Synthesis of an optically active template for 4′-substituted nucleoside analogs via chromium carbene complex photochemistry

Article abstract of DOI:10.1021/om970069o

Procedures for the efficient synthesis and handling of [ethoxy((benzyloxy)methyl)carbene]pentacarbonylchromium(0) have been developed. Photolysis in the presence of the optically active ene carbamate 12 followed by Baeyer-Villiger oxidation and oxazolidinone elimination gave (R)-(+)-5-ethoxy-5-((benzyloxy)methyl)-2(5H)-furanone in good yield and with high enantiomeric purity. This compound provides a potential template for the synthesis of 4'-substituted nucleoside analogs.

Full text of DOI:10.1021/om970069o

Organometallics 1997, 16, 2313-2317
2313
Syn th esis of a n Op tica lly Active Tem p la te for
4′-Su bstitu ted Nu cleosid e An a logs via Ch r om iu m
Ca r ben e Com p lex P h otoch em istr y
Aimee D. Reed and Louis S. Hegedus*
Department of Chemistry, Colorado State University, Fort Collins, Colorado 80523
Received J anuary 30, 1997X
Procedures for the efficient synthesis and handling of [ethoxy((benzyloxy)methyl)car-
bene]pentacarbonylchromium(0) have been developed. Photolysis in the presence of the
optically active ene carbamate 12 followed by Baeyer-Villiger oxidation and oxazolidinone
elimination gave (R)-(+)-5-ethoxy-5-((benzyloxy)methyl)-2(5H)-furanone in good yield and
with high enantiomeric purity. This compound provides a potential template for the
synthesis of 4′-substituted nucleoside analogs.
In tr od u ction
Nucleosides bearing substituents at the 4′-position are
relatively uncommon, but several have unusual biologi-
cal activity. For example, nucleodin (1)1 exhibits anti-
biotic activity, 5′-deoxy-4′,5-difluorouridine (2) has an-
ticancer activity,2 and nucleoside 3 displays agonistic
activity for adenosine receptors.3 In addition, a number
of synthetic 4′-substituted nucleosides (4-10)4 are active
against HIV. Interestingly, the axiomatic notion that
nucleosides must be devoid of the 3′-hydroxy group to
inhibit HIV5 was not true in this instance, since the 3′-
hydroxy group was necessary for the activity of com-
pound 4.
Most 4′-substituted nucleosides have been synthesized
from existing nucleosides,6-11 and the syntheses are
often long and relatively nonstereoselective. Further,
only the enantiomer corresponding to the natural
nucleoside precursor is available by these procedures,
a limitation in light of recent evidence suggesting that
unnatural enantiomers of nucleoside antiviral com-
pounds may retain their biological activity but not their
X Abstract published in Advance ACS Abstracts, April 15, 1997.
(1) J enkins, I. D.; Verheyden, J . P. H.; Moffatt, J . G. J . Am. Chem.
Soc. 1971, 93, 4323.
(2) Ajmera, S.; Bapat, A. R.; Stephanian, E.; Danenberg, P. V. J .
Med. Chem. 1988, 31, 1094.
(3) Siddiqi, S. M.; J acobson, K. A.; Esker, J . L.; Olah, M. E.; J i, X.;
Melman, N.; Tiwari, K. N.; Secrist, J . A.; Schneller, S. W.; Cristalli,
G.; Stiles, G. L.; J ohnson, C. R.; IJ zerman, A. P. J . Med. Chem. 1995,
38, 1174.
(4) (a) Prisbe, E. J .; Maag, H.; Verheyden, J . P. H.; Rydzewski, R.
M. in Nucleosides and Nucleotides as Antitumor and Antiviral Agents;
Chu, C. K., Baker, D. C. Eds.; Plenum Press: New York, 1993; pp
101-113. (b) Maag, H.; Rydzewski, R. M.; McRoberts, M. J .; Crawford-
Ruth, D.; Verheyden, J . P. H.; Prisbe, E. J . J . Med. Chem. 1992, 35,
1440.
toxicity.12 Nucleosides substituted at the 4′-position
have also been synthesized from carbohydrates,13,14 as
well as de novo approaches from methallyl alcohol15 and
1,4-diacetoxycyclopent-2-ene.16 However, these latter
approaches were limited in the range of functionality
available at the 4′-position.
