A Study of the Effects of Strain
on the Structure and Reactivity of Bridgehead Olefins

Lease, T.G.; Shea, K.J. "The Type 2 Intramolecular Imino Diels-Alder Reaction. Synthesis and Structural Characterization of Bicyclo[n.3.1] Bridgehead Olefins/Bridgehead Lactams". J. Am. Chem. Soc. 1993, 115, 2248-2260.
 

The influence of the structure of bridgehead pi-bonds on the reactivity and cyclization of bicyclic compounds.

Introduction

A significant number of compounds that show some form of biological activity contain polycyclic ring systems with a double bond that involves one of the bridgehead carbons. Bridgehead carbons (1, 2 in Figure 1) are located at the branching points of the rings. For example, in the structure of taxol, a chemotherapeutic agent shown in Figure 1, the double bond at atom 2 is located at the bridgehead position of rings A and B.

Figure 1. Taxol.1

As a result of the pharmaceutical importance of such compounds, the synthesis and reactivity of bridgehead p bonds has been studied with intense interest by the chemical community.3 Examples of molecules containing bridgehead p bonds are shown in Figure 2.

Figure 2. Bridgehead olefins.2

Bredt and coworkers were the first to recognize the synthetic impossibility of including double bonds at bridgehead positions (A, B) in camphane and pinane systems (Figure 3). 4

Figure 3. Compounds related to camphane and pinane.

Based upon this observation, Bredt formulated what came to be known as Bredt‚s rule: "In systems of the camphane and pinane series and related compounds, the branching points A and B of the carbon bridges (the bridgeheads) cannot be involved in a double bond (Figure 3)."3 Bredt's rule today is understood as the tendency of double bonds to form away form the bridgehead positions (Figure 4).4

Figure 4. Example of Bredt's rule.

This postulate has been used to justify the failure of certain types of reactions that would have otherwise led to the formation of bicyclic systems with bridgehead p bonds. For example, Bredt's rule explains the failure of some anhydride formations and hydrohalide eliminations (Figure 5).2 This can be explained with the aid of Bredt‚s rule, because the product in this case would contain a bridgehead double bond.

Figure 5. Demonstration of Bredt's rule.2

Bicyclic cycloalkenes with a double bond placed away from the bridgehead position are called Bredt-alkenes. Whereas cyclic systems containing double bonds at the bridgehead are called anti-Bredt alkenes.2

Bredt's rule, however, is limited to bicyclic systems which are not large enough to accommodate the strain that results from a bridgehead double bond. As it can be seen in the case of taxol,1 for example, there exist compounds that do contain a double bond at the bridgehead. The first laboratory synthesis of such molecules was accomplished by Prelog and coworkers in 1940. They managed to isolate bicyclic compounds of the [5.3.1] type with an olefin at the bridgehead position (Figure 6).

Figure 6. Bicyclic compounds of the [5.3.1] type.6
 
 

Based upon this and other similar work that led to the generation of strained bridgehead double bonds, Fawcett et al. formulated the more general "Fawcett Rule". This rule states that for bicyclo[x.y.z]alkenes S should be no smaller than nine, where S = x + y + z. For example, in Figure 5, x = 5, y = 3, z = 1 and therefore S = 9.7 Another example can be found in the successful formation of compound A and the unsuccessful generation of structure B (Figure 7).4

Figure 7. Decarboxylation of b-keto acids.4

Today Fawcett's postulate is primarily of historical significance because in 1967, Wiseman and coworkers4 synthesized bicyclo[3.3.1]non-1-ene (Figure 8), a structure where

S = 7.

Figure 8. Reaction forming bicyclo[3.3.1]non-1-ene.4

This apparent inconsistency with Fawcett's rule was explained by Wiseman et al. by comparing the strain in a bicyclic ring with the strain of a trans-cycloalkene. In Figure 8, for example, the double bond is exocyclic to ring ab but endocyclic to rings ac and bc. In other words the double bond could be viewed as trans in ring ac and cis in ring bc (Figure 9). Based on these findings, Wiseman postulated that a double bond at the bridgehead forms so that it is trans in the larger of the two rings in which it is endocyclic. To date Wiseman's postulate is the accepted reference for predicting where the double bond in a bicyclic system will form.

