1,2,3,4-Tetraethynylbenzene as a template for cobalt-catalyzed alkyne cocyclizations: Synthesis of 2,3,8,9-tetrakis(trimethylsilyl) angular [3]phenylene and bent [5]phenylene (benzo[1″,2″:3,4;3″, 4″:3′,4′]dicyclobuta[1,2-b:1′,2′-b′] bisbiphenylene)

Article abstract of DOI:10.1055/s-2006-947329

An approach to the synthesis of bent phenylenes is described, which features 1,2,3,4-tetraethynylbenzene as starting material. Application of this synthon in CpCo-catalyzed alkyne cocyclizations allows the total synthesis of bent [5]phenylene, the seventh

Full text of DOI:10.1055/s-2006-947329

LETTER
2981
1,2,3,4-Tetraethynylbenzene as a Template for Cobalt-Catalyzed Alkyne
Cocyclizations: Synthesis of 2,3,8,9-Tetrakis(trimethylsilyl) Angular
[3]Phenylene and Bent [5]Phenylene (Benzo[1¢¢,2¢¢:3,4;3¢¢,4¢¢:3¢,4¢]dicyclo-
buta[1,2-b:1¢,2¢-b¢]bisbiphenylene)
T
emplate for Coba
e
lt-Catalyze
b
d
A
lkyne
C
o
r
c
yclizati
a
ons L. Mohler,a Sriram Kumaraswamy,a Amnon Stanger,b K. Peter C. Vollhardt*a
a
Department of Chemistry, University of California at Berkeley, Berkeley, California 94720-1460, USA
Fax +1(510)6435208; E-mail: kpcv@berkeley.edu
b
Department of Chemistry and The Lise Meitner-Minerva Center for Computational Quantum Chemistry,
Technion-Israel Institute of Technology, Haifa 32000, Israel
Received 6 April 2006
This paper is dedicated to Professor Richard F. Heck in recognition of his pioneering contributions to the application of palladium in
organic synthesis.
SiMe3
Abstract: An approach to the synthesis of bent phenylenes is de-
scribed, which features 1,2,3,4-tetraethynylbenzene as starting ma-
Br
Br
SiMe3
SiMe3
terial. Application of this synthon in CpCo-catalyzed alkyne
cocyclizations allows the total synthesis of bent [5]phenylene, the
seventh isomer of the family of twelve [5]phenylenes to be isolated.
Br
Br
i
ii
iii
Key words: alkynes, coupling, cyclizations, palladium, transition
metals
SiMe3
Since the discovery of a CpCo-catalyzed general synthe-
sis of biphenylenes,1 we have exploited this reaction in the
construction of 19 topological variants of the phenylenes,
a new class of cyclohexatrienoid hydrocarbons in which
benzene rings are fused to cyclobutadienes in an alternat-
ing manner.2 In this context, it is particularly instructive to
compare the properties of the various isomers of a family,
as accomplished in the cases of the two [3]phenylenes3
and the five [4]phenylenes.4 However, these two groups
are limited with respect to the number of possible topolo-
gies, and we have therefore embarked on tackling the syn-
thesis of the entire family of twelve [5]phenylenes.5 We
report the construction of the 7th member of this group,
bent [5]phenylene 1, by an approach that exploits the abil-
ity of 1,2,3,4-tetraethynylbenzene (3) to undergo double
CpCo-catalyzed alkyne cycloadditions.
2
3
SiMe3
SiMe3
iv
SiMe3
SiMe3
4
5
Scheme 1
Reagents and conditions: i) Me3SiCCH (11 equiv),
Pd[P(C6H5)3]2Cl2 (0.12 equiv), CuI (0.05 equiv), Et3N, 110 °C, 36 h,
68%; ii) KF·H2O (50 equiv), 18-crown-6 (0.5 equiv), DME, r.t., 15
min (assumed quantitative); iii) CpCo(CO)2 (2 equiv),
Me3SiCCSiMe3 (cosolvent with DME, 1:1), reflux, 18 h, 56%; iv)
CF3COOH (10 equiv), CHCl3, r.t., 18 h, 86%.
Retrosynthetic analysis of 1 suggested building up the
molecule by an iterative cyclization protocol, first em-
ployed in the construction of the series of linear phen-
ylenes using 1,2,4,5-tetraethynylbenzene as the starting
point,1b,2,3 but commencing with the angular isomer 3.
