Polycyclic compounds from aminopolyols and a-dicarbonyls: Structure and application in the synthesis of exoditopic ligands

Article abstract of DOI:10.1039/b500580a

The structure and stereochemistry of the crystalline 2 : 2 condensation product ("glytham") of glyoxal and tris(hydroxymethyl)aminomethane was conclusively determined by X-ray diffractometric analysis. The singular disposition of the heteroatoms suggests the employment of glytham as starting material for the synthesis of ditopic ligands for metal ions. Some derivatives of glytham were prepared and their binding properties towards alkaline metal ions were preliminarily investigated by ESI-MS and NMR. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b500580a

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
Polycyclic compounds from aminopolyols and a-dicarbonyls:
structure and application in the synthesis of exoditopic ligands
Giovanni B. Giovenzana,*a Giovanni Palmisano,*b Erika Del Grosso,a Lorella Giovannelli,a
Andrea Penonib and Tullio Pilatic
a Dipartimento di Scienze Chimiche Alimentari Farmaceutiche Farmacologiche, Via Bovio 6,
28100, Novara, Italy. E-mail: giovenza@pharm.unipmn.it
b Dipartimento di Scienze Chimiche e Ambientali, Via Valleggio 11, 22100, Como, Italy
c CNR-Istituto di Scienze e Tecnologie Molecolari, Via Golgi 19, 20133, Milano, Italy
Received 13th January 2005, Accepted 18th February 2005
First published as an Advance Article on the web 15th March 2005
The structure and stereochemistry of the crystalline 2 : 2 condensation product (“glytham”) of glyoxal and
tris(hydroxymethyl)aminomethane was conclusively determined by X-ray diffractometric analysis. The singular
disposition of the heteroatoms suggests the employment of glytham as starting material for the synthesis of ditopic
ligands for metal ions. Some derivatives of glytham were prepared and their binding properties towards alkaline
metal ions were preliminarily investigated by ESI-MS and NMR.
signals corresponding to two methylenes, one methine and one
Introduction
quaternary carbon atom. All chemical shifts corresponded to
saturated carbons, excluding structure with multiple bonds and
assigning the five unsaturations entirely to rings; furthermore,
the chemical shifts were compatible with the glytham structure,
containing two different N–CH2–O moieties, one N–CH(–C)–
O and a quaternary carbon atom. HSQC-NMR experiments
supported these assignments. The absence of evident couplings
in 1H-NMR spectrum offers no further help for the elucidation
of stereochemistry.
In 1976, a German patent1 described the synthesis of several
condensation products from aldehydes and tris(hydroxymethyl)-
aminomethane (“Tris” or “THAM”) to be used as lubricant
and fuel additives. Among them, particularly intriguing was
the reaction of “Tris” with glyoxal, claimed to give a novel
pentacyclic structure (1, Scheme 1) christened glytham.
A single-crystal X-ray diffractometric analysis was then
performed on glytham, to conclusively resolve the quest for
the relative stereochemistry of the stereogenic centers. The
ORTEP plot of the molecule definitely confirmed the pentacyclic
structure proposed for glytham (Fig. 1) and showed an unusual
all-axial arrangement for the four exocyclic oxygen atoms
Scheme 1
Nevertheless, this interesting structure was not unequivocally
corroborated by spectroscopic data; the patent reported elemen-
tal analysis and IR spectrum only, with no NMR or crystallo-
graphic data, except for a 1H-NMR spectrum of a O,O-diacetyl
derivative, the latter revealing an highly symmetric structure.
Following our interest in processes of covalent self assembly to
give heterocyclic compounds able to complex metal ions,2 we
were prompted to determine the real structure; apart from the
connectivity of the polycyclic compound, particularly impelling
was the determination of the relative (and eventually, absolute)
stereochemistry of the several stereogenic atoms involved, being
essential for the molecule morphology and behaviour towards
metal ions. In addition, we were interested in assessing the
extendibility of this transformation to substrates other than Tris
or glyoxal.
