Photoenolization-Induced Oxirane Ring Opening in 2,5-Dimethylbenzoyl Oxiranes To Form Pharmaceutically Promising Indanone Derivatives
Photoenolization-Induced Oxirane Ring Opening in 2,5-Dimethylbenzoyl Oxiranes To Form Pharmaceutically Promising Indanone Derivatives
Photoenolization-Induced Oxirane Ring Opening in 2,5-Dimethylbenzoyl Oxiranes To Form Pharmaceutically Promising Indanone Derivatives
org/joc
[email protected]
Received August 2, 2010
7300 J. Org. Chem. 2010, 75, 7300–7309 Published on Web 10/01/2010 DOI: 10.1021/jo101515a
r 2010 American Chemical Society
Solomek et al.
JOC Article
SCHEME 1. Photochemistry of o-Alkylphenacyl Derivatives CHART 1. Biologically Active Indanone Derivatives
SCHEME 2. Photochemistry of o-Methylbenzoyl Oxiranes: TABLE 2. Quantum Yields (Φ) of 5a-d Formation from 1a-d
The Strategy compound Φa
5a 0.11 ( 0.01
5b 0.15 ( 0.01 (0.16 ( 0.01b)
5c 0.25 ( 0.01
5d 0.12 ( 0.01
a
Degassed solutions (∼5 10-3 M) in acetonitrile or benzene were
irradiated at λ = 313 ( 5 nm (optical bench). Φ was determined using
valerophenone as an actinometer (Φ of acetophenone formation is 0.33
in hexane42). The results are based on at least five independent measure-
ments; the relative standard deviations of the mean are shown. bBenzene
solution.
SCHEME 3. Preparation of the Epoxides 1a-da
SCHEME 5. Dehydration of 5b
SCHEME 8. Formation of 6a
a
The activation Gibbs energies are shown. The fate of 3E is discussed later.
SCHEME 9. Formation of 5 and 7a population analysis favors the mesomeric form 13b and the
formation of 5g. This is in agreement with our experiments
because we found no evidence that 14 was formed by irradia-
tion of 1. In addition, the production of the benzooxepinones 7
(Table 1) suggests that a ring closure of 12 competes with a
hydrogen shift to give 13b. A complete triplet potential energy
profile for the photochemistry of 1g, calculated at the BMK/
6-311þG(3df,2p) level of theory on B3LYP/6-31G* optimized
geometries, is then depicted in Figure 2. Other reaction path-
ways known from the literature, such as R-cleavage to give the
corresponding acyl and epoxy radicals,32,57,58 heterolytic,34 or
electron-transfer mediated oxirane ring-opening59 intermedi-
ates, are apparently not relevant for our systems.
Finally, we had to decide which of the two possible
mechanisms for the formation of 5, photoenolization or
homolytic CR-O bond scission in 3E, are favored. Regen-
eration of the starting ketone from the (E)-photoenol re-
quires a proton transfer through the solvent. When a suitable
deuteron donor is present in the solution, the incorporation
of deuterium into the o-methyl group can be detected by
NMR measurements.3 The m-methyl group signal in 1H
NMR spectra of 1a-d interferes with those of the products.
Therefore, epoxide 1h (Scheme 10), having only a single
methyl group, was synthesized and irradiated in a degassed
D2O/acetone-d6 mixture (1:20) at >290 nm in an NMR
a
Structures in blue and red denote isolated and spectroscopically cuvette. The ratio of the integrals, corresponding to the
observed species, respectively; a putative (not observed) product is o-methyl and CH (of the oxiranyl ring) signals, decreased
shown in green. The operative pathway (vide infra) of the indanones by ∼25%. We were unable to integrate the peaks properly at
formation is depicted by bold arrows. The activation Gibbs energies are
shown. longer irradiation times because of an overlap of the peaks
with those of the products; nevertheless, it was sufficient
DFT calculations. The primary 1,4-biradical 3E, which was evidence that photoenols are indeed formed in the solution
formed from the triplet excited compound by a hydrogen upon irradiation.
