see: Booth, H., Dixon, J. M., Readshaw, S.A.; Experimental Studies of the Anomeric Effect. Part V. The Influence of Some Solvents on the Conformational Equilibria in 2-Methoxy- and 2-(2', 2', 2'-Trifluoroethoxy)-Tetrahydropyran. Tetrahedron, 1992,48, 6151-6160.
 

Introduction

The anomeric effect, originally defined as the preference of an electronegative substituent at the anomeric position of a carbohydrate to be axially rather than equatorially oriented, is now understood to be the result of multiple steric and stereoelectronic interactions.
In addition, the following subgoal will be addressed: The degree of contribution from dipole-dipole and molecular orbital interactions varies between systems and with solvent.  The anomeric effect was born out of carbohydrate chemistry. Carbohydrates, also known as saccharides, exist as both monomeric sugars and their polymers. Monosaccharides are polyhyroxy aldehydes or ketones with at least three carbons, one of which is a carbonyl and each remaining carbon bares a hydroxyl group. In aqueous solution, carbohydrates exist in an equilibrium between their open chain and cyclic forms (Figure 1). Aldose sugars containing 5 or more carbons or ketoses containing 6 or more carbons cyclize to form hemiacetals in solution.

The carbonyl carbon in the open chain becomes the anomeric carbon (*, Figures 1, 2, and 3) in the cyclized structure, which is the most oxidized carbon of the ring. Cyclized aldose and ketose carbohydrates can adopt either an a or b anomeric configuration depending on the orientation of the group attached to the anomeric carbon.¹ The assignment of configuration is determined by the relationship between the group attached to the anomeric carbon (ie. hydroxyl) and the group attached to the highest numbered chiral carbon in the ring (Figure 2). When the groups are trans the a designation is used and when they are cis, the b designation is used.²

The anomeric effect was initially defined as the preference for an electronegative substituent, at the anomeric carbon in a carbohydrate, to be in an axial rather than equatorial orientation.³ A more modern definition of the anomeric effect has come to include not only carbohydrates, but also saturated heterocycles and acyclic systems containing heteroatoms. In cyclohexane, it is well known that substituents attached to the ring tend to prefer the equatiorial position over the axial position due to unfavorable steric intercactions in the axial conformer. However, contrary to steric predictions, electronegative substituents at the anomeric position of carbohydrates and other heterocycles have been observed to exist predominantly as the axial conformer. This phenomenon is known as the anomeric effect.

In 1955, Chu and Lemieux discovered the first example of the anomeric effect.⁴ They determined that in aldohexopyranoses the conformation with the axial orientation of the acetoxy group at position C(2) is favored. The equilibrium between the axial and equatorial conformers lies in favor of the axial position (Figure 3).⁵ Calculations of D G (kcal/mol) confirmed the

equilibrium shift toward the axial position. Since the equilibrium lies opposite of the outcome predicted by sterics, researchers have used stereoelectronic arguments to rationalize the apparent anomaly in conformational selectivity. Two rationales exist that explain the origin of the anomeric effect, the electrostatic model (also known as the rabbit ear and dipole-dipole model) and the molecular orbital model (also known as the double-bond/no-bond model and hyperconjugative model).⁶

Models

Originally, the anomeric effect was defined in terms of electrostatic interactions alone. It has been stated in the electrostatic model⁶ that there is increased preference for the electronegative group to be axial due to the repulsive dipole-dipole interactions (Figure 4). ⁷

In this example, the structure 1E, with an equatorial C-X bond, has the C-X and net C-O dipole moments pointing in the same direction. This causes the dipoles to be additive, destabilizing the molecule and thus increasing the energy. Structure 1A has dipoles that offset each other (m @ 0).³ This minimizes the destabilization, causing a more stable conformation when the X group is in the axial position.

This model, once thought to be complete, is complicated by effects seen when changing the polarity of the solvent. Using the electrostatic model, one would predict that a polar solvent would stabilize the more polar equatorial conformation to a greater extent than a non-polar solvent. This would drive the equilibrium to shift in favor of the equatorial conformation as you move from non-polar to polar solvents. This result is seen for 2-methoxy tetrahydropyran, where 83% of the molecules adopt the axial conformation in the non-polar solvent CCl₄ and only 52% lie axial in water (Table 1).⁸ However, the predicted result is not seen for many systems.

Juaristi, et al.⁹ have shown that in 2-carbomethoxy-1,3-dithiane at low temperatures the axial conformation is preferred in polar solvents. In changing from methylene chloride to acetone to methanol, the DG for the equilibrium shown in Figure 5 increases from 0.83 to 0.92 to 1.13 kcal/mol, respectively.⁹

As demonstrated above, there are systems in which increasing the polarity of the solvent increases the percent axial conformation. Thus, the electrostatic model is not sufficient for explaining the anomeric effect. In addition to solvent effects, there are many examples where changes in bond-length are also observed, a result not explainable by the electostatic model.¹⁰ These observations indicate that more than just the electrostatic model must be operating and lead to the inception of the MO model. A modern understanding of the anomeric effect includes both molecular orbital and electrostatic interactions as well as solvent effects which all dictate conformational preferences.