The synthesis of glytham was then repeated, following the
reported procedure. Dissolution of a stoichiometric amount of
Tris in 40% aqueous glyoxal resulted in an exothermic reaction
affording a clear yellow solution which, upon standing, left a
white crystalline precipitate of glytham. The solid, recrystallized
from boiling water, gave long colourless needles, with physical
properties identical to literature.1 Mass spectrometric analysis
(ESI) gave a protonated molecular ion at 287 amu, correspond-
ing to [(2 glyoxal + 2 Tris) − 2 H2O]·H+ equivalent to the
formula C12H18N2O6 with five rings or double bonds. 1H-NMR
analysis in D2O showed 3 resonances in the ratio 1 : 2 : 1. In the
13C-NMR spectrum four resonances were present, confirming
high symmetry of the molecule. APT-13C-NMR revealed these
Fig. 1 Glytham: ORTEP plot. Atomic displacement parameters at 50%
probability level. H atoms not to scale.
The formation of the glytham molecule appears as a genuine
covalent self-assembly process, in which two substrates react in
a 2 : 2 ratio in a high-yielding and completely stereoselective
transformation. The latter is extremely remarkable, as in a
single step four C–O bonds, four C–N bonds, five rings and
four stereocentres are efficiently formed, representing one of the
reactions with a very high intricacy quotient.3
To test the generality of this reaction other commercially
available aminopolyols (i.e., serinol 2, 2-amino-2-methyl-1,2-
propanediol 3) were subjected to the optimized reaction with
glyoxal.
As shown in Scheme 2, the substitution of a hydroxymethyl
moiety should lead to the modification of the pendant arms of
the molecule.
Reaction of 3 with glyoxal gave a product in the usual
way, although slower. However, reaction of glyoxal with 2
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 4 8 9 – 1 4 9 4
1 4 8 9
©

View Article Online
Scheme 2 Variation of aminopolyol reactant.
afforded only complex mixtures, containing polymeric products
as did reaction of 1 and 3 with enolizable a-diketones (e.g. 2,3-
butanedione, 1,2-cyclohexanedione).
The behaviour of non-enolizable a-diketones such as 1,2-
diphenyl-1,2-ethanedione (benzil) 6 and acenaphthenequinone
was next investigated. The latter was extremely insoluble in polar
solvents, thereby thwarting successful reaction with aminopoly-
ols. On the other hand, an equimolar mixture of benzil and
Tris was found to be quite soluble in MeOH–H2O mixtures,
yielding after 1 day colourless needles. MS analysis ruled out
the expected 2 : 2 condensation, indicating a molecular formula
compatible with a 1 : 1 condensation product with elimination of
1 H2O. 1H-NMR spectra evidenced a rigid structure, devoid of
symmetry elements, confirmed by 13C-NMR spectrum. COSY
and HETCOR experiments allowed the assignment of the 3,6-
dioxa-8-azabicyclic structure 7 (Scheme 3). This ring system
has few examples in the literature,4 the more comparable being
unexpectedly formed in a similar reaction between Tris and
pyruvaldehyde.5
Fig. 2 ORTEP plot of compound 7. Atomic displacement parameters
at 50% probability level. H atoms not to scale.
with a trigonal bipyramidal and w-octahedral geometries,
respectively (Fig. 3).
Our initial efforts on the functionalization of glytham
proved less trivial than its structure would suggest. At-
tempts to alkylate 1 with either 2-methoxyethyl tosylate
or 2-(2-methoxyethoxy)ethyl tosylate under different reaction
conditions (i.e., NaH–THF, NaH–DMF, KOH–THF, KOH–
Bu4N+HSO4−–PhMe), yielded intractable mixtures showing lit-
tle or no evidence of containing the desired alkylated derivatives.
The failure of this approach was almost certainly due to the
modest solubility of 1 (slightly soluble in cold water; readily
soluble only in boiling water or hot pyridine).
The introduction of side chains bearing the oxygen-donor
groups was then attempted through acylation of the hydroxyl
groups.1 Gratifyingly, treating a solution of glytham in pyridine
at 0 ◦C and then at room temperature overnight with 2.2 equiv. of
methoxyacetyl chloride 9 (prepared by reacting methoxyacetic
acid with thionyl chloride) resulted in an exothermic reaction
and usual workup led to the isolation (56% overall yield) of the
desired O,O-diacylglytham 11 (Scheme 4).
Scheme 3 Condensation of aminopolyols with benzyl.
