A hydrogen atom sandwiched by cyclopropanes

Article abstract of DOI:10.1016/j.tetlet.2005.07.097

2′,3′,4′,5′,6′,7′- Hexahydrodispiro[cyclopropane-1,1′-anthracene-8′, 1″-cyclopropane] (1) was prepared by double olefination (Wittig) and double methylenation (Furukawa) of 1,8-dioxo-1,2,3,4,5,6,7,8-octahydroanthracene (4) that was in turn prepared in two

Full text of DOI:10.1016/j.tetlet.2005.07.097

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 6487–6489
A hydrogen atom sandwiched by cyclopropanes
Robert S. Walters, Douglas M. Ho and Robert A. Pascal, Jr.*
Department of Chemistry, Princeton University, Princeton, NJ 08544, USA
Received 16 June 2005; revised 14 July 2005; accepted 20 July 2005
Available online 9 August 2005
Abstract—20,30,40,50,60,70-Hexahydrodispiro[cyclopropane-1,10-anthracene-80,100-cyclopropane] (1) was prepared by double olefin-
ation (Wittig) and double methylenation (Furukawa) of 1,8-dioxo-1,2,3,4,5,6,7,8-octahydroanthracene (4) that was in turn prepared
˚
in two steps from 1,3-dibromobenzene. The X-ray structure of 1 shows that the C–9-H of its anthracene core is located 2.6 A from
the centroids of each of the flanking cyclopropane rings. The 1H NMR spectrum of 1 shows that the C–9-H resonance (d 5.95) falls
0.84 ppm upfield from the C–10-H resonance (d 6.79).
Ó 2005 Elsevier Ltd. All rights reserved.
Cyclopropane rings display large magnetic anisotropic
effects that are particularly evident in proton NMR
spectra,1–4 and numerous examples of the shielding or
deshielding of protons near cyclopropane rings in a wide
variety of molecules have been cataloged.2–6 The vast
majority of these cases involve the interaction of protons
with a single cyclopropane ring, and even in those mole-
cules containing more than one cyclopropane, the orien-
tation of the cyclopropanes with respect to the affected
hydrogens has not been ÔidealÕ. We wondered if a large
cyclopropane-induced shielding would be observed in
a molecule containing a proton sandwiched between
the faces of two parallel cyclopropane rings.
either C2- or Cs-symmetric conformations that differ in
energy by only 0.02 kcal/mol, and thus the two confor-
mations should be equally populated. GIAO calcula-
tions7 indicated that the H9 resonance should fall
1.09 ppm upfield of the H10 resonance in the C2 confor-
mation and 0.79 ppm upfield in the Cs conformation (see
Table 1).
Given the diketone 4, the synthesis of 1 should be
simple, but the literature syntheses of 48,9 proved chal-
lenging for us. Ultimately, we found that 1,3-bis(3-carb-
oxypropyl)benzene (3) was most efficiently prepared
(54% yield) by means of a double Suzuki coupling reac-
tion of 1,3-dibromobenzene with the adduct of 9-BBN
and methyl 3-butenoate using the method of Esteban
et al.10 With 3 in hand, a double intramolecular
Compound 1 is such a molecule. HDFT calculations at
the B3LYP/6-31G(d) level7 indicate that it may adopt
Br
Br
1) 9-BBN-(CH2)3CO2Me
PdCl2(dppf), NaOMe
2) KOH, then H+
HO2C
CO2H
2
3
PPA
135 ˚C
O
H9
O
H9
CH2 H9 CH2
CH2I2
ZnEt2
Ph3PCH2
H10
H10
H10
1
5
4
Keywords: Cyclopropane; Magnetic anisotropy; NMR.
*
Corresponding author. Tel.: +1 6092 585417; fax: +1 6092 586746; e-mail: snake@chemvax.princeton.edu
0040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.07.097

