Download PDF
Synthesis 2012; 44(16): 2560-2566
DOI: 10.1055/s-0031-1289764
paper
© Georg Thieme Verlag Stuttgart · New York

Microphotochemistry Using 5-mm Light-Emitting Diodes: Energy-Efficient Photooxidations

John M. Carney
Department of Chemistry, Creighton University, 2500 California Plaza, Omaha, NE 68178-0323, USA, Fax: +1(402)2805737   Email: mhulce@creighton.edu
,
Reagan J. Hammer
Department of Chemistry, Creighton University, 2500 California Plaza, Omaha, NE 68178-0323, USA, Fax: +1(402)2805737   Email: mhulce@creighton.edu
,
Martin Hulce*
Department of Chemistry, Creighton University, 2500 California Plaza, Omaha, NE 68178-0323, USA, Fax: +1(402)2805737   Email: mhulce@creighton.edu
,
Chad M. Lomas
Department of Chemistry, Creighton University, 2500 California Plaza, Omaha, NE 68178-0323, USA, Fax: +1(402)2805737   Email: mhulce@creighton.edu
,
Dayna Miyashiro
Department of Chemistry, Creighton University, 2500 California Plaza, Omaha, NE 68178-0323, USA, Fax: +1(402)2805737   Email: mhulce@creighton.edu
› Author Affiliations
Further Information

Publication History

Received: 29 February 2012

Accepted: 07 March 2012

Publication Date:
24 April 2012 (online)

 
 PDF Download Permissions and Reprints All articles of this category


Abstract

Commercial, inexpensive 5-mm milliwatt light-emitting diodes are effective sources for batch microphotochemical oxidations. Using limited quantities of singlet oxygen, these oxidations are atom economical and therefore useful for labeling experiments with rare isotopes.


#

Key words

photooxidation - arenes - peroxides - photoreactors - light-emitting diodes

Oxidations are among the most ubiquitous of chemical transformations: thousands of methods, some of great historical significance in organic synthesis, exist.[ 1 ] Many of these employ well-known, reliable, readily available reagents such as the Jones reagent,[ 2 ] pyridinium chlorochromate,[ 3 ] potassium permanganate,[ 4 ] hypochlorites,[ 5 ] hydrogen peroxide,[ 6 ] and osmium tetraoxide.[ 7 ] Molecular oxygen is among the simplest of oxidizing reagents. It is responsible for inadvertent aerial oxidation of aldehydes[ 8 ] and for autooxidative processes[ 9 ] some of which, like the cumene process[ 10 ] to prepare phenol, are industrially significant.

In its ground state, molecular oxygen exists as a triplet. It has two low-lying excited states, the first excited singlet delta (1Δg) and second excited singlet sigma (1Σg) states, responsible for unique oxidative chemical properties.[11] [12] Most common among these are three characteristic hydrocarbon addition reactions: [2+4] cycloadditions[ 13 ] of O2[1Δg] (hereafter 1O2) to produce endoperoxides, [2+2] cycloadditions[11c] [14] to produce 1,2-dioxetanes, and Schenck ene reactions[ 15 ] to produce allylic hydroperoxides (Scheme [1]).

Singlet oxygen may be produced thermally or photochemically. Well-known thermal methods include production from hydrogen peroxide and hypochlorite[ 16 ] or molybdate,[ 17 ] calcium peroxide diperoxohydrate,[ 18 ] ozone and its derivatives,[ 19 ] and hypervalent iodine compounds,[ 20 ] and by decomposition of dioxetanes[ 21 ] and arene endoperoxides.[13a] [22] Recently, its formation from the fragmentation of certain 1,1-dihydroperoxides has been reported.[ 23 ] Direct photochemical generation[ 24 ] of singlet oxygen from triplet oxygen is inefficient, the transition being formally forbidden. Consequently, efficient preparation requires the use of sensitizer dyes,[ 25 ] for example, methylene blue, Rose Bengal, or tetraphenylporphyrin.