Nucleoside analogs lacking additional functionality at
the 4′-position have been synthesized from an optically
(5) (a) Mitsuya, H.; Broder, S. Proc. Natl. Acad. Sci. U.S.A. 1986,
83, 1911. (b) Nasr, M.; Litterst, C.; McGowen, J . Antiviral Res. 1990,
14, 125. (c) Mahmoudian, M. Pharm. Res. 1991, 8, 43.
(6) (a) O-Yang, C.; Wu, H. Y.; Fraser-Smith, E. B.; Walker, K. A.
M. Tetrahedron Lett. 1992, 33, 37. (b) O-Yang, C.; Kurz, W.; Eugui, E.
M.; McRoberts, M. J .; Verheyden, J . P. H.; Kurz, L. J .; Walker, K. A.
M. Tetrahedron Lett. 1992, 33, 41.
(7) Verheyden, J . P. H.; Moffatt, J . G. J . Am. Chem. Soc. 1975, 97,
4386.
(8) J ones, G. H.; Taniguchi, M.; Tegg, D.; Moffatt, J . G. J . Org. Chem.
1979, 44, 1309.
(12) For examples of different biological activities between enanti-
omers see: (a) Schinazi, R. F.; McMillan, A.; Cannon, D.; Mathis, R.;
Lloyd, R. M.; Peck, A.; Sommadossi, J .-P.; St. Clair, M.; Wilson, J .;
Furman, P. A.; Painter, G.; Choi, W.-B.; Liotta, D. C. Antimicrob.
Agents Chemother. 1992, 36, 2423. (b) Furman, P. A.; Davis, M.; Liotta,
D. C.; Paff, M.; Frick, L. W.; Nelson, D. J .; Dornsife, R. E.; Wurster, J .
A.; Wilson, L. J .; Fyfe, J . A.; Tuttle, J . V.; Miller, W. H.; Condreay, L.;
Averett, D. R.; Schinazi, R. F.; Painter, G. R. Antimicrob. Agents
Chemother. 1992, 36, 2686.
(13) Youssefyeh, R. D.; Verheyden, J . P. H.; Moffatt, J . G. J . Org.
Chem. 1979, 44, 1301.
(9) Codington, J . F.; Fecher, R.; Fox, J . J . J . Org. Chem. 1962, 27,
163.
(10) Haraguchi, K.; Tanaka, H.; Maeda, H.; Itoh, Y.; Saito, S.;
Miyasaka, T. J . Org. Chem. 1991, 56, 5401.
(14) Waga, T.; Nishizaki, T.; Miyakawa, I.; Ohrui, H.; Meguro, H.
Biosci., Biotechnol., Biochem. 1993, 57, 1433.
(15) Lipshutz, B. H.; Sharma, S.; Dimock, S. H.; Behling, J . R.
Synthesis 1992, 191.
(11) Haraguchi, K.; Tanaka, H.; Itoh, Y.; Saito, S.; Miyasaka, T.
Tetrahedron Lett. 1992, 33, 2841.
(16) J ohnson, C. R.; Esker, J . L.; Van Zandt, M. C. J . Org. Chem.
1994, 59, 5854.
S0276-7333(97)00069-1 CCC: $14.00 © 1997 American Chemical Society

2314 Organometallics, Vol. 16, No. 11, 1997
Reed and Hegedus
active butenolide template (eq 1),17 suggesting that 4,4-
disubstituted butenolides might be appropriate precur-
sors to 4′-substituted nucleoside analogs. An efficient
(3)
A second approach to chromium carbene complexes
is via carbonylation of anionic (σ-alkyl)chromium car-
bonyl complexes.22 However, treatment of chloromethyl
benzyl ether with the chromium pentacarbonyl dian-
ion23 failed to produce the desired carbene complex. In
the absence of carbon monoxide pressure, no anionic
acyl complex could be isolated. In the presence of 40
psi of carbon monoxide, only chromium hexacarbonyl
was recovered. Attempts to alkylate the presumed
anionic acyl complex intermediate with Meerwein’s
reagent instead led to the low-yield production of
carbene complexes from the alkylating agent (eq 4).
