Tautomerization-mediated molecular switching between six- and seven-membered rings stabilized by hydrogen bonding

Article abstract of DOI:10.1002/chem.201500524

1,3,4,6-Tetraketones typically undergo keto-enol tautomerism forming bis-enols stabilized by intramolecular hydrogen bonding in two six-membered rings. However, 1,3,4,6-tetraketones derived from the terpene ketone camphor and norcamphor exist as isomers with two distinguishable modes of intramolecular hydrogen bonding, namely, the formation of six- or seven-membered rings. The structural requirements for this so far unknown behavior were investigated in detail by synthesis and comparison of structural analogues. Both isomers of such 1,3,4,6-tetraketones were fully characterized in solution and in the solid state. Intriguingly, they slowly interconvert in solution by means of tautomerism-rotation cascades, as was corroborated by DFT calculations. The influence of temperature and complexation with the transition metals Pd, Rh, and Ir on the interconversion process was investigated.

Full text of DOI:10.1002/chem.201500524

DOI: 10.1002/chem.201500524
Full Paper
&
Tautomerism
Tautomerization-Mediated Molecular Switching Between Six- and
Seven-Membered Rings Stabilized by Hydrogen Bonding
Golo Storch, Markus J. Spallek, Frank Rominger, and Oliver Trapp*[a]
Abstract: 1,3,4,6-Tetraketones typically undergo keto–enol
tautomerism forming bis-enols stabilized by intramolecular
hydrogen bonding in two six-membered rings. However,
1,3,4,6-tetraketones derived from the terpene ketone cam-
phor and norcamphor exist as isomers with two distinguish-
able modes of intramolecular hydrogen bonding, namely,
the formation of six- or seven-membered rings. The structur-
al requirements for this so far unknown behavior were inves-
tigated in detail by synthesis and comparison of structural
analogues. Both isomers of such 1,3,4,6-tetraketones were
fully characterized in solution and in the solid state. Intrigu-
ingly, they slowly interconvert in solution by means of tauto-
merism–rotation cascades, as was corroborated by DFT cal-
culations. The influence of temperature and complexation
with the transition metals Pd, Rh, and Ir on the interconver-
sion process was investigated.
Introduction
chain tautomerism, were observed in deuterated dimethyl sulf-
oxide (Figure 1A).[15–16,17]
1,3-Diketones have drawn attention in many fields of modern
chemistry. They are easily converted to difluoroboron diketo-
nates that exhibit interesting fluorescence and phosphores-
cence properties.[1,2] In addition, they are well known to form
stable complexes with a variety of metals.[3] Chiral 1,3-diketo-
nate complexes[4] proved to be valuable catalysts, for example,
in asymmetric hetero-Diels–Alder reactions.[5]
The multiplicity of coexisting isomeric structures was extend-
ed when Noe et al.[18] and Hart[19] independently reported
1,3,4,6-tetracarbonyl 2 derived from natural occurring (+)-(R)-
camphor. Both reported the simultaneous presence of structur-
ally related isomers that cannot be rationalized by any tauto-
merism known so far. Hart postulated six possible structures
differing in the enol protons position and an s-cis/s-trans iso-
merism of the central single bond and suggested 2s-trans and
2s-cis (Figure 1B) as the most probable structures in solution. In
the context of our investigations of controllable molecular sys-
tems that can coexist in several interconverting states,[20–23] we
became aware of this unexplained phenomenon.
Unlike their smaller counterparts, 1,3,4,6-tetraketones that
consist of two fused 1,3-diketone subunits have been rarely
mentioned. The first description of this structural motif dates
back to Claisen and Stylos in 1887, who reported the conden-
sation of acetone and diethyl oxalate.[6] In sporadic later stud-
ies, tetraketones were used as synthetic intermediates.[7,8] In
contrast, they are well-established building blocks in supra-
molecular coordination chemistry,[9–12] for example, for the for-
mation of metallacrown ethers.[13]
Results and Discussion
We optimized the synthetic protocol to prepare 1,3,4,6-tetra-
ketones by condensation of diethyl oxalate and ketones based
on the results of Hart[19] and Noe et al.[18] using 2 as reference
substance. We found that the deprotonation of (+)-(R)-cam-
phor is the crucial step for the successful conversion to 1,3,4,6-
tetraketones and to obtain a high overall yield. Sodium meth-
oxide did not yield any product, and lithium diisopropylamide
led to undesired side reaction due to nucleophilic attack of di-
ethyl oxalate. By using sodium hydride to generate the enolate
under optimized reaction conditions, the reaction can be per-
formed on a large scale with excellent yield (93%).
In contrast to the smaller 1,3-diketones, tetraketones almost
exclusively form planar bis-enols in solution and in the solid
state.[14] Balenovic and Poje first studied the tautomerism of
various 1,3,4,6-tetraketones in solution by means of NMR spec-
troscopy.[15] They realized that the molecular structure and
ratio of the tautomers present were strongly solvent depen-
dent. In deuterated chloroform, they exclusively observed the
bis-enol structure 1c. All possible keto–enol tautomers 1a–
c and most interestingly the semiacetal 1d, product of a ring–
To investigate a broad range of tetraketones we additionally
prepared compounds 3–7 (Figure 2). Besides (+)-(R)-camphor,
we included (Æ)-norcamphor, which has an unsubstituted
[2.2.1]bicycloheptane core. Compound 3 was prepared in 36%
yield due to tedious purification of the meso isomer. Literature-
known acetone tetraketone 4 was prepared for comparison.
Next, we studied the related [3.1.1]bicycloheptanes. (+)-(R)-No-
[a] G. Storch, Dr. M. J. Spallek, Dr. F. Rominger, Prof. Dr. O. Trapp
Organisch-Chemisches Institut
Ruprecht-Karls-Universitꢀt Heidelberg
Im Neuenheimer Feld 270, 69120 Heidelberg (Germany)
E-mail: trapp@oci.uni-heidelberg.de
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201500524.
