.⁴⁰De Maria and Fontana have referred to K_B(k_cat/k_o) itself

as a metal activating factor,¹⁸but in this paper we follow the

usage of Stewart and Srinivasan

TABLE 6. Binding Constants to Ketone and Enolate Anion and

Calculated and Observed Rate Constant Ratios for Catalysis of

Keto-Enol Tautomerization of Phenylacetylpyrazine and

Phenacylpyridine by Zinc Ions and Protons in Aqueous Solution at

25 °C^a

A Rate-equilibrium Relationship for Catalysis. If k_cat/k_o

is considered a kinetic activating factor, it is convenient to

identify an equilibrium activating factor, represented by the ratio

of equilibrium constants for reaction of the bound and unbound

substrates, which we may denote K_cat/K_o.⁶The magnitude of

K_cat/K_omeasures the increased thermodynamic driving force for

reaction conferred by binding the catalyst. A relationship

between kinetic and equilibrium factors can then be based on

the Bronsted relationship between rate and equilibrium constants,

log k₁/k₂) R log K₁/K₂, in which the coefﬁcient R normally

lies between 0 and 1. This relationship is shown in eq 23 and

expresses the kinetic effect of the thermodynamic advantage

arising from binding the catalyst.

^aBinding constants from this work for phenylacetylpyrazine and ref 6

for phenacylpyridine. ^bThe R (reactant) refers to the ketone and P

(product) to the enolate anion; ∆pK_acorresponds to pK_a^Zn- pK_a^Hand

(∆pK_a^R+ ∆pK_a^P)/2 represents values of log(k_H/k_Zn) predicted from eq

28. ^cThe pK_afor protonation at the 1-position of the pyrazine ring.

k_cat/k_o) (K_cat/K_o)^R

(23)

A simple extension of eq 23 allows k_H/k_oor k_M/k_o() K_Bk_cat/

k_o) to be expressed in terms of the binding constant for the

catalyst-substrate complex and the equilibrium constant ratio

K_cat/K_oas shown.

The approximation is an effective one partly because R is often

not far from 0.5⁴and partly because when K_Band K_Bare

R

P

R

K_Bk_cat/k_o) K_B(K_cat/K_o)^R

(24)

comparable in magnitude K_Bk_cat/k_obecomes insensitive to R.

Failure of the relationship may signal an anomalous value for

the Bronsted coefﬁcient.

This equation is conveniently transformed by expanding

Scheme 7 to include binding constants for the reactant and

For proton catalysis we recognize that the binding constants

for the catalyst are reciprocals of acid dissociation constants. If

P

product (K_B^Rand K_B) and replacing rate constants by equilib-

R

we replace K_Bk_cat/k_oby k_H/k_oand express this logarithmically,

rium constants to form the thermodynamic cycle of Scheme

8.⁶In Scheme 8, the proton catalyst has been replaced by a

metal ion (M²⁺) and again the single arrows represent equilibria.

we can transform eq 26 to eq 27 where (logs of) binding

constants for reactant and products are replaced by the corre-

sponding pK_avalues.

P

From Scheme 8 it is evident that K_cat/K_o) K_B/K_B^R. Making

this substitution in eq 24 we obtain eq 25, in which the

effectiveness of the catalysis is expressed in terms of equilibrium

constants for binding the catalyst to reactants and products, the

relative inﬂuence of which is determined by the Bronsted

coefﬁcient R.

log k_H/k_o) (pK_a^R+ pK_a^P)/2

(27)

Sometimes it is possible to compare catalysis by different

catalysts, such as protons and metal ions, when it is not possible

to measure a rate constant k_ofor the uncatalyzed reaction. The

ratio of rate constants for the two catalyzed reactions (e.g., k_H/

k_M) may then be expressed in terms of differences in binding

R

P

P R

K_Bk_cat/k_o) K_B^R(K_B/K_B^R)^R) (K_B^R)^1-R(K_B) (25)

If the value R is approximated as 0.5 we obtain the following

simple expression which allows the kinetic effect of the catalyst

to be estimated solely from knowledge of the binding constants.