shift (Scheme 8), can undergo either CR-O homolytic clea- In addition, the mechanism of the photoreactions of 1b
vage to 12 or intersystem cross to the ground state to produce was investigated by laser flash photolysis (Table 3; see also
two photoenols (in red). The rate of 3E decay for phenacyl Supporting Information, S41-44). The transient absorption
derivatives is known to depend on the presence of oxygen or spectrum of an aerated solution of 1b in acetonitrile obtained
other triplet quenchers (τ ∼16-300 ns).7,17,18,20 Compound 1 μs after the laser flash is shown in Figure 3. The absorption
12 possesses a reactive primary alkoxy radical center; a spectrum reproduces nicely that measured for both ground
simple H-atom shift between the two oxygen atoms with a state photoenols in previous studies.7,17-20 We were unable
calculated barrier of ΔGq =7.3 kcal mol-1 results in a more to see any significant absorption at ∼340-350 nm characteristic
stable radical (see Figure 2), whose two mesomeric forms, of 3E even at short time delays (15 ns). Such a remarkable short
13a and 13b, are shown in the scheme. This pathway can be lifetime of a 1,4-biradical is consistent with that reported by
considered as a spin-center shift mechanism, recently proposed Caldwell and co-workers for R-benzyloxyacetophenone60 or
by Wessig and co-workers.56 Subsequent radical recombina-
tion and ISC may lead to 5g or 14g (in green). The spin (57) Dunston, J. M.; Yates, P. Tetrahedron Lett. 1964, 505.
(58) Padwa, A. Tetrahedron Lett. 1964, 813.
(59) Hasegawa, E.; Ishiyama, K.; Horaguchi, T.; Shimizu, T. J. Org.
(56) Wessig, P.; Muehling, O. Eur. J. Org. Chem. 2007, 2219. Chem. 1991, 56, 1631.
FIGURE 2. Triplet potential energy surface profile of 1g photoreactions. The Gibbs energies (in kcal mol-1) were calculated at the BMK/
6-311þG(3df,2p) level of theory on B3LYP/6-31G* optimized geometries.
the 1,5-biradical in Scheme 7 found by Park et al.44 A similar SCHEME 10. Deuterium Incorporation in 1h
spectrum has been measured for 1a. The kinetic trace at λ=390
nm contained three exponential components: a very fast singly
exponential rise and a biexponential decay. The lifetime of the
former, short-lived component was sensitive to the presence of
oxygen. Its lifetime of 23 ns in degassed solutions dropped to
<10 ns in air-saturated solutions, which we could not resolve
because of limitations of our apparatus. However, the antici-
TABLE 3. Lifetimes τ of the Transients of 1ba
pated decrease of the rate constant for this component occurred
in less polar cyclohexane: lifetimes of 47 and 12.5 ns were solvent τ (3E) τ (Z)-photoenol τ (E)-photoenol
observed in the degassed and air-saturated solutions, respec- acetonitrile 23 (<10)b ns 8.8 μs 5 ms
tively. This transient intermediate was assigned to the triplet enol cyclohexane 47 (12.5)b ns 2.8 μs >40 msc
(3E) decay. The lifetimes of the other two components were not
a
Degassed solutions (∼1 10-4 M) in acetonitrile or cyclohexane
sensitive to the presence of oxygen. A shorter-lived intermediate measured at λ=390 nm after 266-nm laser flash. bThe lifetimes measured
for aerated solutions are in the parentheses. cThe lower limit estimated
with a lifetime of 8.8 and 2.8 μs in acetonitrile and cyclohexane from the exponential fit of the kinetic trace.