The molecular orbital model presents an alternative explanation for the anomeric effect. ³ Preference for the axial conformation is attributed to stabilization by donation of electron density from the anti-periplanar orbital on oxygen (sp³ or p, represented as n_o) to the antibonding orbital

of the C-X bond (C-Cl antibonding orbital in Figure 6).⁷ This donation can also be represented in an MO diagram (Figure 7). ⁷ The orbital with the lone pair donates electrons to the s* orbital of the C-X bond (in this case C-Cl antibonding orbital). This stabilizing interaction can only occur in the axial conformation due to the requirement for the donating oxygen orbital and the C-X bond to be anti-periplanar, thus providing preference for the axial conformation.
 

In a compound with an alkoxy substituent (or other substituent capable of donating lone pair electrons) at the anomeric carbon, electron donation may proceed from the exo-cyclic oxygen to the antibonding orbital of the bond between the anomeric carbon and the ring oxygen (Figure 8). This process is known as the exo-anomeric effect in contrast to the endo-anomeric effect (refered to simply as the anomeric effect).¹¹ It is noteworthy to mention that the conformation of the substituent at the anomeric carbon must be such that a lone pair orbital is anti-periplanar to the antibonding orbital of the bond between the anomeric carbon and the ring oxygen.

Magnitude

The magnitude of the anomeric effect was originally calculated by comparing the axial versus equatorial equilibrium of a substituted, saturated heterocyclic system (ie. 2-Cl-oxane) to the same equilibrium in a cyclohexyl system with the same substituent. The energy value attributed to stabilization by the anomeric effect (DDG) was assumed to be equal to the DG for the equilibrium of the substituted oxane, less the DG for the equilibrium of the substituted cyclohexane. Several examples of how this was done for substituted oxane systems are given in Table 2.¹²

It is commonly accepted today that comparison of cyclohexyl and saturated heterocyclic systems does not give an accurate measure of the amount stabilization due to the anomeric effect.^12,13 Comparisons of this sort are made under the assumption that the only variable changing between the two systems is the replacement of a cyclohexyl methylene group with an oxygen atom. This is not a valid assumption because other variables also change, such as bond lengths. It is well known that a C-O bond (1.420Å, oxane)¹⁴ is shorter than a C-C bond (1.536Å, cyclohexane)¹⁴. This change in bond length also changes the amount of steric interaction that takes place in 1,3 diaxial interactions. Use of the aforementioned method remained prevalent for numerous years, so one must be cautious when interpreting data from early entries into the literature. Current methods for calculating the magnitude of the anomeric effect involve experimental determination of a correction factor and use of semi-empirical calculations.^7,13 These methods are beyond the scope of this paper, but a good review can be found in reference 7 and the references therein.

Modern Definition

In the modern or more generalized definition of the anomeric effect, it is stated that there is a preference for a substituent on the anomeric carbon to be in the synclinal (gauche) position over the anti-periplanar (anti) (Figure 9).¹⁵ The bond between the electronegative atom X and the anomeric carbon A can be either synclinal (axial in a ring system) to the bond between the atom which possesses lone pairs (the heteroatom in a ring system) and the R group ( a bridge atom in the ring system) or anti-periplanar (equatorial in a ring system).¹⁶ This more generalized definition allows the anomeric effect to be applied to acyclic molecules.

Results

A case study of how the magnitude of the anomeric effect is determined relies on determination of the concentrations of axial (3A) and equatorial (3E) conformers at equilibrium (Figure 10). This study by Booth et. al.¹¹ is unique in its attempt to determine not only the change in free energy of the equilibrium 3A ´ 3E (DG), but also the changes in enthalpy (DH) and entropy (DS). One method for the determination of the relative amounts of 3A and 3E utilizes ¹³C NMR.¹¹ At room temperature, the conformers are not resolvable by NMR. Thus, the NMR must be run at low temperature to "freeze out" the conformers, Table 3. At 294 K all the signals for both conformers of the equilibrium, 3A ´ 3E, are coincident. However, at 150 K, the signals for all of the carbons of both isomers are separated cleanly. The equilibrium constant was determined from the integration of the peaks in the ¹³C spectrum.

