Endoperoxidation of 9,10-bis(1-hydroxyalkyl)-anthracenes and successive formation of 9,10-anthraquinone under Grignard reaction conditions

Article abstract of DOI:10.3987/COM-03-S(P)24

The reaction of anthracene-9,10-carboaldehyde with various Grignard reagents under ambient conditions gave 9,10-bis(1-hydroxyalkyl)anthracene-9,10-peroxides and 9,10-anthraquinone as the first example. The endoperoxidation of 9,10-bis(1-hydroxyalkyl)anthr

Full text of DOI:10.3987/COM-03-S(P)24

HETEROCYCLES, Vol. 62, 2004, pp. 153 - 159
Received, 18th July, 2003, Accepted, 12th September, 2003, Published online, 4th November, 2003
ENDOPEROXIDATION OF 9,10-BIS(1-HYDROXYALKYL)-
ANTHRACENES AND SUCCESSIVE FORMATION OF 9,10-
ANTHRAQUINONE UNDER GRIGNARD REACTION CONDITIONS
Shigeyasu Kuroda,* Mitsunori Oda,* Hir oki Syumiya, Shah M. I. Shaheen,
Ryuta Miyatake, Teruhiko Nishikawa, Akiko Yoneda, Tokiko Tanaka, Masaru
Mouri, and Mayumi Kyogoku
Department of Applied Chemistry, Faculty of Engineering, Toyama University,
Gofuku 3190, Toyama 930-8555, Japan
E-mail address: kuro@eng.toyama-u.ac.jp
Abstract-The reaction of anthracene-9,10-carboaldehyde with various Grignard
reagents under ambient conditions gave 9,10-bis(1-hydroxyalkyl)anthracene-9,10-
peroxides and 9,10-anthraquinone as the first example. The endoperoxidation of
9,10-bis(1-hydroxyalkyl)anthracenes was assisted by the lone paired electron of
oxygen in the side chain, and the endoperoxides were transformed to
anthraquinone by the Grignard reagents.
INTRODUCTION
The endoperoxidation of anthracene and its derivatives by molecular oxygen (1O2) has been widely and
actively investigated from the viewpoint of their utility as synthetic substrates and photosensitizers.1, 2
The endoperoxides rearranged to various types of compounds under photochemical, ionic, and thermal
conditions,3 though the formation of 9,10-anthraquinone under the Grignard reaction conditions has not
been reported so far. During the course of our research on the synthesis of new conjugated aromatic
compounds containing an anthracene moiety, we found that the new reactions of anthracene-9,10-
carboaldehyde (1) with Grignard reagents under ambient conditions gave 9,10-anthraquinone (4) along
*) Dedicated to Prof. Dr. L. A. Paquette on occasion of his 70th birthday

with 9,10-bis(1-hydroxyalkyl)anthracenes (2)
and
9,10-bis(1-hydroxyalkyl)anthracene-9,10-
endoperoxides (3) in excellent yields. The mechanistic studies on the quinone formation revealed that
the lone paired electron of oxygen in the side chain assists the endoperoxidation to themselves, and the
quinone (4) was formed by the action of the Grignard reagents on endoperoxides.
O
CHO
CH(OH)R
CH(OH)R
O
CH(OH)R
O
5 eq. RMgX
THF-ether
+
+
CHO
R(HO)HC
O
1
2a; R = Me
2b; R = Et
3a; R = Me
4
3b; R = Et
3c; R = p-MeC6H4
3d; R =C 6H5
2c; R = p-MeC6H4
2d; R = C6H5
CHO
CH2OH
+
+
R'
R'
6a; R' = Me
6b; R' = H
5a; R' = Me
5b; R' = H
Scheme 1.
Table 1. The Grignard reaction conditions and yields.
Condition
Yield (%)
Entry No. and
Grignard reagent
atmosphere
light
time
2
3
4
6
5
2a
3a
room light
room light
1. R= Me
2. R = Me
air
O2
2 h
62
42
18
25
12
25
0
0
0
0
18 h
N2
N2
room light
dark
2 h
92
0
0
0
0
0
0
0
0
3. R = Me
4. R = Me
quant.
18 h
2b
a)
a)
room light
dark
5. R = Et
6. R = Et
air
N2
18 h
18 h
60
0
0
34
0
0
0
0
0
43
b)
b)
5a
6a
2c
c)
7. R = p-MeC6H4
8. R = p-MeC6H4
O2
O2
lamp
18 h
60 h
51
52
41
47
b)
45
51
b)
46
41
0
0
lamp
6b
2d
5b
O2
lamp
9. R = Ph
18 h
24
52
46
49
0
a) Crude yields.
b) The relative ratio is given compared to 4, which was determined by 1H NMR spectra.
c) The reaction mixture in flask was irraditated with a 100W tungsten lamp from a distance of 20 cm.