Zoom Image
Scheme 1 a) [4+2] Cycloaddition of 1O2 with a conjugated diene; b) [2+2] cycloaddition of 1O2 with an alkene; c) ene reaction of the allylic positions of an alkene with 1O2

A variety of biochemical roles have been identified for this reactive oxygen species (ROS).[ 26 ] It is generated in many biological systems, including peroxidase and lipogenase oxidations,[ 27 ] the decomposition of lipid peroxides,[ 28 ] photosynthesis,[ 29 ] plant defense,[ 30 ] and spore germination induction.[ 31 ] Exceptionally toxic, it is the primary agent of photodynamic therapy.[ 32 ]

With an interest in studying the thermal reversion[13a] [22] of photochemically generated endoperoxides using 17O NMR spectroscopy,[ 33 ] we sought a simple, high efficiency, small-scale photochemical apparatus for the [2+4] cycloaddition of various conjugated alkadienes and arenes[ 34 ] using near stoichiometric amounts of isotopically labeled 1O2.[27b] [28b] [35] Standard photoreactors using immersion wells are too large for such an application; photoreactions run in them require careful attention to cooling of the source and monitoring evaporation of the solvent.[ 36 ] Regardless, photooxidations typically bubble excess oxygen through the reaction medium. This can lead to significant solvent loss[ 37 ] even when the oxygen stream is solvent-presaturated.

Dye-sensitized production of 1O2 with methylene blue uses visible light, so use of an external source has been demonstrated to be effective. For example, using a 5-mL round-bottom flask equipped with a magnetic stirrer bar, an ice water cooled condenser, and an oxygen inlet capillary tube running down the condenser into the flask, ca. 40 mg (240 μmol) of 1,4-dimethylnaphthalene in a 0.3 mol% solution of methylene blue in dichloromethane (4 mL) could be completely photooxidized over 72 hours in a 3 °C cold room when a 500-W incandescent lamp, IR-filtered by passing its output through a 10-cm recirculating refrigerated water wall, was projected onto the flask (Scheme [2]). The inconvenience of adding dichloromethane (2 mL) to the flask every 4 hours to compensate for evaporation and the nonstoichiometric use of oxygen make this approach to endoperoxide synthesis unappealing. Thus, attention was turned to the use of static, pressurized atmospheres of oxygen and cooler sources.

Zoom Image
Scheme 2 WX = tungsten halogen lamp; MB = methylene blue

Using 9,10-dimethylanthracene, which undergoes photooxidation much more rapidly than 1,4-dimethylnaphthalene, the 5-mL flask and condenser were replaced with a 10-mL hypo-vial equipped with a stirrer bar. This hypo-vial was charged with ca. 45 μmol of arene and was crimped shut using a silicone septum stopper and aluminum seal with a tear-away opening. After the tear-away opening was removed, a 14/20 septum stopper was placed on top of the sealed hypo-vial to insure a gas-tight seal. Then, the hypo-vial was wrapped in aluminum foil to exclude light, and was charged with a 0.3 mol% solution of methylene blue in dichloromethane (5 mL). After two –78 °C freeze–aspirator pump–N2 purge–thaw cycles, the hypo-vial was evacuated at –78 °C, then injected with 5–15 mL of oxygen, corresponding to a 2.2–10-fold molar excess. After the aluminum foil was removed, the contents of the hypo-vial were stirred rapidly while irradiated with the IR-filtered 500-W source as before. After 20–25 minutes, analysis by silica gel TLC using petroleum ether as eluent indicated no starting arene remained (Rf  = 0.63); the corresponding endoperoxide was the only observable product. No endoperoxide formation was observed if the hypo-vial remained wrapped in foil.

Having established that static, pressurized atmospheres of oxygen were sufficient for small-scale endoperoxidation, it was clear that use of a bulky source requiring aggressive IR filtering was not ideal. These reasons supported the assessment: failure of the IR filter would lead to dangerous heating of a small, pressurized, closed system; safety aside, any heating of the reaction vessel by the source would be counterproductive for the synthesis of thermally labile endoperoxides, and the total energy input required for this type of source significantly offsets the atom economy gained by the use of near stoichiometric amounts of oxygen. Knowing that even relatively cool light sources produce thermal emissions sufficient to make them unsatisfactory for stand-alone use to prepare some endoperoxides,[ 38 ] our search for an effective alternative source concentrated on highly energy-efficient, milliwatt light-emitting diodes (LEDs) with narrow viewing angles.

Multiwatt LED sources have found application in photochemistry: a commercial photoreactor has appeared,[ 39 ] and multiwatt LED arrays have been used in photodynamic therapy[32b] [40] and various microfluidic and batch photoreactors.[ 41 ] Use of standard 5-mm milliwatt LEDs is more limited;[ 42 ] however, their inherently narrow emissions (FWHM ≤30 nm), low thermal output, narrow viewing angles (30° or less), availability of emission frequencies from the UV into the near-IR, and low cost make them very attractive for small-scale investigations.