H

constants ∆pK_a() log K_B- log K_B^M) for the protons and

metal ions complexed to reactants and products as shown in eq

28.

R

P 0.5

K_Bk_cat/k_o) (K_B^RK_B)

(26)

log(k_H/k_M) ) (∆pK_a^R+ ∆pK_a^P)/2

(28)

(33) Tee, O. S. AdV. Phys. Org. Chem. 1994, 29, 1–85.

(34) Kurz, J. L. Acc. Chem. Res. 1972, 5, 1–9.

(35) Kraut, J. Science 1988, 242, 533–540.

(36) Schneider, M. J.; Gaunitz, S.; Ridgeway, C.; Short, S. A.; Wolfenden,

R. Biochemistry 2000, 39, 9746–9753.

(37) Schowen, R. L. Transition States of Biochemical Processes; Schowen,

R. L., Gandour, R. L., Eds,; Plenum Press: New York, 1978; pp 77-117.

(38) Fersht, A. R. Structure and Mechanism in Protein Science, 3rd ed.; W

H. Freeman and Co.: New York, 1998. Anslyn, E.; Dougherty, D. Modern

Physical Organic Chemistry; University Science Books: Sausalito, 2005.

(39) Stewart, R.; Srinivasan, R. Acc. Chem. Res. 1978, 11, 271–277.

(40) Jones, J. R.; Taylor, S. E. Chem. Soc ReV. 1981, 329–344.

Phenylacetylpyrazine and Phenacylpyridine. Returning to

the comparison of phenylacetylpyrazine and phenacylpyridine,⁶

we can now determine how well the differences in catalytic

effectiveness of protons and metal ions are represented by their

binding constants. Table 6 lists equilibrium constants (shown

as pK_avalues) for binding protons and zinc ions to the reactants

and products of proton transfer from the methylene carbon atoms

of the two substrates, with a water molecule acting as a base.

3366 J. Org. Chem. Vol. 74, No. 9, 2009

Enolization of Phenylacetylpyrazine

transitions states”.^6,33-38This arises from writing a “kinetic-

equilibrium box” in analogy with the thermodynamic box of

Scheme 8 (Scheme 9). If the rate constants k_catand k_oare expressed

in terms of the pseudoequilibrium constants K^qfor formation of

their respective transition states, based on the relationship k ) (k_BT/

CHART 1

q

h)K^q, then a binding constant for the transition state K_Bmay be

identiﬁed with k_H/k_othrough eqs 29 and 30.

q

R

K_B/K_B) k_cat/k_o

(29)

(30)

K_B) k_catK_B^R/k_o) k_H/k_o

q

SCHEME 9

q

It is clear that K_Bprovides an economical characterization

of catalysis. However, to interpret this catalysis normally it must

be related to binding of the catalyst to the reactant and product

of the catalyzed step, e.g. through the Bronsted relationship of

eq 31. In this relationship the binding constant K_B^Pusually refers

to a reactive intermediate: in the case of phenylacetylpyrazine

for example it is an enolate anion. Effective catalysis is then a

result of strong binding of the catalyst to a reactant and a reactive

intermediate.

From these binding constants values of k_Zn/k_Hare calculated

from eq 28 and compared with experiment for both substrates

in Table 6.

In practice, as seen in the right-hand two columns of Table

6, the calculated values of k_Zn/k_Hfor both heterocyclic ketones

agree well with experiment. Likewise they account for the great

part of the variation between the ketones (1.5 × 10⁶compared

with 6.3 × 10⁶). The values do not depend on assignment of R

) 0.5 in eq 28 because ∆pK_a^Rand ∆pK_a^Pare nearly equal (Table

6) and the calculated values of k_Zn/k_Hare quite insensitive to R.

