Crystal structures of conformationally locked cyclitols: An analysis of hydrogen-bonded architectures and their implications in crystal engineering

FULL PAPER

DOI: 10.1002/ejoc.200600610

Crystal Structures of Conformationally Locked Cyclitols: An Analysis of

Hydrogen-Bonded Architectures and their Implications in Crystal Engineering

Goverdhan Mehta,*^[a]Saikat Sen,^[a]and Senaiar S. Ramesh^[a]

Keywords: Crystal engineering / Conformational locking / Cyclitols / Fused-ring systems / Hydrogen bonds

A qualitative study has been carried out on selected polycy-

clitols to evaluate the potential of conformational locking of

polycyclitols to be described in terms of a small group of in-

tramolecularly bonded molecular motifs linked to their re-

hydroxy groups in lending predictability to the O–H···O hy- spective neighbors by four O–H···O bonds. By using this

drogen-bonding network observed in the crystal structures

of such compounds. The polycyclitols employed in this study

are conformationally locked with all the hydroxy groups des-

model and the results of CSD analyses of polyols as a guide,

the O–H···O hydrogen-bonded packing motifs most likely to

be observed in the crystal structure of each polycyclitol were

tined to be axial owing to the trans ring fusion(s) in the poly- proposed and compared with those obtained experimentally.

cyclic carbon framework. The consequent formation of intra-

molecular O–H···O hydrogen bonds between the 1,3-syn di-

axial hydroxy groups now permits any packing pattern in the

Germany, 2007)

of empirical rules that govern the packing of polyols in the

solid state. An attempt in this direction led Jeffrey and

Saenger to point out that hydrogen bonding in carbo-

hydrates follows certain patterns that are based on two pri-

mary concepts:^[3a,3b]a) maximization of the total number

of hydrogen bonds per molecule using as many donor/ac-

ceptor oxygen atoms as possible and b) maximization of the

cooperativity by forming as many finite and infinite chains

of hydrogen bonds as possible.

Introduction

Crystal engineering is being increasingly recognized as

the supramolecular equivalent of organic synthesis.^[1,2]The

rational design of the crystal structure of an organic com-

pound depends heavily on the predictability of the molecu-

lar self-organization through noncovalent interactions.

Though sufficiently weaker than covalent bonds, these

supramolecular interactions are directional enough to co-

operatively assemble molecules in a manner that can be rea-

sonably predicted.^[1,2]Among the variety of intermolecular

interactions that have been recognized in the crystal struc-

tures of organic compounds, the O–H···O hydrogen bond is

among the most commonly encountered, thoroughly

studied and extensively documented noncovalent bond.^[3]In

fact, O–H···O hydrogen bonds appear in a large number of

supramolecular synthons that are routinely utilized in crys-

tal engineering.^[1,2]

Polyhydroxylated compounds, which include many biolo-

gically important molecules such as sugars and inositols,

have long been used as model systems for the systematic

study of O–H···O hydrogen bonds.^[3a,b,d]The hydrogen-

bonding patterns in the crystal structures of these com-

pounds, as determined by neutron diffraction and high-pre-

cision X-ray analyses, have been incisively studied and ex-

tensively reported. Not surprisingly, the large database of

these accurately determined crystal structures has, over the

years, constantly stimulated research into formulating a set

The former of the two concepts highlights a fundamental

feature of any self-assembling process, striving to optimize

all the interactions available at its disposal. Not surpris-

ingly, it has been found to hold good in the crystal struc-

tures of not only carbohydrates, but also a number of other

mono- and polyhydroxylated species as well. The second

concept, namely the occurrence of cooperative O–H···O hy-

drogen-bonding chains, has however been shown recently

by Taylor and Macrae to be strongly dependent on not only

the number of OH groups in a particular polyhydroxylated

species, but also the steric environment (the degree of sub-

stitution) around each hydroxy functionality.^[4]Based on a

CSD survey encompassing 144 monoalcohol (C_nH_mOH)

and 101 dialcohol [C_nH_m(OH)₂] crystal structures, the au-

thors were further able to lay down certain empirical rules

governing the O–H···O hydrogen-bonding motif preferred

in a particular mono- or dialcohol packing assembly. The

pivotal role played by steric effects in the crystal packing of

a polyol was again underlined in a contemporary CSD

study on vic-diols by Brock, wherein it was observed that

the extent of O–H···O bond formation itself depends on the

[a] Department of Organic Chemistry, Indian Institute of Science,

Bangalore 560012, India

degree of substitution of the vic-diol, an R²(10) dimer be-

2

Fax: +91-80-2360-0936

E-mail: gm@orgchem.iisc.ernet.in

ing the preferred motif in fully hydrogen-bonded crystal

Eur. J. Org. Chem. 2007, 423–436

G. Mehta, S. Sen, S. S. Ramesh

FULL PAPER

structures.^[5]Bishop and co-workers also concluded that similar lines to build O–H···O hydrogen-bonded molecular

molecular bulk or shape is important in dictating the choice ensembles from the motifs 4–8, then use the results of the

of hydrogen-bonded motifs adopted by alcohols from a CSD analyses described above^[4–6]to propose the most

separate CSD analysis of the occurrence of ladder-like likely packing modes in the crystal structures of 1–3 and

supramolecular architectures in certain diols.^[6]

compare the results with those obtained experimentally.

Though based on investigations carried out on specific The principal goal of the study was to examine the influ-

classes of alcohols, it is reasonable to believe that the gener- ence of conformational locking in facilitating the prediction

alizations put forth in these CSD analyses can be used as a of O–H···O hydrogen-bonding patterns in polycyclitols and

preliminary guide to understanding the selection of O– this paper details our efforts along these lines.

H···O hydrogen-bonding motifs observed in crystal struc-

tures of polyols structurally and constitutionally dissimilar

to the coterie studied. In keeping with our ongoing interest

Results and Discussion

in the study of the supramolecular assemblies adopted by

conformationally constrained polycyclitols,^[7,8]we were

Synthesis of the Polycyclitols 1–3

specifically interested in utilizing these rules as the basis for

The polycyclitols 1–3 were synthesized starting from

readily available aromatic precursors (tetralin, naphthalene

and anthracene, respectively) by sequential epoxidation and

acid-catalyzed ring-opening of their Birch reduction prod-

ucts (Scheme 1, Scheme 2 and Scheme 3).^[10]Thus, the un-

saturated trans-diols 9 and 10, obtained from tetralin and

naphthalene, respectively, as described previously,^[9a,11]were

subjected to epoxidation with MCPBA to afford the mono-

epoxide 11 and the anti-diepoxide 12, respectively. The

stereoselective formation of 12, whose structure was unam-

biguously assigned by single-crystal X-ray diffraction analy-

sis, can be attributed to the formation of a pre-reaction hy-

drogen-bonded complex between two molecules of the per-

acid and the trans hydroxy groups in 10.^[12]This evidently

leads to anti-selective oxygen delivery from MCPBA to the

two double bonds in 10 (Figure 3). The desired polycyclitols

proposing the various possible modes in which polyols such

as 1, 2 and 3 are likely to pack in the solid state (Figure 1).

Figure 1. The conformationally locked polycyclitols used in this

study.

In each of the three molecules 1–3, the hydroxy groups

are destined, owing to the trans ring fusion(s), to be locked

in an axial conformation^[9]and are thus favorably posi-

tioned (on account of their 1,3-syn diaxial relationship) to

participate in intramolecular O–H···O hydrogen bonding.

