Mechanism of acid-catalyzed photoaddition of methanol to 3-alkyl2

Mechanism of acid-catalyzed photoaddition of methanol to 3-alkyl-2-cyclohexenones

Can. J. Chem. Downloaded from www.nrcresearchpress.com by 218.92.227.165 on 06/03/13 For personal use only.

David I. Schuster, Jie-Min Yang, Jan Woning, Timothy A. Rhodes, and Anton W. Jensen

Abstract: Contrary to a previous report, it is concluded that formation of methanol adducts to 3-methyl-2cyclohexenones and of deconjugated enones on irradiation of the enones in acidified solutions proceeds via protonation of the intermediate enone T,T*triplet excited state and not by protonation of a relatively long-lived ground state trans-cyclohexenone. A rate constant for protonation of the triplet state of 3-methyl-2cyclohexenone by sulfuric acid of 1.7 x lo9 M-I s-' w as determined by laser flash photolysis in ethyl acetate. Based on quantum efficiencies of product formation, a rate constant of ca. 10' M-I s-' was estimated for protonation of the enone triplet by acetic acid, which is too small to cause measurable reduction in the triplet state lifetime in the mM concentration range used in the preparative studies. The intermediate carbocation can be trapped by methanol, or revert to starting enone or the exocyclic deconjugated enone by loss of a proton. Since products revert to starting materials in an acid-catalyzed process, there is an acid concentration at which the yields of products are optimal. This concentration is ca. 6 m M for acetic acid, but is only 0.1 mM for p-toluenesulfonic or sulfuric acids. Product formation could be quenched using I-methylnaphthalene and cyclopentene as triplet quenchers; in the latter case, formation of [2 + 21 photoadducts was observed to compete with formation of methanol adducts. Quenching rate constants were determined by laser flash studies. Key words: laser flash photolysis, kinetic absorption spectroscopy (KAS), photoacoustic calorimetry (PAC), protonation of triplet states, trans-cyclohexenones.

RCsum6 : Contrairement a ce qui a CtC publiC anterieurement, on montre que la formation d'adduits du mCthanol avec les 3-mCthylcyclohexCn-2-ones et les Cnones dCconjuguCes, par irradiation des &nonesen solutions acidifites, ne se produisent pas par le biais d'une protonation de 1'Ctat fondamental de la trans-cyclohexCnone qui possede une durCe de vie relativement longue. Faisant appel h la photolyse laser Cclair dans l'acCtate d'Cthyle, on a dtterminC que la constante de vitesse de protonation de 1'Ctat triplet de la 3-mCthylcyclohexCn-2one, par de l'acide sulfurique, est de 1.7 x lo9 M-I s-I. En se basant sur les efficacitts quantiques de formation du produit, on a Cvalut que la constante de vitesse pour la protonation du triplet de 1'Cnone par l'acide acCtique est d'environ lo8 M-' s-I qui est beaucoup trop faible pour provoquer une reduction mesurable du temps de vie de 1'Ctat triplet dans la zone de concentration (mM) utilisC dans les Ctudes prkparatives. On peut pitger le carbocation intermtdiaire par le mCthano1 ou le laisser retourner a 1'Cnone de dCpart ou 2 l'tnone exocyclique dCconjuguCe par perte d'un proton. Puisque les produits retournent aux produits de dtpart dans un processus acidocatalyst, il existe une concentration d'acide h laquelle les rendements en produits sont optimaux. Cette concentration est d'environ 6 m M pour l'acide acCtique, mais elle n'est que de 0.1 mM pour les acides sulfurique et p-toluenesulfonique. La formation de produit peut &tredCsactivCe par l'utilisation du I-mCthylnaphtalkne ou du cyclopentene comme dtsactivants des triplets; dans le dernier cas, on a observC la formation de photoadduits [2 + 21 qui sont en competition avec la formation d'adduits du mtthanol. On a determine les constantes de vitesse de dksactivation h l'aide dlCtudes au laser tclair. Mots clds : photolyse laser Cclair, spectroscopie d'absorption cinetique, calorimktrie photoacoustique, protonation des ttats triplets, trans-cyclohexCnones. [Traduit par la rCdaction]

Received April 10, 1995. This paper is dedicated to Professor J.E. (Jim) Guillet in recognition of his corltributions to Canadian chemistry.