Chem. Eur. J. 2015, 21, 1 – 8
1
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Figure 1. Molecular structures of 1,3,4,6-tetraketones reported in literature. A) Tautomers of 1 in deuterated dimethyl sulfoxide solution identified by NMR
spectroscopy.[15] B) Additional s-cis/s-trans isomerism of 2 proposed by Hart.[19]
meric ratio observed by X-ray
crystallography was highly re-
producible when crystalline ma-
terial prepared under the same
conditions was used and is inde-
pendent of the isomeric ratio of
the solid material prior to
crystallization.
Samples of 2, 3, and 5 were
subsequently studied by NMR
spectroscopy. As expected, 5 ex-
hibited one distinct signal set in
solution irrespective of the sol-
vent used. However, 2 exhibited
signal sets for two isomers. They
differ most prominently in the
chemical shift of the enol proton
resonances
at
12.53
and
15.54 ppm (in [D6]benzene), re-
spectively. This behavior can
easily be rationalized by the si-
multaneous occurrence of six-
and seven-membered ring iso-
Figure 2. Tetraketones investigated in the present study. General reaction conditions: 1.0 equiv diethyl oxalate,
2.0 equiv ketone, 5.5 equiv sodium hydride, THF, reflux, 1–7 d (see Experimental Section for details).
pinone smoothly yielded 5 in 63% yield. The methylated ana-
logue 6 exclusively formed 7 in 70% yield due to increased
steric hindrance, which makes nucleophilic attack of a second
enolate impossible.
In a first step we investigated the behavior of the obtained
1,3,4,6-tetraketones in the solid state, to have a well-defined
starting point for investigations in the solution phase. There-
fore, we analyzed crystals of 2, 3, and 5 by means of X-ray dif-
fraction. In accordance with 4 and the tetraketone derived
from acetophenone,[14] compound 5 exists exclusively as a bis-
enol with two six-membered rings stabilized by hydrogen
bonding in the solid state (Figure 3A). Surprisingly, and in
sharp contrast to this, we observed two independent mole-
cules as overlapping structures for 2 and 3. Both are bis-enols,
yet a major isomer forms hydrogen-bond-stabilized six-mem-
bered rings, while a minor isomer is made up of seven-mem-
Figure 3. X-ray crystal structures of tetraketones 2, 3, and 5. Thermal dis-
placement ellipsoids (30% probability) are shown. Hydrogen atoms except
enol protons are omitted for clarity.
bered rings (Figure 3B and C). In the case of 2, the minor
isomer refines to about 13% of the measured crystal. The iso-
&
&
Chem. Eur. J. 2015, 21, 1 – 8
www.chemeurj.org
2
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
mers caused by different intramolecular hydrogen bonding in
solution. This isomerism can be described as a coupled cis/
trans isomerism of the keto–enol C=C bond. Although the two
isomers are formed by tautomerism (see below), they are not
normal tautomers,[24] because there is no rapid interconversion
between the isomers due to a migrating group.
lead to any isomerization. Unfortunately, crystalline 3 dissolves
very slowly in CDCl3 or C6D6, and hence reliable comparison to
the solid-state results impossible.
In contrast, crystalline 2 dissolves very rapidly in CDCl3 and
C6D6. This allowed us to record NMR spectra within 1 min after
adding the sample. To confirm that the tautomeric ratio was
not significantly altered within this first minute, we repeated
the measurement after another minute. Intriguingly, we ob-
served two signal patterns without any change of relative in-
tensity within the first two minutes irrespective of whether
CDCl3 or C6D6 was used. The initial tautomeric ratios were de-
termined as 15:85 (CDCl3) and 18:82 (C6D6), respectively. It thus
can be concluded that the two species observed by NMR spec-
troscopy are those overlapping in the X-ray structure. In addi-
tion, with respect to the isomeric ratio in the solid state, the
signals of the minor species in NMR spectroscopy can be iden-
NMR spectroscopic analysis of 3 was hampered by the co-
existence of ketone tautomers in addition to the six- and
seven-membered ring isomers in deuterated chloroform or
benzene solution. However, in deuterated dimethyl sulfoxide,
compound 3 can almost exclusively (>90%) be transformed
into the cyclic semiacetal 3a. Analysis of 2D NMR spectra indi-
cated that 3a and not 3b is the observed isomeric structure
(Figure 4,A). This assignment is further corroborated by isola-
tion of the N-substituted semiaminal 3c (Figure 4B, C) on
treatment with ammonia.
tified as 2seven membered
.
Surprisingly, recording NMR spectra after 4 h, 24 h and 6 d
revealed that the isomeric ratio slowly changed over time and
reached solvent-dependent equilibria at 35:65 (C6D6) and 29:71
(CDCl3) after several days (Figure 5). Thus, a solution of 2 con-
tains significantly larger amounts of 2seven membered
.
The slow adjustment of the equilibrium isomeric ratio can
be drastically accelerated by heating the solution. Thus,
a sample of crystalline 2 was dissolved in 1,2-dideuterotetra-
chloroethane and an initial 14:86 ratio was determined. NMR
spectra were recorded while stepwise heating to 1308C and
after subsequent cooling to room temperature. The equilibri-
um ratio of 34:66 was reached within less than 3 h (time re-
quired to conduct the experiment).
To fully understand the new type of tautomerism-mediated
isomerism, we investigated the structures of all above-men-
tioned tetraketones and mechanistic aspects by DFT calcula-
tions at the B3LYP/cc-pvTZ level of theory. First, we compared
the relative energies of the hydrogen-bond-mediated six- and
seven-membered ring isomers of 2, 4 and 5. Preliminary inves-
tigation of the relative stabilities of six- and seven-membered
ring structures revealed that the latter is less stable for acetone
tetraketone 4 (DG= +26.5 kJmolÀ1) and nopinone tetraketone
5 (DG= +4.9 kJmolÀ1), yet favored for camphor derivative 2
(DG=À9.3 kJmolÀ1). Although the overall stability of seven-
membered ring isomers seems to be overestimated, this gives
a first hint at the anomalous behavior of [2.2.1]bicycloheptane-
derived camphor tetraketone 2. Next, we observed that the
s-cis isomer 2s-cis proposed by Hart[19] is already
DG=17.8 kJmolÀ1 higher in energy than s-trans isomer 2s-trans
and therefore is not further taken into account.