Unfortunately, it was impossible to establish the relative
stereochemistry of the carbon atom bearing the tertiary alcohol
by spectrometric analysis. A single crystal X-ray diffractometric
analysis was then performed on the needles of 7 obtained in the
reaction, and the result is depicted in Fig. 2, showing the exo
configuration of the tertiary hydroxyl group on C(2).
As shown in Scheme 3, 2-amino-2-methyl-1,3-propanediol 3
and benzil 6 afforded in good yield the corresponding bicyclic
adduct as a single product 8. As previously noted, serinol
failed to give a clean reaction with benzil, too; once again,
no crystalline product was obtained from the complex (as
ascertained by HPLC) mixture.
The preliminary investigations on the condensation reaction
of aminopolyols and a-dicarbonyls show a marked influence
of the substituents on the products obtained. The pentacyclic
adducts are obtained only when glyoxal is the a-dicarbonyl
component, while less complex bicyclic structures were obtained
when the steric hindrance of the a-dicarbonyl was increased.
The structure of glytham and its dideoxyderivative 5 appeared
to be a promising building block for the synthesis of ligands.
In particular, the symmetric disposition of the donor atoms,
organized in two distinct N–O3 sets with electron pairs directed
towards the outer surface, suggests its employment in the syn-
thesis of exoditopic ligands. The latter are attracting increasing
attention in view of their potential in metallosupramolecular
chemistry,6 but the scarce reports about them reflect the difficult
task of their lengthy syntheses.7
We were then prompted to convert the glytham molecule (and
its dideoxyderivative 5) into exoditopic ligands. When one or
two donor atoms are appended to the hydroxymethyl arms of
glytham scaffold, polydentate coordination would be accessed
Scheme 4 O,O-Diacyl glytham-based exoditopic ligands.
1 4 9 0
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 4 8 9 – 1 4 9 4

View Article Online
Fig. 3 Putative coordination modes for glytham and its derivatives.
Similarly, 1 also gave the corresponding exoditopic lig-
and 12 (44% overall yield) in the presence of the higher
homolog 2-(2-methoxyethoxy)acetyl chloride 10 (from 2-(2-
methoxyethoxy)acetic acid and thionyl chloride).
low to appreciate higher adducts, it is possible to detect small
signals of 1 : 2 adducts [11·Li2]2+ and [12·Li2]2+.
Affinity determination was not possible for 4: although
solutions of the ligand are indefinitely stable, the addition of
trace metal ions catalyzes its rapid decomposition. However,
from ESI-MS infusion experiments, it is possible to assess that:
decomposition takes place within seconds at lM concentra-
tion;
The functionalization of compound 4 appeared more difficult,
lacking the primary alcoholic groups. Nevertheless, the tertiary
amino groups offer a chance for linking further donor atoms,
through their oxidation to N-oxide. N-Oxides are known to act
efficiently as coordinating groups, binding metal ions through
their anionic oxygen atom.8 Reaction of 4 with hydrogen
peroxide in water solution produced a complex mixture of
partially oxidized and decomposed 4. However, oxidation of 4
with 2 equiv. of m-chloroperbenzoic acid in dichloromethane at
room temperature, resulted in clean conversion to N,Nꢀ-dioxide
13 (75%)(Scheme 5).
decomposition involves a rapid deoxygenation to give a mono
N-oxide, fairly stable in methanol–water solution;
the mass spectra of solutions of 13 in the presence of alkaline
metal ions prior to decomposition show a marked affinity for
lithium rather than sodium or potassium ions.
NMR spectra of 13 in the presence of alkaline metal ions,
run in D2O solution, confirm the rapid decomposition of the
molecule.
The decomposition of 13 to monoxide may also be induced
thermally, as ascertained by TGA-DSC (Fig. 5).
From thermogravimetric run it can be observed that the
degr◦adation of 13 is a two-step process, the first (onset at about
140 C) corresponding to the loss of one oxygen atom (5.02%
of the total weight) and the second to the incoherent decom-
position of the molecule starting from about 170.0 ◦C. These
reactions are confirmed by exothermic peaks shown by the DSC
thermograms.
Scheme 5 Exoditopic ligand from oxidation of 4.