(1)
asymmetric synthesis18 and functionalization19 of 4,4-
disubstituted butenolides has recently been reported
from these laboratories (eq 2). The absolute configura-
(2)
tion of the chiral center in these butenolides is deter-
mined by the absolute configuration of the phenyl
glycine derived oxazolidinone, and both enantiomers are
available with equal facility, giving access to both the
natural and the unnatural series of nucleoside analogs.
The application of this methodology to the synthesis of
a butenolide template for 4′-substituted nucleoside
analogs is reported below.
(4)
Either chloromethyl benzyl ether was not attacked by
chromium pentacarbonyl dianion or, more likely, the
resulting anionic alkyl complex failed to undergo carbon
monoxide insertion. Electron-withdrawing groups such
as alkoxy groups dramatically decrease the migratory
aptitude of such substituted alkyl groups.24 The pro-
duction of small amounts of carbene complex from the
Meerwein alkylating agent confirms that carbene com-
plexes are available by this approach if both alkylation
and CO insertion are favorable.
A third approach to chromium carbene complexes is
the reaction of chromium pentacarbonyl dianion with
acid chlorides, followed by alkylation of the resulting
anionic acyl complex with Meerwein’s reagent.25 The
reported yields were only modest for aroyl halides (40-
65%), which were the best substrates, and were sub-
stantially lower (14-37%) for other classes of acid
chlorides. Since the ((benzyloxy)methyl)carbene com-
plex was to be required in large amounts, a high-yield
synthesis was important, and the acid chloride approach
appeared to be too inefficient. Indeed, when the ap-
proach was tried with R-(benzyloxy)acetyl chloride
Resu lts a n d Discu ssion
The critical element for nucleoside analog biological
activity is the presence of the 4′-hydroxymethyl group,
which translates, in the context of eq 2, into the
requirement for R to be a protected hydroxymethyl
group (PGOCH2) in the starting chromium carbene
complex. The most commonly used synthesis of chro-
mium alkoxycarbene complexes involves the reaction of
organolithium reagents with chromium hexacarbonyl
followed by O-alkylation with a hard alkylating agent.20
Dilithiation of tri-n-butyl(hydroxymethyl)stannane or
monolithiation of its silylated derivatives produced the
requisite organolithium reagents.21 Unfortunately these
were unreactive toward chromium hexacarbonyl at low
temperatures and decomposed upon warming, making
the desired chromium carbene complex unavailable by
this route (eq 3).
(17) (a) Agyei-Aye, K.; Baker, D. C. Carbohydrate Res. 1988, 183,
261. (b) Chu, C. K.; Beach, J . W.; Ullas, G. V.; Kosugi, Y. Tetrahedron
Lett. 1988, 29, 5349. (c) Okabe, M.; Sun, R-C.; Tam, S. Y.; Todero, L.
J .; Coffen, D. L. J . Org. Chem. 1988, 53, 4780.
(18) Miller, M.; Hegedus, L. S. J . Org. Chem. 1993, 58, 6779.
(19) Reed, A. D.; Hegedus, L. S. J . Org. Chem. 1995, 60, 3787.
(20) Fischer, E. O.; Maasbo¨l, A. Angew. Chem., Int. Ed. Engl. 1964,
3, 580.
(22) Stadtmu¨ller, H.; Knochel, P. Organometallics 1995, 14, 3163.
(23) Schwindt, M. A.; Lejon, T.; Hegedus, L. S. Organometallics
1990, 9, 2814.
(24) Collman, J . P.; Hegedus, L. S.; Norton, J . R.; Finke, R. G.
Principles and Applications of Organotransition Metal Chemistry;
University Science Books: Mill Valley, CA, 1987; pp 366-372.
(25) Semmelhack, M. F.; Lee, G. R. Organometallics 1987, 6, 1839.