The relative concentrations and thus the equilibrium constants were then determined at different temperatures. A plot of lnK versus T^-1 was then used to calculate the thermodynamic parameters DG, DH, and DS.¹⁷ (See Table 4)¹¹

Discussion

In molecules such as 3 (Figure 10), the preference for the axial conformation has been determined to be the result of competing endo- and exo-anomeric effects. Both oxygens of compound 2 have the potential to donate their electrons into the s* of the adjacent C-O bond. However, in the equatorial conformation, only the exo-anomeric effect is possible. In the axial conformer, both exo- and endo-anomeric effects are possible (Figure 8).¹¹ It has been determined that the endo-anomeric effect is stronger than the exo-anomeric effect in axial conformations.⁷

The anomeric effect has thus far been shown to have significant consequences with respect to molecular conformation. Molecules such as 4 exist predominantly as the conformation with the electronegative group (Cl) axial, even when this forces three other groups axial (Figure 11).⁷

Booth et al. propose that there is competing exo- versus endo-anomeric effects in axial compounds (4A). The effect that is predominantly observed is governed by the solvent. To explain how the solvent affects the equilibrum the thermodynamic components of free energy (ÆG) must be examined (D H and D S). Booth et al. predicted a decrease in ÆH when moving from a non-polar (ether/toluene) to a polar solvent (ether/methanol).¹¹ This assumption was based on the idea that a more polar solvent would stabilize the more polar equatorial conformation (electrostatic model). However, experimentation the ÆH increased (3A« 3E, Table 4). This is a result of hydrogen bonding between the methanol and the axial alkoxy group. This hydrogen bonding causes a decrease in the enthalpy of the axial conformation relative to the equatorial conformation.

Another parameter which was not predicted well by the electrostatic model is ÆS. The axial conformation has both the endo- and exo-anomeric effects. It is the endo-anomeric effect that has the largest contribution to the ÆS parameter. In Figure 12, the hydrogen bonding of methanol to the alkoxy group is facilitated by a shift in electron density toward the alkoxy group. The hydrogen bonding causes a more ordered system thus decreasing entropy for the axial conformation. Overall, both ÆH and ÆS of the equilibrium increase when changing to a more polar solvent. When the endo-anomeric effect is acting, the lone pairs on the ring oxygen (or heteroatom) are less able to hydrogen bond with solvent. However, the lone pairs on the exocyclic oxygen are able to hydrogen bond. In other words, when the substituent is axial, hydrogen bonding from polar solvents greatly reduces the exo-anomeric effect. When the substituent is equatorial, the opposite seems to hold and the exo-anomeric predominates.

Conclusion

It has been shown that steric, electrostatic, and stereoelectronic factors are all important forces which influence the anomeric effect. Experimentally, researchers have seen that the ratio of axial and equatorial isomers at equilibrium depends mainly on the type of substituent at the anomeric center, the other substituents on the ring, the solvent, and the temperature. These effects are measurable using low temperature NMR methods in which both axial and equatorial compounds are resolved. Originally, electrostatic arguments were employed to explain the experimental results. Current reports seem to favor MO arguments although new reports continue to be published favoring both rationales. The debate continues as to which model is more accurate and the conclusion seems to be that neither is sufficient, but rather there is an interplay between aspects of both models.

The case study by Booth et al. was chosen because it provides an example of the current focus in the investigation of the anomeric effect. While investigations into temperature dependence and efforts to rationalization the thermodynamic parameters continue, much remains obscure. For example, methods are needed to better resolve the contributions of the endo- and exo-anomeric effects. Based on the knowledge already gained, a solid basis has been formed onto which further investigations can occur.

