Nonplanar aromatic compounds. 8.1 synthesis, crystal structures, and aromaticity investigations of the 1,n-dioxa[n](2,7)pyrenophanes. How does bending affect the cyclic π-electron delocalization of the pyrene system?

Article abstract of DOI:10.1021/jo0206059

A series of 1,n-dioxa[n](2,7)pyrenophanes (n = 7-12) with increasingly nonplanar pyrene moieties was synthesized by a 9-10 step sequence starting from 5-hydroxyisophthalic acid. The crystal structure of each member of this series was determined crystallographically. Several spectroscopic properties were found to vary with the extent of the nonplanarity of the pyrene unit. The way in which the distortion from planarity of the pyrene system influences its π-electron delocalization was investigated by using two quantitative descriptors of aromaticity based on geometry (HOMA) and magnetism (magnetic susceptibility and NICS). Both methods suggest that the aromaticity of the pyrene moiety is diminished only slightly upon increasing the bend angle θ from 0° to 109.2°.

Full text of DOI:10.1021/jo0206059

Non p la n a r Ar om a tic Com p ou n d s. 8.1 Syn th esis, Cr ysta l Str u ctu r es,
a n d Ar om a ticity In vestiga tion s of th e
1,n -Dioxa [n ](2,7)p yr en op h a n es. How Does Ben d in g Affect th e
Cyclic π-Electr on Deloca liza tion of th e P yr en e System ?
Graham J . Bodwell,*,† J ohn N. Bridson,† Michal K. Cyran˜ski,*,§ J ason W. J . Kennedy,†
Tadeusz M. Krygowski,§ Michael R. Mannion,† and David O. Miller†
Department of Chemistry, Memorial University of Newfoundland, St. J ohn’s, NL, Canada, A1B 3X7, and
Department of Chemistry, University of Warsaw, Pasteura 1, 02-093 Warsaw, Poland
gbodwell@mun.ca
Received September 16, 2002
A series of 1,n-dioxa[n](2,7)pyrenophanes (n ) 7-12) with increasingly nonplanar pyrene moieties
was synthesized by a 9-10 step sequence starting from 5-hydroxyisophthalic acid. The crystal
structure of each member of this series was determined crystallographically. Several spectroscopic
properties were found to vary with the extent of the nonplanarity of the pyrene unit. The way in
which the distortion from planarity of the pyrene system influences its π-electron delocalization
was investigated by using two quantitative descriptors of aromaticity based on geometry (HOMA)
and magnetism (magnetic susceptibility and NICS). Both methods suggest that the aromaticity of
the pyrene moiety is diminished only slightly upon increasing the bend angle θ from 0° to 109.2°.
In tr od u ction
character clearly manifests itself at points on an energy
surface other than the bottom of the well and the
question becomes one of degree.
Aromaticity is a very important concept of “immense
practical importance”,2b as judged by the intense scrutiny
it has received directly and in association with a very
broad range of chemical investigations.2
Within this context, considerable effort has been
invested over the past few decades in attempting to
understand how deviation from planarity might affect the
π-electron delocalization commonly described by aroma-
ticity.4 [n]Paracyclophanes,5a,6 [n]metacyclophanes,5b,6
and other small cyclophanes5c,6 have played a large role
in such studies because they provide access to series of
compounds that contain increasingly distorted benzene
rings. A shortcoming of this work is that experimentally
determined structural information is not available for
most of the more distorted members of these series. As a
result, it has been necessary to make recourse to calcula-
tions for structural information. Furthermore, work in
Planarity is a structural feature that students of
organic chemistry are taught at an early stage to associ-
ate with systems that are dubbed “aromatic”. This
geometrical arrangement normally corresponds to the
bottom of an energy well and, sadly, this is often the limit
of discussion at the introductory level. As with many
other fundamentals of organic chemistry, such focus on
the most energetically favorable geometry can lead to
confusion and misconceptions at later stages. Thus the
danger exists that the term “nonplanar aromatics” might
be viewed as an oxymoron. In fact, a good number of
systems containing nonplanar benzenoid rings are known
and the overwhelming majority of these exhibit proper-
ties that are routinely linked to aromaticity.3 Aromatic
(4) (a) Dijkstra, F.; van Lenthe, J . H. Int. J . Quantum Chem. 1999,
74, 213-221. (b) Grimme, S. J . Am. Chem. Soc. 1992, 114, 10542-
10547. (c) J enneskens, L. W.; Louwen, J . N.; de Wolf, W. H.; Bickel-
haupt, F. J . Phys. Org. Chem. 1990, 3, 295-300.
(5) (a) Ma, B.; Sulzbach, H. M.; Remington, R. B.; Schaefer, H. F.,
III J . Am. Chem. Soc. 1995, 117, 8392-8400 and references therein.
(b) van Eis, M. J .; de Wolf, W. H.; Bickelhaupt, F.; Boese, R. J . Chem.
Soc., Perkin Trans. 2 2000, 793-801. (c) Schuchmann, P.; Hafner, K.
Tetrahedron Lett. 1995, 36, 2603-2606.
(6) (a) Smith, B. H. Bridged Aromatic Compounds; Academic
Press: New York, 1964. (b) Cyclophanes; Keehn, P. M., Rosenfeld, S.
M., Eds.; Academic Press: New York, 1983; Vols. 1 and 2. (c) Top.
Curr. Chem. 1983, 113, 115. (d) Diederich, F. Cyclophanes; Royal
Society of Chemistry: London, UK, 1991; (e) Vo¨gtle, F. Cyclophane
Chemistry; Wiley: New York, 1993. (f) Top. Curr. Chem. 1994, 172.
(g) Kane, V. V.; de Wolf, W. H.; Bickelhaupt, F. Tetrahedron 1994, 50,
4575-4622. (h) Bodwell, G. J . Angew. Chem., Int. Ed. Engl. 1996, 35,
2085-2088. (i) de Meijere, A.; Ko¨nig, B. Synlett 1997, 1221-1232. (j)
Bodwell G. J . In Organic Synthesis Highlights; Schmalz, H. G., Ed.;
Wiley-VCH: New York, 2000;,Vol. IV, pp 289-300. (k) Hopf, H.
Classics in Hydrocarbon Chemistry; Wiley-VCH: Weinheim, Germany,
2000.
* To whom correspondence should be addressed. G.J .B.: fax (709)-
737-3702. M.K.C: fax (+48) 22 822 28 92; e-mail chamis@chem.
uw.edu.pl.
† Memorial University of Newfoundland.
§ University of Warsaw.
(1) Paper 7 in this series: Bodwell, G. J .; Bridson, J . N.; Chen, S.-
L.; Li, J . Eur. J . Org. Chem. 2002, 243-249.
(2) (a) Krygowski, T. M.; Cyran˜ski, M.; Czarnocki, Z.; Ha¨felfinger,
G.; Katritzky, A. R. Tetrahedron 2000, 56, 1783-1796. (b) Chem. Rev.
2001, 101, 1115-1566, thematic issue on aromaticity. (c) Schleyer, P.
v. R.; J iao, H. Pure Appl. Chem. 1996, 68, 209-218. (d) Minkin, V. I.;
Glukhovtsev, M. N.; Simkin, B. Ya. Aromaticity and Antiaromaticity-
Electronic and Structural Aspects; Wiley: New York, 1994. (e) Garratt,
P. J . Aromaticity; McGraw-Hill: London, UK, 1971. (f) Cyran˜ski, M.
K.; Krygowski, T. M.; Katritzky, A. R.; Schleyer, P. v. R. J . Org. Chem.