Hence, the packing pattern that is eventually observed in

the fully hydrogen-bonded structures of 1–3 will essentially

be determined by the manner in which intramolecularly hy-

drogen-bonded molecular motifs 4–8 (Figure 2) choose to

be linked to its neighbors through the four intermolecular

Scheme 1. Reagents and conditions: a) MCPBA, CH₂Cl₂, 0 °C Ǟ

room temp., 30 min, 85%; b) 10% AcOH, room temp., 4 h, 90%.

O–H···O hydrogen bonds. We intended to proceed along MCPBA: m-chloroperbenzoic acid.

Figure 2. The conformationally locked, intramolecularly O–H···O hydrogen-bonded molecular motifs.

424

www.eurjoc.org

Eur. J. Org. Chem. 2007, 423–436

Crystal Structures of Conformationally Locked Cyclitols

FULL PAPER

1 and 2 were obtained as the sole products upon mild acid-

catalyzed hydrolysis of the oxirane functionalities in 11 and

12, respectively (Scheme 1 and Scheme 2).

Figure 3. Proposed hydrogen-bonded complex between MCPBA

and the diol 10.

configuration of the OH groups, rather than the trans-syn-

trans configuration observed in 1–3.^[15]The mechanism pro-

posed by Craig et al. for the nucleophilic ring-opening reac-

tions of 1,4-cyclohexane dioxide suggests that the configu-

ration of the functional groups observed in the products

arises from an initial axial attack of the nucleophile to open

the first epoxide ring, followed by the inversion of the diaxi-

ally substituted epoxide intermediate to the diequatorial

isomer, which then suffers a second axial nucleophilic at-

tack to furnish the product.^[15a]The methodology followed

in the synthesis of the polycyclitols 1–3 thus utilizes the

stability and conformational rigidity of the trans-decalin

framework to establish the desired all-axial stereochemistry

of the trans-hydroxy groups in the epoxide ring-opening

steps.

Scheme 2. Reagents and conditions: a) MCPBA, CH₂Cl₂, 0 °C Ǟ

room temp., 3 h, 87%; b) i) pTSA, moist DCM, room temp.;

ii) Ac₂O, pyridine, room temp., 20 h (65% over two steps);

c) NaOMe, MeOH, 0 °C, 16 h, quant.

The Most Probable O–H···O Hydrogen-Bonded Packing

Motifs in Crystal Structures of the Polycyclitols 1–3

In order to propose the O–H···O hydrogen-bonded pack-

ing motif most likely to be observed in crystal structures of

the polycyclitols 1–3, it was necessary to identify first the

most likely ways by which an intramolecularly bonded mo-

lecular motif 4–8 may be bonded to its immediate neighbors

by the four intermolecular O–H···O hydrogen bonds.

Hence, among the large variety of packing patterns that

may be speculated for the polycyclitols 1–3, we decided to

limit our choices to the well-documented preferences of hy-

drogen-bonding motifs in solid-state aggregates of polyhy-

droxylated molecules (vide supra).^[3a,4–6]In other words, it

was assumed that in the crystal structures of 1–3, the for-

mation of infinite –OH···OH– chains would be preferred if

steric factors permit, intermolecular –OH···OH– rings be-

ing the next favored hydrogen-bonding motif.^[4]However,

even with this restriction, it was still possible to conceive

Scheme 3. Reagents and conditions: a) Na, liq. NH₃, EtOH,

THF, –78 °C, 12 h, 80%; b) MCPBA, CH₂Cl₂, –20 °C, 5 min,

syn/anti = 10:3, 89% overall yield; c) 10% AcOH (aq), 50–60 °C,

6 h, 95%; d) 5% Pd-C, H₂, MeOH, 1 h, 95%.

To synthesize 3, anthracene-derived tricyclic tetraene of a significantly large number of packing patterns for the

14^[13]was subjected to regioselective epoxidation of the in- polycyclitols 1–3 by varying ZЈ (the number of molecules

ternal double bonds to furnish a mixture (10:3) of syn and per asymmetric unit) and the symmetry elements relating a

anti diastereomers of the diepoxide 15.^[14]Subsequent hy- molecular motif to its immediate neighbors.

drolysis of the mixture of diepoxides 15 afforded a single

unsaturated tetrol 16,^[14c]which on catalytic hydrogenation packing of the tetrol 1, ZЈ Յ 1 and the highest possible

furnished the required tetrol 3. rotation symmetry in the crystal is limited to a two-fold

In such a case, if one assumes, as for example in the

Note that cyclohexanetetrol, obtained upon acid- or symmetry, following the space group statistics for 2°-3° di-

base-catalyzed epoxide ring-opening of both syn- and anti- alcohols by Taylor and Macrae, it is possible to recognize

1,4-cyclohexane dioxide, displays solely a trans-anti-trans 12 modes of molecular packing, all embodying infinite co-

Eur. J. Org. Chem. 2007, 423–436

www.eurjoc.org

425

G. Mehta, S. Sen, S. S. Ramesh

FULL PAPER

operative –OH···OH– chains between the intramolecularly clohexanetetrol moieties, on account of the steric bulk of

hydrogen-bonded 2°-3° OH pairs (Figure 4).^[4]Each of the intervening cyclohexane rings, and therefore render un-

these four packing modes can, in turn, be constructed from tenable the formation of infinite intermolecular O–H···O

pairs of molecular motifs 4–6 related by two-fold, mirror, hydrogen bonds. Therefore it would appear that the five

inversion, 2₁screw and/or glide operations.

patterns of molecular packing envisaged in Figure 4 (a–e)

However, on closer inspection, it becomes apparent that may be the ones most likely to be observed in the supramo-

the modes of molecular assembly depicted in parts f–l in lecular assembly of the tetrol 1. It is pertinent to emphasize

Figure 4 (those shown in Figure 4, j–l in particular) will at this point that supramolecular assemblies of 1 incorpo-

lead to unfavorable spatial segregation of the interacting cy- rating the packing modes shown in Figure 4 (f–l) may still

Figure 4. The packing modes proposed for the tetrol 1 (assuming ZЈ Յ 1 and limiting the highest possible rotation symmetry in the

crystal to a two-fold symmetry). All of them incorporate infinite cooperative –OH···OH– chains between the intramolecularly hydrogen-

bonded 2°-3° OH pairs. The respective molecular motifs 4–6 have been highlighted with a gray background.

426

www.eurjoc.org

Eur. J. Org. Chem. 2007, 423–436

Crystal Structures of Conformationally Locked Cyclitols

FULL PAPER

Figure 4. (continued).

Eur. J. Org. Chem. 2007, 423–436

www.eurjoc.org

427

G. Mehta, S. Sen, S. S. Ramesh

FULL PAPER

Figure 5. The packing mode proposed for the tetrol 1 assuming ZЈ = 2. The two crystallographically independent molecules constitute

the molecular motifs 4 (gray background) and 5 (encircled).

Figure 6. The four packing modes proposed for the hexol 2 (assuming ZЈ Յ 1 and limiting the highest possible rotation symmetry in the

crystal to a two-fold symmetry). All of them incorporate infinite cooperative –OH···OH– chains. The centric motif 7 has been highlighted

with a gray background, while the noncentric one has been encircled.

428

www.eurjoc.org

Eur. J. Org. Chem. 2007, 423–436

Crystal Structures of Conformationally Locked Cyclitols

FULL PAPER

be realized if one relaxes the constraint on ZЈ and the al-

lowed space group symmetry. For example, by allowing two

molecules of 1 to reside in the asymmetric unit (i.e. ZЈ =

2), one may possibly tune the spatial relationship among

the crystallographically independent molecules suitably

enough to offset the steric constraints on the intermolecular

O–H···O hydrogen bonding encountered in such molecular

packing (Figure 5).