Figure 4. Tautomerism of 3. A) In contrast to dissolution in deuterated
chloroform or benzene, 3 is almost exclusively converted to semiacetal 3a
in dimethyl sulfoxide solution. B) Treatment of the 1,3,4,6-tetraketone with
ammonia leads to the formation of cyclic semiaminal 3c, which can be iso-
lated and investigated, for example, in chloroform solution. C) X-ray struc-
ture of 3c. Since meso-3 was used, the depicted structure only represents
one stereoisomer. Thermal displacement ellipsoids (30% probability) are
shown. Non-acidic hydrogen atoms are omitted for clarity.
To compare the solid-state results with the experiments in
solution, our idea was to monitor potential isomerization pro-
cesses after dissolving crystalline material of known isomeric
composition. Thus, larger amounts of crystalline tetraketones
2, 3, and 5 were prepared by applying exactly the same condi-
tions that were used to obtain the samples for X-ray crystallog-
raphy. This material was dissolved in an appropriate deuterated
solvent and NMR spectra were recorded at given time inter-
vals. These intervals had to be short in the beginning to quan-
tify any possible isomerization between sample preparation
and the first NMR measurement.
With regard to the isomerization mechanism, we note that
no mixed intermediate with one hydrogen-bond-stabilized six-
and one seven-membered ring isomer can be formed. Two in-
dependent rotations around formal C=C bonds are ultimately
required to transform 2six membered into 2seven membered (Figure 6A).
In principle, this can be achieved by two mechanistic options:
with or without breaking the protochelate structures by keto–
enol tautomerism. Since the barrier for keto–enol tautomerism
is highly solvent dependent and many different sorts of mech-
anisms are known and plausible,[25–27] we did not study this
As expected, a solution prepared from a crystalline sample
of 5 showed only one signal pattern when analyzed after
1 min. Heating of the solution or prolonged waiting did not
Chem. Eur. J. 2015, 21, 1 – 8
www.chemeurj.org
3
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
1
Figure 5. Partial H NMR spectra at given times after dissolution of a crystalline sample of 2 in C6D6. A) Enol proton region. B) Camphor-backbone bridgehead-
proton resonances. The relative integrals are given as ratios on the right. They can be determined for the resonances in A) or B) without divergence.
plexation.[28–30] As expected, 5 readily forms C2-symmetric ho-
mobimetallic complexes 8 and 9 with [{Rh(nbd)Cl}2] (nbd=nor-
bornadiene) and [{Rh(CO)2Cl}2], in which the metal atoms are
chelated in six-membered rings. Complexation with
[{Pd(C3H5)Cl}2] resulted in statistical formation of endo and exo
isomers, and thus 10 was obtained as a 1/2/1 ratio of endo/
endo, endo/exo, and exo/exo isomers (Figure 7A). The geometri-
Figure 6. Mechanistic investigation of the interconversion of the hydrogen-
bond-stabilized six- and seven-membered ring isomers of 2. A) Isomerization
overview. B) Reaction profile. All calculations were performed with Gaussi-
an 09 (Revision D.01). The B3LYP functional was used with a cc-pVTZ basis
set. All given values are D3-corrected. Standard ambient temperature and
pressure (298 K, 1.0 atm) were applied.
Figure 7. Transition metal complexes of 1,3,4,6-tetraketones. A) Homobime-
tallic complexes of 5 and illustration of the coordination isomers of 10.
B) Molecular structures of {2[Rh(cod)]2}[19] and 9 in the solid state. Thermal
displacement ellipsoids (30% probability) are shown and hydrogen atoms
are omitted for clarity. C) Rhodium and iridium complexes of 3.
mechanism, which in fact is very similar to that investigated
and discussed below.
In a first step, tautomerization of 2six membered to 2int1 lowers
the bond order of the formal C=C bond, which facilitates rota-
tion around this bond in a second step. Intermediate 2int2 is
stabilized by two intramolecular hydrogen bonds and can
readily tautomerize to 2int3, which has an elongated second
former C=C double bond. Another rotation to give 2int4 and fi-
nally consecutive tautomerization of both enol protons results
in 2seven membered (Figure 6,B).
cal parameters of this class of complexes were studied in detail
by means of X-ray crystallographic investigation of
{2[Rh(cod)]2}[19] (cod=1,5-cyclooctadiene) and 9 (Figure 7,B).
Since 2 and 5 exclusively form six-membered metal-chelate
complexes, we were interested in the properties of meso-3,
which exists as complex mixture of several tautomers. Interest-
ingly, treatment of meso-3 with [{Rh(nbd)Cl}2] or [{Ir(cod)Cl}2]
resulted in significant simplification of the NMR spectra by ex-
clusive formation of analogous Ci-symmetric complexes 11 and
12 (Figure 7C).
Various cases are known in which transition metals have sig-
nificant influence on tautomerism. Thus, we were interested in
the behavior of the tautomers towards transition metal com-
&
&
Chem. Eur. J. 2015, 21, 1 – 8
www.chemeurj.org
4
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
with Et2O three times. These organic extracts were dried over
Na2SO4 and concentrated.
Interestingly, on decoordination of the tetraketone ligand 2,
for example, by heating the complex in a protic solvent, both
hydrogen-bond-mediated six- and seven membered ring iso-
mers are again observed in the described ratios. Thus, metal
coordination and decoordination can also be used to control
the isomeric equilibrium.