We may conclude that the greater susceptibility of phen-

acylpyridine than phenylacetylpyrazine to acid catalysis stems

from the stronger binding of the proton to the pyridine than

pyrazine ring in both reactants (pK_a) 5.03 compared with

-1.40) and products (pK_a) 12.22 compared with 2.80) and

that the difference in the products is augmented by stronger

binding of the proton in the resonance-stabilized enaminone than

zwitterion (or enol). For zinc ions the difference in Lewis

basicity between the two rings is moderated and there seems to

be no advantage (over and above that of basicity) of binding

the metal to an enaminone than a zwitterion. No doubt this

reﬂects the ability of zinc to coordinate both oxygen and nitrogen

basic sites of the zwitterions, whereas the proton is constrained

to the latter.

In principle, these conclusions are endorsed by extending

comparisons of k_H/k_Znto 2-phenylacetylpyridine and phenacyl-

pyrazine (5),⁶which show values intermediate between those

for phenylacetylpyrazine and phenacylpyridine (Chart 1). How-

ever, while the agreement between calculated and observed

values is good for phenylacetylpyridine (k_H/k_Zn) 8.2 compared

with 2.5 observed) the same is not true of phenacylpyrazine,

for which 2.5 is calculated but 133 is observed. In the latter

case it is possible that other factors, such as a difference in

intrinsic barriers for protonation of zwitterions and enamin-

ones,²⁸or the hard or soft character of keto binding sites,¹⁸affect

relative magnitudes. However, it is also true that the metal

binding constant for phenacylpyrazine (keto tautomer) had to

be estimated,⁶and it is possible that this is in error. It seems

reasonable to conclude therefore that thermodynamic driving

force, as expressed in the binding constants for protons and metal

ions in reactants and products, is the dominant inﬂuence on the

rates of these reactions.

q

P R

K_B) (K_B

)

^{R 1-R}(K_B)

(31)

A further implication of eq 31, already discussed, is that the

binding formalism it represents is equivalent to the “conven-

tional” treatment of catalysis in terms of rate and equilibrium

constants, based on the identity k_H/k_o) k_catK_B^R/k_oand the more

usual form of the Bronsted relationship k_cat/k_o) (K_catK_o)^R.

However, as discussed elsewhere,⁶there may be differences of

focus, e.g., on the substrate in the binding analysis and on acid

or base catalysts in a Bronsted-Marcus treatment.

The tautomerization of heterocyclic ketones lends itself to

application of a binding analysis partly because metal-

substrate interactions are naturally expressed in terms of binding

constants but also because the binding constants to the reactant

and product of the rate-determining step are accessible experi-

mentally. In cases where these constants are not readily

accessible, the catalysis may still be interpreted qualitatively if

the structures of the reactant and product (reactive intermediate)

are known or can be inferred. However, for more complex

catalysts, including enzymes, the situation may be less straight-

forward. One complication, which can then arise, is that a

conformation change or other equilibrium intervenes between

the binding equilibrium of the catalyst and the rate-determining

step of the reaction. Alternatively, a covalent reaction may take

place between the substrate and the catalyst,⁴¹or indeed, it may

be difﬁcult to identify an uncatalyzed reaction.^6,37In these cases,

the near-to-ideal behavior of the enolization of phenylacetylpyra-

zine provides a useful comparison for assessing the applicability

of (or need to modify) a transition-state-binding treatment.

Conclusions

Keto-enol tautomerism is the dominant tautomerism of

phenylacetylpyrazine, in contrast to phenacylpyridine for which

an enaminone is the preferred minor tautomer. This reﬂects a

lack of conjugation between the keto group and ring nitrogen

atoms of the phenylacetylpyrazine as well as the low basicity

of the nitrogens. This low basisicty is also responsible for the

Binding of Catalysts to Transition States. As mentioned

above, rate constant ratios comparable to k_H/k_oand k_M/k_ohave been

characterized as equilibrium constants for “binding catalysts to their

(41) Lienhard, G. E. Science 1973, 180, 149–154. Bruice, T. C. Acc. Chem.

Res. 2002, 35, 139–148. Braaun-Sand, S.; Olsson, M. H. M.; Warshel, A. AdV.

Phys. Org. Chem. 2005, 40, 201–245.