D.I. ~chuster,'J.-M. Yang, J. Woning, T.A. Rhodes, and A.W. Jensen. New YorkUniversity, Faculty of Arts and Sciences, Department of Chemistry, New York, NY 10003, U.S.A. Author to whom correspondence may be addressed. Telephone: (21 2) 998-8447. Fax: (212) 260-7905. E-mail: [email protected] Can. J. Chem. 73: 20042010 (1995). Printed in Canada / ImprimC au Canada

Schuster et al.

Scheme 1.

Fig. 1. Relative yield of methanol adduct of Pummerer's ketone (compound 6) in benzene containing 10%methanol as a function of acetic acid concentration.


Scheme 2.

0

MeOH

Introduction It was reported by Rudolph and Weedon (1) that small amounts of acetic acid catalyze the photochemical addition of methanol to isophorone ( l a ) in benzene or ethyl acetate solvents (see Scheme 1). The limiting quantum yield of this reaction (4ca. 0.04) was reached at ca. 0.004 M acetic acid. The reaction is accompanied by deconjugation to enone 4a. Quenching of the reaction by isoprene (a common triplet quencher) gave a linear Stern-Volmer plot, implicating the T, state of l a as the reactive intermediate. Formation of neither 3a nor 4a proceeds efficiently in the absence of acetic acid, or in neat methanol. The mechanism therefore appears to involve interaction of acetic acid with either the T , state of l a or a subsequent intermediate to give carbocation 2a (or its en01 tautomer), which then either deprotonates to give 4a or is trapped by methanol to give adduct 3a. Based on an estimated triplet lifetime of 10 ns for l a , Rudolph and Weedon (1) ruled out direct trapping of the triplet by 0.004 M acetic acid, since they estimated that a protonation rate constant of ca. 10" M-' s-' was required, which would be higher than the diffusion-controlled rate. By analogy with the proposed mechanism for acid-catalyzed photochemical reace ~ it was suggested that the reaction tions of c y c l ~ h e x e n '(2); intermediate trapped by acetic acid was a ground state trans isomer of l a with an estimated lifetime of 1 ps. Although trans-cyclohexenones have never been directly observed to date (3), stereochemical data consistent with the intermediacy of a trans-cyclic enone have been reported for the photochemical addition of methanol to Pummerer's ketone 5 (a 2-cyclohexenone) (Scheme 2) and for the photochemical addition of nucleophiles to 2-cycloheptenones and 2-cyclooctenones (4, 5). Based on kinetic data, Schuster et al. suggested that photochemical [2 + 21 cycloadditions of 2-cyclohexenones to alkenes might occur in some cases via transient twisted ground state trans-2-cyclohexenones (6). In response to the report by Rudolph and Weedon, we now report the results of steadystate and nanosecond flash photolysis studies of isophorone ( l a ) and 3-methyl-2-cyclohexenone (lb) in neutral and acidic media (7).