Assuming identical constraints on ZЈ and crystal sym-

metry, one may propose, by following almost the same line

of arguments as put forward in the case of 1, that each

intramolecularly hydrogen-bonded molecule of 2 can be

linked to its nearest neighbors according to the four dif-

ferent O–H···O hydrogen-bonding schemes depicted in Fig-

ure 6 (a–d). As shown in the case of 1 above (Figure 4), all

four hydrogen-bonding patterns proposed for 2 follow the

same trend observed by Jeffrey and Saenger in as much as

they maximize not only the number of O–H···O bonds,^[3a,3b]

but also the hydrogen-bond cooperativity owing to the for-

mation infinite –OH···OH– chains. However, it is well

known that an achiral molecule like the C_2hsymmetric

hexol 2 would prefer to crystallize in a centrosymmetric

space group^[16]and usually occupy, according to a CSD

study by Brock and Dunitz, the crystallographic inversion

centers (ZЈ = ½) in such a case.^[17]Hence, the centrosym-

metric molecular packings (Figure 6, b and d) generated

from the centric molecular motif 7 would appear to be

among the four depicted in Figure 6 more likely to be ob-

served in the crystal structure of the hexol 2.

Figure 7. A centrosymmetric packing mode proposed for the tetrol

3 (assuming ZЈ = ½ and limiting the highest possible rotation sym-

metry in the crystal to a two-fold symmetry) that is capable of

incorporating, in principle, infinite cooperative –OH···OH– chains.

The centric molecular motif 8 has been highlighted with a gray

background.

lar motif 8 if one relaxes the constraint on ZЈ and the al-

lowed space group symmetry. As an illustration, one may

follow the example of the packing pattern shown in Fig-

ure 5 and possibly overcome the steric constraints to inter-

molecular O–H···O hydrogen bonding encountered in the

molecular ensemble depicted in Figure 7 by allowing mole-

cules of 3 to occupy two sets of inversion centers (i.e. ZЈ =

2ϫ½ = 1) instead of one (Figure 8).

In comparison to the polycyclitols 1 and 2, the four hy-

droxy groups in the achiral C_2hsymmetric tetrol 3 are terti-

ary. The space group and ZЈ statistics in the CSD study

by Taylor and Macrae reveal that while 89% of the 3°-3°

dialcohols studied crystallized in the top 15 space groups,

82% of them adopted structures with ZЈ = 1. In addition,

among the 10 3°-3° diols that adopted packing motifs in-

volving intramolecular O–H···O hydrogen bonds, five

formed –OH···OH– chains, two preferred rings and three

favored isolated O–H···O hydrogen bonds.^[4]In the light of

these observations, one may be tempted to extend to the

C_2hsymmetric tetrol 3 the same line of arguments as

adopted for the isosymmetric hexol 2, subject to identical

constraints on the highest-allowed crystal symmetry and ZЈ

(= ½), and propose a centrosymmetric model of molecular

packing, as shown in Figure 7. The supramolecular

aggregate can be generated through suitable symmetry op-

erations on the centric molecular motif 8 and can, in prin-

ciple, result in the formation of infinite –OH···OH– chains

among the intramolecularly bonded pairs of tertiary hy-

droxy groups.

Figure 8. A centrosymmetric packing mode proposed for the tetrol

3 (assuming ZЈ = 2ϫ½ and limiting the highest possible rotation

symmetry in the crystal to a two-fold symmetry) that is capable of

incorporating infinite cooperative –OH···OH– chains. The centric

molecular motif 8 occupies two sets of inversion centers (marked

with a dot, and labeled 1 and 2).

However, as argued in the case of 1, the steric hindrance

caused by the intervening cyclohexane rings in the molecu-

It is also possible for the self-assembling process to over-

lar assembly depicted in Figure 7 will lead to weakening of come the steric inhibition to molecular packing in 3 by opt-

the intermolecular O–H···O hydrogen bonds between the ing for the formation of –OH···OH– rings, as a second alter-

interacting cyclohexanetetrol moieties in 3 (Figure 7). As native, through a “broadside” approach of the interacting

noted in the case of 1 and quite applicable to 2, viable cyclohexanetetrol moieties. In such a case, even with the

supramolecular assemblies of the tetrol 3, embodying constraint of 3 preferring to crystallize in a low symmetry

–OH···OH– chains, may still be generated from the molecu- centrosymmetric space group with ZЈ = ½, it is easily pos-

Eur. J. Org. Chem. 2007, 423–436

www.eurjoc.org

429

G. Mehta, S. Sen, S. S. Ramesh

FULL PAPER

sible to conceive of a molecular packing, as depicted in Fig- ture and pressure from their solutions in 1:1:2 dry meth-

ure 9, generated from translationally related molecular mo- anol/ethanol/ethyl acetate. None of the polycyclitols showed

tifs 7 linked to each other through four intermolecular O– any polymorphic or pseudopolymorphic modification un-

H···O hydrogen bonds. Interestingly, this proposed packing der the crystallization conditions described. The details of

pattern, consisting of alternate regions of organic cycles the packing patterns in the polycyclitols 1–3, as gleaned

(defined by the carbocyclic framework) and centrosymmet- from an analysis of their respective crystal data (see

ric (OH)₄rings, is quite reminiscent of the step-ladder archi- Table 4), are discussed below.

tectures adopted by certain classes of dialcohols.^[6]

Crystal Structure of the Polycyclitol 1

The crystal structure of the bicyclic C₂symmetric tetrol

1 (Figure 10) was solved and refined in the noncentrosym-

metric orthorhombic space group Fdd2 (Z = 16).^[18]The

two pairs of 1,3-syn diaxial hydroxy groups participate in

intramolecular hydrogen bonding in a manner exhibited by

Figure 9. A centrosymmetric packing mode proposed for the tetrol

3 (assuming ZЈ = ½ and limiting the highest possible rotation sym-

metry in the crystal to a two-fold symmetry), incorporating

–OH···OH– rings.

X-ray Crystallographic Studies on the Polycyclitols 1–3

Figure 10. ORTEP diagram of 1 with the atomic numbering scheme

Crystals of the polycyclitols 1–3 suitable for single-crystal

for the asymmetric unit. Displacement ellipsoids for non-hydrogen

X-ray crystallography were grown under ambient tempera- atoms have been drawn at the 50% probability level.

Figure 11. Crystal packing in the tetrol 1, showing details of the hydrogen-bonded tapes, viewed along a) the a axis and b) the c axis.

The hydrogen atoms connected to the carbon atoms have been omitted for clarity. A portion of the packing diagram in (a) has been

encircled to show its resemblance to part a of Figure 4. The packing diagram in (b) shows the typical layered appearance of the supramo-

lecular assembly.

430

www.eurjoc.org

Eur. J. Org. Chem. 2007, 423–436

Crystal Structures of Conformationally Locked Cyclitols

FULL PAPER

the molecular motif 4. Four intermolecular O–H···O hydro- cules of the hexol 2 participate in four intermolecular

gen bonds, involving O2–O4 and O3–O1, link each tetrol O–H···O hydrogen bonds involving O2 and O3 to form

molecule to its nearest neighbors to form hydrogen-bonded O–H···O hydrogen-bonded tapes perpendicular to the (101)

tapes perpendicular to the (100) direction (part a of Fig- direction (part b of Figure 12, Table 2).

ure 11, Table 1).