2002, 67, 1333-1338. (g) Schleyer, P. v. R. Chem. Rev. 2001, 101,
1115-1117.
(3) See e.g.: Buhl, M.; Hirsch A. Chem. Rev. 2001, 101, 1153-1184.
10.1021/jo0206059 CCC: $25.00 © 2003 American Chemical Society
Published on Web 02/15/2003
J . Org. Chem. 2003, 68, 2089-2098
2089

Bodwell et al.
this area has been limited almost exclusively to systems
that contain just a lone benzene ring. To the best of our
knowledge, there has been no prior comprehensive,
systematic experimental or theoretical study of the effect
of distortion from planarity on the aromatic character of
any polycyclic aromatic system.7 In view of the strong
current interest in nonplanar polycyclic aromatic mol-
ecules such as Buckybowls8 and fullerenes,3,9 the results
of such a study could be particularly relevant, especially
if X-ray crystal structures could be obtained.
Very recently we reported the syntheses of a variety
of [n](2,7)pyrenophanes (1b,c, 2b-d , 3, 4) in which the
pyrene moiety is strongly distorted from planarity.10
Pyrenophanes 1a and 2a have not yet been isolated.
X-ray crystal structures were determined for most of
these pyrenophanes and this opened the door to an
investigation of the effect of bending on the aromaticity
of the pyrene system, which manifests itself in the
surfaces of several fullerenes (e.g. D5h-C70,11 D5d-C80,12 D6h-
C8413) and Buckybowls (e.g. pinakene14). Having observed
that the parent [n](2,7)pyrenophanes are the most dif-
ficult pyrenophanes to crystallize in a form suitable for
X-ray crystallography,10c the 1,n-dioxa[n](2,7)pyrenophanes
1 were selected for the present investigation. We now
report the synthesis of a series of 1,n-dioxa[n](2,7)-
pyrenophanes 1b-g and their X-ray crystal structures.
The structural information serves as an excellent starting
point for further analysis of the extent of changes of the
cyclic π-electron delocalization in the pyrene system as
it becomes increasingly nonplanar. This is done here
through the application of quantitative measures of
aromaticity: HOMA15 (Harmonic Oscillator Model for
Aromaticity, a geometry-based index) and NICS16 (Nuc-
leus-Independent Chemical Shift, a magnetism-based
index). Of the many easily accessible quantitative defini-
tions of aromaticity,2d it was anticipated2f,g that these two
models would be the most efficient in accurately deducing
the extent of stabilization due to the cyclic delocalization
effect. Additionally, the values of magnetic susceptibility
calculated for deformed pyrenes can also be used as an
independent indicator of the extent of the changes of
π-electron delocalization involved.17
(7) Except for naphthalene, for which geometry-based investigations
have been performed (see: (a) Krygowski, T. M. Chem. Inf. Comput.
Sci. 1993, 33, 70-78. (b) Krygowski, T. M.; Cyran˜ski, M. K. In
Theoretical and Computational Chemistry; Parkanyi, C, Ed.; Elsevi-
er: New York, 1998; Vol. 5) and fullerenes (see (c) Reference 3. (d)
Krygowski, T. M.; Ciesielski, A. J . Chem. Inf. Comput. Sci. 1995, 35,
1001-1003).
(8) (a) Rabideau, P. W.; Abdourazek, A. H.; Folsom, H. E.; Marcinow,
Z.; Sygula, A.; Sygula, R. J . Am. Chem. Soc. 1994, 116, 7891-7892.
(b) Mehta, G.; Rao, H. S. P. Tetrahedron 1998, 54, 13325-13370. (c)
Mehta, G.; Rao, H. S. P. In Advances in Strain In Organic Chemistry;
Halton, B., Ed.; J AI: London, UK, 1997; Vol. 6, pp 139-187. (d)
Rabideau, P. W.; Sygula, A. Acc. Chem. Res. 1996, 29, 235-242. (e)
Siegel, J . S.; Seiders, T. J . Chem. Br. 1995, 313-316. (f) Faust, R.
Angew. Chem., Int. Ed. Engl. 1995, 34, 1424-1432. (g) Rabideau, P.
W.; Sygula, A. In Advances in Theoretically Interesting Molecules;
Thummel, R. P., Ed.; J AI: Greenwich, CT, 1995; Vol. 3, pp 1-36. (h)
Scott, L. T.; Bronstein, H. E.; Preda, D. V.; Ansems, R. B. M.; Bratcher,
M. S.; Hagen, S. Atual. Fis.-Quim. Org. 1999, 11, 138-150. (i) Scott,
L. T.; Bronstein, H. E.; Preda, D. V.; Ansems, R. B. M.; Bratcher, M.
S.; Hagen, S. Pure Appl. Chem. 1999, 71, 209-219. (j) Reisch, H. A.;
Bratcher, M. S.; Scott, L. T. Org. Lett. 2000, 2, 1427-1430. (k) Seiders,
T. J .; Baldridge, K. K.; Siegel, J . S. J . Am. Chem. Soc. 1996, 118, 2754-
2755. (l) Seiders, J . T.; Elliott, E. L.; Grube, G. H.; Siegel, J . S. J . Am.
Chem. Soc. 1999, 121, 7804-7813. (m) Sygula, A.; Rabideau, P. W. J .
Am. Chem. Soc. 1999, 121, 7800-7803. (n) Rabideau, P. W.; Sygula,
A. J . Am. Chem. Soc. 1998, 120, 12666-12667.
(9) (a) Kroto, H. W.; Heath, J . R.; O’Brien, S. C.; Curl, R. F.; Smalley,
R. E. Nature 1985, 318, 162-163. (b) Kra¨tschmer, W.; Lamb, L. D.;
Fostiropoulos, K.; Huffman, D. R. Nature 1990, 347, 354-358. (c)
Scrivens, W. A.; Bedworth, P. V.; Tour, J . M. J . Am. Chem. Soc. 1992,
114, 7917-7919. (d) Hirsch, A. The Chemistry of the Fullerenes;
Thieme: New York, 1994.
(10) (a) Bodwell, G. J .; Miller, D. O.; Vermeij, R. J . Org. Lett. 2001,
3, 2093-2096. (b) Bodwell, G. J .; Fleming, J . J .; Miller, D. O.
Tetrahedron 2001, 57, 3577-3585. (c) Bodwell, G. J .; Fleming, J . J .;
Mannion, M. R.; Miller, D. O. J . Org. Chem. 2000, 65, 5360-5370. (d)
Bodwell, G. J .; Bridson, J . N.; Houghton, T. J .; Kennedy, J . W. J .;
Mannion, M. R. Chem. Eur. J . 1999, 5, 1823-1827. (e) Bodwell, G. J .;
Bridson, J . N.; Houghton, T. J .; Kennedy, J . W. J .; Mannion, M. R.
Angew. Chem., Int. Ed. Engl. 1996, 35, 1320-1321.
Discu ssion
Syn th esis. Our approach to the [n](2,7)pyrenophane
skeleton is based on the comprehensively studied valence
isomerization between [2.2]metacyclophane-1,9-dienes 5
and 10b,10c-dihydropyrenes 6 (Scheme 1)18 and the
observation that when the internal substituents are
hydrogen atoms (R ) H), dehydrogenation of trans-6
occurs readily to give pyrene 7.19 Presumably due to the
length of the standard synthetic routes to [2.2]metacy-
clophane-1,9-dienes,20 this valence isomerization/dehy-
drogenation (VID) protocol does not appear to have been
utilized in any target oriented synthesis of a pyrene
derivative prior to our involvement in this area. By
introducing a third bridge of variable length between the
5 and 13 positions of the metacyclophanediene system
5, we have been able to demonstrate that the parent
(11) (a) Nikolaev, A. V.; Dennis, T. J . S.; Prassides, K.; Soper, A. K.