1,2-Bis[(1R,4R)-4,7,7-trimethyl-3-oxobicyclo[2.2.1]heptan-2-yl]-
ethane-1,2-dione (2):[18,19] Purification by flash chromatography
(silica, hexanes/ethyl acetate 50/1); bright yellow solid; 93%;
1H NMR (C6D6, 300.51 MHz): seven-membered ring protochelate:
d=0.58 (s, 6H, CH3), 0.68 (s, 6H, CH3), 0.91 (s, 6H, CH3), 1.19–1.49
3
(m, 6H, CH2), 1.70–1.90 (m, 2H, CH2), 3.21 (d, JH,H =3.8 Hz, 2H, CH),
15.54 ppm (s, 2H, OH); six-membered ring protochelate: d=0.61
(s, 6H, CH3), 0.68 (s, 6H, CH3), 0.93 (s, 6H, CH3), 1.19–1.49 (m, 6H,
CH2), 1.70–1.90 (m, 2H, CH2), 3.36 (d, 3JH,H =3.9 Hz, 2H, CH),
12.53 ppm (s, 2H, OH); 13C{1H} NMR (C6D6, 75.56 MHz): seven-mem-
bered ring protochelate: d=9.7 (2C, CH3), 18.5 (2C, CH3), 20.4 (2C,
CH3), 26.3 (2C, CH2), 31.8 (2C, CH2), 47.9 (2C, Cq), 49.5 (2C, CH),
59.7 (2C, Cq), 123.8 (2C, C=CÀO), 158.1 (2C, OÀC=C), 211.7 ppm
(2C, C=O); six-membered ring protochelate: d=9.0 (2C, CH3), 18.5
(2C, CH3), 20.6 (2C, CH3), 27.4 (2C, CH2), 30.7 (2C, CH2), 48.9 (2C,
Cq), 48.9 (2C, CH), 58.0 (2C, Cq), 121.1 (2C, C=CÀO), 155.9 (2C,
OÀC=C), 214.6 ppm (2C, C=O); MS (EI+): m/z (%): 179 (100,
Conclusion
We have described a novel type of tautomerism-mediated iso-
merism caused by hydrogen-bond-stabilized six-membered
and seven-membered cyclic bis-enol structures. Inspired by
camphor derived 1,3,4,6-tetraketone 2, we identified both iso-
mers in solution and in the solid state. Study of several analo-
gous structures revealed that a [2.2.1]bicycloheptane backbone
is crucial for the occurrence of the isomerism, as was proven
by experimental data and DFT calculations. One key factor of
this new phenomenon is the exceptionally slow rate of isomer-
ization in solution. This process can be accelerated by heating
or treatment with transition metals, and the latter form exclu-
sively six-membered metal chelates. Further studies may aim
to use the hydrogen-bond-stabilized six-and seven-membered
ring isomers as switching units that can be controlled by crys-
tallization, addition of auxiliaries, temporary metal bonding, or
temperature.
+
+
[C11H15O2]+), 247 (72, [MÀC6H14 ÀCO] ), 275 (9, [MÀC6H14 ] ), 330
C
C
+
+
(34, [MÀCO]+ ), 358 (17, [M] ); HRMS (EI ): m/z calcd for C22H30O4:
358.2144; found: 358.2140; IR (FT-ATR): n˜ =2957, 2870, 1655, 1620,
1574, 1556, 1456, 1377, 1267, 1221, 1145, 1067, 1025 cmÀ1; elemen-
tal analysis (%) calcd for C22H30O4: C 73.71, H 8.44; found: C 73.22,
H 8.42.
C
C
meso-1,2-Bis(3-oxobicyclo[2.2.1]heptan-2-yl)ethane-1,2-dione (3):
Purification by recrystallization from Et2O and washing with ace-
tone; bright yellow solid; 36%; complex NMR spectra in CDCl3 and
C6D6; MS (EI+): m/z (%): 109 (15, [C8H9O2ÀC2H4]+), 137 (100,
+
+
[C8H9O2]+), 246 (30, [MÀCO]+ ), 274 (2, [M] ); HRMS (EI ): m/z
calcd for C16H18O4: 274.1205; found: 274.1232; IR (FT-ATR): n˜ =
3250, 2961, 2873, 1736, 1694, 1620, 1480, 1456, 1325, 1288, 1262,
1189, 1167, 1085, 1067, 1033 cmÀ1; elemental analysis (%) calcd for
C16H18O4: C 70.06, H 6.61; found: C 70.03, H 6.32. NMR characteriza-
C
C
Experimental Section
Syntheses were carried out under an atmosphere of argon (5.0
grade) with exclusion of air. All glassware was heated prior to use,
and standard Schlenk techniques were applied. THF and CH2Cl2
were dried in an MB SPS-800 system and stored under argon. NMR
spectra were recorded on Bruker Avance 600, 500, and 300 MHz
spectrometers. The residual proton signals of the solvent were
used as internal standards.[31] Mass spectrometric measurements
were performed on a JEOL JMS-700 Magnetic Sector or a Finnigan
MAT TSQ700 spectrometer. IR spectra were recorded on a Thermo
Scientific Nicolet 6700 FTIR spectrometer. X-Ray crystallographic
analysis was done on a Bruker Smart CCD or a Bruker APEX diffrac-
tometer. Microanalyses were performed on an Elementar vario
MIRKO cube instrument. (+)-(R)-Nopinone,[32] 4,4,6,6-tetramethylbi-
cyclo[3.1.1]heptan-2-one,[33] and 2,4,5,7-octanetetraone[34] were pre-
pared by following literature procedures. All other chemicals were
obtained from Aldrich, Acros, TCI, or Alfa Aesar and used without
further purification. IUPAC nomenclature for 2, 3, 5, and 7 was
based on the keto species.