(21) Meyer, N.; Seebach, D. Chem. Ber. 1980, 113, 1290.

Template for 4′-Substituted Nucleoside Analogs
Organometallics, Vol. 16, No. 11, 1997 2315
under standard conditions,25 yields varying from 20 to
40% were obtained. Careful observation indicated that
decomposition was occurring on prolonged contact with
the acid chloride and upon warming of the initial
reaction mixture to room temperature. By adding the
acid chloride26 to the dianion slowly at -78 °C, pouring
this cold slurry over ice after only 0.25 h, and adding
Et3OBF4 to this slurry, reproducible 70-80% yields of
the desired carbene complex 11 were obtained on a 5 g
scale (eq 5). (The ethoxycarbene complex was more
stable than the methoxy complex and was used for the
further studies.)
(6)
(5)
With a ready source of carbene complex 11 at hand,
photocycloaddition studies commenced. Photolysis of
chromium carbene complex 11 in the presence of ene
carbamate 12 under standard reaction conditions
(CH2Cl2, 50 °C, 80 psi of CO, 18-30 h, 450 W Hanovia
lamp, Pyrex) resulted in poor yields of the desired
cyclobutanone, along with a number of identified and
unidentified byproducts (eq 6). The relative instability
of chromium carbene complex 11 had already been
noted in its preparation and was a suspected source of
the inefficiency of the photoreactions. Indeed, simply
stirring complex 11 with substrate 12 at 50 °C without
irradiation produced substantial amounts of 2-ethoxy-
3-(benzyloxy)propene (15), probably by a thermal me-
tathesis reaction (eq 7). Electron-rich olefins are known27
to undergo metathesis with chromium carbene com-
plexes under thermal conditions, and the presence of
the electron-withdrawing benzyloxy group apparently
facilitates this process.
Carrying the photolysis out at 0 °C increased the yield
of the desired cyclobutanone 13 by about 10% and
markedly decreased the production of thermal decom-
position product 15. However, the yields remained
unacceptably low and the desired product was still
accompanied by substantial amounts of unidentified
decomposition products.
Most simple chromium alkoxycarbene complexes are
stable to irradiation at wavelengths above 300 nm
(through Pyrex) and can be recovered unchanged even
after prolonged (days) exposure to light of the above
wavelength.28 In contrast, irradiation of complex 11 in
the absence of substrate, even at -35 °C, resulted in
complete decomposition to unidentified products within
18 h. An examination of the UV-visible spectrum of
complex 11 showed substantial absorption between the
(7)
300 nm cutoff of Pyrex and the metal-to-ligand charge
transfer band (MLCT) responsible for CO insertion at
380 nm. This absorption is likely due to low-lying
ligand field (LF) transitions, excitation of which weak-
ens both metal-carbonyl and metal-carbene carbon
bonds, resulting in photolytic decomposition of the
carbene complex.29
The photochemical reaction in eq 6 was carried out
at -35 °C through a series of filter solutions30 with
absorption cutoffs between 300 and 350 nm. The results
are summarized in Table 1. Of the filters studied, a
solution of carbazole in DMF, with an absorption at 348
nm, was by far the most efficient, and reproducibly high
yields of the desired cyclobutanone 13 could be obtained
on a 1 g scale. Baeyer-Villiger oxidation followed by
elimination of the oxazolidinone by treatment with
TBAF completed an efficient approach to the desired
4′-substituted nucleoside template 17 (eq 8). The results
of studies to convert this compound to a series of 4′-
substituted nucleoside analogs as in eq 1 will be
reported in due course.
Exp er im en ta l Section
1
Gen er a l P r oced u r es. The NMR spectra (300 MHz for H
NMR and 75 MHz for 13C NMR) were recorded in CDCl3, and
chemical shifts are given in δ relative to CDCl3 (δ 7.24 for 1H
and δ 77.0 for 13C) unless otherwise indicated. Optical
rotations were obtained using a Perkin-Elmer 24 polarimeter
at a wavelength of 589 nm (sodium D line) in a 1.0 dm cell
with a total volume of 1 mL. Specific rotation is reported in
(26) The corresponding acid anhydride or mixed tert-butyl anhydride
could also be used, but the yields were lower and the isolation more
difficult.