References

1. Moran, Scrimgeour. Ed. In Biochemistry. 2^nd ed.; Niel Patterson: Toronto, 1994, p. 9.8.
2. Mann, J.; Davidson, R.S.; Hobbs, J.B.; Banthorpe, D.V.; Harborne, J.B. In Natural Products: Their Chemistry and Biological Significance. Longman Group: Essex, 1994, Ch.1.
3. Fabian, M.A.; Armstrong, K.B.; Perrin, C.L. The Origin of the Anomeric Effect: Conformational Analysis of 2-Methoxy-1,3-dimethylhexahydropyrimidine. J. Am. Chem. Soc. 1994, 116, 715-722.
4. Salzner, U. Origin of the Anomeric Effect Revisited. Theoretical Conformation Analysis of 2-Hydroxypiperidine and 2-Hydroxyhexahydropyrimidine. J.Org. Chem. 1995, 60, 986-995.
5. Lemieux, R.U.; Chu, P. Conformations and Relative Stabilities of Acetylated Sugars as Determined by Nuclear Magnetic Resonance Spectroscopy and Anomerization Equilibria. Abstracts of Papers; 133^rd National Meeting of the American Chemical Society, San Francisco, CA; American Chemical Society: Washington, DC, 1958; 31N.
6. Edward, J.T. Stability of Glycosides to Acid Hydrolysis. Chem. Ind.1955, 1102-1104.
7. Juaristi, E.; Cuevas, G. Recent Studies of the Anomeric Effect. Tetrahedron, 1992,48, 5019-5087.
8. Lemieux, R.U.; Pavia, A.A.; Martin, J.C.; Wantanabe, K.A. Solvation Effects on Conformational Equilibria. Studies Related to the Conformational Properties of 2-Methoxytetrahydropyran and Related Methyl Glycopyranocides. Can. J. Chem. 1969, 47, 4427-4439.
9. Juaristi, E.; Tapia, J.; Mendez, R. Study of the Anomeric Effect in 2-substituted 1,3-dithianes. Tetrahedron, 1986, 42, 1253.
10. Deslongchamps, P. In Stereoelectronic Effects in Organic Chemistry; Pergamon Press: New York, 1983; Ch.1
11. Booth, H., Dixon, J.M., Readshaw, S.A.; Experimental Studies of the Anomeric Effect. Part V. The Influence of Some Solvents on the Conformational Equilibria in 2-Methoxy- and 2-(2', 2', 2'-Trifluoroethoxy)-Tetrahydropyran. Tetrahedron, 1992,48, 6151-6160.
12. Eliel, E.L.; Giza, C.A. Conformational Analysis. XVII. 2-Alkoxy- and 2-Alkythiotetrahydropyrans and 2-Alkoxy-1,3-dioxanes. The Anomeric Effect. J. Org. Chem. 1968, 33, 3754-3758.
13. Eliel, E.L. and Wilen, S.H. Ed. In Topics in Stereochemistry. Vol. 21. John Wiley and Sons: New York, 1994; p167.
14. Lide, D.R. Ed. In Handbook of Chemistry and Physics. 78^th ed.; CRC Press: New York, 1997, pp. 9-28, 9-39.
15. Lemieux, R.U. In Molecular Rearrangements, de Mayo, P., Ed.; Interscience Publishers: New York, 1964; p 709.
16. Krol, M.C.; Huige, C.J.M.; Altona, C. The Anomeric Effect: Ab-inito Studies of Molecules of theType X-CH₂-O-CH₃. J. Comp. Chem. 1990, 11, 765-790.
17. For a review of thermodynamic parameters refer to: Carey, F.A. and Sundberg, R.J. In Advanced Organic Chemistry, Part A: Structure and Mechanisms. Plenum Press: New York, 1990, Ch. 4.

1. One of the methods used to quantify the anomeric effect is based on a moleculeâs net dipole moment. (1)First, explain why structure ÎBâ has a larger dipole moment. (2)Next, of these two molecules, which would have a larger anomeric effect in a nonpolar solvent? Please explain your answer using a dipole-dipole interactions argument. Use pictures when necessary. (Angew. Chem., Int. Ed. Engl. 1986, 25, 287.)

Answer:

1. Structure ÎBâ has a larger net dipole because more of its dipole-dipole interactions between the two carbon/methoxysilyl bonds and heteroatoms are in the same direction. The heteroatoms net dipoles will add to zero because each atoms dipole faces the opposite direction. The equatorial methoxysilyl groupsâ net dipole does not cancel therefore and larger dipole results. In structure ÎA,â the dipole moment of the heteroatoms negate each other, but with the methoxysilyl groups in anti-axial positions, the dipole moments nearly adds to zero. Thus structure ÎAâ has a much smaller net dipole moment.
2. In a nonpolar solvent structure ÎAâ would have a larger anomeric effect. This would be evident in an equilibrum shift to the left. The larger dipole moment of structure ÎBâ shows that it would more stable in a polar solvent. In a nonpolar solvent, structure ÎAâ would not suffer from the destabilization present in structure ÎBâ arising from its large dipole moment.

2. For the reaction shown below, compound 1 goes through the intermediate (compound 2) to provide the final product, compound 3. Indicate which of the two intermediate structures, 2a or 2b, would be the major conformation and why. Draw the proper orbitals to explain your answer. (J. Org. Chem. 1998, 63, 5144-5153.)

 

 
 


3. 

 
 



Answer: Intermediate 2b is the major conformation because the s* orbital of the ethoxy bond can interact favorably with the lone pair electron orbitals of the O atom in the chair conformation.







4. Provide a rationale for the preference of the axial conformation of the structure shown below. Indicate on both structures the reason for their differences in stability. (J. Org. Chem. 1988, 53, 3334.)

Answer: The axial conformation is more stable based on stereoelectronic effects: a hyperconjugative interaction is responsible for the preferred antiperiplanar (diaxial) orientation of the S=O group and the lone pair of electrons on the sulfur. The anti arrangement of the two lone pairs in the equatorial form destabilizes this conformer.