J. Org. Chem. Vol. 74, No. 9, 2009 3367

Collot et al.

reagent. Anhydrous cadmium chloride was a GPR reagent, as was

copper(II) chloride dihydrate

weak acid catalysis of tautomerization and correspondingly large

advantage of metal ions over protons as catalysts. The com-

parison of proton and metal ion catalysis of enolization (strictly

formation of the enolate anion) provides an “ideal” example

for analysis in terms of binding of the catalysts to reactants,

products, and transition states of a reaction. The close relation-

ship of this analysis to that based on rate-equilibrium relation-

ships and the kinetic effect of thermodynamic driving force is

emphasised.

Equilibrium Measurements. Equilibrium and kinetic measure-

ments were performed at 25 °C in aqueous solution. Ionic strengths

were usually 0.1 M unless indicated otherwise. Measurements of

pK_avalues were based on spectrophotometric measurements,

described in the Supporting Information, or proton NMR.

The NMR measurements of the protonation of phenylacetylpyra-

zine and acetylpyrazine were made in solutions of DCl in D₂O in

the concentration range 0 - 4.2 M, and pK_avalues in D₂O were

obtained from extrapolations based on eq 33 in which X is Cox

and Yates’s free energy/acidity parameter for correlating medium

effects on acid-base equilibria in DCl-D₂O,²⁰m* is the slope of

the plot of pK_aagainst X, K_a) [B][D⁺]/[BD⁺] at the indicated

acid concentration and pK_a(D₂O) is the intercept at X ) 0 (for pure

D₂O).

Experimental Section

Synthesis and Materials. A sample of 9-phenylacetylpyrazine

was prepared from pyrazine carboxylic acid using a modiﬁcation

of the method employed by Bunting and Stefanidis for the

corresponding phenylacetylpyridine.⁷The carboxylic acid was

converted to its methyl ester by Fischer esteriﬁcation and condensed

with ethyl phenylacetate to give the R-carbethoxy ester of phenyl-

acetylpyrazine (15) as shown in eq 29. The ester was then

hydrolyzed and decarboxylated to phenylacetyl pyrazine.

pK_a) pK_a(D₂O) + m*X

(33)

Values of K_awere evaluated from eq 34 below in which δ is the

chemical shift of the 5-hydrogen atom of the acylpyrazine at the

indicated concentration of D₃O⁺, and δ_Pand δ_Dare the limiting

chemical shifts at high and low acid concentrations for the

protonated and unprotonated heterocycles. Chemical shifts are listed

in Tables S1and S2 (Supporting Information), and the data are

plotted as δ against pH - X_oin Figures 1 and S1 (Supporting

Information). The line through the points in Figure 1 represents a

best ﬁt of calculated to experimental chemical shifts for optimized

values of δ_D, δ_P, m*, and pK_a.

2-Phenyl-2-carbethoxy-1-pyrazinylethanone 15. Phenyl-

ethylacetate (5.49 g, 0.033 mol) and methyl pyrazinoate (4.0 g 0.027

mol) in dry THF (30 mL) were added dropwise with stirring to a

freshly prepared solution of sodium ethoxide (0.75 g sodium, 0.033

mol) in 12 mL of ethanol. The solution turned orange and then

dark red during the course of the addition. It was heated to reﬂux

for 3 h and after cooling poured into water (200 mL). Unreacted

starting material was extracted with chloroform and the solution

acidiﬁed with concentrated sulfuric acid to give a brown semi-

solid which was extracted with chloroform (3 × 100 mL). The

product was dried (Na₂SO₄) and evaporated to give a brown oil

which solidiﬁed on standing (4.05 g, 60%). A portion of the product

(2.5 g) was puriﬁed by ﬂash chromatography on silica. Elution with

10:2 hexane/ethyl acetate gave 1.1 g of white solid. ¹H NMR (270

MHz, CDCl₃) δ: 9.24 (d, 1H, J ) 1.5 Hz), 8.76 (d, 1H, J )2.4

Hz), 8.68 (dd, 1H, J ) 1.4, 2.4 Hz), 7.26-7.45 (m, 5H), 6.06 (s,

1H), 4.13 -4.29 (m, 2H, J ) 7.15 Hz), 1.21 (3H, t, J ) 7.15 Hz).