Sources therefore were constructed from three 5-mm, narrow (8–30°) viewing angle LEDs of the same peak emission wavelength, bundled into a 1.5 × 11 cm piece of glass tubing. Power was provided with a 12-V DC power supply or a 9-V DC battery. The total power consumption of the sources ranged from 240 to 375 mW. Using the previously described reaction conditions for the photooxidation of 9,10-dimethylanthracene in a hypo-vial, endoperoxide formation as a function of LED source peak emission wavelength was investigated by monitoring the relative concentrations of the starting arene and product endoperoxide by 1H NMR spectroscopy (methyl resonances at δ 3.15 and 2.19 ppm, respectively). Two sources, those with peak emission wavelengths at 627 and 405 nm, provided ≥98% conversion into endoperoxide (Figure [1]). With the former source, essentially all of the incident light is absorbed by the methylene blue photosensitizer, allowing for very efficient 1O2 production. With the latter source, 9,10-dimethylanthracene rather than methylene blue acts as photosensitizer[ 43 ] (Figure [2]); thus, endoperoxide was produced with equal efficiency when the reaction was performed without methylene blue.

Zoom Image
Figure 1 Methylene blue sensitized photooxidation of 9,10-dimethylanthracene as a function of LED source wavelength
Zoom Image
Figure 2 Solid lines, LED emission spectra; dashed lines, absorbance spectra

The superior energy efficiency of these LED sources was notable: compared with a 250-watt incandescent source, the 627-nm LED source provided an 8700-fold increase in efficiency of watts used per unit conversion (Table [1]).

Table 1 Source Energy Input in watts vs Percent Conversion

Energy input

3 × 627 nm LED

Incandescent WX

Source

 0.36

 250.0

IR filter

Pump

   5.5

Haake K20 bath

1400.0

Total watts

 0.36

1655.5

% Conversion

98

  51

watts/% conversion

 0.37

3246.1

With an atom-economical microphotochemical LED apparatus now available, three classes of dienes were assayed to determine its general utility (Table [2]). Arenes and α-terpinene provided good to excellent yields of the corresponding endoperoxides over times ranging from 23 minutes to 24 hours (entries 1–5). Room-temperature photooxidation of cyclone (2,3,4,5-tetraphenyl-2,4-cyclopentadienone) provided a transient endoperoxide which subsequently decarbonylated to provide (Z)-1,2,3,4-tetraphenyl-2-butene-1,4-dione (entry 6). The analogous furan, 2,3,4,5-tetraphenylfuran, yielded the same 2-butene-1,4-dione,[ 44 ] but as a mixture of E- and Z-isomers (entry 7). Modifying the photooxidation of furfural[37] [45] to provide 5-hydroxy-2(5H)-furanone so as to be amenable to the microscale proved to be straightforward: the lactone was produced in very good yield (entry 8).

Table 2 627-nm LED Photooxidations in 0.3 mol% Methylene Blue Solution in Dichloromethane

Entry

Substrate

Time (min)

Product

Yield (%)

1

  23

>98, 84a

2

  23

>98, 86a

3

 180

92

4

1440

96

5

  23

81

6

  23

89

7

  30

65b

8

1380

98c

a Without methylene blue; 405 nm LED source.

b Z/E = 4.7:1.

c Reaction run in MeOH.

In other, preliminary work, it was discovered that similar to cyclone, 3,4-dimethyl-2,5-diphenyl-2,4-cyclopenta­dienone[ 46 ] provided a single product upon photooxidation, presumably (Z)-2,3-dimethyl-1,4-diphenyl-2-butene-1,4-dione. On the other hand, 2,5-diethyl-3,4-diphenyl-2,4-cyclopentadienone[ 47 ] provided a complex mixture of products, apparently from competing endoperoxidation and ene reactions. In an attempt to generate singlet oxygen by direct excitation from the triplet ground state, a source consisting of three 5-mm, 30° viewing angle, 1300-nm, near-IR LEDs was used to irradiate a 56.2-μmol sample of recrystallized 9,10-dimethylanthracene plus oxygen (15 mL, 670 μmol) in dichloromethane (5 mL) for 48 hours at 3 °C. 1H NMR spectroscopy indicated 1.5% conversion into the endoperoxide.