Table 2. Hydrogen bond geometry in 2.^[a]

Table 1. Hydrogen bond geometry in 1.^[a]

D–H···A

r(D–H) [Å] r(H···A) [Å] r(D···A) [Å] D–H···A [°]

D–H···A

r(D–H) [Å] r(H···A) [Å] r(D···A) [Å] D–H···A [°]

O1–H1O···O2ⁱ

O2–H2O···O3ⁱⁱ

O3–H3O···O1ⁱⁱⁱ

0.82

2.00

1.95

2.04

2.730 (1)

2.756 (2)

2.718 (1)

149

169

140

O1–H1O···O2ⁱ

O2–H2O···O4ⁱⁱ

O3–H3O···O1ⁱⁱⁱ

O4–H4O···O3ⁱ

0.82

1.98

1.99

2.01

2.710(2)

2.810(2)

2.836(2)

2.735(3)

148

175

177

147

[a] Symmetry codes: i) x, y, z; ii) x+½, –y+½, z+½; iii) –x+1,

–y+1, –z+1.

[a] Symmetry codes: i) x, y, z; ii) –x+¼, y+¼, z+¼; iii) x, y,

z–1.

Crystal Structure of the Polycyclitol 3

These hydrogen-bonded tapes are translated along the a

axis and are held together solely by weak van der Waals in-

teractions between the cyclohexane rings, so that when

viewed perpendicular to the ab plane, the packing pattern

assumes a layered structure reminiscent of graphite (Fig-

ure 11, b).

The tricyclic C_2hsymmetric tetrol 3 (Figure 13, a) crys-

tallized in the centrosymmetric triclinic space group P1 (Z

¯

= 1) with the molecular inversion centers coinciding with

the crystallographic centers of symmetry at (½, 0, ½). Un-

like 1 and 2, the O–H···O hydrogen bonds in the tetrol 3

form a cyclic (OH)₄motif (a quadrilateral four-membered

homodromic hydrogen-bonding cycle),^[3b]so that the mole-

Crystal Structure of the Polycyclitol 2

The bicyclic C_2hsymmetric hexol 2 (Figure 12, a) packed cules are linked by four intermolecular O–H···O hydrogen

in the centrosymmetric monoclinic space group P2₁/n (Z = bonds to form chains growing parallel to the a axis (Fig-

2) with the molecular inversion centers coinciding with the ure 13, b, Table 3). The translationally related molecular

crystallographic centers of symmetry at (0, 0, 0) and (½, ½, chains are held in space solely by weak van der Waals inter-

½). Like 1, the intramolecularly hydrogen-bonded mole- actions between the cyclohexane rings (Figure 13, b).

Figure 12. a) ORTEP diagram of 2 with the atomic numbering scheme for the asymmetric unit. Displacement ellipsoids for non-hydrogen

atoms have been drawn at the 50% probability level. b) Crystal packing in the hexol 2 viewed along the c axis. The hydrogen atoms

connected to the carbon atoms have been omitted for clarity. A portion of the packing diagram in (b) has been encircled to show its

resemblance to part d in Figure 6.

Eur. J. Org. Chem. 2007, 423–436

www.eurjoc.org

431

G. Mehta, S. Sen, S. S. Ramesh

FULL PAPER

Figure 13. a) The ORTEP diagram of 3 with the atomic numbering scheme for the asymmetric unit. Displacement ellipsoids for non-

hydrogen atoms have been drawn at the 50% probability level. b) Crystal packing in the hexol 3 viewed along the b axis. The hydrogen

atoms connected to the carbon atoms have been omitted for clarity. A portion of the packing diagram in (b) has been encircled to show

its resemblance to Figure 9.

Table 3. Hydrogen bond geometry in 3.^[a]

tributed to the self-assembling process, striving to optimize

not only the strong and directional O–H···O hydrogen

bonds, but also the weak and isotropic van der Waals inter-

actions by increasing the area of contact between the paral-

lel cyclohexane rings.

D–H···A

r(D–H) [Å] r(H···A) [Å] r(D···A) [Å] D–H···A [°]

O1–H1···O2ⁱ

O2–H2···O1ⁱⁱ

0.82

2.01

1.94

2.814 (3)

2.670 (3)

167

148

[a] Symmetry codes: i) –x, –y, –z+1; ii) x, y, z.

Conclusions

As a corollary to the view put forth by Jeffrey in his

seminal review “Crystallographic Studies of Carbo-

hydrates”,^[3a]polyhydroxylated molecules, such as the ones

Comparison of the Experimentally Obtained Packing

Motifs with Those Proposed for 1, 2 and 3

A close examination of the crystal structures of the poly- chosen in this study, constitute one of the simplest systems

cyclitols 1–3 (Figure 11, a, Figure 12, b, Figure 13, b) re- for the analysis of hydrogen bonding because they have only

vealed that the O–H···O hydrogen-bonding pattern ob- one functional group, namely the hydroxy group, which

served experimentally for a particular polyol resembled very functions as both donor and acceptor of hydrogen bonds

closely one of those proposed as most probable for it (Fig- in the molecule. On the contrary, prediction of the mode

ure 4, a–e, Figure 6, b,d and Figure 9). As assumed on the of molecular association through intermolecular O–H···O

basis of space group and ZЈ statistics for dialcohols, none hydrogen-bonds in such molecules becomes a difficult prop-

of the polycyclitols 1–3 crystallized in unusually high-sym- osition owing to the orientational flexibility of the C–OH

metry space groups or adopted crystal structures with ZЈ Ͼ groups. Proposing the hydrogen-bonded architecture in po-

1.^[4]The C_2hsymmetric polyols 2 and 3 showed the usual lyols with little or no constraints to the internal degrees of

preference of achiral molecules for centrosymmetric space freedom proves to be even more complicated as the final

groups and indeed occupied one set of inversion centers, spatial disposition of the hydroxy groups realized in the

as proposed on the basis of observations by Brock and crystal structure of such molecules is often largely deter-

Dunitz.^[16,17]The formation of infinite –OH···OH– chains mined by the crystal packing itself. In this study, we have

in the supramolecular assembly of 1 and 2 and the adoption therefore attempted to analyze the influence of the confor-

of O–H···O rings in the crystal structure of 3 as an alterna- mational locking of hydroxy groups in lending greater pre-

tive to the sterically unfavorable chain motif are in conson- dictability to the hydrogen-bonded packing motifs adopted

ance with the order of preference of the basic hydrogen- by polyhydroxylated molecules.

bonding motifs reported for mono- and dialcohols by Tay-

For the polycyclitols 1–3 chosen for this analysis, confor-

lor and Macrae.^[4]Note that the formation of –OH···OH– mational locking and the consequent intramolecular

rings in the crystal structure of the tetrol 3 can also be at- O–H···O hydrogen bonding between the 1,3-syn diaxial OH

432

www.eurjoc.org

Eur. J. Org. Chem. 2007, 423–436

Crystal Structures of Conformationally Locked Cyclitols

FULL PAPER

(465 mg, 85%). ¹H NMR (300 MHz, CDCl₃): δ = 3.56 (s, 2 H,

OH), 3.42–3.39 (m, 1 H), 3.33–3.31 (m, 1 H), 2.21 (½ABq, J =

17 Hz, 1 H), 2.17 (½ABq, J = 15 Hz, 1 H), 1.96–1.77 (m, 3 H),

1.72–1.35 (m, 7 H) ppm. ¹³C NMR (75 MHz, CDCl₃): δ = 70.9 (2

C), 54.7, 51.6, 36.3, 34.6, 34.1, 32.8, 21.1, 20.1 ppm. LRMS (ES,

70 eV): m/z = 207 [M + Na]⁺. HRMS (ES): calcd. for C₁₀H₁₆O₃Na

[M + Na]⁺207.0997; found 207.1002.

groups allowed any packing pattern in 1–3 to be described

in terms of a handful of intramolecularly hydrogen-bonded

molecular motifs 4–8. Thus, despite considering fully hydro-

gen-bonded structures of a tetrol or hexol, it was possible,

in combination with the well-documented results of CSD

analyses of polyols, to narrow down the number of packing

motifs most likely to be observed in the polycyclitols 1–3.