2-Phenyl-1-pyrazinylethanone (phenylacetylpyrazine). 2-Phenyl-

2-carbethoxy-1-pyrazinylethanone (1.0 g) in 60% aqueous ethanol

(100 mL) containing concentrated HCl (1.5 mL) was heated under

reﬂux at 100 °C overnight. Completion of the reaction was checked

by TLC after a “mini-workup”. Saturated aqueous sodium bicar-

bonate was added to the reaction mixture which was then extracted

with dichloromethane (3 × 100 mL), washed with brine, dried

(Na₂SO₄), and evaporated to give a brown oil which solidiﬁed on

cooling (∼700 mg). The product was puriﬁed by ﬂash chromatog-

raphy on silica using 90:10 hexane/ethyl acetate to give white

crystals (∼500 mg). Anal. Calcd for C₁₂H₁₀N₂O: C, 72.72; H, 5.09;

K_a) [D⁺](δ-δ_D)/(δ_P-δ)

(34)

For phenylacetylpyrazine, pK_a) -0.42 ( 0.07 and m* ) 1.0

( 0.25 and for acetylpyrazine pK_a) -0.57 ( 0.07 and m* ) 1.4

( 0.25. The large uncertainty in the m* values reﬂects the low

acid concentrations studied and thus small deviation of pK_avalues

from those in water.

Kinetic Measurements. Kinetic measurements were carried out

using UV-vis spectra to monitor reactions. Directly measured rate

constants are generally subject to random errors of (5%. However,

values dissected from combinations with equilibrium constants,

from pH proﬁles or kinetic saturation plots, may be subject to a

greater uncertainty, usually noted in the text, and were used to

construct the log k_obs- pH proﬁles in Figures 2 and 3 from the

inﬂections of which pK_avalues for mono- and di-protonation of

the enol were inferred.

First-order rate constants for ketonization of the enol tautomer

of phenylacetylpyrazine, k_obs, were measured in HClO₄, HCl and

DCl as described in the Results and Supporting Information. The

values are listed in Tables S6 and S9 (Supporting Information) and

are plotted as log k against pH or pH - X_oin Figures 2 and 3. The

plot against pH - X_o(i.e., for X_o* 0) refers to concentrated

solutions of the acids and is based on modiﬁcation of eq 9 so that

k_AH^H3O+, K_a^EH2+and K_a^EH3++are multiplied by the term 10^m*Xo

,

where m* is the slope of a plot of log k or pK_aagainst X_o.²²As

noted above, spectrophotometric evaluation of m* for the equilib-

rium constants gives values close to 1.0. Likewise, the near

independence of acid concentration of the value of log k_obsat high

acid concentrations (Figure 2, pH - X_o< -2) implies that m₃* for

k_AHis close to 1.0. The lines drawn through the data points in

Figures 2 and 3 are calculated on this basis and correspond to

1

N, 14.15. Found: 72.16; H, 5.08; N 14.15. H NMR (270 MHz,

CDCl₃) δ: 9.3 (m, 1 H), 8.8 (m, 1H), 8.7 (m, 1H,), 7.2 - 7.5 (m,

5H), 4.5 (s, 2H). ¹³C NMR (68 MHz, CDCl₃) δ: 198.5, 147.8, 147.3,

144.2, 143.5, 133.8, 129.9, 129.9, 128.6, 127.0, 44.2. MS m/z: 198

(M⁺, 54), 169 (31), 107 (36), 91(100), 65 (40), 52 (32), 39 (18).

