A New Method for the Preparation of Aluminum and Titanium Tris(2,6-diphenylphenoxide) Reagents and Their Application in Organic Synthesis

Article abstract of DOI:10.1246/cl.2003.1006

A new method for the preparation of aluminum and titanium trisphenoxides was realized using tetraallyltin or α-pinene as a proton-trapping agent. Thus obtained chlorotitanium reagent was converted to the corresponding cation-type reagent, which proved to be an effective catalyst for epoxide rearrangements.

Full text of DOI:10.1246/cl.2003.1006

1006
Chemistry Letters Vol.32, No.11 (2003)
A New Method for the Preparation of Aluminum and Titanium Tris(2,6-diphenylphenoxide)
Reagents and Their Application in Organic Synthesis
Atsushi Sato, Asuka Hattori, Kazuaki Ishihara, Susumu Saito,ꢀy and Hisashi Yamamotoꢀyy
Graduate School of Engineering, Nagoya University, Chikusa, Nagoya 464-8603
yInstitute for Advanced Research & Graduate School of Science, Nagoya University, Chikusa, Nagoya 464-8602
yyDepartment of Chemistry, The University of Chicago, 5735 South Ellis Ave, Chicago, IL 60637, U. S. A.
(Received July 4, 2003; CL-030597)
A new method for the preparation of aluminum and titanium
trisphenoxides was realized using tetraallyltin or ꢀ-pinene as a
proton-trapping agent. Thus obtained chlorotitanium reagent
was converted to the corresponding cation-type reagent, which
proved to be an effective catalyst for epoxide rearrangements.
TiCl4 at room temperature to give the corresponding trisphenox-
ide product 2 in quantitative yield (with respect to the internal
standard, Eq 3), the structure of which was rigorously establish-
ed by 1H NMR and elemental analysis as well as by X-ray single
crystal8 analysis (Figure 1). This is in clear contrast to the results
that afforded the bisphenoxide Ti-reagent 3 by treatment of the
phenol with TiCl4 upon benzene reflux and by sweeping out
HCl under a flow of N2.3f The crystal analysis revealed that 2
has a structure apparently distinct from, though slightly similar
to, that of ATPH; three phenyl rings form a partial cavity, one
of the rings being relatively exposed. The remaining two phenyl
rings have a face-to-face orientation to form a sandwich struc-
ture involving the chloro group (Figure 1b).
Over the last decade, extraordinary organic transformations
have been realized using aluminum tris(2,6-diphenylphenoxide)
(ATPH).1 The preparation of ATPH was routinely carried out by
treatment of Me3Al with 2,6-diphenylphenol at room tempera-
ture (Eq 1). Although Me3Al is a convenient reagent being reac-
tive to phenols, it is extremely pyrophoric2 and thus does not sat-
isfy a demand for a large-scale preparation with easy handling.
We report here a safer and more practical approach to this crit-
ical reagent by the combined use of AlBr3 and tetraallyltin re-
agents (Eq 2). This obviates the need to remove side products
such as alkali metal salts, which are inevitably generated during
usual procedures,3 by cumbersome decantation or other Schlenk
techniques. These practical advantages further extend to the syn-
thesis of a Ti-trisphenoxide reagent.
Ph
OH
Ph
Sn
O
3
Ph
Ph
12
4
Ti
Ph
+
+
4 SnCl4 12
(3)
4TiCl4
+
3
O
O
benzene
Cl
Ph Ph
Ph
rt, 1h
1
2
Ph
Cl
Cl
Ti
O
O
Ph Ph
Ph
3
Ph
OH
Ph
O
Ph
Ph
(a)
(b)
Me3Al
+
Al
Ph
3
+
3CH4
(1)
toluene
O
O
or CH2Cl2
rt, 30 min
Ph Ph
Ph
1
O
Cl
Ti
Cl
ATPH
O
O
O
Ti
O
OH
O
Sn
3
Ph
12
Ph
4
(2)
+
4 SnX4
+
4 AlX3
+
3 ATPH
12
benzene
rt, 1h
1
Treatment of 2,6-diphenylphenol (12 equiv.) with AlBr3 (4
equiv.) in benzene at room temperature was followed by addition
of tetraallyltin (3equiv.) The mixture was stirred at room tem-
perature for 1 h, and the solvent was evaporated in vacuo
(0.05 mmHg, 40 ꢁC) for 3h with concomitant removal of the re-
sulting SnBr4 (bp 202 ꢁC) and propene. The residual yellow solid
was identified as pure ATPH by comparison with the authentic
1H NMR data obtained by treatment of Me3Al with the phenol.4
Figure 1. The X-ray single crystal structure of 2. The (a) OR-
TEP and (b) CPK models.
Based on these results, a plausible mechanism for the reac-
tion could be proposed (Eq 4). We also confirmed that no reac-
tion occurred between the phenol and tetraallyltin alone even at
40 ꢁC for 3h. Thus the complexation-decomplexation equilibri-
um shifted to the right side and activated the release of HX from
coordination complex 4 (Eq 4). Although it was reported that al-
lyltributyltin was protonated by a mixture of binaphthol and
TiCl2(Oi-Pr)2 affording propene and tributyltin chloride,9 in
general, monodentate ligands including 1 are less effective in
forming such a metastable complex. It is also reasonable to sug-
gest a mechanism corroborating a pathway involving the direct
trap of HX from 4. Undoubtedly the acidity of the hydroxy group
of 4 is strengthened; otherwise the formation of metal phenox-
ides would not proceed.10 Whichever mechanism is preferred,
1
Monitoring the reaction by H NMR (benzene-d6) showed that
the peaks corresponding to tetraallyltin had disappeared in a re-
gion ranging from 5.0 to 6.5 ppm. Internal standard5 peaks indi-
cated the quantitative formation of ATPH. The use of AlCl3 also
resulted in the formation of ATPH but was contaminated with
unidentified side products bearing phenoxide components.6 In
both cases, no free phenol remained, as ascertained by 1H
NMR analysis.
This procedure was also applicable to the preparation of Ti-
trisphenoxide 2.7 A similar reaction proceeded smoothly using
Copyright ꢀ 2003The Chemical Society of Japan