RESULTS AND DISCUSSION
The reaction of 1 with freshly prepared 5 equiv. of methylmagnesium iodide in dry ether-THF solution
at rt for 18 h gave diol (2a), 2,4 endoperoxidized diol (3a), and 9,10-anthraquinone (4). The reaction of 1
with the Grignard reagents such as ethyl-, p-tolyl- or phenylmagnesium bromide gave a mixture of diols
(2b-d) and (4), respectively (Table 1), though endoperoxides (3b-d)4 could not be detected. While the
isolated diols (2a-d) were endoperoxideized to give (3a-d) under the room light at room temperature. It
suggests that the endoperoxides as (3b~d) formed in the Grignard reaction were transformed to quinone
under the conditions. The reaction of 2a under the similar conditions as above in the presence of
5
diazabicyclo[2.2.2]octane (dabco) gave no endoperoxide (3a) supporting the formation of singlet
oxygen as a reactant. Furthermore, the endoperoxidation reactions of (2a-d) with chemically induced
singlet oxygen1e gave the corresponding endoperoxides (3a-d) along with a small amount of 4. Since the
formation of endoperoxides (3a) and (4) formed via (3a-d) under ambient conditions is so far
unprecedented, the generality and the reaction mechanism were examined by various conditions shown
in Table 1, and the experiments revealed the following.
Endoperoxidation: (a) The endoperoxidation of 2 was observed under no extra irradiation of light in the
laboratory, but it was not observed under darkness and under nitrogen atmosphere. (b) The preliminary
experiments described as above revealed that the rates of the endoperoxidation of 2a-d bearing
hydroxyalkyl groups at the 9- and 10-positions6 was faster than that of compound bearing phenyl group
at the 9- and 10-positions, which is known as the Steering effect.7~9 And this endoperoxidation
proceeded under Grignard conditions provides an evidence effected by the role of the lone paired
electron of oxygen in the side chain rather than by the hydrogen bonding in the Steering effect.8 (c) The
reaction of 1 with large excess of the Grignard reagent (10~15 mol equiv.) under the same conditions as
with the use of 5 equiv. of MeMgI exclusively gave 2a~d quantitatively, suggesting that oxygen existing
10
in the solution was consumed by the excess Grignard reagents. (d) The pathway from diols (2a, 2b) to
quione (4) by autoxidation was excluded, because the standing of (2a, 2b) or (3a, 3b) dissolved in
solvents such as CH2Cl2, ether, THF, and AcOEt in the dark at room temperature for 18 h recovered the
starting materials. (f) The light-induced dissociation of the O-O bond of the endoperoxides also did not
proceed, since 3a did not change under the irradiation of light.
Quinone formation: The roles of the Grignard reagent and the hydroxyl groups in the side chain for the
quinone formation were revealed as follows. (a) The reaction of 3a with 2.0 or more equiv. of MeMgI
for 2 h gave 4 along with a small amount of 8, which may be formed from 4 (Table 2). The reaction of