Chem. Phys. Lett. 1994, 223, 143-148. (b) van Smaalen, S.; Petricek,
V.; de Boer, J . L.; Dusek, M.; Verheijen, M. A.; Meijer, G. Chem. Phys.
Lett. 1994, 223, 323-328. (c) Balch, A. L.; Catalano, V. J .; Lee, J . W.;
Olmstead, M. M.; Parkin, S. R. J . Am. Chem. Soc. 1991, 113, 8953-
8955. (d) Bu¨rgi, H. B.; Venugopalan, P.; Schwarzenbach, D.; Diederich,
F.; Thilgen, C. Helv. Chim. Acta 1993, 76, 2155-2159.
(15) (a) Krygowski, T. M.; Cyran˜ski, M. Tetrahedron 1996, 52, 1713-
1722. (b) Krygowski, T. M. Chem. Inf. Comput. Sci. 1993, 33, 70-78.
(16) (a) Schleyer, P. v. R.; Maerker, C.; Dransfield, A.; J iao, H.;
Hommes, N. v. E. J . Am. Chem. Soc. 1996, 118, 6317-6318. (b)
Subramanian, G.; Schleyer, P. v. R.; J iao, H. Angew. Chem., Int. Ed.
Engl. 1996, 35, 2638-2641.
(12) Wang, C.-R.; Sugai, T.; Kai, T.; Tomiyama, T.; Shinohara, H.
J . Chem. Soc., Chem. Commun. 2000, 557-558.
(17) (a) Dauben, H. J .; Wilson, J . D.; Laity, J . L. J . Am. Chem. Soc.,
1968, 90, 811-813. (b) Krishnan, K.; Bannerjee, Philos. Trans. R. Soc.
London, Ser. A 1935, 234, 265.
(13) (a) Negri, F.; Orlandi, G.; Zerbetto, F. J . Am. Chem. Soc. 1991,
113, 6037-6040. (b) Manolopoulos, D. E.; Fowler, P. W. J . Chem. Phys.
1992, 96, 7603-7614. (c) Avent, A. G.; Dubois, D.; Pe´nicaud, A.; Taylor,
R. J . Chem. Soc., Perkin Trans. 2 1997, 1907-1910. (d) Cioslowski,
J .; Rao, N.; Moncrieff, D. J . Am. Chem. Soc. 2000, 122, 8265-8270.
(14) J emmis, E. V.; Sastry, G. N.; Mehta, G. J . Chem. Soc., Perkin
Trans. 2 1994, 437-441.
(18) (a) Mitchell, R. H. In Advances in Theoretically Interesting
Molecules; Thummel, R. P., Ed.; J AI Press: London, UK, 1989; Vol. 1,
pp 135-199. (b) Mitchell, R. H. Eur. J . Org. Chem. 1999, 2695-2703.
(19) Mitchell, R. H.; Boekelheide, V. J . Am. Chem. Soc. 1970, 92,
3510-3512.
(20) Mitchell, R. H. Heterocycles 1978, 11, 563-586.
2090 J . Org. Chem., Vol. 68, No. 6, 2003

How Does Bending Affect Cyclic π-Electron Delocalization?
SCHEME 1. P yr en es fr om Meta cyclop h a n ed ien es
SCHEME 2 a
transformation (3 to 5) is well suited to the synthesis of
[n](2,7)pyrenophanes, i.e., 8 to 9.
The syntheses of 1b and 1c have been previously
reported10d,e and pyrenophanes 1d -g were prepared
along the same lines (Scheme 2). The synthesis of a
potential precursor to 1a is also described here. Tet-
raesters 11a ,d -g were prepared from diester 10 and
could be isolated in good yield (64-89%). However, it
proved to be more practical to subject the crude reaction
mixtures to reduction with LiAlH4 and then treat the
resulting crude aluminum salts with HBr/H2SO4 to afford
the tetrabromides 12a ,d -g in 42-65% overall yield from
10. Formation of the dithiacyclophanes 13 (49-66%) was
then achieved with Na2S/Al2O3.21 The yields of 13 are
consistently within the normal range10,21 and there does
not appear to be a detrimental effect on the yield as the
length of the tether increases from six to twelve atoms.
Twofold S-methylation of 13a ,d -g with (MeO)2CHBF4
(Borch reagent22), followed by Stevens rearrangement
gave isomer mixtures 14a ,d -g (22-65%). These products
were not characterized, but rather subjected to immedi-
ate bis(S-methylation) with Borch reagent. In contrast
to the syntheses of 1b and 1c, subsequent reaction of the
bis(methylsulfonium) salts derived from 14d -g with
t-BuOK did not afford the cyclophanedienes 15d -g, but
1
rather gave what appeared (by H NMR analysis of the
crude reaction mixtures) to be ca. 1:1 mixtures of
pyrenophanes 1d -g and dihydropyrenophanes 19d -g.
These compounds coeluted on silica under various condi-
tions and were not separated. Instead the mixtures were
treated with DDQ to give pyrenophanes 1d -g (23-50%).
The formation of 19d -g can be explained by the initial
formation of cyclophanedienes 15d -g, which, due to the
greater flexibility of the cyclophane moiety allowed by
the longer tethers, valence isomerized under the condi-
tions of their formation to give dihydropyrenophanes
16d -g. Three consecutive [1,5]-H shifts lead to the
generation of 19d -g. Loss of H2 from any of the dihy-
dropyrenophanes 16-19d -g results in the production of
1d -g. With its much shorter tether, cyclophanediene 15a
formed in excellent yield (99%) and showed no inclination
toward valence isomerization even when heated in mesi-
tylene at reflux in the presence of DDQ.
a
Reagents and conditions: (a) NaH, R,ω-dibromoalkane, TBAI,
THF, 3 h, reflux; (b) LiAlH4, THF, 14 h, rt; (c) 2:1 48% HBr(aq):
H2SO4, 1 h, 70-90 °C; (d) Na2S/Al2O3, 18% EtOH/CH2Cl2, 14 h,
rt; (e) (MeO)2CHBF4, CH2Cl2,3 h, rt; (f) t-BuOK, THF, 3 h, rt; (g)
t-BuOK, 1:1 THF:t-BuOH, 6 h, rt, then (for 1d -g) DDQ, benzene,
1-6 h, reflux.
X-r a y Cr ysta l Str u ctu r es. The X-ray crystal struc-
tures of 1d -g (Figure 1, Tables 1-3) were determined
and this completed a series of 1,n-dioxa[n](2,7)pyren-
ophanes with bridge lengths ranging from 7 to 12 atoms
(1b-g). Disorder in the bridge was observed for pyreno-
phanes 1e and 1g, which have even numbers of atoms
(21) Bodwell, G. J .; Houghton, T. J .; Koury, H. E.; Yarlagadda, B.
Synlett 1995, 751-752.
(22) For the original preparation of this reagent see: Borch, R. F.
J . Org. Chem. 1969, 34, 627-629. Our procedure is given in the
Supporting Information.
J . Org. Chem, Vol. 68, No. 6, 2003 2091

Bodwell et al.
F IGURE 1. ORTEP representations of 1d -g in the crystal. Thermal ellipsoids are set at 50%.