This, coupled with the fact that the experimentally observed

packing motifs corroborate well those proposed for 1–3,

provides room for speculation on the possibility of confor-

mational locking facilitating prediction of hydrogen bond-

ing in molecules having not only hydroxy groups, but also

other functionalities (such as NH₂and COOH). Hence the

results presented in this study are significant from the

standpoint of crystal engineering and may provide leads for

the identification of new supramolecular synthons in con-

formationally constrained systems.

(2R*,3R*,4aS*,8aS*)-Perhydro-2,3,4a,8a-naphthalenetetrol (1): A

suspension of the monoepoxide 11 (200 mg, 1.087 mmol) in acetic

acid (10% solution in water, 5 mL) was stirred at room temperature

for 4 h. The volatiles were then removed under vacuum and subse-

quent purification of the residue by column chromatography using

50% ethyl acetate/hexane furnished the required tetrol 1 (198 mg,

90%). M.p. 171.1–171.8 °C (dec.). IR (KBr): ν = 3310 cm^–1. ¹H

˜

NMR (300 MHz, D₂O): δ = 3.88 (s, 2 H), 1.96 (½ABq, J = 15 Hz,

2 H), 1.62–1.53 (m, 2 H), 1.43 (½ABq, J = 15 Hz, 2 H), 1.40–1.22

(m, 6 H) ppm. ¹³C NMR (75 MHz, CDCl₃): δ = 74.9 (2 C), 71.3

(2 C), 33.9 (2 C), 33.5 (2 C), 20.1 (2 C) ppm. LRMS (ES, 70 eV):

m/z = 225 [M + Na]⁺. HRMS (ES): calcd. for C₁₀H₁₈O₄Na [M +

Na]⁺: 225.1103; found 225.1099.

Perhydrooxireno[2Ј,3Ј:6,7]naphtho[2,3-b]oxirene-2a,5a-diol (12): A

solution of MCPBA (1.485 g, 6.024 mmol, 70% purity) in dichloro-

methane (5 mL) was added dropwise to a solution of the trans-diol

10 (500 mg, 3.012 mmol) in dichloromethane (5 mL) at 0 °C. The

white suspension thus formed was stirred at room temperature for

3 h and then quenched by addition of a saturated solution of so-

dium sulfite in water. The product was extracted with dichloro-

methane (3ϫ30 mL); the combined extracts were washed success-

ively with saturated sodium hydrogen carbonate solution and brine,

and dried with anhydrous sodium sulfate. Removal of the solvent

and subsequent purification by column chromatography with 35%

Experimental Section

General: Melting points were recorded with a Büchi B-540 appara-

tus and are uncorrected. Infrared spectra were recorded with a JA-

SCO FT-IR 410 spectrometer. H and ¹³C NMR spectra were re-

1

corded with a JEOL JNM-LA 300 spectrometer. Chemical shifts

are reported with respect to tetramethylsilane (Me₄Si) as the in-

1

ternal standard (for H NMR) and the central line (δ = 77.0 ppm)

of CDCl₃(for ¹³C NMR). The chemical shifts are expressed in

parts per million (ppm, δ values) downfield from Me₄Si. The stan-

dard abbreviations s, d, t, q and m refer to singlet, doublet, triplet,

quartet and multiplet, respectively. Coupling constant (J), when-

ever discernible, are reported in Hz. Both low-resolution (LRMS)

and high-resolution mass spectra (HRMS) were recorded with a

Q-TOF Micromass mass spectrometer. Reactions were monitored

by thin-layer chromatography (TLC) performed either on

(10ϫ5 cm) glass plates or on microscopic slides coated with silca

gel G (Acme) containing 13% calcium sulfate as a binder. Visual-

ization of the spots on the TLC plates was achieved by exposure

to iodine vapor, by using UV radiation or by spraying with either

ethanolic vanillin or 30% methanol-sulfuric acid solution and heat-

ing the plates at 120 °C. Commercial silica gel (Acme, 100–

200 mesh particle size) was used for column chromatography. The

columns were usually eluted with ethyl acetate/hexane mixtures. All

solvent extracts were washed with water and brine, dried with anhy-

drous sodium sulfate and then concentrated under reduced pressure

on a rotary evaporator unless specified otherwise. Yields reported

are isolated yields of materials judged homogeneous by TLC and

NMR spectroscopy.

ethyl acetate/hexane afforded the diepoxide 12 (519 mg, 87%). IR

1

(KBr): ν = 3487 cm^–1. H NMR (300 MHz, CDCl ): δ = 3.61 (s, 2

˜

3

H, OH), 3.40–3.38 (m, 2 H), 3.31–3.28 (m, 2 H), 2.14 (d of ABq,

J = 15, 2 Hz, 8 H) ppm. ¹³C NMR (75 MHz, CDCl₃): δ = 69.7 (2

C), 54.4 (2 C), 51.1 (2 C), 35.5 (2 C), 32.4 (2 C) ppm. LRMS (ES,

70 eV): m/z = 221 [M + Na]⁺. HRMS (ES): calcd. for C₁₀H₁₄O₄Na

[M + Na]⁺221.0785; found 221.0790.

Perhydro-2,3,4a,6,7,8a-naphthalenehexol (2): A solution of the diep-

oxide 12 (200 mg, 1.010 mmol) in moist dichloromethane (10 mL)

was stirred at room temperature in the presence of a catalytic

amount of pTSA. After completion of the reaction, the volatiles

were removed under vacuum and the residue, thus obtained, was

acetylated at room temperature in presence of acetic anhydride and

pyridine. The reaction takes 20 h to complete as indicated by TLC

analysis. At the end of this period, the reaction was quenched with

water and the product extracted with dichloromethane (3ϫ20 mL).

The combined extracts were washed with saturated sodium hydro-

gen carbonate solution and brine, and dried with anhydrous so-

dium sulfate. Removal of the solvent and subsequent purification

by column chromatography with 50% ethyl acetate/hexane afforded

the tetraacetate 13 (264 mg, 65% over two steps) as a colorless

(1aR*,2aS*,6aS*,7aS*)-Perhydronaphtho[2,3-b]oxirene-2a,6a-diol

(11): MCPBA (734 mg, 2.976 mmol, 70% purity) was added to a

solution of the trans-diol 9 (500 mg, 2.976 mmol) in dichlorometh-

ane (10 mL) at 0 °C. The reaction was stirred at room temperature

for 30 min and then quenched by addition of a saturated solution

of sodium sulfite in water. The product was extracted with dichlo-

romethane (3ϫ20 mL); the combined extracts were washed suc-

cessively with saturated sodium hydrogen carbonate solution and

brine, and then dried with anhydrous sodium sulfate. Removal of

the solvent and subsequent purification by column chromatography

with 25 % ethyl acetate/hexane afforded the monoepoxide 11

solid. IR (KBr): ν = 3585, 1732 cm^–1. ¹H NMR (300 MHz, CDCl ):

˜

3

δ = 5.07 (s, 4 H), 3.14 (s, 2 H), 2.27 (½ABq, J = 15 Hz, 4 H), 2.08

(s, 12 H), 1.77 (½ABq, J = 15 Hz, 4 H) ppm. ¹³C NMR (75 MHz,

CDCl₃): δ = 169.2 (4 C), 72.16 (2 C), 69.9 (4 C), 33.2 (4 C), 21.2

(4 C) ppm. LRMS (ES, 70 eV): m/z = 425 [M + Na]⁺. HRMS (ES):

calcd. for C₁₈H₂₆O₁₀Na [M + Na]⁺425.1424; found 425.1429.