In conclusion, the compounds 11, 12 and 13 synthesized
are rare examples of multidentate exoditopic ligands, whose
preparation is short, simple and cheap, involving a self-assembly
key step from readily available starting materials; their coor-
dination behaviour will be analyzed in more detail further.
The easily obtained pentacyclic structure of glytham and its
dideoxyderivative 5 represent rigid polyfunctional scaffolds with
a virtually unexplored chemistry; moreover, the noteworthy
covalent self-assembly process leading to the construction of
the pentacyclic structure need to be investigated further to
exploit its potential en route to novel complex polycyclic
molecules.
Compound 13 embodies two O3-subsets, suited to bind two
metal ions. The number of donor atoms and the opposite
orientation of the two subsets provide a rare example of linear
multidentate exoditopic ligand, suitable for the preparation of
infinite chain monodimensional polymer.
In order to exploit optimally their task, preliminary ex-
periments were performed, using electrospray ionization mass
spectrometry (ESI-MS) for obtaining an initial assessment of
the metal binding affinities of new ligands. ESI-MS has proved
invaluable for qualitative and quantitative evaluation of host
binding selectivities in weakly bound noncovalent host–guest
complexes, requiring only negligible amounts of material.9,10 In
view of the denticity and nature of donor atoms, binding exper-
iments for 11, 12 and 13 were limited to alkali metal ions. ESI-
MS spectra were acquired for MeOH solutions containing the
ligand (10 lM), LiCl, NaCl, and KCl. Even if the concentration
of metal ions is in 1 : 1 ratio, the protonated molecular peak
is hardly detected (<0.8%), indicating an overwhelming affinity
for alkali ions. At higher metal : ligand ratio (5 : 1), it is possible
to discern a significant selectivity in the order Li+ > Na+ > K+
(Fig. 4) for both 11 and 12, in excellent agreement with MM
predictions performed on similar structures.11
Experimental
Reagents were obtained from Aldrich and used without further
purification. 1H-NMR and 13C-NMR spectra were registered on
a Jeol Eclipse ECP300 spectrometer at 300 MHz and 75.4 MHz,
respectively. ESI-MS were recorded on a Finnigan Mat spec-
trometer. DSC analysis of 13 was performed on a Perkin-Elmer
Pyris 1 DSC, while thermogravimetric study (TGA) was carried
out on a Perkin-Elmer Pyris 1 TGA. The sample (3–5 mg),
subjected to a scanning rate of 20 ◦C min−1 under a 20 ml min−1
nitrogen purge, was heated in vented aluminium pan between 25
and 250 ◦C (DSC) or in open platinum pan up to 700 ◦C (TGA).
Although the complexes probably possess limited stability and
the concentration involved in these ESI-MS experiments are too
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 4 8 9 – 1 4 9 4
1 4 9 1

View Article Online
Fig. 4 a ESI-MS determination of relative binding affinities—compound 11. b ESI-MS determination of relative binding affinities—compound 12.
Elemental analyses were carried out with a Perkin Elmer Serie
II CHNS/O 2400 analyzer.
N, 9.79%. Mass spectrum (ESI): m/z 287 (MH+), 257 (MH+ −
(s, 4H), 3.90 (bs, 8H), 4.74 (s, 4H). 13C-NMR (DMSO-d6) 64.3,
72.6, 74.1, 88.4.