TABLE 1. Cr ysta llogr a p h ic Da ta for 1d -g
1d
1e
1f
1g
formula
M
C
23H22O2
C
24H24O2
C
25H26O2
C
26H28O2
330.43
orthorhombic
Pnma
colorless
299
9.081(1)
16.938(1)
11.191(1)
344.45
orthorhombic
Cmcm
colorless
299
17.661(3)
11.144(2)
9.120(3)
358.48
orthorhombic
Pnma
colorless
299
9.122(7)
18.536(6)
11.207(7)
372.51
monoclinic
Cc
colorless
299
23.490(2)
7.445(2)
13.692(1)
120.093(5)
2071.6(4)
4
1.194
1731
1683 (0.010)
1135
251
cryst syst
space group
cryst color
temp (K)
a (Å)
b (Å)
c (Å)
â (deg)
V (Å3)
Z
1721.4(3)
4
1.275
1507
1795(1)
4
1.275
923
1895(3)
4
1.256
1949
D
calc (g cm-3
)
total no. of reflns
no. of unique reflns (Rint
no. of observations
no. of variables
no. of reflns/parameters
R, Rw
)
888
150
5.92
0.043, 0.043
2.47
377
67
5.63
0.074, 0.062
2.63
664
125
5.31
0.046, 0.032
1.50
4.52
0.046, 0.042
3.64
GOF
TABLE 2. An gles (d eg) betw een Lea st-Squ a r es P la n es
of Atom s in 1b-g
in their bridges. In the case of 1e, the disorder was the
result of equal occupancies of two enantiomeric forms,
only one shown in Figure 1. As for 1g, the disorder was
localized at one end of the bridge and appears to be a
consequence of looseness in the bridge. Pyrenophane 1g
exhibited no crystallographic symmetry lying in the
general position, whereas 1d , 1e, and 1f lay on a
symmetry element and hence the two halves were sym-
metry related.
The crystallographic data enabled us to perform an
analysis of the effect of the length of the bridge on the
structural features of the pyrene unit. The angles formed
by adjacent planes of atoms23 in the pyrene systems of
1b-g (Table 2) provided a means of quantifying the
degree of bend. The overall bend in the pyrene skeleton
(θ)10b is defined by the angle formed by the planes A and
plane
compd AB BC CD DE EF FG avg AG (θ) θ calc â (deg)
1a
130.4
1ba 18.2 16.0 20.5 20.5 16.1 17.9 18.2 109.2 113.3 8.2, 8.7
1cb 16.3 13.2 16.1 14.8 13.1 14.3 14.6 87.8 94.9 8.8, 8.8
1d
1e
1f
13.1 9.9 13.5 13.5 9.9 13.1 12.2 72.9 77.3 4.6, 4.6
11.5 8.4 9.0 9.0 8.4 11.5 9.6 57.7 61.2 1.3, 1.3
7.5 5.9 6.6 6.6 5.9 7.5 6.7 39.9 42.2 1.9, 1.9
6.6 7.9 3.7 5.3 3.5 7.9 5.8 34.6 31.1 6.3, 7.1
1g
(23) Where a group of three atoms is used, a plane is necessarily
defined. Where a group of four atoms is used, a least-squares plane
was determined mathematically.
a
b
Reference 10d. Reference 10e.
2092 J . Org. Chem., Vol. 68, No. 6, 2003

How Does Bending Affect Cyclic π-Electron Delocalization?
TABLE 3. Bon d An gles (d eg) in th e Br id ges of 1b -g
compd C-O-C (avg) O/C-C-C (range) O/C-C-C (avg)
TABLE 4. 1H Ch em ica l Sh ifts (δ) for 1b-g a n d 21a
1ba
1cb
1d
1e
1f
1g
110.5
110.6
111.8
112.5
113.7
115.0
114.3-117.6
115.0-119.4
115.4-120.3
c
112.9-116.1
c
116.5
117.2
118.3
c
114.6
c
compd
HA
HB
HR
Hâ
Hγ
Hδ
Hꢀ
a
b
Reference 10d. Reference 10e. c Disorder in the bridge pre-
1bb
7.22
(0.63)
7.44
(0.41)
7.64
(0.21)
7.72
(0.13)
7.83
7.72
3.31 -0.04 -2.10
vented determination of these bond angles.
(0.24) (1.00) (1.45)
7.84 3.59 0.10 -1.46
(0.14) (0.72) (1.31)
7.91 3.76 0.71 -1.89 -0.73
(0.05) (0.55) (0.70)
7.92 4.02 0.93 -1.17 -0.67
(0.04) (0.29) (0.48)
7.98 4.19 1.22 -0.27 -0.64 -1.04
1cc
1d
1e
1f
G. The angle â, which is analogous to that used for [n]-
paracyclophanes,5a,6b,e was determined by measuring the
angles O(1)-C(2)-Du(1) and O(2)-C(7)-Du(2), where
Du(1) is a dummy atom at the C(1)-C(3) centroid and
Du(2) is a dummy atom at the C(6)-C(8) centroid (see
Table 2).
(0.02) (-0.02) (0.12) (0.19)
7.85
7.69
1g
21
7.96
7.97
4.31
4.07
1.41 -0.14 -0.14 -0.63
For each of the pyrenophanes, the distortions from
planarity are spread out quite evenly over the pyrene
surface. As expected, the pyrene system becomes increas-
ingly distorted from planarity as the length of the bridge
decreases, as can been seen by the trends in the values
of every interplane angle and θ (Table 2). The only
anomalies occur with the highest homologue 1g (BC and
FG angles are larger than those of 1f), which exhibits a
less symmetric pattern of bend angles than the other
pyrenophanes. This may be a consequence of the dynamic
disorder that was observed at one end of the bridge. The
â angles (Table 2) exhibit the expected trend, albeit a
rather rough one, of diminishing with the degree of
ranging from 8 to 9° for 1b and 1c down to 1-2° for 1e
and 1f. Pyrenophane 1g has anomalously large â angles
(6.3°, 7.1°), which may well be artifacts originating from
the disorder in the tether.
a
Bracketed numbers are the differences between
a
given
b
chemical shift and the analogous shift for 1g. Reference 10d.
c Reference 10e.
three lowest homologues, decreasing from 118.3° in 1d
to 117.2° in 1c to 116.5° in 1b. The higher homologue 1f
exhibited a smaller average angle of 114.6°, which is more
in line with expectations. Unfortunately, data for the
whole series could not be obtained due to disorder in the
bridges of 1e and 1g. In any event, the counterintuitive
trend for 1b-d would suggest that the relative impor-
tance of the factors that determine the distribution of
strain between the pyrene moiety and the bridge does
not remain constant as the overall strain in the molecule
varies.
Sp ectr oscop ic P r op er ties. The signals in 1H NMR
spectra of 1b-g and the comparison compound 2,7-
dimethoxypyrene 21 were assigned by using standard 1D,
2D, and nOe experiments and are listed in Table 4.
Several chemical shift trends with increasing bend in the
pyrene unit were apparent. By using the least strained
pyrenophane 1g as a point of comparison, it can be seen
that for the protons HA, HB, HR, and Hâ, there is a roughly
linear correlation between ∆δ (δHx - δHx(1g)) and the
bend angle θ. In each case the trend is toward a higher
field chemical shift with an increasing value of θ. The
most pronounced effect is for the proton Hâ, which varies
over a range of 1.45 ppm. The ranges for HR, HA, and HB
are 1.00, 0.63, and 0.26 ppm, respectively. The remaining
bridge protons, Hγ, Hδ, and Hꢀ, also show a tendency
toward higher field with increasing θ, but the trends are
not as well defined as for the other protons. This is likely
a consequence of the positioning of these bridge protons
deeper in the shielding zone of the pyrene nucleus, a
region of high magnetic flux, than HR and Hâ.