Finally, but not unexpectedly, the optimized conditions for microphotochemical oxidation were equally effective when 40 atom% 17O2 (ca. 4 mol equiv) was used to form the endoperoxides of 9,10-diphenylanthracene, 1,4-dimethylnaphthalene, and 1,8-dimethylnaphthalene.[ 48 ] These 17O-labeled peroxides had robust 17O NMR spectra (Table [3]).

Table 3 17O NMR Endoperoxide Chemical Shifts

Endoperoxide

δ (ppm)a

Linewidth (KHz)

107.7

4.7

300.2

1.5

287

1.4

a Relative to H2O.

In conclusion, commercial, inexpensive 5-mm milliwatt light-emitting diodes are effective sources for batch microphotochemical oxidations. Using limited quantities of singlet oxygen, these oxidations are atom economical and therefore useful for isotopic labeling. Extensions to other photochemical reactions using gaseous reactants, such as [2+2] cycloadditions of ethane to conjugated ketones,[ 36b ] are under investigation.

Cyclohexane, CH2Cl2, CHCl3, MeCN, benzene, EtOAc, petroleum ether, MeOH, EtOH, acetone, 9,10-dimethylanthracene, methylene blue, 1,4-dimethylnaphthalene, α-terpinene, neutral alumina, silica gel, and 40 atom% 17O2 were used as received. Anthracene was recrystallized twice from cyclohexane before use. Furfural was distilled before use. 9,10-Diphenylanthracene was prepared from 9,10-dibromoanthracene,[ 49 ] 2,3,4,5-tetraphenyl-2,4-cyclopentadienone from benzil and dibenzyl ketone,[ 50 ] and 2,3,4,5-tetraphenylfuran from 2-bromo-2-phenylacetophenone.[ 51 ] Photooxidations were run in magnetic stirrer bar equipped, 10-mL, silicone septa sealed, 20-mm crimp-top glass vials. Standard 250- and 500-watt WX incandescent sources were used in conjunction with an IR filter made from a 27 × 27 × 10 cm glass tank filled with circulating water cooled to 3 °C with a copper heat-exchanging coil attached to a Thermo Haake K20 refrigerated recirculating bath. The distance from source to reaction vessel was 30 cm. LED sources were constructed of LEDs purchased from Super Bright LEDs, Inc. or Thor Scientific Products and an 800-mW, 12-V DC power supply. A typical source consisted of three narrow viewing angle LEDs of the same emission wavelength inserted in a 1.5 × 11 cm piece of glass tubing so that the LED lenses were flush with the end of the tubing. The distance from source to reaction vessel was 0 cm. 3 °C reactions were performed in a cold room in the dark. Room temperature reactions were run under a double thickness of sueded black cloth. 1H, 13C, and 17O NMR spectra were recorded on a Varian Unity Inova spectrometer at 300, 75, and 40.7 MHz, respectively. IR spectra were acquired on a Thermo Avatar 370 spectrometer equipped with a Pike Technologies MIRacle single reflection ZnSe ATR sampling accessory, and were ATR-corrected after acquisition. EI mass spectra (70 eV) were obtained using an Agilent 6850/5973 GC/MSD system. MALDI mass spectra were acquired with an Applied Biosystems DE Voyager Pro mass spectrometer. HPLC was performed on a Hewlett-Packard 1100 isocratic chromatograph using a 50 × 4.6 mm Alltech Econosphere 3-μ, 60-Å silica column; detection was at 254 nm. GC/MS used a 30-m Alltech Econocap EC-5 column with a helium gas flow rate of 1 mL/min; the injector temperature was 250 °C, and the oven temperature program 60 °C (3 min), 10 °C/min (18 min), 240 °C (9 min). Analytical TLC was performed using Analtech silica gel GHLF 250-μm, 2.5 × 7.5 cm plates, visualized by UV light after development.