1
CH2O). Calc. for C12H18N2O6: 286 amu. H-NMR (D2O) 3.65
(4aR*,4bR*,8aS*,8bS*)-[6a-(Hydroxymethyl)hexahydro-
2H,6H-1,4,5,8-tetraoxa-6b,8c-diazadicyclopent[cd,ij]-s-
indacene-2(3H)-yl]methanol (“glytham”) (1)
2a,6a-Dimethyloctahydro-2H,6H-1,4,5,8-tetraoxa-6b,8c-
diazadicyclopenta[cd,ij]-s-indacene (5)
Tris(hydroxymethyl)aminomethane (50.0 g, 0.413 mol) was
dissolved in water (150 mL) and 50% aqueous glyoxal (24.0 g,
0.414 mol) was added dropwise in 5 min; a mild exothermic
reaction ensued. The yellow solution was stirred for 1 h, during
which colourless crystals separated out. The solid was filtered,
washedtwicewith cold water, recrystallized from water anddried
in vacuo. Yield 53.8 g (91%). Mp 356–358 ◦C (dec.). Found: C,
50.25; H, 6.51, N, 9.62. Calc. for C12H18N2O6: C, 50.35; H, 6.34,
2-Amino-2-methyl-1,3-propanediol (9.04 g, 0.086 mol) was
dissolved in water (20.0 mL). Glyoxal (50% aqueous solution,
10.0 g, 0.086 mol) was added in one portion. The solution was
left standing for 10 days; the crystalline precipitate was filtered,
washed twice with cold water and dried in vacuo. Yield 5.58 g
(51%). Mp 290–294 ◦C (dec.). Found: C, 56.47; H, 7.31, N,
10.87. Calc. for C12H18N2O6: C, 56.68; H, 7.13, N, 11.02%. Mass
1 4 9 2
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 4 8 9 – 1 4 9 4

View Article Online
2-Methoxyacetyl chloride (9)
The crude chloride was purified by d◦istillation, obtaining a clear
colourless liquid. Yield 69%. Bp 98 C. 1H-NMR (CDCl3) 3.47
(s, 3H), 4.34 (s, 2H). 13C-NMR (CDCl3) 59.7, 77.7, 172.0.
2-(2-Methoxyethoxy)acetyl chloride (10)
The crude chloride was used as such in the next step. Yield 51%.
1H-NMR (CDCl3) 3.28 (s, 3H), 3.49 (m, 2H), 3.68 (m, 2H), 4.42
(s, 2H). 13C-NMR (CDCl3) 58.9, 71.7, 71.8, 76.7, 172.0.
{6a-[(2-Methoxyacetoxy)methyl]hexahydro-2H,6H-1,4,5,8-
tetraoxa-6b,8c-diazadicyclopenta[cd,ij]s-indacen-2(3H)-
yl}methyl 2-methoxyacetate (11)
Glytham (1, 5.00 g, 17 mmol) was suspended in anhydrous
pyridine (20 mL). 2-Methoxyacetyl chloride (9, 4.2 g, 38 mmol)
was slowly added dropwise, maintaining the reaction mixture
at 0–5 ◦C with a ice–water bath. The whole was kept 30 min
at 0–5 ◦C and then stirred overnight at room temperature. The
orange reaction mixture was filtered and the filtrate poured into
a 10% aqueous citric acid solution (150 ml) and extracted with
dichloromethane (4 × 15 mL). The organic extracts were washed
in the order with 5% aqueous ammonia and water, then dried
(Na2SO4), filtered and evaporated in vacuo. The crude light-
brown solid was recrystallized from ethyl acetate, obtaining 11
as colourless crystals. Yield 4.1 g (56%). Mp 119–121 ◦C. Found:
C, 49.99; H, 6.22, N, 6.51. Calc. for C18H26N2O10: C, 50.23; H,
6.09, N, 6.51%. Mass spectrum (ESI): m/z 453 (MNa+), 431
(MH+). Calc. for C18H26N2O10: 430 amu. 1H-NMR (CDCl3) 3.43
(m, 8H), 3.84 (s, 6H), 4.04 (s, 4H), 4.25 (s, 4H), 4.75 (s, 4H). 13C-
NMR (CDCl3) 58.5, 66.8, 69.6, 72.0, 72.6, 88.5, 170.1.
Fig. 5 Simultaneous TG-DSC trace of 13 (solid line, TG; dashed line,
DSC signal).
spectrum (ESI): m/z 255 (MH+), 225 (MH+ − CH2O). Calc.
1
for C12H18N2O4: 254 amu. H-NMR (DMSO-d6) 1.26 (s, 6H),
3.47 (d, 4H, J = 8.5 Hz), 3.78 (d, 4H, J = 8.5 Hz), 4.58 (s, 4H).
13C-NMR (DMSO-d6) 23.0, 70.2, 75.5, 88.4.