This alkyne was unknown, but envisaged to be made
readily by Pd-catalyzed ethynylation, as pioneered by
Heck, Cassar, and Sonogashira.6 Indeed, 1,2,3,4-
tetrabromobenzene7 could be coupled efficiently with tri-
methylsilylethyne in a pressure vessel to furnish 2 in 68%
yield (91% per coupling step), as shown in Scheme 1.8
This compound was desilylated in situ to give 3, which
was too unstable to be isolated, but relatively stable in
solution. To test the potential of 3 to undergo double
CpCo-catalyzed cocycloaddition to alkynes, it was con-
verted successfully with bis(trimethylsilyl)acetylene (BT-
MSA) to the new 2,3,8,9,-tetrakis(trimethylsilyl) angular
[3]phenylene (4), a potentially useful synthon for the as-
sembly of larger phenylenes. Protodesilylation furnished
the parent 5.2,3 Overall, this approach to 5 is more conver-
gent and improves on the original synthesis, which fea-
tured an intramolecular triyne cycloisomerization that
generated the central ring.9
On the basis of the preceding model study, the title mole-
cule was targeted through a strategy employed previously
in other phenylene syntheses,2 namely regioselective co-
oligomerization of 3 with the central triple bond of the
SYNLETT 2006, No. 18, pp 2981–2984
Advanced online publication: 04.08.2006
DOI: 10.1055/s-2006-947329; Art ID: S05906ST
© Georg Thieme Verlag Stuttgart · New York
0
3
.1
1
.2
0
0
6

2982
D. L. Mohlera et al.
LETTER
readily available hexatriyne 610 (Scheme 2).8 This process and 529 nm (8), at almost the same energy as those in tet-
created the required tetraethynyl angular [3]phenylene rakis(trimethylsilyl) linear [5]phenylene, and bathochro-
frame in the form of 7 in 37% yield. Desilylation provided mically shifted relative to all other known isomers.2,3 An
the free tetrayne, unlike 3 sufficiently stable to allow its assessment of relative aromaticity can be made on inspec-
1H NMR spectrum to be recorded [(300 MHz, CDCl3): tion of the data in Figure 1, which compares NICS (nucle-
1
d = 6.99 (s, 2 H), 6.94 (s, 2 H), 6.27 (s, 2 H), 3.37 (br s, 4 us independent chemical shift)11 and H NMR data for 1
H)]. Subsequent cocyclization with BTMSA then fur- with those of some illustrative related phenylenes.2,3
nished the desired bent [5]phenylene nucleus in 8 in good
yield. Finally, protodesilylation gave the parent 1, thus
completing the synthesis.
–9.5
6.96
6.89
C
C
A
B
6.24
6.42
B
3.1
–3.3
6.98
6.99
6.63
Si(i-Pr)3
–7.5 7.3 –5.4
6.12
Si(i-Pr)3
Si(i-Pr)3
10
9
6.97
6.90
6.97
D
i
ii
–9.8
2.3
–7.5 7.5 –6.4 2.6 –2.9
6.50
6.46
C
3
+
A
B
6.90
6.68
6.68
6.07
Si(i-Pr)3
6.39
5.94
6.50
Si(i-Pr)3
11
Si(i-Pr)3
6.99
7.01
6 (5 equiv)
7
6.93
6.87
–9.2
4.0
SiMe3
SiMe3
6.29
–4.3
1.5
–6.2
6.51
7.02
iii
–7.5
7.5
–6.5
1.7
–2.5
12
6.50
A
C
6.43
B
6.68
6.68
SiMe3
SiMe3
5.88
6.50
6.40
8
1
1
Figure 1 1H NMR chemical shifts (CDCl3, except for 12, CH2Cl2,
d, ppm) and NICS(1) values (above the respective rings) of 1 and
some reference phenylenes (GIAO-HF/3-21G//B3LYP/6-31G*), and
labeling of the benzene rings (A–D, see discussion).
Scheme 2 Reagents and conditions: i) CpCo(CO)2 (2 equiv), DME,
reflux, 10 h, 37%; ii) a) Bu4N+F– (4 equiv), THF, r.t., 5 min, 100% (1H
NMR), b) CpCo(CO)2 (7 equiv), Me3SiC≡CSiMe3 (cosolvent with
THF, 1:1), reflux, 20 h, 65%; iii) CF3COOH (25 equiv), CHCl3, r.t.,
16 h, 43%.