solution, respectively, was attributed to the ground state (Z)-
photoenol, which is expected to undergo fast intramolecular relevant intermediates shown in Figure 2. The results
reketonization,7,17 while the long-lived component was attrib- are summarized in the Supporting Information (S48). Only
uted to the (E)-photoenol. Its lifetime of 5.0 ms was observed in the photoenols and conformers of the triplet enol, 3E,
acetonitrile, whereas that in cyclohexane (>40 ms) could not showed a strong absorption at ∼390 nm (f ∼0.15-0.19;
be fit properly because of instability of our white light source at Figure 3) and ∼355-363 nm (f ∼0.1), respectively. The
such a long time scale. Our results resemble well those reported transitions for other relevant species are approximately
for the photoenolization in dimethylphenacyl chromophore 10 times weaker in this region. Regarding the absence of
substituted by different leaving groups in the R-position.7,17-20 significant spectral changes within the time of the experi-
Our assignments were also rationalized by performing ment or in the presence of oxygen, we conclude that both
TD-DFT calculations of the electronic transitions for all kinetic traces assigned to the decay of the 390 nm signal
belong to the photoenol intermediates, although the pre-
(60) Caldwell, R. A.; Majima, T.; Pac, C. J. Am. Chem. Soc. 1982, 104, sence of some other relevant species cannot be unambigu-
629. ously ruled out.
J. Org. Chem. Vol. 75, No. 21, 2010 7305
JOC Article Solomek et al.
(m, 3H, J = 7.0 Hz), 7.36 (d, 2H, J = 7.0 Hz), 7.58 (d, 1H, J =
1.5 Hz). 13C NMR (75.5 MHz, CDCl3): δ (ppm) 20.8, 21.0, 48.4,
126.9, 128.6, 129.2, 129.6, 131.9, 132.1, 134.7, 135.1, 135.3, 137.8,
201.6. MS (EI, 70 eV): m/z = 224, 133, 105, 91, 77, 65, 51, 39.
1-(2,5-Dimethylphenyl)propan-1-one (3d). Prepared from
p-xylene and propionyl chloride (17.8 mL, 224 mmol). Yield:
96%; colorless oil. 1H NMR (300 MHz, CDCl3): δ (ppm) 1.08 (t,
3H, J = 7.3 Hz), 2.22 (s, 3H), 2.34 (s, 3H), 2.75 (q, 2H, J = 7.3 Hz),
6.96 (d, 1H, J = 7.8 Hz), 7.01 (d, 1H, J = 7.8 Hz), 7.31 (s, 1H). 13C
NMR (75.5 MHz, CDCl3): δ (ppm) 8.0, 20.3, 20.4, 34.1, 128.5,
131.3, 131.4, 134.2, 134.7, 137.7, 204.1. MS (EI, 70 eV): m/z = 162,
133, 105, 91, 77, 65, 51, 41.
1-(4,5-Dimethoxy-2-methylphenyl)ethanone (3e). Prepared
from 1,2-dimethoxy-4-methylbenzene and acetyl chloride (15.9
mL, 224 mmol). Yield: 84%; white crystals, mp 70.6-72.3 °C.
1
H NMR (300 MHz, CDCl3): δ (ppm) 2.44 (s, 3H), 2.47 (s, 3H),
3.82 (s, 3H), 3.83 (s, 3H), 6.62 (s, 1H), 7.18 (s, 1H). 13C NMR (75.5
MHz, CDCl3): δ (ppm) 21.8, 29.3, 55.8, 56.2, 113.4, 114.6, 129.4,
133.4, 146.3, 151.6, 199.4. MS (EI, 70 eV): m/z = 194, 179, 151, 136,
FIGURE 3. Transient absorption spectrum of 1b. The spectrum 121, 108, 93, 77, 65, 43, 39.