(1R*,2R*,5S*)-1-Hydroxymethyl-4,5-diphenyl-3,6-dioxa-8-
azabicyclo[3.2.1]octan-4-ol (7)
Tris(hydroxymethyl)aminomethane (5.0 g, 41 mmol) was dis-
solved in water (20 mL) and benzil (8.6 g, 41 mmol) was
added in a single portion. Methanol (100 ml) was added and
the suspension was heated to 40 ◦C to give a clear yellow
solution. After 2 h heating, the reaction mixture was left standing
overnight, obtaining crude 7 as colourless needles. The product
was filtered, washed thoroughly with water, recrystallized from
water–acetone and finally dried in vacuo. Yield 9.1 g (71%). Mp
189–191 ◦C (dec.). Found: C, 68.93; H, 6.29, N, 4.37. Calc. for
C18H19NO4: C, 68.99; H, 6.11, N, 4.47%. Mass spectrum (ESI):
m/z 314 (MH+), 296 (MH+ − H2O). Calc. for C18H19NO4: 313
amu. 1H-NMR (CD3OD) 3.65–3.87 (m, 4H), 4.24 (dd, 1H, J1 =
10.3 Hz, J2 = 1.5 Hz), 4.37 (d, 1H, J = 6.9 Hz), 7.03–7.21 (m,
10H). 13C-NMR (CD3OD) 61.6, 66.4, 69.0, 71.4, 99.8, 100.9,
127.2, 127.7, 128.4, 128.6, 129.2, 129.4, 139.6, 140.8.
{6a-{[2-(2-Methoxyethoxy)acetoxy]methyl}hexahydro-2H,6H-
1,4,5,8-tetraoxa-6b,8c-diazadicyclopenta[cd,ij]s-indacen-2(3H)-
yl}methyl 2-(2-methoxyethoxy)acetate (12)
Glytham (1, 5.00 g, 17 mmol) was suspended in anhydrous
pyridine (20 mL). 2-(2-Methoxyethoxy)acetyl chloride (10, 5.9 g,
38 mmol) was slowly added dropwise, maintaining the reaction
mixture at 0–5 ◦C with an ice–water bath. The whole was kept
30 min at 0–5 ◦C and then stirred overnight at room temperature.
The orange reaction mixture was filtered and the filtrate poured
into a 10% aqueous citric acid solution (150 ml) and extracted
with dichloromethane (4 × 15 mL). The organic extracts were
washed in the order with 5% aqueous ammonia and water, then
dried (Na2SO4), filtered and evaporated in vacuo. The crude
reddish oily product was purified by flash chromatography (SiO2,
eluant ethyl acetate–hexane–2-propanol 8 : 1 : 1) and finally
recrystallized from ethyl acetate, obtaining 12 as colourless
1-Methyl-4,5-diphenyl-3,6-dioxa-8-azabicyclo[3.2.1]octan-4-ol (8)
The procedure employed for 7 was adopted for this prepara-
tion, starting from 2-amino-2-methyl-1,3-propanediol (5.0 g,
48 mmol) and benzil (10.1 g, 48 mmol) in water–methanol
(20 mL–130 mL). Recrystallization of the precipitated product
from water–acetone gave pure 8. Yield 11.30 g (80%). Mp
176–178 ◦C (dec.). Found: C, 72.82; H, 6.45, N, 4.57. Calc.
for C18H19NO3: C, 72.71; H, 6.44, N, 4.71%. Mass spectrum
(ESI): m/z 298 (MH+). Calc. for C18H19NO3: 297 amu. 1H-NMR
(CDCl3) 1.26 (s, 3H), 2.54 (bs, 1H), 3.50 (dd, 1H, J1 = 7.1 Hz,
J2 = 1.9 Hz), 3.72 (d, 1H, J = 10.7 Hz), 4.12 (dd, 1H, J1 =
11.1 Hz, J2 = 2.0 Hz), 4.45 (d, 1H, J = 7.1 Hz), 4.99 (bs, 1H),
7.03–7.28 (m, 10H). 13C-NMR (CDCl3) 17.5, 61.6, 71.1, 74.5,
97.9, 100.5, 126.8, 126.9, 127.2, 127.8, 128.0, 128.1, 137.7, 137.8.
◦
crystals. Yield 3.9 g (44%). Mp 89–92 C. Found: C, 50.91; H,
6.79, N, 5.35. Calc. for C22H34N2O12: C, 50.96; H, 6.61, N, 5.40%.