1
tion for semiacetal structure 3a: H NMR ([D6]DMSO, 500.13 MHz):
d=1.18–1.28 (m, 2H, CH2), 1.30–1.43 (m, 3H, CH2), 1.44–1.51 (m,
1H, CH2), 1.51–1.60 (m, 2H, CH2), 1.61–1.70 (m, 2H, CH2), 1.78–1.92
3
(m, 2H, CH2), 2.14 (brs, 1H, CH), 2.37 (d, JH,H =3.1 Hz, 1H, CH), 2.46
3
3
(d, JH,H =3.6 Hz, 1H, CH), 2.55 (d, JH,H =3.5 Hz, 1H, CH), 3.95 (brs,
1H, CH), 7.59 ppm (s, 1H, OH); 13C{1H} NMR ([D6]DMSO,
125.76 MHz): d=22.0 (CH2), 23.5 (CH2), 27.5 (CH2), 28.2 (CH2), 34.8
(CH2), 37.1 (CH), 37.4 (CH2), 42.1 (CH), 45.9 (CH), 49.3 (CH), 57.9
(CH), 115.3 (C(OH)ÀO), 116.5 (C=CÀO), 144.6 (C=CÀO), 205.2 (C=O),
205.4 ppm (C=O). Characterization of 3c, obtained by acidification
of the aqueous phase with saturated NH4Cl solution instead of hy-
drochloric acid and subsequent extraction with Et2O at pH 7 fol-
lowed by crystallization from MeOH/Et2O: 1H NMR (CDCl3,
300.08 MHz): d=1.20–2.10 (m, 12H, CH2), 2.10 (m, 1H, CH), 2.40
(m, 1H, CH), 2.50 (m, 1H, CH), 2.65 (m, 1H, CH), 3.79 (s, 1H, OH),
3.89 (m, 1H, CH), 8.63 ppm (s, 1H, NH); 13C{1H} NMR (CDCl3,
75.46 MHz): d=22.3 (CH2), 24.1 (CH2), 28.2 (CH2), 28.9 (CH2), 35.6
(CH2), 37.2 (CH), 40.6 (CH2), 41.8 (CH), 47.4 (CH), 49.8 (CH), 59.6
(CH), 94.5 (NÀCÀC(O)), 108.5 (C=C), 139.6 (OÀCÀN), 205.6 (C=O),
211.0 ppm (C=O); MS (EI+): m/z (%): 82 (21), 150 (15), 176 (15), 178
General procedure for the preparation of tetracarbonyl
compounds
In a three-necked flask equipped with a dropping funnel and
a reflux condenser, 5.5 equiv of NaH was suspended in THF. The
dropping funnel was charged with 2.0 equiv of the carbonyl com-
pound and 1.0 equiv of (CO2Et)2 dissolved in THF. The NaH suspen-
sion was brought to reflux and the reagent mixture was added
dropwise. Reaction progress was monitored by TLC and NMR
(1–7 d). After complete conversion, the solvent was removed
under reduced pressure and the residue partitioned between H2O
and Et2O. The aqueous phase was washed with Et2O three times,
acidified to a pH of 1 with 3m hydrochloric acid, and extracted
(15), 204 (24), 205 (13), 206 (22), 217 (12, [MÀCOÀC2H4] +), 245 (46,
C
+
[MÀC2H4] ), 245 (15), 273 (100, [M] +); HRMS (EI+): m/z calcd for
C16H19O3N: 273.1365; found: 273.1364; IR (FT-ATR): n˜ =3343, 3296,
2947, 2870, 1737, 1699, 1660, 1603, 1455, 1396, 1299, 1224, 1199,
C
C
1087, 1037, 934 cmÀ1
.
1,2-Bis[(1R,5R)-6,6-dimethyl-2-oxobicyclo[3.1.1]heptan-3-yl]-
ethane-1,2-dione (5): Purification by flash chromatography (silica,
hexanes/ethyl acetate 10/1); colorless solid; 63%; 1H NMR (CDCl3,
300.51 MHz): d=0.94 (s, 6H, CH3), 1.35 (s, 6H, CH3), 1.46 (m, 2H,
Chem. Eur. J. 2015, 21, 1 – 8
www.chemeurj.org
5
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
CH2), 2.26 (m, 2H, CH), 2.58 (m, 4H, CH, CH2), 2.70 (m, 4H, CH2),
14.62 ppm (s, 2H, OH); 13C{1H} NMR (CDCl3, 75.56 MHz): d=21.7
(2C, CH3), 26.0 (2C, CH3), 26.4 (2C, CH2), 27.8 (2C, CH2), 39.4 (2C,
CH), 39.9 (2C, Cq), 55.2 (2C, CH), 107.0 (2C, C=CÀO), 165.4 (2C,
C=CÀO), 210.9 ppm (2C, C=O); MS (EI+): m/z (%): 137 (18,
PdII(C3H5) complex of 5 (10): Bright yellow solid; 93%; air-sensitive;
a 1/2/1 ratio of endo/endo, endo/exo, and exo/exo isomers leads to
a fourfold (1/1/1/1) signal pattern; 1H NMR (CDCl3, 300.08 MHz):
d=0.87/0.88/0.96/0.97 (4s, 4ꢁ6H, CH3), 1.27/1.28/1.28/1.29 (4s, 4ꢁ
6H, CH3), 1.34/1.35/1.43/1.45 (4d, 3JH,H =9.6 Hz, 4ꢁ2H, CH2), 2.19
(m, 4ꢁ2H, CH), 2.30–2.65 (m, 4ꢁ8H, CH2, CH), 2.87 (m, 4ꢁ4H,
CH2allyl), 3.86 (m, 4ꢁ4H, CH2allyl), 5.51 ppm (m, 4ꢁ2H, CHallyl); 13C{1H}
NMR (CDCl3, 75.46 MHz): d=21.5/21.6/21.7/21.7 (4ꢁ2C, CH3), 26.0/
26.0/26.0/26.0 (4ꢁ2C, CH3), 27.7/27.8/27.8/27.8 (4ꢁ2C, CH2), 28.2/
28.2/28.2/28.2 (4ꢁ2C, CH2), 38.9/38.9/39.0/39.0 (4ꢁ2C, Cq), 40.0/
40.0/40.1/40.1 (4ꢁ2C, CH), 55.2/55.2/55.2/55.2 (4ꢁ2C, CH), 55.6/
55.6/55.7/55.8 (4ꢁ2C, CH2allyl), 56.0/56.1/56.2/56.2 (4ꢁ2C, CH2allyl),
98.5/98.6/98.6/98.7 (4ꢁ2C, C=C-O), 112.6/112.6/112.6/112.6 (4ꢁ2C,
CHallyl), 181.2/181.3/181.3/181.3 (4ꢁ2C, O-C=C), 197.5/197.5/197.5/
197.5 ppm (4ꢁ2C, C=O); HRMS (FAB+): m/z calcd for
C26H34O4108Pd2: 626.0548; found: 626.0507.