was taken 1 μs after the laser flash for an acetonitrile solution of 1b. (E)-1-(2,5-Dimethylphenyl)-3-phenylprop-2-en-1-one (4a). A
The red line shows the TD-DFT calculated transition of the E-enol solution of 3a (2.00 g, 13.5 mmol) in ethanol (15 mL) was added
derived from 1g (the intensity is arbitrarily scaled; see Supporting dropwise to aq KOH (40 mL, 10%) at 0 °C. Freshly distilled
Information, S48). benzaldehyde (1.43 g, 13.5 mmol) was then added dropwise to
the mixture in the period of 10 min. The reaction mixture was
In conclusion, four 2,5-dimethylbenzoyl oxiranes were stirred for 2 days at 20 °C and extracted with CH2Cl2 (3
synthesized in high overall chemical yields, and their photo- 50 mL). The organic extracts were washed with brine (50 mL),
chemistry was explored. Irradiation of these epoxy ketones dried with MgSO4, and solvents were evaporated at reduced
resulted in indanone derivatives as major products, which pressure to give 4a. Yield: 94%; yellowish oil. 1H NMR (300
are structurally similar to some known pharmaceutically MHz, CDCl3): δ (ppm) 2.37 (s, 3H), 2.41 (s, 3H), 7.14 (d, 1H,
J = 16.1 Hz), 7.18 (m, 2H), 7.30 (s, 1H), 7.41 (m, 3H), 7.48 (d,
important compounds. In addition, a simple route to a
1H, 1H, J = 16.1 Hz), 7.57 (m, 2H). 13C NMR (75.5 MHz,
2-ethylidene-2,3-dihydro-1H-inden-1-one derivative from CDCl3): δ (ppm) 19.7, 20.9, 126.8, 128.4, 128.5, 128.8, 128.9,
the synthesized indanones was described. The laser flash 130.5, 131.2, 133.7, 134.7, 135.0, 139.1, 145.6, 196.6. MS (EI, 70
photolysis experiments and quantum-chemical calculations eV): m/z = 236, 145, 131, 103, 77.
provided solid evidence that the indanone photoproducts are (E)-1-(2,5-Dimethylphenyl)but-2-en-1-one (4b). The compound
formed via a photoenolization mechanism. A total synthesis was prepared from p-xylene (19.6 g, 185 mmol) and crotonyl
of several biologically active compounds, which is based on chloride (21.2 g, 224 mmol) according to the procedure described
the photochemical approach introduced here, is currently for 3a-e. Yield: 94%; colorless oil. 1H NMR (300 MHz, CDCl3):
under investigation in our laboratory. δ (ppm) 1.95 (dd, 3H, J1 = 6.8 Hz, J2 = 1.4 Hz), 2.33 (s, 3H), 2.34
(s, 3H), 6.49 (dd, 1H, J1 = 15.6 Hz, J2 = 1.4 Hz), 6.73 (qd, 1H,
Experimental Section J1 = 15.6 Hz, J2 = 6.8 Hz), 7.10-7.16 (m, 2H), 7.17 (s, 1H). 13C
NMR (75.5 MHz, CDCl3): δ (ppm) 17.3, 18.7, 19.8, 127.6, 130.1,
General Procedure for the Synthesis of Acylphenones 3a-e. 130.2, 131.3, 132.6, 133.8, 138.1, 144.9, 194.9. MS (EI, 70 eV):
The corresponding acyl chloride (224 mmol) was added drop- m/z = 174, 159, 141, 131, 115, 15, 91, 79, 69, 41.
wise to a mixture of aluminum chloride (27.2 g, 204 mmol) and 1-(2,5-Dimethylphenyl)-2-phenylprop-2-en-1-one (4c). Piperi-
either p-xylene (19.6 g, 185 mmol) or 1,2-dimethoxy-4-methyl- dine (0.13 mL, 1.3 mmol), acetic acid (0.12 mL, 2.1 mmol),
benzene (28.2 g, 185 mmol) in CS2 (100 mL) under nitrogen and aq formaldehyde (37% solution, 4 mL, 50 mmol) were
atmosphere at 0-3 °C in a period of 1.5 h. The reaction mixture successively added to a magnetically stirred solution of 3c