The pure tetraacetate 13 (150 mg, 0.373 mmol) thus obtained was

dissolved in dry methanol (1 mL) and sodium methoxide (85 mg,

1.567 mmol) was added to the resulting solution. The reaction mix-

Eur. J. Org. Chem. 2007, 423–436

www.eurjoc.org

433

G. Mehta, S. Sen, S. S. Ramesh

FULL PAPER

ture was stirred at room temperature under dry nitrogen for 16 h.

The solvent was then completely removed under vacuum and the

residue dissolved in a minimum volume of deionized water. The

solution was passed through a short column of pretreated

DOWEX^50W ion-exchange resin (8–200 mesh, acidic cation) and

washed with deionized water. The aqueous solution of the product

thus obtained was concentrated under vacuum to obtain the hexol

1,4,4a,5,8,8a,9,9a,10,10a-Decahydro-4a,8a,9a,10a-anthracenetetrol

(16): There is a wide difference in the rates of hydrolysis of the two

diepoxides. Thus, while the syn isomer undergoes complete hydroly-

sis in 10% aqueous acetic acid within 12 h at room temperature,

the anti isomer under the same conditions reacts extremely slowly,

with the reaction taking nearly 1.5 d to complete. Since both iso-

mers gave the same tetrol 16, the hydrolytic reaction was carried

out with the mixture of the diepoxides 15. A suspension of the

2 (87 mg) in quantitative yield. M.p. 248.5–249.1 °C (dec.). IR

1

(KBr): ν = 3585, 1732 cm^–1. H NMR (300 MHz, D O): δ = 3.84 mixture (200 mg, 0.926 mmol) in acetic acid (10% solution in water,

˜

2

(s, 4 H), 3.08 (½ABq, J = 15 Hz, 4 H), 1.54 (½ABq, J = 15 Hz, 4

H) ppm. ¹³C NMR (75 MHz, D₂O): δ = 75.9 (2 C), 71.3 (4 C), 33.5

(4 C) ppm. LRMS (ES, 70 eV): m/z = 257 [M + Na]⁺. HRMS (ES):

calcd. for C₁₀H₁₈O₆Na [M + Na]⁺257.1001; found 257.0995.

3 mL) was stirred vigorously at 50 °C for 6 h. The progress of the

reaction was indicated by the slow conversion of the crystalline

diepoxide into a voluminous white suspension. At the end of the

reaction, the reaction mixture was cooled and the volatiles removed

under vacuum to give the crude tetrol as a white powder. The crude

compound was washed repeatedly with dichloromethane and then

ethyl acetate to remove traces of the starting material. The tetrol

thus obtained (221 mg, 95%) was sufficiently pure, as evident from

its spectroscopic data, for use in further transformations. However

further purification can be achieved by crystallization from meth-

1,4,5,8,9,10-Hexahydroanthracene (14): Compound 14 was synthe-

sized following an adapted procedure of Birch et al.^[13a]Anthracene

(1 g, 0.565 mmol) was dissolved in dry THF (80 mL) and dry etha-

nol (25 mL, 0.430 mol). The mixture was added to liquid ammonia

(80 mL) and the vigorously stirred suspension of anthracene was

reduced with sodium (2.6 g, 0.113 mol) at –78 °C. The mixture was

stirred for 12 h after which ammonia was allowed to evaporate.

Water (50 mL) was then added to dissolve the inorganic salts com-

pletely. The product was extracted with diethyl ether (3ϫ50 mL),

and the combined ether extracts washed with brine and dried with

anhydrous sodium sulfate. Removal of the solvent and subsequent

column chromatography through neutral alumina using a 2 %

EtOAc/petroleum ether mixture afforded the product as a colorless,

crystalline solid (800 mg, 80 %). M.p. 146.5–147.0 °C (ref.^[13c]

147.5–148.5 °C). ¹H NMR (300 MHz, CDCl₃): δ = 5.49 (s, 4 H),

2.57 (s, 8 H), 2.43 (s, 4 H) ppm.

anol. M.p. 253.5–255.5 °C (dec.).^[20]IR (KBr): ν = 3256, 3023,

˜

1

1653 cm^–1. H NMR (300 MHz, CD₃OD): δ = 5.55 (s, 4 H), 2.36

(½ABq, J = 17 Hz, 4 H), 2.06 (½ABq, J = 14 Hz, 2 H), 1.94

(½ABq, J = 17 Hz, 4 H), 1.46 (½ABq, J = 14 Hz, 2 H) ppm. ¹³C

NMR (75 MHz, CD₃OD): δ = 124.6 (4 C), 74.3 (4 C), 40.6 (2 C),

37.0 (4 C) ppm. LRMS (ES, 70 eV): m/z = 275 [M + Na]⁺. HRMS

(ES): calcd. for C₁₄H₂₀O₄Na [M + Na]⁺275.1259; found 275.1265.

Perhydro-4a,8a,9a,10a-anthracenetetrol (3): A heterogeneous mix-

ture of the unsaturated tetrol 16 (100 mg, 0.397 mmol) and 5% Pd-

C (10 mg, 10% w/w) in methanol (2 mL) was hydrogenated under

1 Torr pressure for 1 h. After the disappearance of the starting ma-

terial, as indicated by TLC analysis, the reaction mixture was fil-

tered through a small pad of Celite and washed with methanol.

The combined filtrate and washings were concentrated under vac-

uum and the residue purified by column chromatography to afford

4a,9a:8a,10a-Diepoxy-1,4,4a,5,8,8a,9,9a,10,10a-decahydro-

anthracene (15): A solution of MCPBA (1.340 g, 5.434 mmol, 70%

purity) in dichloromethane (40 mL) was added dropwise to a solu-

tion of hexahydroanthracene 14 (500 mg, 2.717 mmol) in dichloro-

methane (20 mL) cooled to –30 °C. The reaction was stirred at the

same temperature for 5 min and was then quenched with a satu-

rated solution of sodium hydrogen carbonate. The product was ex-

tracted with dichloromethane (3ϫ20 mL); the combined extracts

were washed with saturated sodium hydrogen carbonate solution

and brine, and dried with anhydrous sodium sulfate. Removal of

the solvent afforded the crude diepoxide 15 as a stereoisomeric mix-

ture that can be used directly in the preparation of the tetrol 16.

However separation of the isomeric diepoxides by column

chromatography presents no great difficulty as their polarity varies

widely. Usual chromatographic separation of the crude mixture

with 30 % ethyl acetate/hexane yielded successively the anti-

(120 mg) and syn-diepoxides (400 mg) in 89% overall yield.