[C10H13O2ÀCO]+), 165 (100, [C10H13O2]+), 302 (41, [MÀCO]+ ), 330
C
+
(20, [M]+ ); HRMS (EI ): m/z calcd for C20H26O4: 330.1831; found:
C
330.1837, IR (FT-ATR): n˜ =2971, 2956, 2924, 2869, 1607, 1541, 1450,
1337, 1322, 1208, 1192, 1061, 1005 cmÀ1; elemental analysis (%)
calcd for C20H26O4: C 72.64, H 7.93; found: C 72.64, H 8.09.
Ethyl
2-oxo-2-[(1R,5S)-2,2,6,6-tetramethyl-4-oxobicyclo[3.1.1]-
heptan-3-yl]acetate (7): Yellow oil; 70% relative to ethyl oxalate;
1H NMR (CDCl3, 300.51 MHz): d=1.05 (s, 3H, CH3), 1.24 (s, 3H, CH3),
3
1.30 (s, 3H, CH3), 1.36 (s, 3H, CH3), 1.39 (t, JH,H =7.2 Hz, 3H, CH3),
3
1.65 (d, JH,H =10.5 Hz, 1H, CH2), 1.82 (dd, 1H, CH2), 2.47–2.58 (m,
2H, CH), 4.36 (q, 3JH,H =7.2 Hz, 2H, CH2), 15.83 ppm (s, 1H, OH);
13C{1H} NMR (CDCl3, 75.57 MHz): d=14.1 (CH3), 25.0 (CH3), 26.8
(CH3), 27.1 (CH3), 27.7 (CH3), 30.4 (CH3), 35.3 (Cq), 41.2 (Cq), 55.7
(CH), 55.8 (CH), 62.2 (CH2), 113.7 (C=CÀO), 163.7 (CÀO), 166.9
RhI(nbd) complex of 3 (11): Bright yellow solid; quantitative yield;
1H NMR (CD2Cl2, 500.13 MHz): d=1.21 (m, 4H, CH2(nbd)), 1.22–1.28
(m, 6H, CH2), 1.54 (m, 2H, CH2), 1.71–1.76 (m, 4H, CH2), 2.63 (brs,
2H, CH), 2.91 (brs, 2H, CH), 3.80 (brs, 4H, CH(nbd)), 3.87 (brs, 4H,
CH=C(nbd)), 3.90 ppm (brs, 4H, CH=C(nbd)); 13C{1H} NMR (CD2Cl2,
125.76 MHz): d=24.7 (2C, CH2), 29.3 (2C, CH2), 40.8 (2C, CH), 43.4
(2C, CH2), 50.1 (2C, d, 2JC,Rh =2.9 Hz, CH(nbd)), 50.2 (2C, d,
2JC,Rh =2.9 Hz, CH(nbd)), 50.3 (2C, CH), 52.5 (2C, d, 1JC,Rh =11.1 Hz,
C=C(nbd)), 52.6 (2C, d, 1JC,Rh =11.0 Hz, C=C(nbd)), 52.8 (2C, d,
(CÀO), 210.9 ppm (C=O); MS (EI+): m/z (%): 83 (59, [C5H7O]+), 137
+
(73, [C8H9O2]+), 193 (100, [C12H17O2]+), 266 (48, [M]+ ); HRMS (EI ):
C
m/z calcd for C15H22O4: 226.1518; found: 226.1513; IR (FT-ATR): n˜ =
2976, 1739, 1712, 1608, 1472, 1388, 1370, 1280, 1247, 1214, 1189,
1008 cmÀ1; elemental analysis (%) calcd for C15H22O4: C 67.64, H
8.33; found: C 67.81, H 8.49.
1JC,Rh =11.3 Hz,
C=C(nbd)),
53.2
(2C,
d,
1JC,Rh =11.3 Hz,
3
C=C(nbd)), 60.9 (2C, d, 3JC,Rh =6.9 Hz, CH2(nbd)), 113.3 (2C, d, JC,Rh
=
1.9 Hz, C=C-O), 173.3 (2C, OÀC=C), 201.1 ppm (2C, C=O); HRMS
(FAB+): m/z calcd for C30H32O4Rh2: 662.0411; found: 662.0392; IR
(FT-ATR): n˜ =2946, 2866, 1577, 1436, 1381, 1299, 1291, 1270, 1246,
General procedure for the preparation of transition metal
complexes:
1219, 1097 cmÀ1
.
2.0 equiv of KOtBu was added to a solution of 1.0 equiv of tetracar-
bonyl compound in THF and the mixture was stirred for 15 min at
room temperature. Then 1.0 equiv [{MLCl}2] was added and stirring
continued over night at room temperature. The solvent was
evaporated under reduced pressure and the residue suspended in
CH2Cl2, filtered through Celite, and concentrated. The crude prod-
uct was washed with a small amount of pentane.
