CD-spectroscopic investigations for the determination of the absolute configuration of α-alkylated 1,4-cyclohexanedione derivatives

Article abstract of DOI:10.1515/znb-2000-1104

The absolute configuration of the conformationally flexible six membered ring system 2-methyl- and 2,6-dimethyl-1,4-cyclohexanedione monoethylene acetal was determined by comparison of measured and calculated CD spectra. The rotational strengths were calculated by means of the CNDO/S-method assuming R at the stereogenic center. The results were compared with the predictions made by the octant rule. The enantiomerically pure material was synthesized via the corresponding SAMP- and RAMP-hydrazones.

Full text of DOI:10.1515/znb-2000-1104

CD-Spectroscopic Investigations for the Determination of the Absolute
Configuration of a-Alkylated 1,4-CycIohexanedione Derivatives
Jörg Fleischhauer3, Sven Gabriel3, Dieter Enders3, Anja Nühring3, and Axel Wollmerb
3 Institut für Organische Chemie, RWTH Aachen,
Professor-Pirlet-Str. 1, 52074 Aachen, Germany
b Institut für Biochemie, RWTH Aachen, Pauwelsstr. 30, 52057 Aachen, Germany
Reprint requests to Prof. J. Fleischhauer. Fax: +49 (0)241 8888 385.
E-mail: Joerg.Fleischhauer@thc.rwth-aachen.de
Z. Naturforsch. 55 b, 1011-1014 (2000); received July 3, 2000
1,4-Cyclohexanedione Derivatives, Absolute Configuration, CD Spectra
The absolute configuration of the conformationally flexible six membered ring system 2-me-
thyl- and 2,6-dimethyl-l,4-cyclohexanedione monoethylene acetal was determined by compar-
ison of measured and calculated CD spectra. The rotational strengths were calculated by means
of the CNDO/S-method assuming R at the stereogenic center. The results were compared with
the predictions made by the octant rule. The enantiomerically pure material was synthesized
via the corresponding SAMP- and RAMP-hydrazones.
1. Introduction
duce at least qualitatively the longest wave length
part of the spectrum [3-7].
The SAMP/RAMP-hydrazone method has been
used successfully to synthesize enantiomerically
pure cyclohexanone and 1,4-cyclohexanedione
derivatives, which are versatile chiral building
blocks in the synthesis of natural compounds [1 , 2 ],
One of the standard methods to determine the abso-
lute configuration of these products is X-ray crys-
tallography, but that requires single crystals of suffi-
cient quality. There are many compounds which, in
spite of all efforts, crystallize poorly or even not at
all. Consequently, the structure of such compounds
cannot be elucidated by this analytical method.
Therefore, it is highly desirable to have methods
that allow the determination of the absolute con-
figuration in solution. The combination of experi-
mental and theoretical CD spectroscopy provides
such a procedure. Here one compares the mea-
sured CD spectrum of the compound with unknown
configuration with the calculated spectrum for the
compound with known absolute configuration. For
the calculation we used the semiempirical CNDO/S
method. We believe that this method is able to repro-
As a further test, this procedure was now used
to determine the absolute configuration of the 1 ,4-
cyclohexanedione monoethylene acetal derivatives
1 and 2 (Scheme 1).
The compounds examined in the present investi-
gation were optically active (R)- and (5>2-methyl-
cyclohexanedione monoethylene acetals and the
C2-symmetric (R,R)~ and (S,S)-fra;«-2,6-dirnethyl-
cyclohexanedione monoethylene acetals. To deter-
mine the absolute configuration at the ct-carbons
of these cyclohexanone derivatives we also applied
the octant rule for the prediction of the sign of the
Cotton effect of the n —* ir* transition, which de-
termines the spectrum at the longest wave length
[8 , 9],
2. Results and Discussion
For the monomethylated compounds 1 the ORD
values at 546, 436 and 365 nm (mercurylines) are
negative for the R enantiomer and positive for the
S enantiomer (see Table 2). Both measurements
c h 3
o ^ \
CH,
O '-'X
chJ T )
-?H3IT}’
W
h3
Q> CH3
Scheme 1. 1,4-Cyclohexanedione monoethy-
lene acetal derivatives 1 and 2.
{R,R)-2
(R)-1
(sy-i
(S,S)-2
K
0932-0776/00/1100-1011 $ 06.00 © 2000 Verlag der Zeitschrift für Naturforschung, Tübingen • www.znaturforsch.com
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/20/15 12:41 AM