Taking linear [3]phenylene 9 as a reference, it can be seen
that the NICS and NMR values for the terminal rings in 1
stay unperturbed. The biggest change is observed for ring
C, for which angular fusion enforces an extensively cyclo-
hexatrienoid configuration, more than those of the corre-
sponding rings C in 10 and 11. As a consequence, the
attached protons exhibit the lowest chemical shifts (d =
5.88 ppm, NICS = –2.5, versus 6.63/6.42 ppm, NICS =
–7.5 in 9, and 6.12 ppm, NICS = –3.3 in 10). By relay,
Phenylenes 1 and 8 are red crystalline solids, which de-
compose slowly in solution on exposure to air. The parent
system is very insoluble, precluding the recording of a 13C
NMR spectrum. The electronic spectra are typical of the
phenylenes in their pattern of relatively intense high ener-
gy bands followed by lower energy and weaker absorp-
tions. The longest wavelength bands, reflecting the
respective HOMO–LUMO gaps, are centered at 523 (1)
Synlett 2006, No. 18, 2981–2984 © Thieme Stuttgart · New York

LETTER
Template for Cobalt-Catalyzed Alkyne Cocyclizations
2983
this effect reduces the paratropism of the adjacent four- C in 11 appeared more reactive than B,4 reversing the or-
membered ring drastically, allowing B to become com- der found from similar experiments carried out with 9 and
paratively more aromatic, its protons appearing de- 10. In 1 and 8, this selectivity was expected to be more
shielded (d = 6.43/6.40 ppm, NICS = –6.5) relative to pronounced. Surprisingly, however, exposure of 8 to H2
those in 9 (d = 6.24 ppm, NICS = –5.4). Taking angular (1 atm) and Pd/C in THF for 16 hours left it untouched,
[3]phenylene 10 as a reference, double linear benzo- and conversion required increasing the hydrogen pressure
cyclobutadienofusion to give 1 should amplify bond to 7 atm. Under these conditions, a mixture of products
localization in B in the sense indicated by the resonance ensued, the mass spectral analysis of which revealed addi-
structure chosen for 1 in Figure 1. This change would in- tion of 7–11 molecules of hydrogen, most prominently 9
crease the cyclobutadienoid character of the inside four- equivalents (m/z = 680), indicative of simultaneous hy-
membered rings toward a dicyclobutadienobenzene drogenation and hydrogenolysis. Similarly, but now
core,12 in turn accommodated by an extensively bond al- under atmospheric hydrogen (Pd/C, THF, 10 min), 1 gave
ternating ring C. These arguments are nicely validated by several inseparable products of hydrogenation. In another
inspection of 11, in which only one ‘arm’ of 9 and 10, re- attempt to probe the reactivity of the system, photooxida-
spectively, has been elongated, and which can be regarded tion (8, O2, methylene blue, CH2Cl2) was attempted, in a
as lying halfway along the structural alterations to give 1. manner that leads to selective oxidative cleavage of
In this molecule, the protons of ring C have intermediate the center of 10.15 Again, an intractable mixture was
chemical shifts (d = 6.07/5.94 ppm, NICS = –2.9) be- generated.
tween those of the corresponding nuclei in 10 and 1.
Finally, the all-angular 12 is quite different, its topology
It is clear that bent [5]phenylene is qualitatively more re-
active than comparable phenylenes, perhaps requiring
allowing for a resonance form devoid of any cyclobutadi-
much milder conditions for its transformation to defined
enoid circuits, hence relatively more aromatic internal six-
products. Experiments along those lines will be the
membered rings.2,3 As observed previously in other
subject of future experimentation.
phenylenes,2 there appear to be no long-range effects of
the remote rings on each other in 1.
Acknowledgment
Unfortunately, it was not possible to grow crystals of
This study was aided by the NSF (CHE-0451241). A.S. thanks the
VPR fund and the fund for promotion of research at the Technion
for financial support.
either 1 or 8 suitable for an X-ray diffraction analysis.
However, in view of the extraordinary accuracy of DFT
structural calculations in this series,2 a computed structure
(B3LYP/6-311G*) appeared sufficient for the purposes of
further analysis (Figure 2). The system distorts in the
direction expected on the basis of the above arguments,
albeit moderately so. Thus, the central ring C shows more
bond alternation (66%) than that of 10 (62%),13 whereas
A (29%) displays less than that of reference 9 (33%).14 In
addition, ring B distorts from the symmetrical bisallyl
structure in 9 in the direction that appears to amplify the
dicyclobutadienobenzene formulation of the center (vide
supra).