IrI(cod) complex of 3 (12): Orange solid; quantitative yield;
1H NMR (CD2Cl2, 300.51 MHz): d=1.25–1.41 (m, 6H, CH2), 1.57–1.70
(m, 10H, CH2, CH2(cod)), 1.82 (m, 4H, CH2), 2.23 (m, 8H, CH2(cod)), 2.86
(brs, 2H, CH), 3.23 (brs, 2H, CH), 4.00 ppm (m, 8H, CH=C(cod));
13C{1H} NMR (CD2Cl2, 125.76 MHz): d=24.3 (2C, CH2), 29.3 (2C, CH2),
31.3 (2C, CH2(cod)), 31.3 (2C, CH2(cod)), 31.3 (2C, CH2(cod)), 31.4 (2C,
CH2(cod)), 40.9 (2C, CH), 43.3 (2C, CH2), 50.4 (2C, CH), 59.1 (2C,
C=C(cod)), 59.2 (2C, C=C(cod)), 59.4 (2C, C=C(cod)), 59.7 (2C, C=C(cod)),
116.7 (2C, C=C-O), 171.3 (2C, OC=C), 202.2 ppm (2C, C=O); HRMS
(FAB+): m/z calcd for C32H40O4193Ir2: 874.2185; found: 874.2130; IR
(FT-ATR): n˜ =2938, 2867, 2829, 1581, 1444, 1375, 1292, 1272, 1249,
RhI(nbd) complex of 5 (8): Bright yellow solid; 90%; 1H NMR
(CDCl3, 400.18 MHz): d=0.80 (s, 6H, CH3), 1.20 (m, 4H, CH2(nbd)),
3
1.24 (s, 6H, CH3), 1.27 (d, JH,H =9.3 Hz, 2H, CH2), 2.16 (m, 2H, CH),
2.25 (m, 2H, CH2), 2.31–2.44 (m, 6H, CH2, CH), 3.76 (m, 4H, CH (nbd)),
3.88 (m, 4H, CH=C(nbd)), 3.95 ppm (m, 4H, CH=C(nbd)); 13C{1H} NMR
(CDCl3, 100.63 MHz): d=21.5 (2C, CH3), 25.9 (2C, CH3), 27.8 (2C,
CH2), 28.2 (2C, CH2), 38.8 (2C, Cq), 39.9 (2C, CH), 49.6 (4C, d,
1221, 1101, 1000 cmÀ1
.
1
2JC,Rh =2.5 Hz, CH(nbd)), 52.1 (2C, d, JC,Rh =11.1 Hz, C=C(nbd)), 52.3 (2C,
Acknowledgements
d, 1JC,Rh =10.8 Hz, C=C(nbd)), 52.4 (2C, d, 1JC,Rh =10.7 Hz, C=C(nbd)),
53.2 (2C, d, 1JC,Rh =11.0 Hz, C=C(nbd)), 54.6 (2C, CH), 60.7 (2C, d,
This work was supported by the European Research Council
under Grant Agreements No. StG 258740 (AMPCAT). G.S. ac-
knowledges support by the Fonds der Chemischen Industrie
for a Ph.D. fellowship. Support by bwGRiD, member of the
German D-Grid initiative, funded by the Ministry for Education
and Research (Bundesministerium fꢂr Bildung und Forschung)
and the Ministry for Science, Research and Arts Baden-Wꢂrt-
temberg is grateful acknowledged. We thank Thomas Wurm
for fruitful discussions.
3
3JC,Rh =6.8 Hz, CH2(nbd)), 98.6 (2C, d, JC,Rh =2.2 Hz, C=C-O), 179.6 (2C,
OÀC=C), 196.6 ppm (2C, C=O); HRMS (FAB+): m/z calcd for
C34H40O4Rh2: 718.1037; found: 718.1027; IR (FT-ATR): n˜ =2910, 1558,
1444, 1417, 1375, 1372, 1302, 1240, 1209, 1064, 1012 cmÀ1
.
RhI(CO)2 complex of 5 (9): Yellow orange solid; quantitative yield;
1H NMR (CDCl3, 500.13 MHz): d=0.84 (s, 6H, CH3), 1.32 (s, 6H, CH3),
1.34 (m, 2H, CH2), 2.22 (m, 2H, CH), 2.34 (m, 2H, CH2), 2.48 (m, 4H,
CH2), 2.67 ppm (t, 3JH,H =5.1 Hz, 2H, CH); 13C{1H} NMR (CDCl3,
125.76 MHz): d=21.4 (2C, CH3), 25.8 (2C, CH3), 27.2 (2C, CH2), 28.3
(2C, CH2), 39.3 (2C, Cq), 39.7 (2C, CH), 54.3 (2C, CH), 101.4 (2C, d,
3JC,Rh =3.0 Hz, C=C-O), 177.6 (2C, OÀC=C), 183.5 (2C, d,
1JC,Rh =74.1 Hz, RhCO), 183.6 (2C, d, 1JC,Rh =72.9 Hz, RhCO),
198.1 ppm (2C, C=O); HRMS (FAB+): m/z calcd for C24H25O8Rh2
([M+H]+): 646.9660; found: 646.9648; IR (FT-ATR): n˜ =2956, 2922,
2077, 2063, 2018, 1997, 1582, 1559, 1451, 1419, 1367, 1326, 1260,
Keywords: chelates · hydrogen bonds · ketones · molecular
switches · tautomerism
1244, 1209, 1093, 1065, 1016 cmÀ1
.
&
&
Chem. Eur. J. 2015, 21, 1 – 8
www.chemeurj.org
6
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
[1] Y. L. Chow, C. I. Johansson, Y.-H. Zhang, R. Gautron, L. Yang, A. Rassat,
S.-Z. Yang, J. Phys. Org. Chem. 1996, 9, 7–16.
[18] C. R. Noe, M. Knollmꢂller, P. Gꢅrtner, K. Mereiter, G. Steinbauer, Liebigs
Ann. 1996, 1996, 1015–1021.
[2] G. Zhang, J. Chen, S. J. Payne, S. E. Kooi, J. N. Demas, C. L. Fraser, J. Am.
Chem. Soc. 2007, 129, 8942–8943.
[3] P. J. Toscano, C. Dettelbacher, J. Waechter, N. P. Pavri, D. H. Hunt, E. T. Ei-
senbraun, B. Zheng, A. E. Kaloyeros, J. Coord. Chem. 1996, 38, 319–335.