Results and discussion A. Continuous irradiations We first wanted to establish whether Pummerer's ketone (PK, 5) behaved in a manner similar to that reported by Rudolph and Weedon for isophorone (1). Indeed, irradiation of PK in benzene containing 10% methanol and varying amounts of acetic acid led to a methanol adduct, according to GC-MS analysis, presumably the adduct 6 previously described by Hart and co-workers (4). As shown in Fig. 1, the yield of this adduct peaked at about 4 rnM HOAc, consistent with the findings for isophorone (1). Having established this similarity in photochemical behavior, we decided to focus our studies on 3methyl-2-cyclohexenone (lb), which has been the subject of numerous mechanistic investigations in our laboratory (3, 6, 8-1 0). Continuous irradiation of l b in 90: 10 ethyl acetate - methanol containing up to 0.008 M acetic acid gave 3b and 4b. These compounds were characterized by 'H NMR spectroscopy and GC-MS (see Experimental for details). A plot of the yield of these products vs. the concentration of acetic acid is shown in Fig. 2. Analogous to the behavior of isophorone la and PK (see above), the yields of 36 and 4b level off at acetic acid concentrations above ca. 0.006 M. This behavior is probably due to the thermal instability of these products, which both undergo acid-catalyzed reversion to the starting material l b . Upon irradiation of l b in ethyl acetate - methanol containing a stronger acid, p-toluenesulfonic or sulfuric acid, 3b and 4b were again obtained, but in this case their yields sharply declined at acid concentrations above lo4 M (Fig. 3).

Can. J. Chern. Vol. 73, 1995


Fig. 2. Relative yields of methanol adduct of 3-methylcyclohexenone (compound 3b,+) and deconjugated enone (4b.A) in ethyl acetate containing 10% methanol as a function of acetic acid concentration.

photocycloaddition of enones to alkenes occurs via enone triplet excited states. Thus, the steady-state behavior of l b in acidic methanol, which clearly occurs via triplet states of the enone, closely resembles that previously reported (1) for isophorone (la).

B. Laser flash photolysis - kinetic absorption spectroscopy

Fig. 3. (a) Relative yields of compound 3b as a function of the concentration of p-toluenesulfonic acid in ethyl acetate solvent. (b) Same as (a) but using sulfuric acid in place of p-toluenesulfonic acid.

&L

d

[p-Toluenesulfonic acid], mM

0

2

4

6

8

10

1 2 1 4

Using both nanosecond kinetic absorption spectroscopy (KAS) (8, 11) and time-resolved photoacoustic calorimetry (PAC) (9), the triplet lifetime of isophorone (la) in acetonitrile was found to be 79 ns, much higher than previously assumed (1). Thus, direct protonation of triplet excited states of l a by acids is more likely than previously concluded (1). In principle, if the second-order rate constant for protonation of the triplet is sufficiently large, the enone triplet lifetime should decrease as a function of increasing acid concentration. We elected to use enone lb, which, as shown above, behaves similarly to enone l a under the photocheniical reaction conditions. For the laser flash (KAS) studies, l b was excited with a 10 ns pulse of a YAG (355 nm) or nitrogen laser (337 nm). Following excitation, the triplet lifetime (7,) was determined from the decay of the transient absorption at 280 nm (8, 1 I). The values for 7, measured under different conditions were 78 2 2 ns in benzene containing up to 2 x M HOAc; 73 -t 2 ns in neat EtOAc; 79 2 2 ns in EtOAc-MeOH 90: 10 containM HOAc; and 95 -+ 3 ns in MeOH containing up to 2 x 1o - ~ M HOAc (the indicated errors reflect the ing up to 7 x 90% confidence limit). (Values of 7, for enones are consistently higher in methanol than in nonpolar and aprotic polar solvents (3)). Thus, 7, appears to be independent of the acetic acid concentration within experimental error. However, when acetic acid was replaced by sulfuric acid, direct quenching of the triplet of enone l b in ethyl acetate was observed. From the Stern-Volmer plot shown in Fig. 5, a rate constant of 1.7 x 10' M-' s-I for protonation of the triplet of l b was obtained using eq. [I]. This value can then be used to calculate the quantum yield [I]

1/~,),~=l / ~ ~ + k , [ H + ]