Chemistry Letters Vol.32, No.11 (2003)
1007
with other metal (Ti, Zr, Hf, Ln, etc.) halide salts to give metal
trisphenoxides. See for Sm: a) K. Katagiri, M. Kameoka, M.
Nishiura, and T. Imamoto, Chem. Lett., 2002, 426. For various
Ln: b) D. M. Barnhart, D. L. Clark, J. C. Gorden, J. C. Huffman,
R. L. Vincent, J. G. Watkin, and B. D. Zwick, Inorg. Chem., 33,
3487 (1994). For Ti, Zr, Hf: c) S. L. Latesky, J. Keddington, A.
K. McMullen, I. P. Rothwell, and J. C. Huffman, Inorg. Chem.,
24, 995 (1985). d) L. Chamberlain, J. C. Huffman, J. Keddington,
and I. P. Rothwell, J. Chem. Soc., Chem. Commun., 1982, 805.
For TiMe4: e) R. W. Chesnut, L. D. Durfee, P. E. Fanwick,
and I. P. Rothwell, Polyhedron, 6, 2019 (1987). f) M. G. Thorn,
Z. C. Etheridge, P. E. Fanwick, and I. P. Rothwell, J. Organomet.
Chem., 591, 148 (1999).
For the X-ray single crystal structures of various ATPH–carbonyl
complexes, see Refs. 1d, 1e and: S. Saito, T. Nagahara, M.
Shiozawa, M. Nakadai, and H. Yamamoto, J. Am. Chem. Soc.,
125, 6200 (2003).
Trioctylmethylsilane (purchased from Shinetsu) was purified by
column chromatography and used.
either of the reaction sequences is repeated in situ to give ATPH
and Ti-trisphenoxide 2.
Xn-1MO
Ph
Sn
Ph
X
H
O
X
H
O
Ph
Ph
4
Xn-1
M
Xn-1
M
+
MXn
1
(4)
+
Ph
Ph
SnX
M = Al, Ti
X = Br , Cl
n = 3, 4
3
4
In place of tetraallyltin, ꢀ-pinene as a proton scavenger was
also capable of the synthesis of 2 (Eq 5). Although the yield was
moderate (ca. 70%), pure trisphenoxide 2 was obtained through
the Wagner–Meerwein–Whitmore rearrangement of ꢀ-pinene,
followed by recrystallization.
4
5
OH
( 10 mL)
Ph
Ph
+
TiCl4
(5)
+
2
rt
ca. 70 %
(1.2 mmol)
(0.4 mmol)
Cl
6
7
Unidentified tin phenoxides (RnSn(OAr)4-n) might be formed.
Ti-complex 2 is the new compound. Typical procedure: To a so-
lution of 1 (295 mg, 1.2 mmol) in dry benzene (4.0 mL) was add-
ed TiCl4 (44 mL, 0.4 mmol), followed by tetraallyltin (72 mL,
0.3mmol), both being purified by distillation prior to use, at room
temperature under argon. The resulting dark red solution was
stirred for 1 h. The solvents and other volatile side products
(SnCl4: bp 114 ꢁC (770 mmHg)) were evaporated in vacuo
Although the titanium reagent 2 showed relatively modest
catalytic activity, it was used as a cation-type catalyst superior
to the neutral catalyst ATPH (Eq 6). For example, treatment of
a 10 mol% of 2 with AgOTf, followed by addition of 5 gave 6
in 75% yield within 1 h, compared with ATPH that afforded
55% yield after a prolonged reaction time (48 h). Tetra-substitut-
ed epoxide 7 was also compatible with similar conditions to give
allylic alcohol 8 (Eq 7).
ꢁ
(0.03mmHg) at 40 C to give 2 in dark red crops (325 mg,
99%). Anal. Calcd for C54H39O3ClTi: C, 79.17, H, 4.80%;
Found: C, 79.08, H, 4.82%. The single crystals suitable for the
X-ray crystal analysis were grown from hexane–benzene at room
temperature for 1 to 2 days. A crystal was transferred to a glass
capillary tube using a dry-box for the diffraction experiments.
The X-ray single crystal structure data of 2: a ¼ 10:2910 (2),
O
cat. (10 mol%)
OTBS
(6)
OTBS
toluene
25 °C
OH
cat
ATPH, 48 h
2 , 48 h
5
6
: 55%
: 46%
: 75%
8
2 + AgOTf, 1 h
ꢁ
ꢀ