O
O
O
O
R
MeMgI
+
O
Me
HO
R
3a; R=CH(OH)Me
7; R=C6H5
4
8
Scheme 2
Table 2 The reaction of 3a and 7 with various molar ratios of MeMgI
Yield (%)
Conditions
(molar ratio of MeMgI)
Entry
Substrate
3a
4
8
(recovered)
10 equiv.
2 equiv.
0.5 equiv.
0.2 equiv.
1
2
3
4
0
0
56
81
82
37
12
4
3
0
0
3a
3a
3a
3a
75
7
4
8
(recovered)
2 equiv.
5
7
100
0
0
9,10-diphenylanthracene-9,10-endoperoxide (7) with MeMgI under the same conditions gave no
products. This suggests that the hydroxyl groups on the side chain at the 9- and 10-positions are
necessary for the O-O bond fission. The reaction of 1 with p-Me-C6H4MgBr gave no endoperoxide (3c)
or (3d), but it gave 2c, 4, p-methylbenzaldehyde (5a), and p-methylbenzyl alcohol (6a). Similarly,
benzaldehyde (5b) and benzyl alcohol (6b). were obtained by the use of PhMgBr. These afford
significant information on the role of the Grignard reagent and the reaction sequence of the quinone
formation.9 (d) It should be noted that the endoperoxide (3a) was also transformed quantitatively to
quinone (4) by PhMgBr. Although the reaction of 3a with activated magnesium (Mg*)11 obtained by
using 1,2-dibromoethane as entrainer also gave 4 quantitatively, that with Mg* prepared by other
12
methods11,
gave no product. It suggests that some amount of Grignard reagent as 2-
bromoethylmagnesium bromide exists in the former reaction mixture.
Mechanism: Taking the above results into account, the overall plausible mechanism is proposed as
shown in Scheme 3. First, the Grignard reagent added to formyl groups to form intermediate (A).
Subsequently, the room light-induced singlet oxygen (1O2) sensitized by anthracene derivatives reacted
at the 9,10-positions of (A) which was assisted by the interactions of the lone paired electron of oxygen
in the side chain (intermediate B) to form an intermediate (C). Additionally, PM3 calculation showed a

little greater electron density of the HSOMO of 9- and 10-positions of 2a than those of HOMO of 9,10-
dimethylanthracene (9). It suggests the greater interaction of 2a with oxygen than that of 9. Then, the O-
O bond of endoperoxide was broken by the action of the Grignard reagents to give quinone, aldehydes,
and alcohols.
CH(OMgI)R
O O
RMgI
1O2
O
O
MgI
1
IMg
CH(OMgI)R
R
R
A
B
OH
O MgI
O
O
R
O
O
RMgI
R
+
+
6
4
5
R
O-MgI
R
OH
3
C
Scheme 3
AKNOWLEDGEMENT
We gratefully acknowledge financial support by a Grant-in-Aid for Scientific Research (No. 10640513)
from the Ministry of Education, Science, Sports, and Culture, Japan.
REFERENCES AND NOTES
1. a) H. H. Wasserman and R. W. Murray, "Singlet Oxygen, A Series of Monographs", Academic Press:
New York, 1979, 40, pp. 59-114. b) R. W. Denny and A. Nickon, Org. React., 1973, 20, 133, ed. by
W. G. Dauben, John Wiley & Sons Inc., New York. c) M. Balci, Chem. Rev., 1982, 81, 91, and
references cited therein. d) R. G. Zepp, N. L. Wolf, G. L. Baughman, and R. C. Hollis, Nature, 1977, 267,
421. e) N. A. Garcia, J. Photochem. Photobiol., B: Biol., 1994, 22, 185. f) C. S. Foote, S. Waxler, and
W. Ando, Tetrahedron Lett., 1965, 4111. g) C. S. Foote, S. Waxler, W. Ando, and R. Higgins, J. Am.
Chem. Soc., 1968, 90, 975. h) J. M. Aubry, J. Am. Chem. Soc., 1985, 107, 5844. i) C. S. Foote and S.
Waxler, J. Am. Chem. Soc., 1964, 86, 3880. j) C. S. Foote, Science (Washington D. C.) 1970, 162,