[Sulfuric acid], mM

Although linear Stern-Volmer plots were obtained for quenching of both products by the triplet quencher l-methylnaphthalene (I-MeNA), quantitative data in this system were not very reproducible, probably because of competitive light absorption by the quencher at the concentrations required for these experiments (up to 0.3 M) using broad spectral sources. Better data were obtained under similar reaction conditions using cyclopentene as the triplet quencher. In this case, a reduction in the yields of products 3b and 4b was observed concomitant with an increase in yields of the well-known [2+2] cycloadducts 7 (see Fig. 4). It is well established (8) that

of protonation of the triplet (+ ) using eq. [2]. Thus, at a sulfuric acid concentration of 10-J M, where 7, was determined is -0.1. This is slightly greater than the quanto be 58 ns, tum yield of product formation (+ = 0.04), suggesting that deprotonation of carbocation 2b to regenerate the starting material competes with formation of 3b and 4b. Based on our value of 7, for isophorone (la), the quantum yield reported (1) for formation of photoproducts 3a and 4a from l a at 4 x M acetic acid would require a rate constant k, for protonation of the enone triplet by acetic acid of ca. 10'

+,

Schuster et al.


Fig. 4. (a) and (b)Relative yields of 3b and the cis- and trans-fused isomers of 7 upon irradiation of l b in ethyl acetate containing 10% methanol and 10 rnM acetic acid, as a function of the concentration of added cyclopentene. (c) and (4Stern-Volrner plots of the ratio of the yields of 3b and 4b in the absence and presence of cyclopentene as a function of cyclopentene concentration.

[Cyclopentene], M

5 = -a d

0

0.2

0.4 0.6 0.8 [Cyclopentenel, M

[Cyclopentenel, M

1.0

1.2

Fig. 5. Inverse triplet lifetime (in s-I) of l b in ethyl acetate as a function of sulfuric acid concentration.

[Sulfuric acid], N

M-' s-' rather than the previous estimate (1) of loL2M-' s-I. This lower value of k, is realistic given that HOAc is a much weaker acid than H2S04.Using eq. [ I ] it can be calculated that the reduction in T, for l b would be too small to be detected by KAS over the HOAc concentration range (up to 7 mM) utilized in these studies, in accord with our observations. Rudolph and Weedon (1) reported that the Stern-Volmer slope k,~, for quenching of deconjugation of l a by isoprene in benzene solvent was 60 M-', which is consistent with quenching of a triplet state of lifetime about 80 ns since rate constants

0.54 0

I

0.2

0.4

0.6

0.8

1.0

1.2

[Cyclopentene], M

for triplet energy transfer to acyclic 1,3-dienes in related systems are typically about lo8 M-' s-' (8). Rudolph and Weedon (1) also showed a plot of the inverse quantum yield for deconjugation of l a in benzene vs. the inverse concentration of acetic acid. Based on the slope and intercept of this plot, assuming a linear correlation, they concluded that the enone triplet state could not be the species undergoing protonation. In addition to their erroneous assumption regarding the rate of decay of the enone triplet, the significant deviation of their data from the assumed straight-line correlation at high acid concentrations was ignored. Given the kinetic and thermodynamic instability of the deconjugation product 4b observed in the present study (see Experimental), the validity of the quantum yield data reported for formation of 4a in acidic solution as well as the derived kinetic parameters (1) are open to question. In any event, the inconsistency between the kinetic data of Rudolph and Weedon (1) and the results of the present study need to be resolved. As a check of the steady-state quenching studies discussed earlier, rate constants for quenching of triplet states of l b by both 1-MeNA and cyclopentene were directly measured. Stern-Volmer plots of (7,)-' vs. quencher concentrations in ethyl acetate (in the absence of methanol or acetic acid) afford quenching rate constants of 4.5 x lo9 M-' s-' and 5.6 x lo6 M-I s-I, respectively (see Figs. 6 and 7). The latter value is within a factor of 3 of that (1.6 x lo7 M-' s-') calculated from

Can. J. Chem. Vol. 73, 1995

Fig. 6. Inverse triplet lifetime (in s-') of 16 in ethyl acetate in the absence of methanol and acetic acid as a function of l-methyl-


naphthalene concentration.