b ¼ 21:0260 (3), c ¼ 20:191 (4) A, ꢁ ¼ 104:177 (3) , monoclin-
ic, P21/c, Z ¼ 4, ꢂ(Mo) = 0.437 mmÀ1, R ¼ 0:050, Rw
¼
cat. (10 mol%)
cat
OH
Ph
O
: trace
ATPH
0:051, GOF = 1.836, 4989 unique reflections with I > 3:0ꢃðIÞ.
Crystallographic data (excluding structure factors) for the X-
ray crystal structure analysis reported in this paper have been de-
posited with the Cambridge Crystallographic Data Centre
(CCDC) as supplementary publication no. CCDC-184707. Cop-
ies of the data can be obtained free of charge on application to
CCDC, 12 Union Road, Cambrdge CB2 1EZ, UK (fax: +44-
1223-336-033, e-mail: deposit@ccdc.cam.ac.uk).
(7)
tlueneo
25 °C, 48 h
: 58%
2
Ph
7
8
In summary, we achieved an alternative route to ATPH and
the chlorotitanium trisphenoxide 2 by use of metal halide salts
and tetraallyltin or ꢀ-pinene. Since organotin products are not
generated, this method has an environmentally more benign na-
ture. The reaction proceeded effectively with metals bearing
strong coordination capability. Application of this method to
phenols having a bidentate coordination site that would be ex-
pected to form more stable metal complexes9,10 is now in prog-
ress in our laboratory.
9
a) S. Casolari, D. D’Addario, and E. Tagliavini, Org. Lett., 1,
1061 (1999). The TiF4-binaphthol-allyltrimethylsilane system
to produce binaphthol-TiF2 complex was also reported. See: b)
R. O. Duthaler and A. Hafner, Angew. Chem. Int. Ed., 36, 43
(1997). c) D. R. Gauthier, Jr. and E. M. Carreira, Angew. Chem.
Int. Ed., 35, 2363 (1996).
This work was supported by SORST, Japan Science and
Technology Corporation (JST).
10 In cases where ZrCl4ꢂ(THF)2 and HfCl4ꢂ(THF)2 were reacted
with 1, no reaction or unknown product mixtures6 were observed,
respectively. This is probably due to the lack of sufficient Lewis
acidity and/or of steric requirement requisite for the formation of
stable trisphenoxides, which led to undesired disproportion reac-
References and Notes
1
For review of ATPH, see: a) S. Saito and H. Yamamoto, Chem.
Commun., 1997, 2585. Recent application of ATPH, see: b) S.
Saito, S. Yamazaki, and H. Yamamoto, Angew. Chem., Int.
Ed., 40, 3613 (2001). c) S. Saito, M. Shiozawa, T. Nagahara,
M. Nakadai, and H. Yamamoto, J. Am. Chem. Soc., 122, 7847
(2000). d) S. Saito, T. Sone, M. Murase, and H. Yamamoto, J.
Am. Chem. Soc., 122, 10216 (2000). e) S. Saito, M. Shiozawa,
and H. Yamamoto, Angew. Chem., Int. Ed., 38, 1769 (1999).
‘‘Encyclopedia of Reagents for Organic Synthesis,’’ ed. by L. A.
Paquette, John Wiley & Sons, Chichester (1995), Vol. 7, p 5186.
1
tions. Monitoring the reaction by H NMR analysis (300 MHz,
THF-d8) showed a set of peaks corresponding to the free phenol.
This implies that the metastable formation of the coordination
complex (such as 4) should be a critical step to initiate the reac-
tion. Similar to titanium complexes (Ref. 9), a bidentate com-
plexation with zirconium was achieved by treatment of tributy-
lallyltin with binaphthol and ZrCl4ꢂ(THF)2. See: S. Casolari, P.
G. Cozzi, P. Orioli, E. Tagliavini, and A. Umani-Ronchi, Chem.
Commun., 1997, 2123.
2
3Alkali metal (Li, Na, K) phenoxides are frequently used to react
Published on the web (Advance View) October 6, 2003; DOI 10.1246/cl.2003.1006