was stirred for 1 h at 0 °C, warmed to 20 °C, stirred for (2.92 g, 13.0 mmol) in methanol (50 mL), and the resulting
additional 2 h, and then poured on ice (400 g). After the ice mixture was refluxed for 4 h and concentrated under reduced
melted, the aqueous layer was extracted with CH2Cl2 (3 100 mL). pressure. Water (50 mL) was added to the residue, and the
The combined organic extracts were washed with brine (100 mL) mixture was extracted with CH2Cl2 (3 30 mL). Combined
and dried with MgSO4. The solvent was evaporated under reduced organic extracts were washed with water (2 50 mL) and brine
pressure to give the crude title product, which was purified by (30 mL) and dried with MgSO4, and solvent was evaporated
vacuum distillation. under reduced pressure. The crude product was purified by flash
1-(2,5-Dimethylphenyl)ethanone (3a). Prepared from p-xylene column chromatography (silica, petroleum ether/ethyl acetate,
and acetyl chloride (15.9 mL, 224 mmol). Yield: 95%; colorless 8:1) to give the title product. Yield: 97%; white crystals, mp
oil. 1H NMR (300 MHz, CDCl3): δ (ppm) 2.35 (s, 3H), 2.47 178.3-181.0 °C. 1H NMR (300 MHz, CDCl3): δ (ppm) 2.33 (s,
(s, 3H), 2.55 (s, 3H), 7.10 (d, 1H, J = 7.8 Hz), 7.16 (d, 1H, J =7.8 3H), 2.42 (s, 3H), 5.75 (s, 1H), 6.17 (s, 1H), 7.14 (d, 1H, J = 7.8
Hz), 7.47 (s, 1H). 13C NMR (75.5 MHz, CDCl3): δ (ppm) 20.7, Hz), 7.19 (dd, 1H, J1 = 7.8 Hz, J2 = 1.3 Hz), 7.29 (d, 1H, J = 1.3
20.8, 29.3, 129.7, 131.8, 132.0, 135.0, 135.6 137.6, 201.6. MS (EI, Hz), 7.38 (m, 3H), 7.47 (dd, 1H, J1 = 7.9 Hz, J2 = 1.7 Hz). 13C
70 eV): m/z = 148, 133, 105, 79, 77, 51, 43. NMR (75.5 MHz, CDCl3): δ (ppm) 19.9, 21.0, 120.7, 128.1,
1-(2,5-Dimethylphenyl)-2-phenylethanone (3c). Prepared from 128.4, 128.5, 129.9, 131.2, 131.6, 134.4, 135.0, 137.0, 138.7,
p-xylene and phenylacetyl chloride (27.1 mL, 224 mmol). Yield: 149.8, 199.9. MS (EI, 70 eV): m/z = 236, 159, 145, 133, 105,
95%; colorless crystals, mp 30.2-31.6 °C. 1H NMR (300 MHz, 77, 63, 51.
CDCl3): δ (ppm) 2.42 (s, 3H), 2.46 (s, 3H), 4.25 (s, 2H), 7.16 (d, 1-(2,5-Dimethylphenyl)-2-methylprop-2-en-1-one (4d). Acetic
1H, J = 7.8 Hz,), 7.22 (dd, 1H, J1 = 7.8 Hz, J2 = 1.5 Hz), 7.31 acid (10 mL) was added to a magnetically stirred suspension of
J1 = 9.0 Hz, J2 = 8.0 Hz, J3 = 4.5 Hz) 2.75 (dd, 1H, J1 = 17.0 Hz, 1H, J1 = 13.5 Hz, J2 = 5.8 Hz), 3.16 (dd, 1H, J1 = 13.5 Hz, J2 =
J2 = 4.5 Hz), 3.24 (dd, 1H, J1 = 17.0 Hz, J2 = 8.0 Hz), 3.98 (dq, 5.5 Hz), 4.24 (tq, 1H, J1 = 6.3 Hz, J2 = 6.1 Hz,), 4.84 (d, 1H,
1H, J1 = 9.0 Hz, J2 = 6.2 Hz), 4.36 (broad, 1H), 7.36 (d, 1H, J = J = 15.3 Hz), 4.94 (d, 1H, J = 15.3 Hz), 7.12 (d, 1H, J = 7.7 Hz),
7.9 Hz), 7.44 (d, 1H, J = 7.9 Hz), 7.56 (s, 1H). 13C NMR (75.5 7.26 (d, 1H, J = 7.7 Hz), 7.66 (s, 1H). 13C NMR (75.5 MHz,
MHz, CDCl3): δ (ppm) 21.1, 21.6, 29.5, 53.8, 69.2, 124.0, 126.2, CDCl3): δ (ppm) 21.1, 21.3, 50.0, 69.4, 71.4, 127.8, 129.2, 132.9,
136.7, 136.9, 137.8, 151.3, 210.0. Diastereomer B: 1H NMR (300 137.5, 137.9, 139.6, 200.5. MS (EI, 30 eV): m/z = 190, 149, 119,
MHz, CDCl3): δ (ppm) 1.28 (d, 3H, J = 6.4 Hz), 2.40 (s, 3H), 2.80 103, 91, 77, 65, 51, 39. HRMS (EIþ): calcd for C12H15O2 [M þ
(ddd, 1H, J1=8.0 Hz, J2=4.5 Hz, J3=3.6 Hz) 3.07 (dd, 1H, J1 = Hþ] 191.1072, found 191.1062.