#

9,10-Dimethylanthracene 9,10-Endoperoxide[ 52 ]

9,10-Dimethylanthracene (10.3 mg, 49.9 μmol) was sealed in an aluminum foil wrapped, magnetic stirrer bar equipped, 10-mL hypo-vial. The vial was charged with 27 μM methylene blue in CH2Cl2 (5 mL). After two –78 °C freeze–aspirator pump–N2 purge–thaw cycles, O2 (12 mL, ca. 0.5 mmol, 10 equiv) was injected into the vial using a gas-tight syringe. The foil was removed and the contents of the reaction vial were stirred rapidly at 3 °C as the vial was irradiated with three 125-mW, 627-nm LEDs. After 20–25 min, the vial was vented by inserting a syringe needle through the septum. The septum was removed and the contents of the vial were filtered through a Pasteur pipet containing a 1–2-cm plug of neutral chromatographic alumina into a tared, 25-mL round-bottom flask. The alumina plug was washed with CH2Cl2 (3–5 × 1 mL); these washings were combined with the filtrate. Concentration by rotary evaporation afforded 9,10-dimethylanthracene 9,10-endoperoxide as a white solid containing no 9,10-dimethylanthracene (Rf  = 0.63) when analyzed by TLC (petroleum ether); yield: 11.8 mg (99%).

IR: 3041 (w), 2983 (w), 2935 (w), 1463 (m), 1378 (m), 756 (s) cm–1.

1H NMR (CDCl3): δ = 7.44 (m, 4 H), 7.31 (m, 4 H), 2.19 (s, 6 H).

13C NMR (CDCl3): δ = 141.11, 127.64, 120.91, 79.84, 13.95.

MS (MALDI): m/z = 239.12 [M + H+].

Similarly, a sealed 10-mL hypo-vial containing 9,10-dimethylanthracene (9.4 mg, 45.6 μmol), CH2Cl2 (5 mL), and O2 (12 mL, ca. 0.5 mmol, 11 equiv) was stirred rapidly at 3 °C and irradiated with three 80-mW, 405-nm LEDs for 20–25 min. Workup afforded 9,10-dimethylanthracene 9,10-endoperoxide as a white solid in greater than 98% purity, as determined by 1H NMR analysis; yield: 9.1 mg (84%).


#

9,10-Diphenylanthracene 9,10-Endoperoxide[18] [35a] [b] [53]

In a manner similar to the preparation of 9,10-dimethylanthracene 9,10-endoperoxide, a sealed 10-mL hypo-vial containing 9,10-diphenylanthracene (10.4 mg, 31.5 μmol), 27 μM methylene blue in CH2Cl2 (5 mL), and O2 or 40 atom% 17O2 (12 mL, ca. 0.5 mmol, 16 equiv) was stirred rapidly at r.t. and irradiated with three 125-mW, 627-nm LEDs for 20–25 min. Workup afforded 9,10-diphenylanthracene 9,10-endoperoxide as a white solid in 97% purity, as determined by 1H NMR analysis; yield: 11.6 mg (ca. 100%).

HPLC: t R = 7.2 min (CH2Cl2–cyclohexane, 2:8; 0.5 mL/min).

IR: 3074 (w), 1599 (w), 1494 (m), 1458 (m), 920 (m), 764 (s), 744 (s), 703 (m) cm–1.

1H NMR (DMSO-d 6): δ = 7.75–7.66 (m, 4 H), 7.66–7.58 (m, 6 H), 7.34–7.28 (m, 4 H), 7.12–7.04 (m, 4 H).

13C NMR (CDCl3): δ = 140.49, 133.25, 128.61, 128.53, 127.91, 127.80, 123.76, 84.34.

17O NMR (CDCl3): δ = 107.7.

MS (MALDI): m/z = 363.18 [M + H+].

Similarly, a sealed 10-mL hypo-vial containing 9,10-diphenylanthracene (12.6 mg, 38.1 μmol), CH2Cl2 (5 mL), and O2 (12 mL, ca. 0.5 mmol, 13 equiv) was stirred rapidly at 3 °C and irradiated with three 80-mW, 405-nm LEDs for 20–25 min. Workup afforded 9,10-diphenylanthracene 9,10-endoperoxide as a white solid in 98% purity, as determined by 1H NMR analysis; yield: 11.9 mg (86%).


#

Anthracene 9,10-Endoperoxide[ 54 ]

In a manner similar to the preparation of 9,10-dimethylanthracene 9,10-endoperoxide, a sealed 10-mL hypo-vial containing anthracene (17.9 mg, 0.10 mmol), 40 μM methylene blue in CH2Cl2 (5 mL), and O2 (15 mL, ca. 0.67 mmol, 6.7 equiv) was stirred rapidly at 3 °C and irradiated with three 125-mW, 627-nm LEDs for 3 h. Workup afforded anthracene 9,10-endoperoxide as a white solid in greater than 98% purity, as determined by 1H NMR analysis; yield: 19.3 mg (92%).