1012
J. Fleischhauer et al. ■a-Alkylated 1,4-Cyclohexanedione Derivatives
200 220
240 260 280 300 320 340
X. [nm]
A, [nm]
50
0
-SO
§ -100
CH3 I
(RR)-2
)
J
.
1 5 0
o
a
-200
3
-250
200 220 240 260 280 300 320 340
X, [nm]
X [nm]
Fig. 1. Measured CD-spectra of 1,4-cyclohexanedione derivatives.
were carried out in CHCI3 . The circular dichroism
curve was strongly negative for the R enantiomer
and strongly positive for the 5 enantiomer with a
minimum respectively maximum at A = 290 nm.
The extrema for the (R,R)-2 and (S,S)-2 CD-curves
appear at A = 295 nm and are much more inten-
sive then those of their monosubstituted counter-
Fig. 2. The octant rule applied to (/?)-1.
parts (see Fig. 1).
Predictions o f the octant rule
So the /^-configuration is confirmed for (R )-l. In
the same way, it can be concluded that (5)-l has
5-configuration at the a-carbon atom: at carbon 2
the equatorial methyl group lies in the y,z-plane and
does not contribute to the Cotton effect. After flip-
ping to the axial position, it is located at carbon 6 in
the octant with -x , +y and -z, so that its contribution
to the Cotton effect of the n —>w* transition should
be positive.
The fact that the absolute values of the Cotton
effects of (/?,/?)-2 and (5,5)-2 are still larger than in
the (R )-l and (5)-l case is easily understood: dur-
ing the flipping process one methyl group remains
always axial for both molecules.
First, the octant rule was applied to these exper-
imental findings. As was concluded from chemi-
cal synthesis, (R)-1 should be ^-configurated at the
Q-carbon atom (see Fig. 2). In this case, the ax-
ial methyl group is not located in one of the three
Cartesian planes, but in the octant where all Carte-
sian coordinates are negative. Its contribution to the
rotational strength of the n ^ 7r* transition should
therefore be negative [corresponding to the octant
rule: (-)•(-)•(-) = (-)]. The acetal substituent at
carbon 4 contributes nothing to the Cotton effect
because it is located in the jc,z-plane. If the methyl
group flips from the axial to the equatorial posi-
tion, one has to locate it at the a-carbon 6 . Now
its contribution to the Cotton effect should be zero,
because it is located in the >\z-plane. This is in per-
Q uantum chem ical calculations
To calculate the rotational strength of the n —» 7r*
fect agreement with the experimental CD spectrum. transition of (R)-l we first optimized the geometry
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/20/15 12:41 AM

J. Fleischhauer et al. ■Q-Alkylated 1,4-Cyclohexanedione Derivatives
1013
(S,R)-5
(SM
(d)
(d)
Scheme 2. Preparation of ketones (/?)-l
and (R,R)-2. (a) SAMP, benzene, reflux;
(b) LDA, THF, 0 °C; Mel, -78 °C to rt;
(c) TMPLi, THF, 0 °C; Mel, -78 °C to rt;
(d) saturated oxalic acid, ether, rt.
...CH,
...CH,
(R).1
Table l. Calculated transition energies and rotational
strengths. In brackets is given the corresponding value
for the second transition.
CNDO/S-calculations using the programs DZDO
and MCD3SP, [11], taking into account only sin-
gle excitations from the 13 highest occupied into
the 13 lowest unoccupied molecular orbitals, were
performed.
Compound
Transition
Rotational
energy [au]
strength [au]
The transition energies and the rotational
strengths for the longest wave length transitions are
collected in Table 1. From an investigation of the
Cl coefficients it follows that the first excited state
is an n —>ir* state. The results agree, at least quali-
tatively, with the experimental findings and the pre-
dictions of the octant rule. The next transition occurs
at much higher energy (see Table 1) and does not
influence the peak at the longest wave length.
(/?)-l axial
(/?)-1 equatorial
(R,R)-2
0.119(0.300)
0.118 (0.301)
0.119(0.299)
-0.151 (0.401)
-0.054 (0.252)
-0.214(0.485)
Table 2. Optical rotation, enantiomeric purity and CD data
of ketones 1 and 2 .
Com-
pound
Optical rotation
(c in CHCI3)
ee [%] Sign* Amax.
[nm]
(/?)-1
(S)-l
[a]fAlc = -1.3 (c — 1.11)
92a
91a
-
290
290
3. Experimental section
- 12.8 (c= 1.11)
-60.7 (c= 1.11)
+ 1.1 (c= 1.08)
+ 13.4 (c= 1.08)
+60.4 (c = 1.09)
[a]IIT =
[ a l S C =
[Ql546C =
[a]lfeC =
[ a ] S C =
The 1,4-cyclohexanedione derivatives 1 and 2 were
synthesized starting from commercially available 1,4-
cyclohexanedione monoethylene acetal 3 following the
protocol outlined in Scheme 2. In the first step the free ke-
tone function of 3 was converted into SAMP-hydrazone
(5)-4, which was then metalated with LDA and alky-
lated with Mel to give compound (S,R)-5. Cleavage of
the hydrazone with saturated aqueous oxalic acid [12]
gave the desired a-monoalkylated ketone (/?)-l with an
enantiomeric excess of ee = 92 % (see Table 2).
+
(R,R)-2 [ö]ß°c = -93.0 (c= 1.20) 94b
(5,5)-2 [a]2/ c =
-
295
295
+93.2 (c= 1.13) 86b
+
* Sign of the Cotton effect;a determined by Pirkle shift
NMR; b determined by GC on chiral stationary phases
(de > 96%, determined by l3C NMR).
The ketone (R,R)-2 was synthesized by a second me-
talation of monoalkylated hydrazone (S,R)-5 carried out
with LiTMP and further alkylation with Mel to form
hydrazone (S,R,R)-6. Cleavage of (S,R,R)-6 with satu-
rated aqueous oxalic acid [12] gave the desired a a '-
bis-alkylated ketone (R,R)-2 with an enantiomeric excess
of ee = 94 %. The other enantiomers (5)-l and (S,S)-2
were synthesized following the same protocol as out-
with the methyl group in the axial and in the equa-
torial position using the AMI method [10]. The
molecule with the equatorial methyl group is lower
in energy than the molecule with the methyl group
in axial position (« 0.84 kcal/mol), as expected.
The energy of {R,R)-2 was also optimized with the
AMI method. For the three optimized structures
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/20/15 12:41 AM