0

1

2

3

4

5

[MeNA], mM

Fig. 7. Same as Fig. 6, but using cyclopentene as the triplet

quencher. 2.90E+07 T

more, PAC data (7, 9) for isophorone, 3-methyl-2-cyclohexenone, and other 2-cyclohexenones rigorously exclude the formation of a high-energy ground state intermediate with a lifetime in the microsecond range and measurable (>0.05) quantum efficiency. That is, intersystem crossing and radiationless decay of the enone triplet accounts for essentially all (>98%) of the energy of the absorbed photon. Substantial heat discrepancies are seen only in cases where bona fide strained ground state trans isomers are formed, as with l-acetylcyclohexene (13). As noted earlier, some inconsistencies between our present data and the earlier report by Rudolph and Weedon (1) remain to be resolved. Theoretical calculations by ~ohnson'indicate that trans-2-cyclohexenones indeed appear to lie in a potential minimum on the 2-cyclohexenone ground state energy surface. However, under ambient conditions, the lifetimes of such species are expected to be extremely short, probably on the order of nanoseconds or even less. Thus, we conclude that trans-cyclohexenones are not likely intermediates in the photoreactions of cyclohexenones under the acidic conditions described in this paper and in the earlier study (I), and that the only hope of observing such species spectroscopically is by generating them in particularly favorable systems at very low temperatures.

Experimental A. Laser flash photolysis

[Cyclopentene], M

the slope (1.29 M-') of the steady-state Stern-Volmer plot in Fig. 4b under the reaction conditions, assuming a triplet lifetime of 80 ns. However, quenching of product formation by IMeNA is not nearly as efficient as expected from the diffusion-controlled triplet quenching rate constant measured by flash photolysis. A likely explanation is that under the steadystate reaction conditions, where I-MeNA is competing with the enone for light with a broad spectral distribution generated from the Hanovia or Rayonet lamps (see Experimental), 1MeNA may be acting as both a singlet sensitizer and a triplet quencher. This is not unreasonable since the singlet excited state lifetime of 1-MeNA is nearly 100 ns in polar solvents (1 2) and diffusion-controlled singlet energy transfer to l b at a concentration of 0.07 M should be an efficient process.

Conclusions Based on both steady-state and flash kinetic data, we conclude that it is highly probable that the species actually undergoing protonation in these reactions is the enone triplet excited state and not a ground state trans-cyclohexenone. The enone triplet in these systems is sufficiently long lived that direct quenching by acids to give the carbocations 2 is a kinetically viable process. Whether protonation of the triplet may be adiabatically reversible under these conditions remains in question. As far as the intermediacy of trans-cyclohexenones is concerned, we have repeatedly failed to detect such species on laser flash photolysis of a large number of cyclohexenones (8). Further-

General sample preparation Commercially available 3-methyl-2-cyclohexenone, cyclopentene, and 1-methylnaphthalene were purified by careful distillation. All the solvents employed were either HPLC grade or spectrograde. Sample solutions were prepared by weighing the enone in volumetric flasks followed by addition of the proper amount of other chemicals such as methanol or acetic acid into the flask and adding the solvent to the mark. A pipet was used to transfer solutions to the reaction cells. The optical density of the solutions studied ranged from 0.5 to 1.0 at the excitation wavelength. All samples were purged with either argon or oxygen-free nitrogen for at least 10 min prior to the experiments. Quenchers (either neat or in solution) were added using a microlitre syringe. At Columbia University, data was obtained by averaging 10-30 laser pulses in each measurement depending on the signal-to-noise ratio. At the National Research Council in Ottawa, data from 30-200 laser pulses were averaged. Apparatus and procedures The nanosecond laser flash photolysis configurations at Columbia University and at the National Research Council of Canada are very similar, but differ in details such as the type of laser used and in the arrangement of the exciting and analyzing beams. In experiments carried out at Columbia University (14), a Quanta Ray Q-switched Nd-YAG laser was used. After pulse selection, amplification, and frequency conversion, pulses of duration of ca. 6 ns and energies of 2-5 d l pulse, which varied less than 10% in intensity at 355 nm, were provided. The exciting and analyzing beams were oriented at R. Johnson, University of New Hampshire, private

communication of unpublished results.