963. k) C. S. Foote, In "Free radical in biology" W. A. Pryor, Academic Press, New York, 1976, Vol.
2, 419.
2. a) D. R. Kearns, Chem. Rev., 1971, 71, 418. b) G. O. Schenck, Angew. Chem., 1957, 69, 579. c) C. S.
Foote and S. Waxler, J. Am. Chem. Soc., 1964, 86, 3880~3881. d) C. S. Foote and S. Waxler, J. Am.
Chem. Soc., 1964, 86, 3881.
3. B. Stevens, S. R. Perez, and J. A. Ors, J. Am. Chem. Soc., 1974, 96, 6846.
4. The data of the endoperoxides. 3a: colorless crystals, mp 97˚C (benzene-hexane), IR(KBr) n
max
3410brm, 3020w, 2891w, 2872w, 1460m, 1378m, 1252w, 1158w,1112m, 997w, 929m, 907m, 760s,
1
668w, 639w cm-1; H NMR (400 MHz, CDCl3-TMS) d = 7.81 (br, 2H), 7.42 (br, 2H), 7.23 (br, 4H),
13
5.14 (br, 2H), 2.28 (br s, 2H), 1.75 (br, 6H); C NMR (100 MHz, CDCl3-TMS) d = 139.8, 139.7,
127.0, 126.9, 126.8, 126.7, 124.2, 123.9, 122.3, 122.0, 83.6, 83.6, 66.6, 66.5, 19.4; UV-VIS lmax
(MeOH) 361 (log e =1.80), 270 (3.08), 250sh (3.43), 223 nm (3.86); MS m/z 298 (M+, 1%), 208 (M+-
2xCH(OH)CH3, 77%), 180 (M+-{2xCH(OH)CH3 + CO}, 91%), 152 (M+-{2xCH(OH)CH3 + 2xCO},
100%). Anal. Calcd for C18H18O4 : C, 72.47; H, 6.08. Found: C, 72.26; H, 5.99.
3b: Pale yellow crystals, mp 79˚C (benzene-hexane); IR (KBr) n max 3450brs, 3060w, 3040w, 2960s,
2930s, 2870s, 1460s, 1400w, 1240w, 1150w, 1110m, 1060w, 1040w, 980m, 920m, 890w, 760s, 710w,
1
700w, 640w cm-1; H NMR (400 MHz, CDCl3-TMS) d =7.81 (m, 2H), 7.31 (m, 2H), 7.22 (m, 4H),
4.79 (t, J = 9.2Hz, 2H, -CH-), 2.23 (m, 2H, -CH2-), 2.00 (m, 2H, -CH2-), 1.90 (m, 2H, -OH), 1.28 (m,
6H, 2x -CH3); 13C NMR (100 MHz, CDCl3-TMS) d =140.1, 140.0, 126.9, 126.9, 126.7, 124.5, 123.9,
122.2, 121.8, 83.9, 83.8, 72.4, 31.6, 26.1, 11.5; UV-VIS lmax (MeOH) 260sh (log e = 3.30), 250sh
(3.37), 217nm (4.53); MS m/z 326 (M+, 0.2%), 208 (M+ -2xCH(OH)Et, 100%), 180 (M+-
{2xCH(OH)Et + CO}, 86%), 152 (M+-{2xCH(OH)Et + 2xCO}, 67%); Anal. Calcd for C20H22O4: C,
73.60; H, 6.79. Found: C, 73.59; H, 6.90.
3c: Pale yellow powder, mp 135˚C (decomp) (CH2Cl2-hexane); IR (KBr) n max 3399brs, 3031w,
2920w, 1655m, 1613m, 1512s, 1458s, 1387s, 1186s, 1060s, 1017s, 918s, 818s, 758vs, 723m, 696m,
642m cm-1; 1H NMR (400 MHz, CD3CN-TMS) d =7.94 (m, 2H), 7.59 (d, J=7.99, 4H), 7.42 (m, 2H),
7.18 (m, 2H), 7.17 (d, J=7.99, 4H), 7.09 (m, 2H), 5.96 (m, 2H), 4.08 (brs, 2H, -OH), 2.30 (s, 6H, -
13
CH3); C NMR (100 MHz, CDCl3-TMS) d =142.0, 141.1, 138.5, 138,0, 130.0, 129.4, 127.2, 127.0,
126.9, 125.2, 125.1, 123.8, 123.7, 83.9, 83.8, 73.0, 72.9, 20.9; UV-VIS lmax (MeOH) 259 (log
e =4.80), 333 (3.50), 355 (2.59), 374 (1.62), 395 nm (1.58); MS m/z 330 (M+ -p-Me-C6H4CHO, 54%),
296 (100%), 208 (anthraquinone, 17%), 180 (12%); Anal. Calcd for C20H22O4: C, 78.41; H, 5.92.