6488
R. S. Walters et al. / Tetrahedron Letters 46 (2005) 6487–6489
Table 1. Experimental and calculateda chemical shifts of aromatic
protons in compound 1
In the older literature, the largest differential shift for
two chemically similar protons due to cyclopropanes
seems to be 1.80 ppm for HA and HB in compound 6,5
and an extensive (but certainly not exhaustive) search
d (H9)
d (H10
)
Dd (H9–H10)
1 (calcd, C2 conf.)
1 (calcd, Cs conf.)
1 (calcd, average)
1 (exptl, CDCl3)
5.97
6.06
6.01
5.95
7.06
6.85
6.96
6.79
ꢀ1.09
ꢀ0.79
ꢀ0.95
ꢀ0.84
of the modern literature found
a difference of
2.46 ppm for HA and HB in compound 7 at low tempera-
ture.18 However, these large differences are due to a
combination of shielding and deshielding effects.
a GIAO calculations at the B3LYP/6-311+G(2d,p)//B3LYP/6-31G(d)
level.
HA
HB
HA
Friedel–Crafts acylation proceeded exactly as described
by Christopfel and Miller8 to give diketone 4 in 28%
yield.11 The double Wittig olefination of 4 gave 5,12
and methylenation using the Furukawa modification13
of the Simmons–Smith reaction gave the desired biscy-
clopropane 1.14 The last two steps suffered from low
yields (14% and 24%, respectively), but sufficient mate-
rial for NMR and crystallographic studies was obtained.
HB
H
HA
CH2
HB
6
7
8
9
H
OH
OH
H
HO
H
H
HO
Crystals of 1 were grown from ethanol, and the X-ray
structure was determined. Compound 1 adopts the C2
conformation in the crystal, and its molecular structure
is illustrated in Figure 1.15 H9 is located almost exactly
above the centroids, X(1) and X(2), of the two cyclopro-
pane rings at distances of 2.57 and 2.64 A, respectively,
and thus H9 is well positioned in the NMR shielding
regions of the flanking cyclopropanes.
10
11
12
13
In an attempt to find the most extreme cases of pure
shielding, we conducted a search of Cambridge Crystal-
lographic Database19 for protons lying very close to, and
directly above, the centers of cyclopropane rings. Com-
pound 1 seems to have claimed the record for the closest
such approach of one proton to the centers of two differ-
˚
In the 1H NMR spectrum of 1, the H9 resonance (d 5.95)
falls 0.84 ppm upfield from that of H10 (d 6.79), a rea-
sonable reference proton (Table 1). The observed differ-
ence is slightly less than the calculated value of
0.95 ppm, but the overall agreement of experiment and
calculation is very good, given that two interconverting
conformations of 1 must be considered. The observed
difference in the H9 and H10 chemical shifts probably
represents the largest cyclopropane-induced shift of a
proton resonance due purely to shielding (rather than
deshielding) effects, and is due to the fact that H9 is close
to the centers of two cyclopropane rings, a very rare
situation.
˚
ent cyclopropanes (at 2.57 and 2.64 A), but the closest
single approach is found in compound 8,20 where HA
˚
is only 2.36 A from the center of the opposing ring.
However, HA is shielded by only 0.43 ppm with respect
to HB, likely due to a conformational exchange process.
Quite a few molecules with the substructure 9 (or similar
˚
fragments) possess contacts on the order of 2.4–2.6 A,
but the magnitudes of the proton shielding in these mole-
cules (where reasonable reference protons can be found)
are smaller than observed in compound 1. For example,
the chemical shifts of the bold protons in compounds 10,
11, 12, and 13 are d 3.57, 3.60, 4.09, and 4.24, respec-
tively.5 Poulter et al.5 analyzed this series and concluded
that the protons in compounds 10–12 were shielded by
the cyclopropane, but the proton in 13 was deshielded;
however, the entire range of chemical shifts is only
0.67 ppm.
Acknowledgement
This work was supported by National Science
Foundation Grant CHE-0314873, which is gratefully
acknowledged.
References and notes
1. Lacher, J. R.; Pollock, J. W.; Park, J. D. J. Chem. Phys.
1952, 20, 1047–1048.
2. Forsen, S.; Norin, T. Acta Chem. Scand. 1961, 15, 592–
Figure 1. Molecular structure of compound 1.
598.