3 (96 mg, 95%) as a colorless, crystalline solid. M.p. 327.6–327.7 °C

1

(dec.). IR (KBr): ν = 3275 cm^–1. H NMR (300 MHz, 1:1 CDCl /

˜

3

CD₃OD): δ = 1.78 (½ABq, J = 15 Hz, 2 H), 1.66–1.58 (m, 4 H),

1.46–1.40 (m, 4 H), 1.22 (m, 4 H), 1.10–1.06 (m, 4 H), 0.935

(½ABq, J = 15 Hz, 2 H) ppm. ¹³C NMR (75 MHz, CDCl₃): δ =

74.2 (4 C), 40.1 (2 C), 32.7 (4 C), 19.2 (4 C) ppm. LRMS (ES,

70 eV): m/z = 279 [M + Na]⁺. HRMS (ES): calcd. for C₁₄H₂₄O₄Na

[M + Na]⁺279.1272; found 279.1285.

Crystal Structure Analysis: The single-crystal X-ray diffraction data

were collected with a Bruker AXS SMART APEX CCD dif-

fractometer at 296 K. The X-ray generator was operated at 50 kV

and 35 mA using Mo-K_αradiation. Crystal data for the polycycli-

tols 1–3 are given in Table 4. The data was collected with an ω scan

width of 0.3°. A total of 606 frames per set were collected using

SMART^[21a]at three different settings of φ (0, 90 and 180°) or at

four different settings of φ (0, 90, 180 and 270°) for triclinic crystal

systems, keeping the sample-to-detector distance at 6.062 cm and

the 2θ value fixed at –25°. All the data were corrected for Lo-

rentzian, polarization and absorption effects using the SAINT^[21b]

and SADABS^[22]programs. SHELX-97^[23]was used for structure

solution and full-matrix least-squares refinement on F². Hydrogen

atoms were included in the refinement using the riding model. De-

tails of the intra- and intermolecular hydrogen-bonding scheme

were calculated using PARST95^[24]and PLATON.^[25]

anti-Diepoxide: M.p. 186.9–187.6 °C (dec.). ¹H NMR (300 MHz,

CDCl₃): δ = 5.40 (s, 4 H), 2.49 (½ABq, J = 18 Hz, 4 H), 2.28

(½ABq, J = 18 Hz, 4 H), 2.19 (s, 4 H) ppm. ¹³C NMR (75 MHz,

CDCl₃): δ = 122.4 (4 C), 59.7 (4 C), 36.0 (2 C), 31.3 (4 C) ppm.

LRMS (ES, 70 eV): m/z = 239 [M + Na]⁺. HRMS (ES): calcd. for

C₁₄H₁₆O₂Na [M + Na]⁺239.1048; found 239.1052.

syn-Diepoxide: M.p. 186.5–187.0 °C (dec.). ¹H NMR (300 MHz,

CDCl₃): δ = 5.42 (s, 4 H), 2.57 (½ABq, J = 16 Hz, 4 H), 2.48

(½ABq, J = 16 Hz, 4 H), 2.24 (½ABq, J = 17 Hz, 2 H), 2.06 CCDC-273131 (for tetrol 1), -273132 (for hexol 2), -250461 (for

(½ABq, J = 18 Hz, 2 H) ppm. ¹³C NMR (75 MHz, CDCl₃): δ =

122.1 (4 C), 58.4 (4 C), 33.4 (2 C), 31.1 (4 C) ppm. LRMS (ES,

70 eV): m/z = 239 [M + Na]⁺. HRMS (ES): calcd. for C₁₄H₁₆O₂Na

[M + Na]⁺239.1048; found 239.1056.

tetrol 3) and -601706 (for diepoxide 12) contain the supplementary

crystallographic data for this paper. These data can be obtained

free of charge from The Cambridge Crystallographic Data Centre

via www.ccdc.cam.ac.uk/data_request/cif.

434

www.eurjoc.org

Eur. J. Org. Chem. 2007, 423–436

Crystal Structures of Conformationally Locked Cyclitols

FULL PAPER

Table 4. Summary of the crystal data, data collection, structure solution, and refinement details for 1–3.

1

2

3

Formula

M_r

C₁₀H₁₈O₄

202.12

C₁₀H₁₈O₆

234.25

C₁₄H₂₄O₄

256.34

Crystal size [mm]

Crystal system

Space group

a [Å]

b [Å]

c [Å]

0.30ϫ0.27ϫ0.20

orthorhombic

Fdd2

20.432(5)

17.320(7)

7.0558(17)

90

0.46ϫ0.33ϫ0.21

monoclinic

P2₁/n

5.7859(12)

14.326(3)

6.5982(14)

90

106.683(3)

90

523.90(19)

2

0.14ϫ0.10ϫ0.09

triclinic

P1

¯

5.958(5)

5.994(5)

9.867(8)

86.177(13)

82.547(13)

73.200(12)

334.3(5)

1

α [°]

β [°]

90

γ [°]

V [Å³]

3938.6(17)

16

Z

F (000)

1760

1.364

0.104

252

1.485

0.123

140

1.273

0.092

ρ_calcd.[gcm^–3]

µ [mm^–1]

Absorption correction

Min./max. transmission

Reflrctions collected

No. of l.s. parameters

Unique reflections

Observed reflections

Index range

multiscan

0.9694/0.9795

7591

131

1090

multiscan

0.9387/0.9747

3785

76

960

multiscan

0.9620/0.9918

2431

84

1213

999

894

1038

–25 Յ h Յ 25

–33 Յ k Յ 33

–8 Յ l Յ 8

0.0426

–7 Յ h Յ 7

–15 Յ k Յ 17

–7 Յ l Յ 7

0.0379

–7 Յ h Յ 7

–6 Յ k Յ 7

–11 Յ l Յ 11

0.0504

R₁[I Ͼ 2σ(I)]

wR₂

0.0914

0.1160

0.1405

Goodness of fit

1.331

0.303/–0.220

1.023

0.284/–0.188

1.058

0.315/–0.203

∆ρ_max/min[eÅ^–3]

ing; Oxford University Press, Oxford, 1997; e) A. Bonnet, J.

Chisholm, W. D. S. Motherwell, W. Jones, CrystEngComm

2005, 7, 71–75.

R. Taylor, C. F. Macrae, Acta Crystallogr., Sect. B 2001, 57,

815–827.

Acknowledgments

We thank Prof. K. Venkatesan for his critical comments and helpful

advice in preparing the manuscript and DST, India for the CCD

facility at IISc, Bangalore.

[4]

[5]

[6]

C. P. Brock, Acta Crystallogr., Sect. B 2002, 58, 1025–1031.

V. T. Nguyen, P. D. Ahn, R. Bishop, M. L. Scudder, D. C.

Craig, Eur. J. Org. Chem. 2001, 4489–4499.

The term “polycyclitol” is derived from “polycyclic cyclitols”,

and as the name suggests, refers to any polyhydroxylated mole-

cule consisting of one or more cyclitol units embedded in a

polycyclic carbon framework; see: a) G. Mehta, S. S. Ramesh,

Chem. Commun. 2000, 2429–2430; b) G. Mehta, S. S. Ramesh,

Tetrahedron Lett. 2001, 42, 1987–1990.

a) G. Mehta, S. Sen, K. Venkatesan, CrystEngComm 2005, 7,

398–401; b) G. Mehta, S. Sen, S. S. Ramesh, CrystEngComm

2005, 7, 563–568; c) G. Mehta, S. Sen, CrystEngComm 2005,

7, 656–663.

a) G. Mehta, S. S. Ramesh, M. K. Bera, Chem. Eur. J. 2003, 9,

2264–2272; b) G. Mehta, S. S. Ramesh, Tetrahedron Lett. 2003,

44, 3105–3108; c) G. Mehta, S. S. Ramesh, Eur. J. Org. Chem.