[4] B. M. Bocknack, L.-C. Wang, F. W. Hughes, M. J. Krische, Tetrahedron
2005, 61, 6266–6275.
[19] I. J. Hart, Polyhedron 1992, 11, 729–734.
[20] F. Maier, O. Trapp, Angew. Chem. Int. Ed. 2012, 51, 2985–2988; Angew.
Chem. 2012, 124, 3039–3043.
[21] F. Maier, O. Trapp, Angew. Chem. Int. Ed. 2014, 53, 8756–8760; Angew.
Chem. 2014, 126, 8901–8905.
[22] M. Kamuf, O. Trapp, Chirality 2013, 25, 224–229.
[23] G. Storch, O. Trapp, Angew. Chem. Int. Ed. 2015, 54, 3580–3586; Angew.
Chem. 2015, 127, 3650–3656.
[5] A. Togni, G. Rist, G. Rihs, A. Schweiger, J. Am. Chem. Soc. 1993, 115,
1908–1915.
[6] L. Claisen, N. Stylos, Ber. Dtsch. Chem. Ges. 1887, 20, 2188–2191.
[7] J. H. M. Hill, D. M. Berkowitz, K. J. Freese, J. Org. Chem. 1971, 36, 1563–
1565.
[8] B. Unterhalt, U. Pindur, Arch. Pharm. 1977, 310, 264–268.
[9] C. Pariya, C. R. Sparrow, C.-K. Back, G. Sandꢃ, F. R. Fronczek, A. W. Maver-
ick, Angew. Chem. Int. Ed. 2007, 46, 6305–6308; Angew. Chem. 2007,
119, 6421–6424.
[10] R. W. Saalfrank, N. Lçw, B. Demleitner, D. Stalke, M. Teichert, Chem. Eur.
J. 1998, 4, 1305–1311.
[11] R. W. Saalfrank, C. Spitzlei, A. Scheurer, H. Maid, F. W. Heinemann, F.
Hampel, Chem. Eur. J. 2008, 14, 1472–1481.
[24] P. J. Taylor, G. van der Zwan, L. Antonov in Tautomerism (Ed.: L. Anto-
nov), Wiley-VCH, Weinheim, 2014, pp. 1–20.
[25] G. da Silva, Angew. Chem. Int. Ed. 2010, 49, 7523–7525; Angew. Chem.
2010, 122, 7685–7687.
[26] B. Bandyopadhyay, P. Pandey, P. Banerjee, A. K. Samanta, T. Chakraborty,
J. Phys. Chem. A 2012, 116, 3836–3845.
[27] V. B. Delchev, Monatsh. Chem. 2004, 135, 371–384.
[28] J. M. O’Connor, R. Uhrhammer, J. Am. Chem. Soc. 1988, 110, 4448–4450.
[29] G. Manca, M. Caporali, A. Ienco, M. Peruzzini, C. Mealli, J. Organomet.
Chem. 2014, 760, 177–185.
[30] J. Ruiz, B. F. Perandones, Chem. Commun. 2009, 2741–2743.
[31] G. R. Fulmer, A. J. M. Miller, N. H. Sherden, H. E. Gottlieb, A. Nudelman,
B. M. Stoltz, J. E. Bercaw, K. I. Goldberg, Organometallics 2010, 29, 2176–
2179.
[12] R. W. Saalfrank, C. Schmidt, H. Maid, F. Hampel, W. Bauer, A. Scheurer,
Angew. Chem. Int. Ed. 2006, 45, 315–318; Angew. Chem. 2006, 118,
322–325.
[13] R. W. Saalfrank, N. Lçw, F. Hampel, H.-D. Stachel, Angew. Chem. Int. Ed.
Eng. 1996, 35, 2209–2211; Angew. Chem. 1996, 108, 2353–2354.
[14] N. Nawar, Qatar Univ. Sci. J. 1994, 14, 105–107.
[15] M. Poje, K. Balenovic, J. Heterocycl. Chem. 1979, 16, 417–420.
[16] V. O. Kozminykh, L. O. Konshina, N. M. Igidov, J. Prakt. Chem. 1993, 335,
714–716.
[32] D. G. Lee, T. Chen, Z. Wang, J. Org. Chem. 1993, 58, 2918–2979.
[33] K. J. Moriarty, R. D. Rogers, L. A. Paquette, Organometallics 1989, 8,
1512–1517.
[34] T. M. Shironina, N. M. Igidov, E. N. Koz’minykh, L. O. Kon’shina, Y. S. Ka-
satkina, V. O. Koz’minykh, Russ. J. Org. Chem. 2001, 37, 1486–1494.
[17] E. N. Koz’minykh, N. M. Igidov, E. S. Berezina, G. A. Shavkunova, I. B. Ya-
kovlev, S. A. Shelenkova, V. ꢄ. Kolla, ꢄ. V. Voronina, V. O. Koz’minykh,
Pharm. Chem. J. 1996, 30, 458–462.
Received: February 8, 2015
Published online on && &&, 0000
Chem. Eur. J. 2015, 21, 1 – 8
www.chemeurj.org
7
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
FULL PAPER
Hydrogen bonding makes the differ-
ence! 1,3,4,6-Tetraketones with unantici-
pated and so far unknown tautomer-
ism-mediated isomerism due to differ-
ent modes of intramolecular hydrogen
bonding (i.e., six- and seven-membered
rings; see figure) are described. Struc-
tural requirements and options for
switching of the equilibrium isomeric
ratio were studied.
&
Tautomerism
G. Storch, M. J. Spallek, F. Rominger,
O. Trapp*
&& – &&
Tautomerization-Mediated Molecular
Switching Between Six- and Seven-
Membered Rings Stabilized by
Hydrogen Bonding
&
&
Chem. Eur. J. 2015, 21, 1 – 8
www.chemeurj.org
8
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!