References and Notes
(1) (a) Berris, B. C.; Lai, Y.-H.; Vollhardt, K. P. C. J. Chem.
Soc., Chem. Commun. 1982, 953. (b) Berris, B. C.;
Hovakeemian, G. H.; Lai, Y.-H.; Mestdagh, H.; Vollhardt,
K. P. C. J. Am. Chem. Soc. 1985, 107, 5670.
(2) For a recent review, see: Miljanič, O. Š.; Vollhardt, K. P. C.
In Carbon-Rich Compounds: From Molecules to Materials;
Haley, M. M.; Tykwinski, R. R., Eds.; Wiley-VCH:
Weinheim, 2006, 140.
(3) Mohler, D. L.; Vollhardt, K. P. C. In Advances in Strain in
Organic Chemistry, Vol. 5; Halton, B., Ed.; JAI Press:
London, 1996, 121.
1
.386
(4) Bong, D. T.-Y.; Gentric, L.; Holmes, D.; Matzger, A. J.;
Scherhag, F.; Vollhardt, K. P. C. Chem. Commun. 2002, 278.
(5) See: Bong, D. T.-Y.; Chan, E. W. L.; Diercks, R.; Dosa, P.
I.; Haley, M. M.; Matzger, A. J.; Miljanič, O. Š.; Vollhardt,
K. P. C.; Bond, A. D.; Teat, S. J.; Stanger, A. Org. Lett.
2004, 6, 2249; and references therein.
(6) (a) Dieck, H. A.; Heck, R. F. J. Organomet. Chem. 1975, 93,
259. (b) Cassar, L. J. Organomet. Chem. 1975, 93, 253.
(c) Sonogashira, K.; Tohda, Y.; Hagihara, N. Tetrahedron
Lett. 1975, 4467.
1.420
1.370
1.506
29%
1.426
1.370
1.385
1.398
1.491
1.400
1.492
1.359
66%
(7) Collins, I.; Suschitzky, H. J. Chem. Soc. C 1969, 2337.
(8) 1,2,3,4-Tetrakis(trimethylsilylethynyl)benzene (2).
Trimethylsilylethyne (900 mg, 9.16 mmol), 1,2,3,4-
tetrabromobenzene (317 mg, 0.805 mmol),
Figure 2 Calculated structure of 1, with bond lengths in Å and the
percent bond alternation in the terminal and central six-membered
rings.
Pd[P(C6H5)3]2Cl2 (69 mg, 0.098 mmol), CuI (7 mg, 0.04
mmol), and Et3N (12 mL) were combined in a glass bomb
with a teflon stopcock. The reaction mixture was placed in a
110 °C sand bath for 36 h, after which it was cooled and the
solvent removed under reduced pressure. Chromatography
of the residue on silica gel with 5% CH2Cl2–hexane followed
With quantities of 1 and 8 in hand, preliminary experi-
ments were executed, aimed at probing their reactivity. In
this respect, catalytic hydrogenation was thought to be
particularly illuminating, in view of the finding that ring
Synlett 2006, No. 18, 2981–2984 © Thieme Stuttgart · New York

2984
D. L. Mohlera et al.
LETTER
by recrystallization from EtOH gave 2 (252 mg, 68%).
Tetrakis(triisopropylsilylethynyl) Angular [3]Phenyl-
ene 7.
287 (4.67), 292 (4.65), 301 (4.88), 310 (4.54), 316 (4.54),
326 (4.90), 366 (3.84), 388 (3.82), 410 (3.73) nm. IR
(CHCl3): 2949, 1402, 1253, 1054, 851, 823, 684, 636 cm–1.
MS (70eV): m/z (rel intensity) = 514(100) [M+], 499(7),
483(9), 442(6), 411(7), 74(8), 73(97). 1H NMR (300 MHz,
CDCl3): d = 7.25 (s, 2 H), 7.18 (s, 2 H), 6.16 (s, 2 H), 0.37
(s, 18 H), 0.35 (s, 18 H). 13C{H} NMR (75 MHz, CDCl3):
d = 150.2, 149.8, 148.6, 148.5, 147.9, 137.3, 125.1, 124.0,
114.6, 2.64, 2.58.