Although the principal focus of this study is the effect
of the bridge on the pyrene unit, the effect of the pyrene
unit on the bridge should not be overlooked. Since the
strain in the pyrenophanes would be expected to be
distributed over the entire molecular framework, it would
seem reasonable to expect that the bridge should become
increasingly strained as the pyrene unit becomes more
distorted. Examination of models led to the prediction
that the bond angles (C-O-C) about the terminal oxygen
atoms of the bridge should become smaller as the pyrene
system becomes more bent and that the remaining
skeletal bond angles of the bridge (C-C-C) should
become larger. In fact the C-O-C bond angles followed
the expected trend (Table 3), as they decreased from an
average of 115.0° in 1g to an average of 110.5° in 1b. By
comparison, Caryl-O-Calkyl bond angles are normally in
the range of 117 ( 0.7°.24 Surprisingly, the O/C-C-C
angles (angles around the remaining bridge atoms) did
not follow the expected trend. In fact, the average
C-C-C bond angles showed the opposite trend for the
It is particularly interesting to note that the trends
for the aromatic protons of 1b-g (higher field with
increasing bend) are opposite to the situation for the [n]-
paracylophanes (lower field with increasing bend).25
Arguments based on diminished ring current26 and
rehybridization of the aromatic carbon atoms27 with
increasing bend would lead to the prediction of the
(24) Based on 1183 molecular geometries of 885 systems (where R
is <5% and the mean standard deviation of the bond length is <0.005
Å) retrieved from the Cambridge Structural Database. See also: (a)
Allen, F. H.; Davies, J . E.; Galloy, J . J .; J ohnson, O.; Kennard, O.;
McRae Mitchell, E. M.; Mitchell, G. F.; Smith, J . M.; Watson, D. G. J .
J . Chem. Inf. Comput. Sci. 1991, 31, 187-204. (b) Allen, F. H.; Bellard,
S.; Brice, M. D.; Cartwright, B. A.; Doubleday: A.; Higgs, H.; Hum-
melink, T.; Hummelink-Peters, B. G.; Kennard, O.; Motherwell, W.
D. S.; Rodgers, J . R.; Watson, D. G. Acta Crystallogr. 1979, B35, 2331-
2339. (c) For 1,3,5-trimethoxybenzene the C(Phenyl)-O-CH3 angles are
117.4, 117.4, and 118.3°. Howard, S. T.; Krygowski, T. M.; Glo´wka,
M. L. Tetrahedron 1996, 52, 11379-11384.
(25) Agarwal, A.; Barnes, J . A.; Fletcher, J . L.; McGlinchey, M. J .;
Sayer, B. G. Can. J . Chem. 1977, 55, 2575-2581.
J . Org. Chem, Vol. 68, No. 6, 2003 2093

Bodwell et al.
TABLE 5. 13C Ch em ica l Sh ifts (δ) for 1b-g a n d 21
compd
CA
CB
CC
CD
CE
CR
1ba
1cb
1d
1e
1f
1g
21
152.0
153.6
153.6
154.8
155.1
156.9
157.1
123.1
123.2
122.0
121.0
120.3
118.7
110.6
133.4
132.7
132.7
132.3
132.0
131.9
131.5
126.7c
126.9
127.0
127.1
127.2
127.2
127.5
126.3c
127.4
125.3
123.6
122.5
121.8
120.1
76.3
77.7
75.8
76.3
75.3
74.2
55.8
a
b
Reference 10d. Reference 10e. c These signals could not be
assigned unambiguously.
observed trends for the pyrenophanes. Why the observed
trends for the two systems should oppose one another is
unclear. It is, however, important to stress that bending
does not involve dramatic changes of the chemical shifts
of the protons attached to the pyrene moietysthe differ-
ences between the most bent pyrene derivative and the
parent system are at most only 0.63 ppm. This suggests
that the cyclic π-electron delocalization pattern is modi-
fied only moderately. Moreover, the extent of changes of
proton shifts might imply that the π-electron structure
of the outer rings is more influenced by strain as
compared with the structure of the central ones.
13C NMR data for the aromatic and OCH2 carbons of
1b-g are presented in Table 5. These were assigned by
using standard 1D and 2D experiments. Most of the
remaining bridge carbons could not be assigned due to
the near degeneracy of their chemical shifts. It can be
seen that CD is essentially insensitive to changes in the
degree of distortion from planarity of the pyrene system
and that CR shows no clear trend. Movement to lower
field with an increase in bend of the pyrene unit can be
seen for CB (range ) 4.5 ppm), CC (range ) 2.5 ppm),
and CE (range ) 5.6 ppm). In the cases of CB and CE, the
trend appears to taper off, or even reverse direction, as
the bend angle reaches its maximum. Contrary to the
other aromatic carbons, the bridgehead carbon atom (CA)
exhibits a trend to higher field with increasing bend
(range ) 4.9 ppm). We previously observed that CE are
the most pyramidalized carbon atoms in 1b10d and 1c,10e
so it is perhaps not surprising that the widest range of
chemical shifts is observed at this position. The paucity
of 13C NMR data for the [n]paracyclophanes26 rules out
a comparison to the trends observed for pyrenophanes
1b-g.
F IGURE 2. UV/vis spectra of 1b-g.
1
TABLE 6. J C-H Va lu es (Hz) a n d P er cen t s Ch a r a cter
for Som e Ca r bon Atom s of 1c-e
compd
CB
CD
CR
Câ
Cγ
1c
161.0
32.2%
163.7
32.7%
160.5
32.1%
162.7
32.5%
161.9
32.4%
161.1
32.2%
146.3
29.3%
145.8
29.2%
146.4
29.3%
128.0
25.6%
124.3
24.9%
123.8
24.8%
125.3
25.1%
124.9
25.0%
128.8
25.8%
1d
1e
experience significant pyramidalization/rehybridization.
Thus the normal coupling constants and lack of trends
for these carbons is understandable. On the other hand,
bond angles in the bridge do appear to change with the
length of the bridge (Table 3). It is therefore surprising
to see that CR, Câ, and Cγ show little or no variance with
bridge length and that only CR exhibits increased s
character.
As for benzene, Clar has classified the absorption
bands of pyrene according to their behavior with changes
in temperature and solvent.28 A strong absorption at 242
nm (ꢀ ) 8.84 × 104) is called the â′ band and a second
strong absorption at 273 nm (ꢀ ) 5.36 × 104) is called
the â band. A series of three bands at 306, 320, and 336
nm (ꢀ ) 1.25 × 104, 3.23 × 104, 5.58 × 104) are termed
the p bands. Finally, the R bands (352-372 nm) are very
difficult to observe due to their very low extinction
coefficients. The UV/vis spectra of less strained pyreno-
phanes (Figure 2) bear a strong resemblance to that of
pyrene itself, with all of the bands having undergone a
small red shift that would be expected to occur upon
substitution at the 2 and 7 positions. As the bridge
becomes shorter and the pyrene unit becomes more
distorted, a number of systematic changes were observed.