Schuster et al

right angles (crossed-beam arrangement). The analyzing light source was a 150 W xenon lamp (Oriel). The samples (2 mL) were contained in Super cells with a 10 mm optical path. At the NRC in Ottawa (15), a Molectron UV-24 nitrogen laser, providing pulses of 8-10 ns and energies up to 10 mJ at 337.1 nm, was employed. A crossed-beam arrangement with an angle of 15" was used throughout. The use of this small angle configuration considerably decreases the volume irradiated and is more convenient for lower energy lasers (16). A pulsed 200 W xenon lamp was employed in the monitoring system and a high-intensity B & L monochromator and an RCA-4840 photomultiplier tube were used at the detection end. The photomultiplier response was fed to a 7A16A plugin used in a R79 12-Tektronix transient digitizer that contains a 4K local memory in which each trace was stored. The samples (1 rnL) were contained in Suprasil cells made of 7 x 3 mm rectangular tubing (Vitro Dynamics). The temperature of the cell was controlled by a constant-temperature nitrogen flow and was measured with a copper-constantan thermocouple. At both laboratories, the lifetime of the optically detected transient is determined from the slope of a plot of In (IIIo) as a function of time. Although the data collection and analysis were controlled by the computer using programs developed individually in each lab, the baseline and region of the trace to be analyzed is determined manually by the operator. The error in the triplet lifetimes evaluated by the present analyzing method is estimated to be f10%.

B. Steady-state photolyses All preparative reactions were done using a Rayonet 350 nm Srinivasan-Griffin photochemical reactor, which has a broad range from 300 to 400 nm, with a maximum at 350 nm, a Hanovia 550 W high-pressure mercury arc lamp, or a Hanovia 450 W medium-pressure mercury arc lamp. The latter two required a water-cooled jacket and the samples were suspended in Pyrex test tubes outside the immersion well. All ihree reactors were equipped with a merry-go-round rotating turntable for even irradiation of each sample. Samples in Pyrex tubes that filter out light below 280 nm were purged with argon for 10 min prior to irradiation. All reactions were monitored by capillary GC analysis, using either a HewlettPackard model 5890 gas chromatograph equipped with a Hewlett-Packard 3396A integrator or a Hewlett-Packard 5710A gas chromatograph equipped with an 18740B capillary column control, 5702A oven temperature programmer, and a 3390A integrator. The same type of capillary column (Alltech, 30 m X 0.25 mm Heliflex, RSL-150) was used throughout. nDodecane was used as the internal standard. The major products formed in the reaction were analyzed by GC-MS, using a Hewlett-Packard model 5992 GC-MS system equipped with a Hewlett-Packard 20 m x 0.25 mm cross-linked methyl silicone capillary column. Ultraviolet absorption spectra were obtained with a Cary 210 double-beam spectrophotometer. Preparative cycloaddition of methanol to Pummerer's ketone 5 Pummerer's ketone (PK) was synthesized and purified as described in the literature (4). A series of benzene solutions containing 0.145 M PK and 10% MeOH and various concentrations of acetic acid were prepared. The final concentrations of acetic acid were 0.0, 0.25, 0.50, 1.00, 2.00,4.00, 6.00, and