Other organic materials were used without puriﬁcation, including

2-pyrazinecarboxylic acid and 2-acetylpyrazine. Inorganic reagents

used for kinetic or equilibrium measurements were generally AR

grade (NaCl, NaBr, KI, nickel(II) chloride hexahydrate, and

copper(II) nitrate trihydrate). Iodine was GPR grade (resublimed),

and bromine was Aristar grade. Zinc(II) nitrate tetrahydrate was a

“proanalysi” reagent and Cobalt(II) nitrate trihydrate was a “reinst”

replacing [H⁺] in eq 9 by 10^(pH-Xo)

.

It is noteworthy that the pK_avalues derived from Figures 2 and

3 have lower values than those determined spectrophotometrically

from intial absorbances of the enol or protonated enol from

quenching the enolate anion in acid media. For pK_a^EH2+and

pK_a^EH3++in HClO₄the spectrophotometric values are, respectively,

0.45 (0.42 in HCl) and -4.8 compared with kinetic values of -0.58

(-0.41 in HCl) and -7.9. The discrepancy for pK_a^EH3+2reﬂects

3368 J. Org. Chem. Vol. 74, No. 9, 2009

Enolization of Phenylacetylpyrazine

the failure of the rate constants in Figure 2 to show a distinct

downturn at very negative pH - X_ovalues and might be dismissed

as an unexplained feature of strong acid solutions such as that

leading to an increase in rate at very negative pH - X_ovalues in

HCl. Similar differences in values for ionization of the enol to its

anion, i.e. 9.0 from combining pK_E(2.9) with the spectrophoto-

metrically determined value for ionization of the keto tautomer

(pK_a^KH) 11.9) and 7.7 from the pH proﬁle, may reﬂect uncertainty

in kinetic measurements in borate buffers. However, the values for

pK_a^EH2+show a consistent discrepancy of one unit both in HClO₄

and HCl. It is difﬁcult to ﬁnd an explanation for this behavior other

perhaps than that E to Z isomerization of an initially formed enol

conﬁguration from quenching the enolate anion occurs in competi-

tion with reaction of the enol to form its keto tautomer. Such E to

Z isomerization has been observed for enaminone tautomers of other

heterocylic ketones.⁴²Differences in pK_avalues and reactivities of

E and Z enols of R-phenylacetophenone have been reported by

Kresge.⁴³

Measurements of rate constants for enolization were based on

trapping the enol with iodine or bromine in the presence excess

iodide or bromide ions as described previously for other heterocyclic

ketones.³Iodinations were conducted under zero order conditions

with a large excess of substrate over the iodine. The standard

expression for the rate constant k^Ebased on measurement of the

slope of the linear dependence of absorbance of I₃^-upon time (dA/

dt) is given by eq 35, in which [PPz] is the concentration of the

phenylacetylpyrazine, ε ) 2.6 × 10⁴is the extinction coefﬁcient

of I₃^-and K_I) 713 is the association constant between I₂and I^-.

Inclusion of the term (1 + K_I[I^-])/K_Icorrects for dissociation of a

fraction of the triiodide ion into iodine and iodide ions.

the measured rate constants in dilute HClO₄show a mild increase

with acid concentration. The increase is too large to be attributed

to enolization and, as discussed above, may be a result of acid-

catalyzed bromination of the pyrazine ring. However, extrapolation

to zero acid concentration gave a rate constant k^E) 1.37 × 10^-5

s^-1in satisfactory agreement with the iodination measurements. A

single measurement in DCl and D₂O at an acid concentration too

low to affect the rate (0.005M) gave k^E(D₂O) ) 0.57 × 10^-5

corresponding to an approximate solvent isotope effect k^E(H₂O)/

k^E(D₂O) ) 2.4.

Rate constants for iodination were also measured in 4:1 acetic

acid buffers ([AcO^-]/ [AcOH]). At the prevailing pH there was no

problem with reversibility and the kinetics were consistently zero

order. The derived ﬁrst order rate constants are listed in Table S11

(Supporting Information) and when plotted against concentration

of acetate ion gave a second order rate constant for catalysis of

enolization by acetate ion of k_GB) 2.33 × 10^-3M^-1s^-1

.