The â′ band is markedly red shifted and the absorption
becomes less intense. The â band experiences only a very
slight red shift. The result of this is that the two bands
1J C-H values were measured for CB, CD, CR, Câ, and Cγ
of 1c, 1d , and 1e (Table 6), but these did not show any
clear trends. Furthermore, the percent s character cal-
culated for these carbon atoms was normal, except at CR,
which exhibited greater s character than expected for an
sp3-hybridized carbon atom. Being at the periphery of the
pyrene system, CB and CD would not be expected to
(26) (a) Allinger, N. L.; Freiberg, L. A.; Hermann, R. B.; Miller, M.
A. J . Am. Chem. Soc. 1963, 85, 1171-1176. (b) Allinger, N. L.; Sprague,
J . T.; Liljefors, T. J . Am. Chem. Soc. 1974, 96, 5100-5104. (c) Mori,
N.; Takemura, T. J . Chem. Soc., Perkin Trans. 2 1978, 1259-1262.
(27) Haddon, R. C. Acc. Chem. Res. 1988, 21, 243-249.
(28) Clar, E. Spectrochim. Acta 1950, 4, 116-121.
2094 J . Org. Chem., Vol. 68, No. 6, 2003

How Does Bending Affect Cyclic π-Electron Delocalization?
F IGURE 4. Cyclic voltammograms of 1b-g.
F IGURE 3. UV/vis spectra of 2b-d .
move closer together such that the â band is observed
only as a shoulder on the â′ band in the spectrum of 1c
and the two bands overlap completely in the spectrum
of 1b. The absorption maximum for this band (280 nm)
is 23 nm red shifted from the â′ band of 1g (257 nm).
The p bands do not appear to be significantly affected by
the changes in the geometry of the pyrene system until
the distortion becomes more pronounced, at which point
there is a small blue shift. There is also a steady decrease
in the intensity of the p bands with increasing distortion
so that, in the spectrum of 1b, they appear merely as
poorly resolved humps. Similar trends are observed for
the [n](2,7)pyrenophanes 2b-d (Figure 3).
The electrochemistry of both pyrene29 and D5h-C70,30
which has a repeating pyrene unit around its equator,
has been studied, so cyclic voltammetry was employed
to ascertain whether fullerene-like behavior would begin
to manifest itself as the degree of bend in the pyrene unit
approached that of the pyrene substructures in fullerenes
such as D5h-C70,11 D5d-C80,12 and D6h-C84.13 Both the
dioxapyrenophanes 1b-g and the pyrenophanes 2b-d
show an irreversible first oxidation, which moves to lower
potential as the bend in the pyrene system increases, i.e.,
oxidation occurs more easily as the aromatic nucleus
becomes more distorted (Figures 4 and 5). This is
consistent with the HOMO becoming higher in energy.
A second oxidation wave is apparent, but there is no clear
trend. By comparison, pyrene undergoes an irreversible
oxidation at +1.54 V29 and D5h-C70 undergoes an oxida-
tion at +1.66 V,which is reversible under some conditions,30b
but irreversible under others.30c Other than a marked
increase in potential close to the switching potential (-2
V), no evidence of reduction is apparent in the negative
scan region. On the other hand, it has been reported that
F IGURE 5. Cyclic voltammograms of 2b-d .
D5h-C70 is already reduced four times by the time a
potential of -1.9 V has been reached.30 On the whole,
none of these results points to the emergence of fullerene-
like behavior as the pyrene unit becomes more distorted.
Comparisons to the [n]paracyclophanes were ruled out
because there does not appear to be any reported study
of their electrochemistry.
Ar om a ticity of th e P yr en op h a n es. To what extent
can strain influence cyclic π-electron delocalization? The
crystallographically determined molecular geometries of
1b,10d 1c,10e 1d -g, pyrene 20,31 and [8](2,7)pyrenophane
2c10c were used as the starting points for optimizations
of the pyrene moieties. To cut off possible experimental
errors and to eliminate the diversity of the data due to
(29) Peover, M. E.; White, B. S. J . Electroanal. Chem. 1967, 13, 93-
99.
(30) (a) Echegoyen, L.; Echegoyen, L. E. Acc. Chem. Res. 1998, 31,
593-601. (b) Xie, Q.; Aria, F.; Echegoyen, L. J . Am. Chem. Soc. 1993,
115, 9818-9819. (c) DuBois, D.; Kadish, K. M.; Flanagan, S.; Wilson,
L. J . J . Am. Chem. Soc. 1991, 113, 7773-7774.
(31) (a) Kai, Y.; Hama, F.; Yasuoka, N.; Kasai, N. Acta Crystallogr.
1978, B34, 1263-1270. (b) Frampton, C. S.; Knight, K. S.; Shankland,
S.; Shankland, K. J . Mol. Struct. 2000, 520, 29-32.
J . Org. Chem, Vol. 68, No. 6, 2003 2095

Bodwell et al.
describes lowering of aromaticity due to elongation of the
bond lengths, whereas the GEO describes the lowering
as a result of an increase of bond alternation. The
advantage of HOMA is that it may be applied as a
descriptor of both local and global aromaticity.36 The
models used here are easily accessible and have been
proved as very efficient in the description of cyclic
π-electron delocalization of many diverse π-electron
systems.2b,f,15,35,36,37
F IGURE 6. Distances and bond designations in the pyrene
As a result of optimizations for each pyrenophane 1b-
g, 2c, and pyrene 20 the total energies of the systems,
Et, were calculated. These model values (Et) varying by
210 kJ /mol are mostly affected by strain, and this is very
well reflected in the good correlation with the bending
parameter h (see Figure 6 -, the bending parameter h
is almost perfectly equivalent to θ) with a correlation
coefficient of R ) 0.962. It is not an easy task to separate
strain from the contribution due to the stabilization
energy attributed to π-electron structure. Therefore the
total energy values in fact are not very useful for the
detailed analysis of the π-electron structure.
Table 7 contains values of HOMA, EN, GEO, and NICS
for the terminal (a) and central (b) rings (Figure 6) of
the pyrene systems, which are ordered in terms of
decreasing distortion from planarity. The HOMA(a)
values of the pyrenophanes span a range of 0.053 (0.078
if pyrene is included), whereas the corresponding HOMA-
(b) values span a considerably larger range (0.146 for the
pyrenophanes and 0.174 if pyrene is included). For both
types of rings, the GEO term is dominant in determining
the HOMA value, which indicates that changes in bond
alternation are more significant than bond elongation as
the degree of bend increases. The π-electron structure of
the rings is not influenced much by strain despite the
fact that the HOMA model may even slightly exaggerate
the changes. This is due to the fact that, in strained
systems, bond lengths calculated as the shortest dis-
tances between the nuclei are more adequately described
by bond paths.38 Moreover, the geometry-based indices
describing π-electron effects may be much more strongly
influenced by the σ-electron framework than the mag-
netic descriptors.39
unit.