(27) (a) Do¨tz, K. H.; Fischer, E. O. Chem. Ber. 1972, 105, 1356. (b)
Soderberg, B. C.; Hegedus, L. S. Organometallics 1990, 9, 3113.
(28) Hegedus, L. S.; deWeck, G.; D’Andrea, S. J . Am. Chem. Soc.
1988, 110, 2122.
(29) (a) Foley, H. C.; Strubinger, L. M.; Targos, T. S.; Geoffroy, G.
L. J . Am. Chem. Soc. 1983, 105, 3064. (b) Servaas, P. C.; Stufkens, D.
J .; Oskam, A. J . Organomet. Chem. 1990, 390, 61.
(30) Laporta, P.; Zaraga, F. Appl. Opt. 1981, 20, 2946.

2316 Organometallics, Vol. 16, No. 11, 1997
Reed and Hegedus
Ta ble 1. Effect of F ilter s on P h otoch em ica l
Cycloa d d ition (eq 8)
filter
cutoff, nm
yield of 3, %
aq Zn(NO3)2
aq Cu2SO4
aq K4Fe(CN)6
DMF/carbazole
335
∼310
∼400
348
29
34
16
79
adjacent Hanovia lamp at this temperature for 1 day. The
pressure tube was removed from the solution, and the CO was
released. More carbene complex 11 (977 mg, 2.64 mmol) was
added to the reaction mixture, and the pressure tube was
flushed and filled with CO as before. This mixture was
irradiated in the carbazole/DMF solution at -35 °C for 1 day.
The CO was released from the tube, and the solvent was
removed. Chromium hexacarbonyl was sublimed under vacuum
at 50 °C. Purification by column chromatography (10:2:1
hexanes/ethyl acetate/CH2Cl2) gave 988 mg (79%) of the
product. 1H NMR: δ 1.14 (t, 3H, J ) 7.0 Hz), 2.43 (dd, 1H, J
) 10.6, 18.0 Hz), 2.87 (dd, 1H, J ) 10.0, 18.0 Hz), 3.55-3.67
(m, 2H), 3.78 (d, 1H, J ) 9.5 Hz), 3.92 (d, 1H, J ) 9.5 Hz),
3.97 (dd, 1H, J ) 3.6, 8.5 Hz), 4.38 (t, 1H, J ) 8.5 Hz), 4.49 (d,
1H, J ) 11.3 Hz), 4.55 (d, J ) 11.3 Hz), 4.59 (t, 1H, J ) 10.3
Hz), 4.89 (dd, 1H, J ) 3.6, 8.5 Hz), 6.95 (m, 2H), 7.25-7.39
(m, 8H). 13C NMR: δ 206.3, 158.3, 140.0, 137.0, 129.3, 128.7,
128.6, 128.2, 128.1, 125.8, 97.7, 74.3, 70.3, 68.7, 61.5, 59.7, 47.7,
45.2, 15.6. Anal. Calcd for C22H25NO5: C, 69.85; H, 6.36; N,
(8)
degrees per decimeter at the specified temperature and the
concentration (c) given in grams per 100 mL in the specified
solvent. Chromatography was performed using silica gel 60
PF254 (with gypsum, E. Merck Science), and flash chromatog-
raphy was performed with ICN 32-63 µm, 60 A silica gel.
Elemental analyses were performed by M-H-W Laboratories,
Phoenix, AZ. Photolysis reactions were performed in Ace
pressure tubes with size #15 Ace-Thread, placed at a distance
of 10 cm from a Conrad-Hanovia 7825 medium-pressure
mercury lamp operating at 450 W, which was placed in a
water-cooled immersion well (Pyrex). A Conrad-Hanovia
7830-C power supply was used for the mercury lamp. Reac-
tions run under CO pressure were saturated with CO (3 cycles,
80 psi) and were photolyzed under 70-90 psi of CO.
25
3.54. Found: C, 69.65; H, 6.46; N, 3.53. [R]D ) +59.9° (c )
0.715 in ether).