1014
J. Fleischhauer et al. • Q-Alkylated 1,4-Cyclohexanedione Derivatives
lined in Scheme 2 using (i?)-l-amino-2-methoxymethyl-
pyrrolidine (RAMP) instead of (S)-l-amino-2-methoxy-
methylpyrrolidine (SAMP) as the chiral auxiliary for the
ketone 3. The enantiomeric excesses of the monomethy-
lated ketones 1were determined by Pirkle shift NMR [13].
The ee's of the bismethylated ketones 2 were determined
by GC on chiral stationary phases (chiral column: Chi-
rasil dex 25m, Chrompack). These results are collected in
Table 2.
(A < 260 nm) and a JASCO (Japan Spectroscopic Co.,
Ltd., Tokyo, Japan) J-600 (A > 260 nm), both calibrated
according to Chen and Yang [14], The spectral bandwidth
/ path length of the optical cells used were 2 nm / 20 mm
(A < 260 nm) and 1 nm / 20 mm (A > 260 nm) for the
monoalkylated compounds and 2 nm / 5 mm (A < 260 nm)
and 1 nm / 5 mm (A > 260 nm) for the bisalkylated
compounds. The AMI calculations were performed on
an IBM 43P/150 computer, using the program SPARTAN
5.1 (Wavefunction, Inc. ; Irvine, CA 92612 USA).
Measurements
Acknowledgements
All measurements were performed in spectrograde or
purified solvents. The ORD spectra were measured on a
Perkin Elmer polarimeter. For the CD measurements a
solution of the ketone was prepared at a concentration of
0.200 mmol / 5 ml acetonitrile. CD spectra were collected
on an AVIV (Lakewood, NJ, USA) 62 DS spectrometer
This work was supported by the Deutsche Forschungs-
gemeinschaft (SFB 380 and Normalverfahren Fl 142/3-3)
and the Fonds der Chemischen Industrie. We thank
Degussa-Hüls AG, BASF AG, Bayer AG and the former
Hoechst AG for the donation of chemicals.
[8] W. Moffitt, R. B. Woodward, A. Moscowitz,
W. Klyne, C. Djerassi, J. Am. Chem. Soc. 83, 4013
(1961).
[1] D. Enders, A. Niihring, J. Runsink, Chirality 12, 374
(2000).
[2] D. Enders, H. Eichenauer, Chem. Ber. 112, 2933
[9] A. Gergely, F. Zsila, P. Horvath, G. Szasz, Chirality
11, 741 (1999).
(1979).
[3] J. Fleischhauer, C. Jansen, A. Koslowski, G. Raabe,
J. Schiffer, A. Wollmer, Z. Naturforsch. 53a, 704
(1998).
[4] J. Fleischhauer, A. Koslowski, J. Schiffer, A. Woll-
mer, D. Enders, A. Zamponi, G. Raabe, Z. Natur-
forsch. 53a, 891 (1998).
[10] M. J. S. Dewar, E. G. Zoebisch, E. F. Healy and
J. J. P. Stewart, J. Am. Chem. Soc. 107, 3902 (1985).
[11] The programs DZDO and MCD3SP were written by
J. Downing and J. Michl, University of Colorado,
Boulder, USA.
[12] D. Enders, T. Hundertmark, R. Lazny, Synlett, 721
(1998).
[13] W. H. Pirkle, D. L. Sikkenga, J. Org. Chem. 42,
1370(1977).
[5] J. Fleischhauer, G. Raabe, A. G. Santos, J. Schiffer,
A. Wollmer, Z. Naturforsch. 53a. 896 (1998).
[6] J. Fleischhauer, A. Koslowski, Ch. Repges, K.-P.
Gulden. G. Bringmann, Z. Naturforsch. 53a, 993
(1998).
[14] G. C. Chen, J. T. Yang, Anal. Lett. 10, 1195 (1977).
[7] K. Woydowski, J. Fleischhauer, J. Schiffer, J. Lieb-
scher. J. Chem. Soc., Perkin Trans. 1, 149 (1999).
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/20/15 12:41 AM