Found: C, 78.88; H, 5.97.
3d: Pale yellow powder, mp 90˚C (decomp) (CH2Cl2-hexane); IR (KBr) n max 3393brs, 3065m, 3030m,
2921m, 2850w, 1698m, 1589m, 1455s, 1317m, 1204m, 1171s, 1121s, 1090m, 1042m, 932m, 828m,
1
747s, 648m, 613m cm-1; H NMR (400 MHz, CD3CN-TMS) d =7.96 (m, 2H), 7.72 (d, J=7.99, 2H),
7.42 (m, 2H), 7.36 (m, 4H), 7.30 (m, 2H), 7.19 (m, 2H), 7.11 (m, 2H), 6.00 (m, 2H), 4.18 (brs, 2H, -
13
OH); C NMR (100 MHz, CDCl3-TMS) d =142.3, 141.3, 141.2, 130.4, 129.1, 128.9, 127.6, 127.3,
127.2, 125.5, 125.4, 124.1, 124.0, 84.1, 73.5; UV-VIS lmax (MeOH) 230 (log e =4.92), 260 (4.38),
356 (3.48), 376 (2.57), 395 nm (2.55); MS m/z 316 (M+ -C6H5CHO, 3%), 299 (4%), 208 (100%), 180
(83%), 67%); Anal. Calcd for C20H22O4: C, 79.60; H, 5.25. Found: C, 79.40; H, 5.55.
5. E. A. Orryzlo and C. W. Tang, J. Am. Chem. Soc., 1970, 92, 5034, I. Saito, T. Matsuura, and K. Inoue,
J. Am. Chem. Soc., 1981, 103, 188.
6. The rates of endoperoxidation of anthracenes bearing one hydroxlyl group on the side chain are fairly
greater than that of 9,10-phenylanthracene under the room light.
7. M. Prein and W. Adam, Angew. Chem., Int. Ed. Engl., 1996, 35, 6635.
8. M. Matsumoto, Photochem., 1998, 28, 31.
9. W. Adam and M. Prein, J. Am. Chem. Soc., 1993, 115, 3766.
10. C. Walling and S. A. Buckler, J. Am. Chem. Soc., 1955, 77, 6032, and literatures cited therein.
11. Activated magnesium; a) D. E. Pearson, D. Cowan, and J. D. Beckler, J. Org. Chem., 1959, 24, 504,
and Y.-H Lai, Synthesis, 1981, 585, and references cited therein.
12. R. D. Rieke and M. V. Hanson, Tetrahedron Lett., 1997, 53, 1925.