La cton e 16. m-CPBA (986 mg, 3.37 mmol, 60% mixture)
was added to a CH2Cl2 solution of the cyclobutanone 13 (740
mg, 1.87 mmol) and Na2HPO4 (2.12 g, 14.9 mmol). The
reaction was stirred at room temperature for 15 h. An aqueous
solution of Na2S2O3 was added, and the mixture was stirred
vigorously for 1 h. The layers were separated, and the water
layer was extracted with CH2Cl2. The organics were combined
and washed with a saturated aqueous solution of NaHCO3.
The organics were dried over MgSO4 and filtered, and the
solvents were removed. Purification by column chromatogra-
phy (1:3 ethyl acetate/hexanes) gave the product (724 mg, 94%)
as a white foam. 1H NMR: δ 1.14 (t, 3H, J ) 7.0 Hz), 2.32
(m, 2H), 3.61 (m, 1H), 3.72 (d, 1H, J ) 11.1 Hz), 3.81 (m, 1H),
3.89 (dd, 1H, J ) 3.1, 7.9 Hz), 4.18 (t, 1H, J ) 8.4 Hz), 4.54 (d,
1H, J ) 1.1 Hz), 4.62 (d, 1H, J ) 11.1 Hz), 4.75 (dd, 1H, J )
3.1, 8.4 Hz), 4.78 (t, 1H, J ) 7.9 Hz), 6.97 (dd, 2H, J ) 1.5, 7.5
Hz), 7.21-7.37 (m, 8H). 13C NMR: δ 172.1, 158.3, 140.2,
136.6, 129.3, 129.0, 128.6, 128.3, 125.8, 109.1, 75.5, 70.7, 68.9,
58.8, 58.3, 54.8, 32.4, 15.1. Anal. Calcd for C22H25NO6: C,
67.24; H, 6.12; N, 3.41. Found: C, 67.13; H, 6.17; N, 3.41.
[E t h oxy((b en zyloxy)m et h yl)ca r b en e]p en t a ca r b on yl-
ch r om iu m (0) (11). Potassium metal (1.17 g, 30.0 mmol) was
added to flame-dried graphite (2.95 g, 245 mmol), and the
mixture was heated at 120 °C under vacuum until a bronze
laminate formed (1 h). The flask was cooled to room temper-
ature and filled with argon, and argon-saturated THF (40 mL)
was added. After the mixture was cooled to -78 °C, chromium
hexacarbonyl (3.00 g, 13.6 mmol) was then added. The
reaction mixture was stirred at this temperature for 1.5 h and
then warmed to 25 °C. After it was recooled to -78 °C,
R-(benzyloxy)acetyl chloride31 (2.64 g, 14.3 mmol) in 5 mL of
THF was slowly added to the K2Cr(CO)5 solution. The reaction
mixture was stirred at -78 °C for 20 min, after which it was
poured into a beaker. Ice (50 mL) and water (50 mL) were
added, and Et3OBF4 was added to the aqueous solution until
a pH of ∼2 was obtained. The mixture was filtered through
Celite and washed several times with ether. The filtrate was
separated, and the aqueous layer was extracted with ether.
The organic layers were combined and dried over MgSO4.
Purification by flash column chromatography (6:1 hexanes/
ether) gave an orange solid (3.82 g, 76%). Note: the workup
procedure and purification were performed in the dark, and
argon was used for the flash column chromatography. 1H
NMR: δ 1.63 (t, 3H, J ) 7.1 Hz), 4.41 (s, 2H), 4.59 (s, 2H),
25
[R]D ) +74.7° (c ) 0.825 in ether).