R. S. Walters et al. / Tetrahedron Letters 46 (2005) 6487–6489
6489
3. Wiberg, K. B.; Nist, B. J. J. Am. Chem. Soc. 1961, 83,
1226–1230.
4. Forsen, S.; Norin, T. Tetrahedron Lett. 1964, 5, 2845–
2849.
5. Poulter, C. D.; Boikess, R. S.; Brauman, J. I.; Winstein, S.
J. Am. Chem. Soc. 1972, 94, 2291–2296.
6. Hahn, R. C.; Howard, P. H. J. Am. Chem. Soc. 1972, 94,
3143–3148.
5.95 (s, 1H, H9), 6.79 (s, 1H, H10); 13C NMR (benzene-d6)
d 18.6, 19.2, 23.0, 30.6, 35.9, 114.7, 129.2, 133.8, 139.2; MS
(EI) m/z 238 (M+, 54), 223 (MꢀCH3, 53), 209 (MꢀC2H5,
100); exact mass 238.1716, calcd for C18H22 238.1722.
15. Crystal data for 1: C18H22; orthorhombic, space group
˚
˚
P212121 (No. 19); a = 7.4834(4) A, b = 10.0592(5) A,
3
˚
˚
c = 17.7749(7) A, V = 1338.0(1) A , Z = 4. Data were
˚
collected at 200 K with k = 0.71073 A; a total of 7758
7. Frisch, M. J. et al. All molecular orbital calculations were
performed with GAUSSIAN 98; Gaussian, Inc.: Pittsburgh,
PA, 1998. B3LYP/6-311+G(2d,p)//B3LYP/6-31G(d)
wave functions were employed in the GIAO calculations
to give the absolute 1H chemical shifts for 1, and these
were converted to ordinary chemical shifts by comparison
to the 1H chemical shift for tetramethylsilane computed at
the same level.
reflections (hmax = 27.47°) were processed to give 3058
unique reflections (R = 0.045) by using the program
DENZO-SMN.16 The structure was solved and refined using
Siemens SHELXTL.17 All atomic coordinates were refined;
carbon atoms were refined anisotropically and hydrogen
atoms isotropically. The refinement converged to
R(F) = 0.0503, wR(F2) = 0.0986, and S = 1.119 for 2664
reflections
with
I > 2r(I),
and
R(F) = 0.0623,
8. Christopfel, W. C.; Miller, L. L. J. Org. Chem. 1984, 49,
5198–5202.
wR(F2) = 0.1092, and S = 1.149 for 3057 unique reflec-
tions and 251 parameters (one reflection was suppressed).
Crystallographic data (excluding structure factors) for this
structure have been deposited with the Cambridge Crys-
tallographic Data Centre as supplementary publication
number CCDC 275196. Copies of the data can be
obtained, free of charge, on application to CCDC, 12
Union Road, Cambridge, CB2 1EZ, UK [fax: +44 (0)
1223 336033 or e-mail: deposit@ccdc.cam.ac.uk].
16. Otwinowski, Z.; Minor, W. Methods Enzymol. 1997, 276,
307–326.
9. Caluwe, P.; Pepper, T. J. Org. Chem. 1988, 53, 1786–1790.
10. Esteban, G.; Lopez-Sanchez, M. A.; Martinez, M. E.;
Plumet, J. Tetrahedron 1998, 54, 197–212.
11. Compound 4: mp 177–179 °C (lit.8 198–199.5 °C); 1H
NMR (CDCl3) d 2.12 (quintet, J = 6 Hz, 4H), 2.65 (t,
J = 6 Hz, 4H), 2.96 (t, J = 6 Hz, 4H), 7.13 (s, 1H, 10-H),
8.68 (s, 1H, 9-H).
1
12. Compound 5: H NMR (CDCl3) d 1.84 (m, 4H), 2.50 (m,
4H), 2.77 (m, 4H), 4.90 (d, J = 1 Hz, 2H), 5.46 (d,
J = 1 Hz, 2H), 6.84 (s, 1H, 10-H), 7.92 (s, 1H, 9-H); 13C
NMR (CDCl3) d 24.1, 30.4, 33.6, 107.4, 119.9, 129.7,
132.8, 137.2, 143.7; MS (EI) m/z 210 (M+, 100), 195
(MꢀCH3, 53); exact mass 210.1408, calcd for C16H18
210.1409.
17. Sheldrick, G. M. SHELXTL Version 5.04. Siemens Analyt-
ical X-ray Instruments, Madison, WI, USA, 1996.
18. Fitjer, L.; Klages, U.; Wehle, D.; Giersig, M.; Schormann,
N.; Clegg, W.; Stephenson, D. S.; Binsch, G. Tetrahedron
1988, 44, 405–416.
13. Furukawa, J.; Kawabata, N.; Nishimura, J. Tetrahedron
Lett. 1966, 7, 3353–3354.
19. Allen, F. H.; Kennard, O.; Taylor, R. Acc. Chem. Res.
1983, 16, 146–153.
20. de Meijere, A.; Wenck, H.; Kopf, J. Tetrahedron 1988, 44,
2427–2438.
14. Compound 1: 1H NMR (CDCl3) d 0.73 and 0.86 (AA0BB0
system, 8H), 1.61 (m, 4H), 1.86 (m, 4H), 2.80 (t, J = 6 Hz),