A solution of tetrakis(trimethylsilylethynyl)benzene (2; 221
mg, 0.477 mmol), KF·2H2O (1.95 g, 22.6 mmol), and 18-
crown-6 (50 mg, 0.19 mmol) in DME (20 mL) was allowed
to stir for 15 min at r.t. The green mixture was decanted from
the solids into a 100-mL pear-shaped flask and diluted with
additional DME (30 mL). After flushing with argon,
CpCo(CO)2 (120 mL, 0.942 mmol) was added via syringe.
This solution was injected over an 8 h period to
Compound 7: orange crystals, mp 233 °C. UV/Vis (hexane):
l
max (log e) = 234 (4.41), 260 (sh, 4.64), 279 (4.78), 288
bis(triisopropylsilyl)hexatriyne (6, 853 mg, 2.21 mmol) in
boiling toluene (50 mL). During the addition and for an
additional 10 h, the reaction mixture was kept at reflux and
irradiated with a slide projector lamp. The solution was
cooled to r.t., the toluene removed by rotary evaporation, and
the residue subjected to chromatography on silica gel,
starting with hexane and gradually changing to 4% CH2Cl2–
hexane, eluting initially some 6. Further elution of an orange
band, removal of the solvents, and recrystallization from
CH2Cl2–EtOH (2:1) gave 7 (165 mg, 37%).
(4.80), 314 (4.73), 329 (4.94), 361 (4.91), 412 (3.91), 430
(3.76), 440 (3.75) nm. IR (CHCl3): 2930, 2872, 2149, 1470,
1418, 1321, 1020, 993, 891, 685, 650 cm–1. MS (70 eV):
m/z (rel. intensity) = 948(16) [M+], 947(16), 401(16),
400(35), 399(38), 323(18), 322(45), 321(100). 1H NMR
(300 MHz, CDCl3): d = 7.00 (s, 2 H), 6.92 (s, 2 H), 6.22 (s,
2 H), 1.12 (br s, 42 H), 1.11 (br s, 42 H). 13C{H} NMR (75
MHz, CDCl3): d = 148.9, 148.5, 147.3, 135.7, 126.4,125.7,
123.3, 122.3, 115.8, 106.1, 106.0, 96.9, 96.6,
18.81,18.79,11.4, 11.3.
Tetrakis(trimethylsilyl) Bent [5]Phenylene 8.
Compound 8: square red plates, mp >250 °C. UV/Vis
(hexane): lmax (log e) = 229 (4.48), 283 (4.83), 302 (4.85),
362 (4.34), 379 (4.48), 403 (4.70), 430 (5.11), 495 (4.07),
529 (3.72) nm. IR (CHCl3): 2942, 2864, 1339, 1251, 1084,
996, 836 cm–1. MS (70 eV): m/z (rel. intensity) = 662(61)
[M+], 590(12), 460(10), 400(36), 399(47), 323(20), 322(42),
321(100), 262(22), 261(20). 1H NMR (400 MHz, CDCl3):
d = 6.84 (s, 4 H), 6.47 (d, J = 1.5 Hz, 2 H), 6.44 (d, J = 1.5
Hz, 2 H), 5.89 (s, 2 H), 0.314 (s, 18 H), 0.311 (s, 18 H).
13C{H} NMR (125 MHz, CDCl3): d = 153.7, 153.2, 150.7,
149.44, 149.36 (2 C), 148.4, 148.3, 146.5, 131.0, 121.7,
121.6, 111.3, 110.7, 110.4, 2.10 (2 C).
To a solution of 7 (116 mg, 0.122 mmol) in THF (15 mL)
was added Bu4N+F– (0.52 mL of a 1.0 M solution in THF,
0.52 mmol). The mixture was allowed to stir for 30 min, then
treated with activated charcoal and filtered through silica
gel, washing the plug with additional THF (10 mL). The
combined solutions were diluted with more THF (25 mL),
and the mixture was placed in a 100-mL pear-shaped flask
and flushed with argon. To this was added CpCo(CO)2 (110
mL, 0.863 mmol) and the solution injected over an 8 h period
into a 250 mL round-bottom flask containing boiling
bis(trimethylsilyl)acetylene (50 mL) under argon. During
the addition and for an additional 12 h, heat and irradiation
with a slide projector lamp were maintained. After cooling to
r.t., the THF was removed by rotary evaporation, and the
excess bis(trimethylsilyl)acetylene was recovered by
vacuum transfer. The remaining solid was subjected to
chromatography on silica with 5% CH2Cl2–hexane,
followed by recrystallization from THF (by evaporation at
2 °C) to give 8 (52 mg, 65%).