17.0 Hz, J2 = 4.5 Hz), 3.17 (dd, 1H, J1 = 17.0 Hz, J2 = 8.0 Hz), 4,7-Dimethyl-3,4-dihydrobenzo[c]oxepin-5(1H)-one (7d). Ob-
4.46 (dq, 1H, J1 = 6.4 Hz, J2 = 3.6 Hz), 7.36 (d, 1H, J = 7.9 Hz), tained from 1d.Yield: 15%; colorless oil. 1H NMR (300 MHz,
7.44 (d, 1H, J = 7.9 Hz), 7.56 (s, 1H). 13C NMR (75.5 MHz, CDCl3): δ (ppm) 1.14 (d, 3H, J=6.7 Hz), 2.37 (s, 3H), 3.26-3.38
CDCl3): δ (ppm) 20.8, 21.1, 27.3, 54.3, 67.4, 123.7, 126.4, 136.3, (m, 1H), 3.65 (dd, 1H, J1 = 10.7 Hz, J2 = 10.7 Hz), 4.00 (dd, 1H,
136.6, 137.4, 152.0, 208.4. MS (EI, 70 eV, m/z): 190, 175, 133, 129, J1 = 10.7 Hz, J2 = 6.8 Hz), 4.83 (d, 1H, J = 15.2 Hz), 4.94 (d,
105, 91, 77, 69, 43. UV-vis (acetonitrile): ε313 = 1340 dm3 mol-1 1H, J = 15.2 Hz), 7.10 (d, 1H, J = 7.7 Hz), 7.24 (d, 1H, J = 7.7
cm-1, ε254 = 8030 dm3 mol-1 cm-1. HRMS (EIþ): calcd for Hz), 7.58 (s, 1H). 13C NMR (75.5 MHz, CDCl3): δ (ppm) 11.6,
C12H15O2 [M þ Hþ] 191.1072, found 191.1056. 21.1, 46.6, 71.9, 72.8, 127.4, 129.3, 132.3, 137.7, 138.5, 138.7,
2-(Hydroxymethyl)-6-methyl-2-phenyl-2,3-dihydro-1H-inden- 204.1. MS (EI, 30 eV, m/z): 190, 149, 119, 103, 91, 77, 65, 51, 39.
1-one (5c). Obtained from 1c. Yield: 82% (at a 41% conversion HRMS (EIþ): calcd for C12H15O2 [M þ Hþ] 191.1072, found
of 1c); colorless oil. 1H NMR (300 MHz, CDCl3): δ (ppm) 2.43 191.1053.
(s, 3H), 3.97 (d, 1H, J = 17.0 Hz), 4.13 (d, 1H, J = 17.0 Hz), 4.15 (E)-2-Ethylidene-6-methyl-2,3-dihydro-1H-inden-1-one (8).