10.00 mM. These solutions in Pyrex tubes were purged with argon for 10 min and irradiated with a Hanovia 450 W lamp for 45 h. One new product was formed in the reaction, which was confirmed by GC-MS to be the PK-methanol adduct. The mass spectrum of the adduct, mlz (relative intensity), was 246 (42.7) M', 160 (14.9), 159 (79.9), 147 (11.4), 146 (loo), 145 (49.0), 115 (15.5), 91 (10.0). Optimization of photoaddition of methanol to 3-methyl-2cyclohexenone (Ib) This experiment was carried out in ethyl acetate solvent with a series of solutions containing 0.145M lb, 10% methanol, and acetic acid concentrations of 0.00, 2.00, 4.00, 8.00, 16.00, 32.00, and 64.00 mM, respectively. The samples were purged with argon for 10 min and then irradiated for 24 h. Several new species appeared on GC analysis, which corresponded to compounds 3b, 4b, and trace amounts of 3-methyl-3-cyclohexenone (8) and enone photodimers. The mass spectra of each component, mlz (rel. intensity), were as follows: 3b: 142 (10.2) M', 110 (1 1.8), 85 (83.1), 72 (40.6), 54 (loo), 42 (86.0); 4b: 110 (20.2) M', 82 (loo), 54 (35.2), 41 (14.5); 8: 110 (10.3), 82 (20.1), 55 (25.0), 43 (100.0). Similar results were obtained when benzene was employed as the solvent. When sulfuric acid orp-toluenesulfonic acid was used as the catalyst, the acid concentrations in the tubes were 0.0,0.025,0.05,0.1, 0.2, 0.4, 0.8, 1.6, 3.2, 6.4, and 12.8 mN, respectively. The results otherwise were similar to those obtained using acetic acid as the catalyst. Preparative scale irradiation of 3-methyl-2-cyclohexenone (Ib) in methanol To 335 pL (4.5 mmol) of l b in a 35 mL Pyrex test tube were added 3.1 mL MeOH, 28 mL EtOAc, and lastly 17 pL of HOAc. The final concentrations of l b and HOAc were 0.14 M and 0.01 M, respectively. Each solution was then sealed with a rubber septum and purged with argon for 30 min. Each sample was then irradiated with a 450 W medium-pressure Hg arc lamp fitted with a water-cooled jacket at a distance of about 1 cm from the jacket for 18 h. GC-MS analysis of the irradiated mixture indicated conversion to two major photoproducts. The crude mixture was then washed (NaHCO,, water, and brine), dried over anhydrous MgSO,, gravity filtered, and concentrated in vacuo. When washings were not performed and the sample was concentrated immediately after irradiation, the deconjugated product 4b decomposed. The crude 'H NMR spectrum (200 MHz, CDC1,) of the washed photoproduct indicated the presence of some unreacted l b along with methanol adduct 3b ((s, 3.15 ppm, 3H, -OCH,; m, 1.5-2.6 ppm, -CH2-; s, 1.21 ppm, 3H, -CH3), lit. NMR in CCl, (17) (s, 3.00 ppm, 3H, -OCH,; m, 1.5-2.5 ppm, 8H, -CH2-; s, 1.15, 3H, -CH3)) and deconjugated product 4b (m, 4.77 ppm, lH, vinyl H, J = 1 Hz; m, 4.69 ppm, lH, vinyl H, J = 1 Hz; d, 3.08 ppm, 2H, allylic H a to the carbonyl, J = 1.14 Hz; m, 1.5-2.5 ppm, -CH2-). Attempts to purify previously uncharacterized 4b were not successful since 4b slowly decomposed on distillation (55"C, 1.5 Torr; 1 Torr = 133.3 Pa), and completely decomposed on chromatography on silica (1:l pentane:EtOAc) or activated basic Brockmann 1 alumina (EtOAc). As 4b decomposed, the ratio between the 'H NMR peaks at 4.69 and 4.77 to the allylic proton a to the carbonyl (3.08) remained constant.