TLC: Rf  = 0.22 (benzene–petroleum ether, 2:1).

IR: 3046 (w), 2963 (w), 1462 (w), 769 (s), 752 (s), 720 (m) cm–1.

1H NMR (CDCl3): δ = 7.43–7.40 (m, 4 H), 7.27–7.24 (m, 4 H), 6.02 (s, 2 H).

13C NMR (CDCl3): δ = 138.30, 128.19, 123.86, 79.64.

MS (MALDI): m/z = 211.02 [M + H+].


#

1,4-Dimethylnaphthalene 1,4-Endoperoxide[35a] [b] [55]

In a manner similar to the preparation of 9,10-dimethylanthracene 9,10-endoperoxide, a sealed 10-mL hypo-vial containing 1,4-dimethylnaphthalene (26.4 mg, 0.17 mmol), methylene blue (1.4 mg, 4.4 μmol, 2.6 mol%), CH2Cl2 (5 mL), and O2 or 40 atom% 17O2 (15 mL, ca. 0.67 mmol, 3.9 equiv) was stirred rapidly at 3 °C and irradiated with three 125-mW, 627-nm LEDs for 24 h. Crude product (46.8 mg) was recovered as a white solid after workup at 3 °C. 1H NMR analysis indicated 97% conversion into 1,4-dimethylnaphthalene 1,4-endoperoxide.

TLC: Rf  = 0.11 (benzene–petroleum ether, 1:1).

IR: 3044 (w), 2982 (m), 2934 (w), 1459 (m), 1377 (m), 758 (s) cm–1.

1H NMR (CDCl3): δ = 7.40–7.25 (m, 4 H), 6.73 (s, 2 H), 1.93 (s, 6 H).

13C NMR (CDCl3): δ = 141.26, 139.48, 127.00, 120.37, 78.93, 16.41.

17O NMR (CDCl3): δ = 300.2.

MS (MALDI): m/z = 189.04 [M + H+].


#

Ascaridole[18] [20b] [41b] [d] [56]

In a manner similar to the preparation of 9,10-dimethylanthracene 9,10-endoperoxide, a sealed 10-mL hypo-vial containing α-terpinene (16.3 mg, 0.12 mmol), 40 μM methylene blue in CH2Cl2 (5 mL), and O2 (15 mL, ca. 0.67 mmol, 5.2 equiv) was stirred rapidly at 3 °C and irradiated with three 125-mW, 627-nm LEDs for 20–25 min. Workup afforded ascaridole as a clear, colorless liquid; yield: 16.3 mg (81%).

GC: t R = 10.4 min; TLC: Rf  = 0.50 (petroleum ether–EtOAc, 95:5).

IR: 3051 (w), 2965 (s), 2932 (s), 2877 (w), 1470 (w), 1451 (m), 1378 (m), 729 (m), 697 (m) cm–1.

1H NMR (CDCl3): δ = 6.52 (d, J = 8.7 Hz, 1 H), 6.44 (d, J = 8.7 Hz, 1 H), 2.06 (m, 2 H), 1.95 (m, 1 H), 1.55 (apparent d, J = 9.9 Hz, 2 H), 1.40 (s, 3 H), 1.03 (d, J = 7.2 Hz, 6 H).

13C NMR (CDCl3): δ = 136.65, 133.31, 80.04, 74.61, 32.38, 29.77, 25.85, 21.65, 17.48, 17.40.

MS (EI): m/z (%) = 168 (5), 136 (71), 121 (100), 93 (62).


#

(Z)-1,2,3,4-Tetraphenyl-2-butene-1,4-dione[18] [57]

In a manner similar to the preparation of 9,10-dimethylanthracene 9,10-endoperoxide, a sealed 10-mL hypo-vial containing 2,3,4,5-tetraphenyl-2,4-cyclopentadienone (17.2 mg, 44.7 μmol), 27 μM methylene blue in CH2Cl2 (5 mL), and O2 (12 mL, ca. 0.5 mmol, 11.2 equiv) was stirred rapidly at r.t. and irradiated with three 125-mW, 627-nm LEDs for 20–25 min. Workup afforded (Z)-1,2,3,4-tetraphenyl-2-butene-1,4-dione as a white solid in greater than 98% purity, as determined by 1H NMR analysis; yield: 15.4 mg (89%).