(d, 1H, J = 10.9 Hz), 4.27 (d, 1H, J = 10.9 Hz), 7.27-7.46(m, KOH (0.10 g, 1.7 mmol) was added to a stirred solution of 2b
7H), 7.63 (s, 1H). 13C NMR (75.5 MHz, CDCl3): δ (ppm) 21.2, (50 mg, 0.26 mmol) in methanol (30 mL). The reaction mixture
38.8, 58.6, 68.0, 124.5, 126.2, 127.0, 127.3, 128.9, 136.6, 136.9, was heated to 40 °C in the period of 20 min and then extracted
137.9, 140.4, 150.7, 208.3. MS (EI, 70 eV): m/z = 238, 119, 91, with CH2Cl2 (3 10 mL). The combined organic layers
77, 65. UV-vis (acetonitrile): ε313 = 2270 dm3 mol-1 cm-1, ε254 were washed with brine (30 mL) and dried with MgSO4, and
= 10 140 dm3 mol-1 cm-1. HRMS (EIþ): calcd for C17H17O2 the solvent was evaporated under reduced pressure. The
[M þ Hþ] 253.1229, found 253.1223. crude product was purified by flash chromatography (silica,
2-(Hydroxymethyl)-2,6-dimethyl-2,3-dihydro-1H-inden-1-one petroleum ether/ethyl acetate, 10:1) to give 8. Yield: 97%; color-
(5d). Obtained from 1d. Yield: 66%; colorless oil. 1H NMR (300 less oil. GC showed a mixture of the (E)- and (Z)-isomers in a
MHz, CDCl3): δ (ppm) 1.24 (s, 3H), 2.39 (s, 3H), 2.84 (d, 1H, 94:6 ratio. 1H NMR (300 MHz, CDCl3): δ (ppm) 1.97 (dt, 3H,
J = 17.0 Hz), 3.18 (d, 1H, J = 17.0 Hz), 3.62 (d, 1H, J = 10.7 J1 = 7.2 Hz, J2 = 2.1 Hz), 2.43 (s, 3H), 3.62 (broad s, 2H), 6.94
Hz), 3.81 (d, 1H, J = 10.7 Hz), 7.34 (d, 1H, J = 7.8 Hz), 7.43 (d, (tq, 1H, J1 = 7.1 Hz, J2 = 2.1 Hz), 7.38-7.43 (m, 2H), 7.67
1H, J = 7.8 Hz), 7.53 (s, 1H). 13C NMR (75.5 MHz, CDCl3): δ (s, 1H). 13C NMR (75.5 MHz, CDCl3): δ (ppm) 15.54, 21.38,
(ppm) 14.2, 20.7, 37.7, 51.1, 67.9, 124.1, 126.3, 136.0, 136.5, 29.70, 124.6, 126.2, 132.9, 135.8, 137.6, 138.3, 139.3, 146.8, 193.5.
137.4, 150.6, 211.1. MS (EI, 70 eV): m/z = 190, 172, 159, 145, MS (EI, 70 eV): m/z = 172, 157, 143, 129, 115, 102, 77, 63, 51.
132, 129, 115, 105, 91, 77, 51, 32. UV-vis (acetonitrile): ε313 = HRMS (EIþ): calcd for C12H13O [M þ Hþ] 173.0966, found
880 dm3 mol-1 cm-1, ε254 = 6560 dm3 mol-1 cm-1. HRMS 173.0962.
(EIþ): calcd for C12H15O2 [M þ Hþ] 191.1072, found 191.1067. Irradiation Experiment in an NMR Cuvette. D2O (25 μL) was
1-(2,5-Dimethylphenyl)-3-hydroxybut-2-en-1-one (6b). Ob- added to a solution of 1h (20 mg) in acetone-d6 (500 μL) in an
tained from 1b. Yield: 7%; colorless oil. 1H NMR (300 MHz, NMR cuvette. The resulting solution was purged with argon for
CDCl3): δ (ppm) 2.17 (s, 3H), 2.34 (s, 3H), 2.45 (s, 3H), 5.84 (s, 10 min. 1H NMR was measured in 15-min intervals of irradia-
1H), 7.11 (d, 1H, J = 7.8 Hz), 7.15 (d, 1H, J = 7.8 Hz), 7.28 (s, tion by a 125-W medium pressure Hg UV lamp through a Pyrex
1H), 15.96 (broad, 1H). 13C NMR (75.5 MHz, CDCl3): δ (ppm) filter (λ < 290 nm).