2005, 2225–2238.

[1] a) C. B. Aakeröy, K. R. Seddon, Chem. Soc. Rev. 1993, 22, 397–

407; b) G. R. Desiraju, Angew. Chem. Int. Ed. Engl. 1995, 34,

2311–2327; G. R. Desiraju, Angew. Chem. 1995, 107, 2541–

2558; c) G. R. Desiraju, Crystal engineering. The design of or-

ganic solids; Elsevier, Amsterdam, 1989; d) C. B. Aakeröy, Acta

Crystallogr., Sect. B 1997, 53, 569–586.

[7]

[2] For some selected recent references on crystal engineering and

its applications, see: a) C. B. Aakeröy, A. M. Beatty, Aust. J.

Chem. 2001, 54, 409–421; b) M. D. Hollingsworth, Science

2002, 295, 2410–2413; c) O. Almarsson, M. J. Zaworotko,

Chem. Commun. 2003, 1889–1896; d) D. Braga, Chem. Com-

[8]

[9]

ˇ

mun. 2003, 2751–2754; e) B. Kojic´-Prodic´, Z. Stefanic´, M.

ˇ

Zinic´, Croat. Chem. Acta 2004, 77, 415–425; f) P. Erk, H. Hen-

gelsberg, M. F. Haddow, R. van Gelder, CrystEngComm 2004,

6, 474–483; g) M. D. Ward, M. J. Horner, CrystEngComm

2004, 6, 401–407; h) S. George, A. Nangia, C.-K. Lam, T. C. W.

Mak, J.-F. Nicoud, Chem. Commun. 2004, 21, 1202–1203; i) D.

Schlatter, R. Thoma, E. Kueng, M. Stihle, F. Mueller, E. Bor-

roni, A. Cesura, M. Hennig, Acta Crystallogr., Sect. D 2005,

61, 513–519; j) K. Kinbara, Synlett 2005, 732–743; k) A. V.

Trask, W. D. S. Motherwell, W. Jones, Cryst. Growth Des. 2005,

5, 1013–1021.

[10]

The details of the synthetic procedures followed in the prepara-

tion of the polycyclitols 1–3 have been included in the Exp.

Sect.

A. Shani, F. Sondheimer, J. Am. Chem. Soc. 1967, 89, 6310–

6317.

[11]

[12]

Crystal data for 12: C₁₀H₁₄O₄, M = 252.31, monoclinic, space

group P2₁/n, a = 6.557(1), b = 6.240(1), c = 10.861(2) Å, β =

[3] a) G. A. Jeffrey, Acta Crystallogr., Sect. B 1990, 46, 89–103;

b) G. A. Jeffrey, W. Saenger, Hydrogen bonding in biological

structures, Springer, Berlin, 1991; c) C. André, P. Lurger, J.-H.

Fuhrhop, B. Rosengarten, Acta Crystallogr., Sect. B 1993, 49,

375–382; d) G. A. Jeffrey, An Introduction to Hydrogen Bond-

90.319(3)°, V = 444.40(2) Å³, Z = 2, F(000) = 212, ρ_calcd.

1.48 gcm^–3, µ = 0.114 mm^–1, T = 293 K, R₁= 0.045, wR₂

=

0.103, GOF = 1.118 for 724 reflections with IϾ2σ(I). An OR-

TEP plot of 12, with displacement ellipsoids for non-hydrogen

atoms drawn at the 50% probability level, is shown below.

Eur. J. Org. Chem. 2007, 423–436

www.eurjoc.org

435

G. Mehta, S. Sen, S. S. Ramesh

FULL PAPER

[18] Owing to the absence of any significant anomalous scatterers

(Z Ͼ Si), attempts to confirm the absolute structure by refine-

ment of the Flack parameter led to an inconclusive value of

0.4(16) (see ref.^[19]). Therefore the intensities of the Friedel pairs

(883) were averaged prior to merging of data in Fdd2 and the

absolute configuration was assigned arbitrarily. The reported

value of R_intcorresponds to subsequent merging of equivalent

reflections in this space group.

[19] a) H. D. Flack, Acta Crystallogr., Sect. A 1983, 39, 876–888;

b) H. D. Flack, G. Bernardinelli, J. Appl. Crystallogr. 2000, 33,

1143–1148.

[20] The tetrol 16 crystallizes from methanol in two concomitant

polymorphic modifications (see ref.^[8a]). The melting point re-

ported corresponds to that of the major polymorph of the te-

trol.

[21] a) SMART (V6.028), Bruker AXS Inc., Madison, Wisconsin,

USA, 1998; b) SAINT (V6.02), Bruker AXS Inc., Madison,

Wisconsin, USA, 1998.

[22] G. M. Sheldrick, SADABS, University of Göttingen, Germany,

1996.

[23] G. M. Sheldrick, SHELXL97, University of Göttingen, Ger-

many, 1997.

[24] M. Nardelli, J. Appl. Crystallogr. 1995, 28, 659.

[25] A. L. Spek, Acta Crystallogr., Sect. A 1990, 46, C34.

Received: July 17, 2006

[13] a) A. J. Birch, P. Fitton, D. C. C. Smith, D. E. Steere, A. R.

Stelfox, J. Chem. Soc. 1963, 2209–2216; b) J. J. Runge, Prakt.

Chem. 1966, 31, 280–292; c) R. G. Harvey, J. Org. Chem. 1967,

32, 238–240.

[14] a) E. Vogel, M. Biskup, A. Vogel, H. Günther, Angew. Chem.

Int. Ed. Engl. 1973, 12, 989–991; b) E. Vogel, F. Kuebart, J. A.

Marco, R. Andree, J. Am. Chem. Soc. 1983, 105, 6982–6983;

c) P. J. Garatt, F. Sondheimer, J. Chem. Soc., C 1967, 565–568.

[15] a) T. W. Craig, G. R. Harvey, G. A. Berchtold, J. Org. Chem.

1967, 32, 3743–3749; b) G. Mehta, S. Sen, S. Dey, Acta Crys-

tallogr., Sect. C 2005, 61, o358–360.

[16] a) C. P. Brock, B. Schweizer, J. D. Dunitz, J. Am. Chem. Soc.

1991, 113, 9811–9820; b) M. C. Etter, K.-S. Huang, Chem. Ma-

ter. 1992, 4, 824–827; c) A. Gavezzotti, Synlett 2002, 201–214;

d) M. S. Hendi, P. Hooter, R. E. Davis, V. M. Lynch, K. A.

Wheeler, Cryst. Growth Des. 2004, 4, 95–101.

Published Online: October 18, 2006

[17] C. P. Brock, J. D. Dunitz, Chem. Mater. 1994, 6, 1118–1127.

436

www.eurjoc.org

Eur. J. Org. Chem. 2007, 423–436

Article Doi

Crystal structures of conformationally locked cyclitols: An analysis of hydrogen-bonded architectures and their implications in crystal engineering

DOI: 10.1002/ejoc.200600610

Source and publish data:

Authors:

Article abstract of DOI:10.1002/ejoc.200600610

Full text of DOI:10.1002/ejoc.200600610

Products guided by the article

R&D Labs maybe for 5910-28-1

Relevant to this article

Hot Product