Mass spectrum (ESI): m/z 541.7 (MNa+), 519.4 (MH+). Calc.
for C22H34N2O12: 518 amu. 1H-NMR (CDCl3) 3.38 (m, 6H), 3.58
(m, 4H), 3.71 (m, 4H), 3.86 (m, 8H), 4.05 (s, 4H), 4.18 (s, 4H),
4.26 (s, 4H). 13C-NMR (CDCl3) 58.6, 67.0, 70.0, 71.6, 71.9, 72.1,
72.5, 88.4, 170.2.
2a,6a-Dimethyloctahydro-2H,6H-1,4,5,8-tetraoxa-6b,8c-
diazoniadicyclopenta[cd,ij]s-indacene-6b,8c-diolate (13)
Compound
4 (0.257 g, 1.01 mmol) was dissolved in
dichloromethane (30 mL). m-Chloroperoxybenzoic acid
(0.800 g, 4.34 mmol) was added in four portions during 30 min
with vigorous stirring. After the addition a voluminous white
precipitate separated out and the reaction mixture was stirred at
room temperature for 3 h. The precipitate was then filtered out
and the filtrate was evaporated in vacuo. The solid residue was
purified by column chromatography (SiO2, eluant methanol–
28% aqueous ammonia 9 : 1 → 8 : 2). The purified product
was redissolved in water, washed twice with dichloromethane to
General procedure for the preparation of acyl chlorides
The acid (50 mmol) was slowly dropped in a round bottomed
flask containing thionyl chloride (150 mmol). The resulting
clear solution was stirred at room temperature for 15 min, then
refluxed for 1 h. Thionyl chloride and volatiles are removed by
distillation obtaining the crude acyl chloride.
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 4 8 9 – 1 4 9 4
1 4 9 3

View Article Online
remove the last traces of m-chlorobenzoic acid and the aqueous
layer was then evaporated to give pure 13 as amorphous off-
white solid (0.217 g, 75%). Mp 120 ◦C (sint) 150 ◦C (dec). Mass
spectrum (ESI): m/z 287 (MH+), 271 (MH+ − O). Calc. for
conformation, while the five-membered one shows an envelope
˚
conformation with N5 out of plane of about 0.60 A. The crystal
density, q = 1.320 g cm−3, is lower than in 1, despite the O9–
H9 · · · O23ii H bond [O9 · · · O23ii 2.780(2), H9 · · · O23ii 1.93(3)
1
iii
iv
˚
C12H18N2O6: 286 amu. H-NMR (D2O) 5.14 (s, 4H), 4.56 (d,
A, ii = x, y, 1 − z] and the bifurcated H bond N5 · · · H23 · · · O9
4H, J = 9.9 Hz), 4.01 (d, 4H, J = 9.9 Hz). 13C-NMR (D2O)
[N5iii · · · H23 2.24(2), H23 · · · O9iv 2.29(2), N5iii · · · O23 2.927(2),
iii
˚
96.8, 82.8, 77.9, 20.6.
O23 · · · O9 2.982(2) A, iii = x, 1 − y, 1/2 − z]; this is probably
due to the two phenyl rings that show very weak packing
interaction.†
Crystal structure analysis
Crystal data of compounds 1 and 7 were collected at room tem-
perature on a Bruker SMART-APEX diffractometer equipped
with CCDC device. The structures were solved by direct methods
using the SIR-9212 program and refined by SHELX-97 full-
matrix least-squares.13
Acknowledgements
Two of us (G.P., A.P.) are grateful for the financial support
(Progetto d’Eccellenza per la Ricerca d’Ateneo 2002) by the
Universita` dell’Insubria. We thank the reviewers for fruitful
suggestions and for ref. 11.
Crystal data of compound 1. C12H18N2O6, M = 286.28, a =
◦
˚
6.2236(12), b = 9.575(2), c = 10.455(2) A, b = 92.630(10) , V =
3
˚
References
622.4(2) A , space group P21/n (no. 14), Z = 2, l(Mo-Ka) =
0.123 mm−1, 11124 measured reflections, 2200 unique (Rint
=
1 S. J. Brois, J. Ryer and E. Winans, (Exxon Research and Engineering),
Ger. Offen., DE 2617680, 1976 (Chem. Abstr., 1977, 87, 455620).
2 S. Aime, C. Cavallotti, E. Gianolio, G. B. Giovenzana, G. Palmisano
and M. Sisti, Tetrahedron Lett., 2002, 43, 8387–8389.