20.3, 21.0, 25.8, 101.0, 129.0, 131.5, 131.6, 134.1, 135.5, 136.0, Quantum Yield Measurements. The quantum yield measure-
188.4, 193.2. MS (EI, 70 eV): m/z = 190, 172, 132, 129, 115, 104, ments were performed on an optical bench consisting of high
77, 43, 32. HRMS (EIþ): calcd for C12H15O2 [M þ Hþ] pressure 350- or 450-W UV lamps, a 1/8 m monochromator with
191.1072, found 191.1059. 200-1600 nm grating, set to 313 nm, and a 1.0-cm matched
1-(2,5-Dimethylphenyl)-3-hydroxy-2-methylprop-2-en-1-one (6d). quartz cell containing a solution degassed by purging with argon
Obtained from 1d. Yield: 7%; colorless oil. 1H NMR (300 MHz, for 10 min. The sample temperature was maintained using a
CDCl3): δ (ppm) 1.68 (s, 3H), 2.26 (s, 3H), 2.33 (s, 3H), 6.99 (s, 1H), Peltier thermo block set to 20 °C. The light intensity was
7.13 (s, 2H), 8.54 (d, 1H, J = 4.8 Hz), 14.94 (d, 1H, J = 4.8 Hz). 13C monitored by a Si photodiode detector (UV enhanced) with a
NMR (75.5 MHz, CDCl3): δ (ppm) 13.2, 18.8, 21.1, 127.3, 130.5, multifunction optical power meter. The concentration of all
130.7, 135.4, 136.3, 184.4, 208.1. MS (EI, 70 eV): m/z = 190, 172, sample solutions was in the interval of 1 10-3 and 5 10-4 M,
159, 145, 132, 129, 115, 104, 91, 77, 32. HRMS (EIþ): calcd for containing hexadecane (10-3 M; GC) or acetone (10-2 M;
C12H15O2 [M þ Hþ] 191.1072, found 191.1067. HPLC) as internal standards. Valerophenone was used as an
1-(4,5-Dimethoxy-2-methylphenyl)-3-hydroxy-3-phenylprop-2- actinometer in all cases.42 The reaction conversions were always
en-1-one (6e). Obtained from 1e. Yield: 70%; colorless oil. 1H kept below 10% to avoid the interference of photoproducts. The
NMR (300 MHz, CDCl3): δ (ppm) 2.54 (s, 3H), 3.91 (s, 3H), 3.92 relative standard deviations of the mean for multiple (at least
(s, 3H), 6.51 (s, 1H), 6.74 (s, 1H), 7.15 (s,1H), 7.44-7.54 (m, 3H), five) samples were found below 10% in all analyses.
7.95 (m, 2H). 13C NMR (75.5 MHz, CDCl3): δ (ppm) 21.0, 56.2, Steady-State Quenching Experiments. Stern-Volmer anal-
56.4, 97.2, 112.0, 114.5, 127.2, 128.9, 131.3, 132.5, 135.7, 147.1, ysis was performed with naphthalene as a triplet quencher in a
151.3, 184.3, 190.0 (see an HMBC spectrum in the Supporting merry-go-round apparatus (see also the Supporting Informa-
Information, S30). MS (EI, 70 eV): m/z = 298, 283, 267, 221, 192, tion, S41). Samples in Pyrex tubes (13 120 mm) were degassed
179, 150, 136, 121, 105, 91, 77, 69, 51. HRMS (EIþ): calcd for by purging with argon for 10 min. The concentration of all
C18H19O4 [M þ Hþ] 299.1283, found 299.1278. sample solutions was approximately 5 10-3 M, and they
3,7-Dimethyl-3,4-dihydrobenzo[c]oxepin-5(1H)-one (7b). Ob- contained hexadecane (10-3 M) as an internal standard for
tained from 1b. Yield: 12%; colorless oil. 1H NMR (300 MHz, GC measurements. Three degassed tubes for a given naph-
CDCl3): δ (ppm) 1.33 (d, 3H, J = 6.3 Hz), 2.38 (s, 3H), 2.85 (dd, thalene concentration were irradiated using a medium pressure