Bu ten olid e 17. Tetrabutylammonium fluoride (1.0 M, 2.28
mmol) was added to a THF solution of the lactone 16 (522 mg,
1.27 mmol). The solution was stirred at room temperature
for 1 h. It was then poured into water and extracted with
ether. The ether layers were dried over MgSO4, and the
solvent was removed. Purification by column chromatography
(1:4 ethyl acetate/hexanes) gave the product (300 mg, 95%) as
an oil. 1H NMR: δ 1.18 (t, 3H, J ) 7.0 Hz), 3.42 (m, 1H), 3.53
(m, 1H), 3.61 (d, 1H, J ) 10.8 Hz), 3.81 (d, 1H, J ) 11.1 Hz),
4.56 (s, 2H), 6.22 (d, 1H, J ) 5.7 Hz), 7.22 (d, 1H, J ) 5.7 Hz),
7.25-7.35 (m, 5H). 13C NMR: δ 169.6, 152.6, 137.2, 128.4,
127.8, 127.7, 125.3, 109.2, 73.9, 71.5, 59.7, 15.1. Anal. Calcd
for C14H16O4: C, 67.73; H, 6.50. Found: C, 67.76; H, 6.30.
5.12 (q, 2H, J ) 7.1 Hz), 7.29-7.36 (m, 5H). 13C NMR:
353.3, 223.4, 216.3, 137.2, 128.5, 128.0, 87.7, 73.6, 30.9, 15.5.
IR (thin film): ν 2063, 1922 cm-1
The complex lacked
δ
.
sufficient stability to permit elemental analysis.
Cyclobu ta n on e 13. The carbene complex 11 (1.17 g, 3.17
mmol) and ene carbamate 12 (600 mg, 3.17 mmol) were
combined in a pressure tube. Degassed CH2Cl2 was added to
the tube via a cannula. The pressure tube was flushed with
CO (3 × 80 psi) and filled with 80 psi of CO. The pressure
tube was placed in a larger tube containing a saturated
solution (2 g/L) of carbazole in DMF, and this entire unit was
placed in an unsilvered Dewar containing 1:1 ethylene glycol/
water and cooled to -35 °C by a Cryocool unit. The reaction
mixture was irradiated through the unsilvered Dewar by an
25
[R]D ) +64.2° (c ) 0.600 in ether).
2-Eth oxy-3-(ben zyloxy)p r op en e (15). This is the main
byproduct of the photochemical reaction described above. 1H
NMR: δ 1.27 (t, 3H, J ) 7.2 Hz), 4.07 (s, 2H), 4.21 (q, 2H, J
) 7.2 Hz), 4.62 (s, 2H), 7.30-7.35 (m, 7H). 13C NMR: δ 170.3,
137.1, 128.5, 128.0, 73.3, 67.2, 60.8, 14.2. HRMS: m/e calcd
192.1150, found 192.1147.
(31) Manhas, M. S.; Amin, S. G.; Chawla, H. P. S.; Bose, A. K. J .
Heterocycl. Chem. 1978, 15, 601.

Template for 4′-Substituted Nucleoside Analogs
Organometallics, Vol. 16, No. 11, 1997 2317
(Ben zyloxy)a ceton e. This product was often isolated from
the photochemical reaction. Its source is the hydrolysis of enol
ether 15. 1H NMR: δ 2.14 (s, 3H), 4.04 (s, 2H), 4.58 (s, 2H),
7.34 (m, 5H). 13C NMR: δ 206.7, 137.1, 128.5, 128.0, 127.9,
75.2, 73.3, 26.4.
Eth yl r-(ben zyloxy)a ceta te (14). This was the main
oxidation product of 1 in the presence of light. 1H NMR: δ
1.27 (t, 3H, J ) 7.2 Hz), 4.07 (s, 2H), 4.21 (q, 2H, J ) 7.2 Hz),
4.62 (s, 2H), 7.30-7.35 (m, 5H). 13C NMR δ 170.3, 137.1, 128.5,
128.0, 73.3, 67.2, 60.8, 14.2. MS (CI): m/e 195 (M + 1), 181
(37), 91 (100). HRMS: m/e calcd 195.1021, found 195.1011.
Ack n ow led gm en t. Support for this research under
Grant No. GM26178 from the National Institutes of
General Medical Sciences (Public Health Service) is
gratefully acknowledged. A.D.R. acknowledges the
Department of Education Patricia Roberts Harris Doc-
toral Fellowship. Mass spectra were obtained on in-
struments supported by the National Institutes of
Health shared instrumentation Grant No. GM49631.
OM970069O