E

Rate constants for zinc-catalyzed iodination in the absence of

added acid or buffer are recorded in Table S15 (Supporting

Information) and show a linear dependence on concentration of

zinc ions corresponding to a rate constant k_Zn^E(H₂O) ) 3.68 × 10^-3

M^-1s^-1. However, as described in the Results, a similar plot for

the Zn²⁺and cacodylate buffer-base catalyzed reaction in 60%

TFE-H₂O mixtures ([base] ) 8 × 10^-4M, [I^-] ) 0.005 M) showed

distinct curvature with increasing concentration of zinc ions up to

the solubility limit of 0.35 M in the mixed solvent. The measured

ﬁrst-order rate constants are shown in Table S16 (Supporting

Information) and plotted against [Zn²⁺] in Figure S3 (Supporting

Information). The cacodylate buffer (with 4:1 ratio of buffer base

to buffer acid) was chosen to maintain a high enough pH to prevent

reversibility of the iodination (cf. eq 33). Rate constants for the

Zn²⁺-catalyzed iodination of acetylpyridine in 60% TFE-H₂O in

the absence of buffer are shown in Table S17 (Supporting

Information). Comparison of binding constants for the two sub-

strates allowed extrapolation of a value of K_B) 6.9 M^-1for

phenylacetylpyrazine in aqueous solution as already described.

∆A(1 + K_I[I])

ε∆t[PPZ]K_I[I]

k^E)

(35)

Reversibility of iodination reactions was observed at higher

concentrations of hydrogen and iodide ions under the inﬂuence of

the equilibrium shown in eq 36 and leads to mixed zero- and ﬁrst-

order kinetics. Measurements in 0.008 and 0.018 M HClO₄with

2.0 × 10^-3M phenylacetylpyrazine and 0.02 M KI showed some

curvature of plots of absorbance against time but allowed rate

constants k^E) 1.26 and 1.29 × 10^-5s^-1to be determined from

measurements of initial slopes.

Acknowledgment. Synthesis and preliminary kinetic and

equilibrium measurements carried out by Laura Macon and Eoin

Clarke are gratefully acknowledged.

Supporting Information Available: Analysis of O- versus

N-protonation of phenylacetylpyrazine, experimental details of

kinetic and equlibrium measurements, Tables S1-S17 contain-

ing additional details of rate and equilibrium data referred to in

the text; Figure S1, a plot of chemical shifts of a ring proton

against acidity for the protonation of acetylpyrazine, Figure S2,

a plot of absorbance against acidity for the second protonation

of phenylacetylpyrazine enol, Figure S3, a plot of rate constants

for iodination of phenylacetyl pyrazine against concentration

of zinc ions, and Figures S4 and S5, NMR spectra of pheny-

lacetylpyrazine and 2-phenyl-2-carbethoxy-1-pyrazinylethanone.

This material is available free of charge via the Internet at

http://pubs.acs.org.

I^-₃+ P_Z–H h P_Z–I + H⁺+ 2I^-

(36)

Bromination is less susceptible to reversibility than iodination.

Moreover, because of the high rate of reaction and relatively low

-

extinction coefﬁcient of Br₃it was possible to work under ﬁrst

order conditions, with a small excess of bromine over substrate

(e.g., 4 × 10^-3compared with 2 × 10^-3M), thus avoiding a need

to know the concentration of halogen as required of zero order

kinetics (eq 32). As shown in Table S12 (Supporting Information),

(42) De Maria, P.; Fontana, A.; Spinelli, D. J. Chem. Soc., Perkin Trans. 2

1991, 1067–1070.

(43) Chiang, Y.; Kresge, A. J.; Walsh, P. A.; Yin, Y. J. Chem. Soc., Chem.

Commun. 1989, 869–871.

JO802650S

J. Org. Chem. Vol. 74, No. 9, 2009 3369

Article Doi

Binding catalysis and inhibition by metal ions and protons in the enolization of phenylacetylpyrazine

DOI: 10.1021/jo802650s

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jo802650s

Full text of DOI:10.1021/jo802650s

Products guided by the article

R&D Labs maybe for 1137718-82-1

Relevant to this article

Hot Product