-

-

Can. J. Chem. Vol. 73, 1995

2010

Quenching of the photoaddition of methat201 to 3-methyl-2cyclohexenone by cyclopentene


Cyclopentene was added using a microlitre syringe to solutions containing 0.145 M l b , 10% MeOH, and 10 m M HOAc, to give final cyclopentene concentrations of 0.00, 0.020, 0.035, 0.070, 0.14, 0.28, 0.56, and 1.12 M , respectively. The solutions were purged with argon for 10 min, then irradiated through Pyrex using a 350 nm Rayonet Lamp for 17 h, and were then analyzed directly using capillary GC.

Quenching of the photoaddition of methat201 to 3-methylcyclohexenone by 1-methylnaphthalene Benzene solutions were prepared containing 0.07 M l b , 10% MeOH, 10 m M HOAc, and n-dodecane as an internal standard for G C analyses. Varying amounts of 1-methylnaphthalene were added to give final concentrations of 0.020,0.035,0.070, 0.14, and 0.28 M , respectively. The solutions in Pyrex tubes were purged with argon, irradiated for 2 0 h using a 550 W Hanovia lamp, and then analyzed for products 3b and 4b using capillary GC.

Acknowledgments This work was supported in part by grants CHE-890009 and CHE-9400666 from the National Science Foundation of the United States. W e also thank Dr. J. C. Scaiano at the National Research Council of Canada in Ottawa and Professor Nicholas J. Turro of Columbia University for allowing us to use their nanosecond laser flash equipment, and for their invaluable assistance in these experiments. W e also thank Dr. Monica Barra at the NRC for her help with the laser flash studies. W e finally thank George Lem for helpful suggestions in the course of preparation of this manuscript.

References 1. A. Rudolph and A.C. Weedon. J. Am. Chem. Soc. 111, 8756 (1989). 2. R. Bonneau, J. Joussot-Dubien, L. Salem, and A. Yarwood. J. Am. Chem. Soc. 98,4329 (1976). 3. D.I. Schuster. In The chemistry of enones. Part 2. Edited by S. Patai and Z. Rappoport. John Wiley and Sons, Ltd., Chichester, U.K. 1989. pp. 623-756. 4. E. Dunkelblum, H. Hart, and M. Jeffares. J. Org. Chem. 43, 3409 ( 1 973). 5. H. Hart and E. Dunkelblum. J. Am. Chem. Soc. 100, 541 (1978). 6. D.I. Schuster, P.B. Brown, L.J. Capponi, C.A. Rhodes, J.C. Scaiano, P.C. Tucker, and D. Weir. J. Am. Chem. Soc. 109, 2533 (1987). 7 . Presented at the 201st National Meeting of the American Chemical Society, Atlanta, Ga. April 14-19, 1991. Abstract ORGN 303. 8. D.I. Schuster, D.A. Dunn, G.E. Heibel, P.B. Brown, J.M. Rao, J. Woning, and R. Bonneau. J. Am. Chem. Soc. 113,6245 (199 1). 9. D.I. Schuster, G.E. Heibel, R.A. Caldwell, and W. Tang. Photochem. Photobiol. 52,645 (1990). 10. D.I. Schuster, G.E. Heibel, and J. Woning. Angew. Chern. Int. Ed. Engl. 30, 1345 (1991). 1 1 . R. Bonneau. J. Am. Chern. Soc. 102,3816 (1980). 12. J.C. Scaiano (Editor). Handbook of organic photochemistry. Vol. I. CRC Press, Inc., Boca Raton, Fla. 1989. p. 409. 13. R. Bonneau and P. Fornier de Violet. C.R. Seances Acad. Sci. Ser. 3: 284,63 1 (1977). 14. G.E. Heibel. Ph. D. Dissertation, New York University, 1990. 15. J.C. Scaiano. J. Am. Chern. Soc. 102,7747 (1980). 16. R.D. Small, Jr. and J.C. Scaiano. J. Phys. Chem. 81, 828 (1977). 17. T. Sato, S. Yoshiie, T. Imamura, K. Hasegawa, M. Miyahara, S. Yamarnura, and 0. Ito. Bull. Chem. Soc. Jpn. 50,2714 (1977)