Chemistry Letters 2000
547
time to prepare alkoxides such as lithium 1-hexanolate. iii) If
the resulting pcs are soluble in THF, they can be separated easi-
ly from sodium methoxide and/or metal salts present in solution
by filtration. Conversely, if the pcs are insoluble in THF, they
can be collected by filtration and the residual sodium methoxide
and/or metal salts can be easily washed out with water.
VNN is grateful to JSPS for financial support (grant
P98418). This work was partially financed by a Grant-in-Aid
for Scientific Research (B) No. 11440192 and that on Priority
area "Creation of Delocalized Electron Systems" 11133206
from the Ministry of Education, Science, Sports, and Culture,
Japan to NK.
References and Notes
1
2
C. C. Leznoff, in "Phthalocyanines: Properties and
Applications," ed. by C. C. Leznoff and A. B. P. Lever,
VCH, NY (1989), Vol. 1, Chap. 1.
H. Tomoda, S. Saito, and E. Hibiya, Chem. Lett., 1976,
1003; S. Greenberg, A. B. P. Lever, and C. C. Leznoff,
Can. J. Chem., 66, 1059 (1988); J. G. Young and W. Onye-
buagu, J. Org. Chem., 55, 2155 (1990).
spectrometry, as well as gel permeation chromatography. The
MALDI-TOF MS spectrum shown in Figure 1 clearly indicates
that essentially only one macrocyclic product, i.e. pc(CN)8H2 is
present. Signals of planar di-, tri-, or higher condensation prod-
ucts of pc(CN)8H2, as well as carboxylic acid-containing pcs,
were absent or negligibly small.6 The results of gel-permeation
chromatography also support the MS data: only one blue-green
pc band was observed on Bio-beads, SX-2 column (Bio-rad),
which was pc(CN)8H2. This high selectivity for the formation
of pc(CN)8H2 can be explained by taking into consideration the
solubility of sodium methoxide in THF. Thus, in a separate
experiment, we found that the solubility of sodium methoxide
in dry THF was ca. 10-4 mol/L at 20 °C and therefore the ratio
of 1,2,4,5-tetracyanobenzene : sodium methoxide in THF solu-
tion was ca. 1000:1 (mol/mol). Such a low concentration of the
base is obviously not enough for nucleophilic attack on cyano
groups not only in pc(CN)8H2, but also in 1,2,4,5-tetra-
cyanobenzene. On the other hand, it is considered that the pc
core formation is probably occurring on the surface of solid
sodium methoxide, since the yields of pc formation decreased
when the stirring speed was reduced.7
Now we may need to explain why the sodium methoxide
surface is much more selective for the activation of cyano
groups in 1,2,4,5-tetracyanobenzene than those in pc(CN)8H2
(present in the reaction solution as a dianion). This kind of acti-
vation can be explained as a multistep process which consists,
at least, of the adsorption of 1,2,4,5-tetracyanobenzene or
pc(CN)8H2 onto the negatively charged surface of a methoxide
anion and the nucleophilic attack of the methoxide anion on the
cyano groups. The adsorption energy between the negatively
charged surface (in this case sodium methoxide) and polar
organic molecules depends predominantly on coulombic attrac-
tive interaction, which is roughly proportional to the positive
values of the molecular electrostatic potential (VMEP) of organic
molecues. The VMEP values for 1,2,4,5-tetracyanobenzene and
the pc(CN)8H2 dianion calculated at the PM3 semi-empirical
level,8 are 16.95 and 10.23 eV, respectively, indicating that the
adsorption energy of the pc(CN)8H2 dianion is probably less
than that for 1,2,4,5-tetracyanobenzene.
3
4
5
C. C. Leznoff, M. Hu, and J. M. Nolan, Chem. Commun.,
1996, 1245; C. C. Leznoff, A. M. D'ascanio, and S. Z. Yil-
diz, J. Porphyrins Phthalocyanines, 4, 103 (2000).
H. Tomoda, S. Saito, S. Ogawa, and S. Shiraishi, Chem.
Lett., 1980, 1277; T. G. Linssen, K. Dürr, M. Hanack, and
A. Hirsch, J. Chem. Soc., Chem. Commun., 1995, 103.
Typical procedure: To a solution of 100 mg (5.62 x 10-4
mol) of 1,2,4,5-tetracyanobenzene in 5 mL of dry THF,
was added 90 mg (1.66 x 10-3 mol) of sodium methoxide.
The resulting suspension was stirred for 5 days at room
temperature, and then the solution was acidified by acetic
acid to neutral. The residue of pc(CN)8H2 was filtered,
washed with 2-propanol and precipitated from DMF solu-
tion by adding water. Yield 51 mg (50%). UV-VIS (λ, nm,
only peak positions are given because of the strong aggre-
gation of pc(CN)8H2 in solution): 712, 662, 428sh, 406
(acetone); 696, 648, 629, 426, 406 (DMF); IR (KBr, cm-1):
2210 (νCN); m/z: (MALDI-TOF+, dithranol): 715 ([M+1]+,
100%).
In Figure 1, small mass peaks are observed at ca. m/z =
580, 950, and 1100. These are corresponding to the frag-
mentation of pc(CN)8H2, M+ + dithranol (matrix), and M+
+ 2 x dithranol. The degree of metallation may depend on
the amount of metal salt present in solution. When ca.
25–30 mg of cuprous chloride or zinc acetate was added to
the above pc(CN)8H2 solution in ref. 5, no metal-free
species was detected from the solution after a 3-day reac-
tion by mass spectroscopy.
6
7
8
One referee pointed out that the solubility of NaOMe may
increase during the reaction, since the reaction time is long.
We recovered, however, almost all NaOMe used for the
reaction when the reaction solution was filtered through a
membrane filter. Since the pore size of this filter was 3 µm,
it may be said that most NaOMe was present as solid in the
reaction.
The room-temperature pc synthesis in THF has several
advantages over previously known methods. i) This method is
applicable to precursors containing nucleophile-labile and/or
high temperature-labile substituent groups. ii) We can spare
All calculations were performed using the HyperChem 5.1
program (HyperCube Inc.).