Mp 204–206 °C (Lit.[ 58 ] 210.5–211.5 °C).

TLC: Rf  = 0.54 (petroleum ether–EtOAc, 9:1).

IR: 3062 (w), 2920 (w), 2850 (w), 1660 (s), 1596 (m), 1580 (w), 1262 (s), 695 (s) cm–1.

1H NMR (CDCl3): δ = 7.86 (m, 4 H), 7.42 (tt, J = 7.4, 2.2 Hz, 2 H), 7.31 (tt, J = 7.8, 2.5 Hz, 4 H), 7.18 (s, 10 H).

13C NMR (CDCl3): δ = 197.18, 144.82, 136.66, 135.53, 133.24, 130.28, 130.10, 128.91, 128.63, 128.55.

MS (MALDI): m/z = 389.13 [M + H+].


#

(EZ)-1,2,3,4-Tetraphenyl-2-butene-1,4-dione

In a manner similar to the preparation of (Z)-1,2,3,4-tetraphenyl-2-butene-1,4-dione, a sealed 10-mL hypo-vial containing 2,3,4,5-tetraphenylfuran (32.4 mg, 87.0 μmol), 29 μM methylene blue in CH2Cl2 (5 mL), and O2 (15 mL, ca. 0.67 mmol, 7.7 equiv) was stirred rapidly at r.t. and irradiated with three 125-mW, 627-nm LEDs for 30 min. Workup afforded (EZ)-1,2,3,4-tetraphenyl-2-butene-1,4-dione (32.3 mg) as a white solid in a Z/E ratio of 4.7:1, as determined by 1H NMR and HPLC analyses [HPLC: t R = 2.6 (E), 3.2 (Z) min (EtOAc–cyclohexane, 7.7:92.3; 0.5 mL/min)]. The crude product was recrystallized (EtOH–H2O) to give the product as a white solid, with no change in the ratio of diastereomers; yield: 21.9 mg (65%).


#

5-Hydroxy-2(5H)-furanone[37] [45]

In a manner similar to the preparation of 9,10-dimethylanthracene 9,10-endoperoxide, a sealed 10-mL hypo-vial containing furfural (41.2 mg, 0.42 mmol), methylene blue (1.3 mg, 4.1 μmol, 1 mol%), MeOH (5 mL), and O2 (15 mL, ca. 0.67 mmol, 1.6 equiv) was stirred rapidly at r.t. and irradiated with three 125-mW, 627-nm LEDs for 23 h. The contents of the hypo-vial were concentrated by rotary evaporation, then taken up in acetone (5 mL). The acetone solution was filtered through a Pasteur pipet containing a 1–2-cm plug of silica gel into a tared, 25-mL round-bottom flask. The silica gel plug was washed with acetone (3 × 1 mL); these washings were combined with the filtrate. Concentration by rotary evaporation afforded crude 5-hydroxy-2(5H)-furanone (46.3 mg) as a clear yellow liquid in greater than 98% purity, as determined by 1H NMR analysis.

1H NMR (CDCl3): δ = 7.33 (dd, J = 1.2, 5.7 Hz, 1 H), 6.30 (t, J = 1.2 Hz, 1 H), 6.25 (dd, J = 1.2, 5.7 Hz, 1 H), 5.2–4.1 (1 H).

A sample was crystallized (CHCl3, –78 °C); mp 52–54 °C (Lit.[ 45 ] 52–55 °C).


#
#

Acknowledgment

Funding from the Creighton University Department of Pathology and the Creighton University Graduate School is gratefully acknowledged. James T. Fletcher, Robert C. Allen, and Ahsan U. Khan are thanked for valuable insights.



   Permissions and Reprints All articles of this category
Zoom Image
Scheme 1 a) [4+2] Cycloaddition of 1O2 with a conjugated diene; b) [2+2] cycloaddition of 1O2 with an alkene; c) ene reaction of the allylic positions of an alkene with 1O2
Zoom Image
Scheme 2 WX = tungsten halogen lamp; MB = methylene blue
Zoom Image
Figure 1 Methylene blue sensitized photooxidation of 9,10-dimethylanthracene as a function of LED source wavelength
Zoom Image
Figure 2 Solid lines, LED emission spectra; dashed lines, absorbance spectra