Compound 1: red powder, mp >250 °C. UV/Vis (CH2Cl2):
l
max (log e) = 280 (4.78), 293 (4.80), 354 (4.26). 373 (4.34),
398 (sh, 4.64), 424 (5.09), 488 (4.04), 523 (3.86) nm. IR
(KBr): 3040, 2935, 1245, 807, 722 cm–1. MS (70 eV): m/z
(rel. intensity) = 374(100) [M+], 373(13), 372(34), 371(11),
370(20), 187(19), 186(15), 185(13). 1H NMR (400 MHz,
CDCl3): d = 6.68 (m, 4 H), 6.50 (m, 4 H), 6.43 (s, 2 H), 6.40
(s, 2 H), 5.88 (s, 2 H).
Bent [5]Phenylene 1.
(9) Diercks, R.; Vollhardt, K. P. C. Angew. Chem., Int. Ed. Engl.
1986, 25, 266.
In a 25-mL round-bottom flask, 8 (24.2 mg. 0.0366 mmol),
CF3COOH (100 mL, 0.88 mmol), and CHCl3 (5 mL) were
combined and allowed to stir for 16 h, generating an initial
crop of 1 (3.5 mg) as a red precipitate that was filtered off.
The filtrate was diluted with hexane (30 mL) and Et2O (10
mL), washed with 10% aq NaOH (30 mL) and H2O (50 mL),
and dried over MgSO4 to give an additional crop of 1 (total
of 5.9 mg, 43%).
(10) Rubin, Y.; Lin, S. S.; Knobler, C. B.; Anthony, J.; Boldi, A.
M.; Diederich, F. J. Am. Chem. Soc. 1991, 113, 6943.
(11) For a recent review, see: (a) Chen, Z.; Wannere, C. S.;
Corminboeuf, C.; Puchta, R.; Schleyer, P. v.R. Chem. Rev.
2005, 105, 3842. For relevant NICS data, see:
(b) Schulman, J. M.; Disch, R. L.; Jiao, H.; Schleyer, P. v.R.
J. Phys. Chem. A 1998, 102, 8051. (c) See ref. 5.
(12) For a recent reference to this problem, see: Soncini, A.;
Havenith, R. W. A.; Fowler, P. W.; Jenneskens, L. W.;
Steiner, E. J. Org. Chem. 2002, 67, 4753.
All new compounds gave satisfactory analytical and spectral
data, with the exception of unstable 3. For selected data,
compound 2: white flakes, mp 128 °C; UV/Vis (hexane):
l
max (log e) = 227 (3.83), 258 (sh, 4.71), 265 (4.75), 273
(13) As defined in: Beckhaus, H.-D.; Faust, R.; Matzger, A. J.;
Mohler, D. L.; Rogers, D. W.; Rüchardt, C.; Sawhney, A. K.;
Verevkin, S. P.; Vollhardt, K. P. C.; Wolff, S. J. Am. Chem.
Soc. 2000, 122, 7819.
(14) Schleifenbaum, A.; Feeder, N.; Vollhardt, K. P. C.
Tetrahedron Lett. 2001, 42, 7329.
(15) Kumaraswamy, S.; Jalisatgi, S. S.; Matzger, A. J.; Miljanič,
O. Š.; Vollhardt, K. P. C. Angew. Chem. Int. Ed. 2004, 43,
3711.
(4.86), 287 (4.69), 296 (4.47), 307 (4.74) nm. IR (CHCl3):
3004, 2957, 2891, 2141, 1461, 1387, 1251, 1013, 945, 873,
833, 695, 641 cm–1. MS (70 eV): m/z (rel. intensity) = 462(5)
[M+], 428(6), 355(9), 327(9), 239(20), 131(20), 73(100). 1H
NMR (300 MHz, CDCl3): d = 7.29 (s, 2 H), 0.27 (s, 18 H),
0.24 (s, 18 H). 13C{H} NMR (75 MHz, CDCl3): d = 131.3,
128.4, 125.9, 103.6, 102.6, 101.1, 100.6, 0.09, –0.03.
Compound 4: bright yellow crystals, mp 190 °C. UV/Vis
(hexane): lmax (log e) = 228 (4.61), 245 (4.52), 277 (4.33),
Synlett 2006, No. 18, 2981–2984 © Thieme Stuttgart · New York