the influence of the oxygen neighborhood,32 the pyrene
systems were optimized with the bending parameters (d1,
d2, and d3, see Figure 6) taken from X-ray experiments
to maintain the proper shape of the moiety and implying
C2v and D2h symmetry for pyrene derivatives and the
parent system, respectively. The model geometries were
optimized at the HF/6-31G** level of theory33 and the
resulting structures were used to calculate NICS values,16a
which were computed at the HF/6-31G* level of theory,
magnetic susceptibilities17 at HF/6-311+G** using the
CSGT method, and HOMA values.15 NICS is a magnetism-
based index, which is by definition16a the negative value
of the absolute magnetic shielding in the center of the
ring in question. A negative value of NICS is indicative
of an aromatic system and a positive value points to an
antiaromatic one. In principle, the model describes local
aromaticity, although some attempts toward the global
estimations have also been made.16b,34 The exaltation of
magnetic susceptibility, defined as a difference between
the magnetic susceptibility of a given system and of a
reference nonaromatic model one, is another well-known
descriptor of global aromaticity.2c,d,f,17 In the present work
we compare the extent of changes of magnetic suscepti-
bility with respect to pyrene. On the other hand, HOMA
is a geometry-based index, which is described by eq 1
HOMA ) 1 - R(Ropt - R )2 - (R/n) (R - Ri)2 )
∑
av
av
1 - EN - GEO (1)
, where n is the number of bonds taken into the sum-
mation, R is an empirical constant fixed to give HOMA
) 0 for a model nonaromatic system15 and HOMA ) 1
for the corresponding system in which all bonds are equal
to the optimal value Ropt, which is assumed to be realized
when full delocalization of the π electrons occurs (1.388
Å for CC bonds), Rav is the average bond , and Ri (i ) 1
- n) are the individual bond lengths. The EN term
Interestingly, in contrast to the trends observed for the
HOMA values, the NICS values for the pyrenophanes
vary substantially less in the central (b) rings (range )
1.05 ppm) than in the terminal (a) rings (range ) 1.69
ppm). Both NICS(a) and NICS(b) values exhibit a general
trend of decreasing aromaticity with increasing bend, and
they both correlate well with the degree of bend (h) with
correlation coefficients 0.939 and 0.885, respectively.
Impressively good correlations exist between NICS(a) and
NICS(b), NICS(a) and GEO(b), and NICS(b) and GEO-
(b) (correlation coefficients R ) 0.985, 0.950, and 0.934,
(32) (a) Hammett, L. P. Physical Organic Chemistry, 2nd ed.;
McGraw-Hill: London, UK, 1970. (b) Chapman, N. B.; Shorter, J ., Eds.
Correlation Analysis in Chemistry, Plenum Press: London, UK, 1978.
(33) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgomery, J . A.,
J r.; Stratmann, R. E.; Burant, J . C.; Dapprich, S.; Millam, J . M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J .;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J .; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J . B.; Cioslowski, J .; Ortiz, J . V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J .; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; J ohnson, B. G.; Chen,
W.; Wong, M. W.; Andres, J . L.; Head-Gordon, M.; Replogle, E. S.;
Pople, J . A. Gaussian 98, revision A.7; Gaussian, Inc.: Pittsburgh, PA,
1998.
(35) Patchkovskii, S.; Thiel, W. J . Mol. Model. 2000, 6, 67-75.
(36) Cyran˜ski, M. K.; Krygowski, T. M.; Wisiorowski, M.; Hommes,
v. E. N. J . R. Angew. Chem., Int. Ed. 1998, 37, 177-180.
(37) Krygowski, T. M.; Cyran˜ski, M. In Advances in Molecular
Structure Research; Hargittai, I., Hargittai, M., Eds.; J AI Press:
London, UK, 1997; Vol. 3, pp 227-268.
(38) (a) Bader, R. F. W. Atoms in Molecules: a Quantum Theory;
Oxford University Press: Oxford, UK, 1990. (b) Boese, R.; Blaser, D.;
Billups, W. E.; Haley, M. M.; Maulitz, A. H.; Mohler, D. L.; Vollhardt,
K. P. C. Angew. Chem., Int. Ed. Engl. 1994, 33, 313-317.
(39) (a) Krygowski, T. M.; Pindelska, E.; Cyran˜ski, M. K.; Ha¨felinger,
G. Chem. Phys. Lett. 2002, 359, 158-162. (b) Fowler, P. W.; Hevenith,
R. W. A.; J enneskens, L. W.; Soncini., A.; Steiner, E. Chem. Commun.
2001, 2386-2387.
(34) Schleyer, P. v. R.; Manoharan, M.; J iao, H. J .; Stahl, F. Org.
Lett. 2001, 3, 3643-3646.
2096 J . Org. Chem., Vol. 68, No. 6, 2003

How Does Bending Affect Cyclic π-Electron Delocalization?
TABLE 7. Ar om a ticity In d ices for Com p ou n d s 1b-1g, 2c, a n d 20
HOMA(a)
EN(a)
GEO(a)
NICS(a)
HOMA(b)
EN(b)
GEO(b)
NICS(b)
1b
1c
2c
1d
1e
1f
0.929
0.917
0.906
0.913
0.886
0.876
0.884
0.962
0.001
0.000
0.000
0.000
0.000
0.000
0.000
0.014
0.070
0.083
0.094
0.087
0.114
0.124
0.116
0.034
-12.59
-13.21
-13.46
-13.63
-13.96
-14.27
-14.28
-14.14
0.305
0.392
0.373
0.451
0.399
0.422
0.441
0.479
0.250
0.214
0.227
0.183
0.232
0.222
0.207
0.176
0.444
0.393
0.400
0.366
0.369
0.356
0.353
0.344
-3.77
-4.20
-4.25
-4.47
-4.69
-4.89
-4.82
-4.63
1g
20
TABLE 8. Bon d Len gth s of P yr en e Moiety
TABLE 10. Ben d in g P a r a m eter (h ), En er gies, Globa l
HOMA Va lu es, a n d Ma gn etic Su scep tibilities for 1b-g, 2c
a n d 20a
system
b1
b2
b3
b4
b5
b6
1b
1c
2c
1d
1e
1f
1.375
1.373
1.374
1.372
1.371
1.369
1.372
1.384
1.382
1.380
1.375
1.379
1.376
1.377
1.372
1.391
1.455
1.450
1.451
1.447
1.448
1.446
1.446
1.446
1.335
1.336
1.337
1.336
1.338
1.338
1.338
1.339
1.413
1.414
1.415
1.414
1.418
1.419
1.417
1.411
1.444
1.437
1.438
1.430
1.438
1.436
1.434
1.432
magnetic
h
Et
rel E
suscept.
[Å]
[hartree] [kJ /mol] HOMA EN
GEO [cgs‚ppm]
1b 1.571 611.705797
1c 1.248 611.735838
2c 1.107 611.747755
1d 0.994 611.753260
1e 0.711 611.766989
1f 0.298 611.778325
1g 0.266 611.779988
20 0.000 611.785868
210.2
131.4
100.1
85.6
49.6
19.8
15.4
0.0
0.593 0.042 0.366
-143.8
-148.3
-149.7
-150.8
-153.0
-154.5
-154.5
-154.2
0.633 0.030 0.337
0.618 0.028 0.354
0.659 0.023 0.318
0.631 0.027 0.342
0.640 0.024 0.336
0.648 0.020 0.333
0.696 0.051 0.253
1g
20
TABLE 9. Cor r ela tion Coefficien ts betw een h a n d
b1-b6
a
The experimental value for pyrene (20) is -154.9 (ref 17).
h
b1
b2
b3
b4
b5
b1 -0.2456
b2 -0.1383
0.8495
0.0222
0.2454
b3
0.8793
0.0387
0.1579 -0.8196
b4 -0.9604
b5 -0.2223 -0.8397 -0.7916 -0.3245
b6 0.6179 -0.1673 -0.1330
0.2623
0.8455 -0.4944 0.0849
respectively), indicating that despite different sensitivi-
ties of the rings to changes in the π-electron structure
implied by strain, the aromaticity of one ring follows the
changes in the other in a regular way. No other important
correlation between the indices of the (a) and (b) rings
was observed, which may suggest that the bond lengths
in the two types of rings vary independently with
bending. A Factor Analysis40 (a statistical tool that
decreases the number of variables and transforms them
into orthogonal, i.e. independent vectors) of the bond
lengths (b1-b6, Figure 6, Table 8) and the bending
parameter h was performed and it showed that three
factors describe 97.0% of the total variance. The first is
loaded mostly with bonds constituting the (a) rings
(48.8%), the second is loaded with b3, b4, and h (38.4%),
and the third (9.8%) is loaded primarily with b6 and
slightly with b3. A very similar picture emerged from a
correlation analysis of the bond lengths (b1-b6, Figure
6) and the bending parameter h (Table 9). The bond types
can be clustered into two groups, namely those of the (a)
rings (b1, b2, and b5) and those of the (b) rings (b3, b4,
and b6). As is evident from Table 9, correlations between
bond lengths within clusters are good and those between
bond lengths from different clusters are considerably
worse. The bending parameter h correlates very well with
b3 and b4 and to a lesser extent with b6, showing that
these bond lengths regularly respond to changes in
strain.