3 (a) P. L. Fuchs, Tetrahedron, 2001, 57, 6855–6875; (b) W. H. Whitlock,
J. Org. Chem., 1998, 63, 7982–7989.
4 (a) G. Gaudiano, E. Frank, M. S. Wysor, D. Averbuch and T. H.
Koch, J. Org. Chem., 1993, 58, 7355–7363; (b) G. Just and W.
Zehetner, J. Chem. Soc., Chem. Commun., 1971, 81–82; (c) G.
Gaudiano and T. H. Koch, J. Org. Chem., 1993, 52, 3073–3081.
5 W. A. Bubb, Magn. Reson. Chem., 1997, 35, 191–198.
6 S. Leininger, B. Olenyuk and P. J. Stang, Chem. Rev., 2000, 100,
853–908.
0.0210) which were used in all calculations. The final wR(F2) was
0.0967 (all data). The molecule lies on a crystallographic centre
of symmetry, but its symmetry is near to C2h, being 0.0307 the
molecular RMS14, including the H atoms. Molecular geometry
does not show any particular feature, with the exclusion of
˚
the quite strained N1–C5 bond [1.506(1) A]. The central six-
membered ring has pressed chair conformation, being the
angle between the planes through N1,C2,C8 and C2,C8,C2ꢀ,C8ꢀ
147.93(5)◦ to be compared with 120◦ for ideal cyclohexane ring.
The five membered rings show envelope conformation with the
˚
oxygen atom out of plane by 0.589 and 0.544 A for O3 and
7 L. A. Paquette, J. Tae and J. C. Gallucci, Org. Lett., 2000, 2, 143–146
and references cited therein.
O7, respectively. The molecular packing is characterized by
˚
a medium nearly linear H bond [2.07(2) and 2.945(1) A for
8 D. Bayot, B. Tinant, B. Mathieu, J.-P. Declercq and M. Devillers,
Eur. J. Inorg. Chem., 2003, 737–743.
9 (a) S. Inokuma, T. Sakaizawa, T. Funaki, T. Yonekura, H. Satoh, S.-I.
Kondo, Y. Nakamura and J. Nishimura, Tetrahedron, 2003, 59, 8183–
8190; (b) A. Szczepa´nska, P. Sałan´ski and J. Jurczak, Tetrahedron,
2003, 59, 4775–4783.
10 D.-S. Young, H.-Y. Hung and L. K. Liu, Rapid Commun. Mass
Spectrom., 1997, 11, 769–773.
11 (a) R. D. Hancock and K. Hegetschweiler, J. Chem. Soc., Dalton
Trans., 1993, 14, 2137–2140; (b) K. Hegetschweiler, R. D. Hancock,
M. Ghisletta, T. Kradolfer, V. Gramlich and H. W. Schmalle, Inorg.
Chem., 1993, 32, 5273–5284.
N1 · · · H10i and N1 · · · O10i, respectively (i = 3/2 − x, −1/2 +
y, 3/2 − z)] that is the main responsible for the high crystal
density, q = 1.528 g cm−3.†
Crystal data of compound 7. C18H19NO4, M = 313.34, a =
◦
˚
15.6300(19), b = 12.6665(15), c = 7.9832(10) A, b = 93.946(3) ,
3
˚
V = 1576.7(3) A , space group Cc (no. 9), Z = 4, l(Mo-
Ka) = 0.093 mm−1, 27040 measured reflections, 1826 unique
(Rint = 0.0237) which were calculated in all calculations. The final
wR(F2) was 0.0845 (all data). The only unusual bond distance
˚
is C1–C6 [1.574(2) A]. The six-membered hetero-ring has chair
12 G. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi, M. C.
Burla, G. Polidori and M. Camalli, J. Appl. Crystallogr., 1994, 27,
435.
† CCDC reference numbers 260606 (1) and 260607 (7). See
http://www.rsc.org/suppdata/ob/b5/b500580a/ for crystallographic
data in .cif or other electronic format.
13 G. M. Sheldrick, 1997. SHELX 97. University of Go¨ttingen,
Germany.
14 T. Pilati and A. Forni, J. Appl. Crystallogr., 1998, 31, 503–504.
1 4 9 4
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 4 8 9 – 1 4 9 4