F IGURE 7. (a) Dependence between magnetic susceptibility
and bending parameter h (correlation coefficient R ) 0.935).
(b) Dependence between HOMA and bending parameter h
(correlation coefficient R ) 0.783).
in Table 10, along with the bending parameter h and the
total and relative energies. These provide the most
important information concerning the changes in aro-
maticity as a function of bending, which is best illustrated
by the plot of magnetic susceptibility against h, for which
the correlation coefficient is 0.935 (Figure 7). HOMA
correlated markedly worse, probably due to shortcomings
mentioned above, but the correlation coefficient of
0.783 clearly indicates statistical significance41 at the
level of 0.05. This demonstrates that the extent of the
cyclic π-electron delocalization of pyrene drops down in
a regular way with an increase of strain over a broad
The global HOMA values and magnetic susceptibilities
for the pyrene moieties of 1b-g, 2c, and 20 are presented
(40) U¨ berla, K. Faktorenanalyse: Springer-Verlag: Berlin, Ger-
many, 1977.
J . Org. Chem, Vol. 68, No. 6, 2003 2097

Bodwell et al.
range of bend. However, the most important point is that
the π-electron structure is almost untouched by the
substantial increase in strain that accompanies the
increase in bend in going from planar pyrene 20 to the
most bent pyrenophane 1b. HOMA changes by only 0.1
of a unit over this range, while the magnetic susceptibil-
ity varies by 10.4 units, which is even more striking when
considering that the exaltation of magnetic susceptibility
of planar pyrene is 57.3.17 Moreover, insignificantly
higher magnetic susceptibilities for slightly deviated
pyrenes may be a sign that small deviations from
planarity do not involve a dramatic drop in the π-electron
delocalization pattern and weak intermolecular interac-
tions may easily deform it, as is well-known in the case
of benzene.42
Exp er im en ta l Section
Cr ysta l Str u ctu r e Deter m in a tion s. Crystals suitable for
X-ray diffraction experiments were obtained by crystallization
from xylenes solutions. The measurements were performed at
299 ( 2 K on a Rigaku AFC6S diffractometer with graphite-
monochromated Mo KR (0.71073 Å) or Cu KR (1.54056 Å)
radiation. Data reduction and analysis were carried out with
the TEXSAN or teXsan for Windows programs. The data were
corrected for Lorentz and polarization effects. Absorption
correction was applied where appropriate. The structures were
solved by direct methods44 and refined by using TEXSAN or
teXsan for Windows.45 The refinements were based on F for
all reflections with I > 2.00σ(I). The weighted R factor, wR,
and all goodness-of-fit S values are based on F. All non-
hydrogen atoms were refined anisotropically, except for 1g,
where some were refined isotropically. H-atoms were placed
in calculated positions with isotropic thermal parameters set
20% greater than those of their bonding partners and were
not refined, except for 1d , in which they were optimized by
positional refinement. The atomic scattering factors were
taken from the International Tables.46
Thus the answer to the question posed in the title is
that despite substantial changes in the hybridization of
carbon atoms involving changes in the σ-electron struc-
ture of pyrene, the aromaticity of the system decreases
slightly and regularly with increasing bend. However the
changes are not dramatic, even with the great increase
in strain (approximately 210 kJ /mol). That aromaticity
drops off with increasing departure from planarity is to
be expected.43 However, our results provide, for the first
time, a clear indication of the extent to which aromaticity
decreases with increasing nonplanarity.
Cyclic Volta m m etr y. Cyclic voltammograms vs SSCE
were obtained at a scan rate of 100 mV/s from Ar-degassed
CN3CN (distilled from CaH2 directly prior to use) solutions
containing tetraethylammonium perchlorate.
Ackn owledgm en t. This work was supported (G.J.B.)
by the Natural Sciences and Engineering Research
Council (NSERC) of Canada. The Interdisciplinary
Center for Mathematical and Computational Modelling
(ICM, University of Warsaw) is kindly acknowledged for
computational facilities. M.K.C. gratefully acknowledges
Polityka Weekly and PKN Orlen S. A. for a stipendium
(2001/2002).
Con clu sion s
The valence isomerization/dehydration approach was
successfully applied to the synthesis of a series [n](2,7)-
pyrenophanes 1b-g, further establishing its scope. Sev-
eral trends in the spectroscopic properties of the pyreno-
phanes were observed to accompany the increasing
deviation from planarity of the pyrene system, but no
evidence was found to suggest that increasing the degree
of nonplanarity resulted in fullerene-like behavior of the
pyrene moiety. HOMA, NICS, and magnetic susceptibil-
ity data all led to the conclusion that increasing the
degree of bend in the pyrene system results in only a
small, but regular, loss of aromaticity over a wide range
of bend.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures and characterization data for all new compounds and
copies of 1H and 13C NMR spectra for all new compounds;
absolute electronic energies, Cartesian coordinates, and mag-
netic susceptibilities for pyrene fragments of 1b-g, 2c, 20;
tabulated UV/Vis and CV data for 1b-g and 2b-d . This
material is available free of charge via the Internet at http://
pubs.acs.org. Crystallographic data for the structure have been
deposited with the Cambridge Crystallographic Data Centre
as supplementary material No. CCDC 203281 (1d ), 203282
(1e), 203283 (1f), 203284 (1g). Copies of the data can be
obtained on application to CCDC, 12 Union Road, Cambridge
CB2 1EZ, UK (email: deposit@ccdc.cam.ac.uk).
J O0206059
(41) Snedecor, G. W.; Cochran, W. G. Statistical Methods; The Iowa
State University Press: Ames, IA, 1973; p 184.
(42) J effrey, G. A.; Ruble, J . R.; McMullan, R. K.; Pople, J . A. Proc.
R. Soc. (London) 1987, A414, 47.
(43) See for example: (a) Haddon, R. C. Acc. Chem. Res. 1988, 21,
243-249. (b) Haddon, R. C. J . Am. Chem. Soc. 1987, 109, 1676-1685.
(c) Haddon, R. C. J . Am. Chem. Soc. 1990, 112, 3385-3389.
(44) Sheldrick G. M. Acta Crystallogr. 1990, 46, 467-473.
(45) Sheldrick G. M. Program for the Refinement of Crystal Struc-
tures. University of Go¨ttingen, Germany, 1993.
(46) International Tables for X-ray Crystallography; Kynoch Press:
Birmingham, UK, 1974; Vol. IV.
2098 J . Org. Chem., Vol. 68, No. 6, 2003