Figure 9. Example of generic cycloalkene (A) and bicyclo[3.3.1]non-1-ene (B).8

The physical reason for the difficulty associated with the synthesis of anti-Bredt molecules arises from the preferred geometry of the double bond. Ideally, two sp2-hybridized carbons and their substituents lie in one plane (Figure 10).
 
 




Upon inclusion in a ring system, however, this planarity is disrupted as a result of the topological constraints of the ring. For example, in the case of bicyclo[3.3.1]non-1-ene (Figure 8), the only possible geometry that allows for the formation of the bicyclic system disrupts the planar structure of the double bond at the bridgehead. This disruption forces the substituents (W, X, Y, Z, Figure 11) of the olefinic carbons out of the plane of the p bond. The strain caused by the distortion of the double bond becomes apparent if one looks down the bond axis. The topology of the ring forces a misalignment of the non-hybridized p atomic orbitals by an angle t (Figure 11). This decreases the overlap between these two orbitals, thereby destabilizing the bond which causes an increase in the reactivity of the double bond.

In addition to the magnitude of the torsional angle distortion, the degree of rehybridization of p-orbitals to include s-character must be quantified. The c angle that results from a projection of the s -bond connecting an arbitrary member of the W, X, Y, Z substituents to the geminal substituent through the central atom is used to determine rehybridization (Figure 11). An sp2 hybridized atom has a c angle of 00, while for an sp3 hybridized atom c = 60o.9
 
 




Strained bridgehead p bonds are highly reactive.4 Therefore, reagents will add easily across a bridgehead double bond. Olefinic bridgehead compounds with a small S number (S<7) are unstable and highly reactive. For example, the perfluoronorbor-1-nene intermediate (A) has only been characterized by the subsequent Diels-Alder reaction it undergoes (Figure 12).

Figure 12. Reactivity of perfluoronorbor-1-ene.2

1-lithioperfluoronorbornane, a Bredt-alkene of S=5, eliminates to form (A) which then reacts with furan to form the two possible stereoisomers.2 In accordance with Bredt's rule this compound is highly unstable and its existence is fleeting.

Not only does strain affect reactivity, but it also affects cyclization. The strain energy of the bridgehead olefin is due to the strain from the double bond and the size of the ring. Cyclization is preferred in the least strained bicyclic compounds; in other words, isolation of anti-Bredt compounds is more likely when the double bond is located in the larger ring. For example, as shown in Figure 13, the smaller bridgehead olefin product from the intramolecular cyclopropanation of unsaturated carbenes could not be isolated.2
 
 

Results

The effect of bridgehead p bonds on structure and reactivity is exemplified by a case study of five novel anti-Bredt compounds. Compounds 1-3 synthesized by Lease and Shea4 contain both a bridgehead olefin and a bridgehead amide (Figure 14). Specifically, structural determination was performed to evaluate strain and distortion on the olefin and amide occupying the bridgehead positions of 1-3. Additionally, the structures of bicyclo[3.3.1]non-1-ene (4) and bicyclo[4.3.1]dec-1(9)-ene (5) were studied (Figure 14). , 4






Figure 14. Compounds Studied.

The bridgehead double bonds in compounds 1-5, characterized by X-ray crystallography, revealed two deviations from the optimal planar geometry. These two deviations, angles c and t, represent the torsional distortion and rehybridization values for the olefin and amide systems, Tables II and III, respectively.4 In Table II, cB is the rehybridization of the bridgehead carbon, cE represents the angle of rehybridization at the exocyclic carbon, while t represents the torsional angle between the p atomic orbitals and the sp2 hybridized carbon. An increase in tether length results in a net decrease of cB, cE, and t. Similarly, this trend is observed for bridgehead amides (Table III).

Table II. Distortions in Bridgehead Olefins 1-5.

Bridgehead
Tether
cB
cE
t
Olefin
Length
(deg)
(deg)
(deg)
14
3
39.0(2)
17.9(1.1)
10.8(5)
24
4
22.7(3)
8.2(1.8)
6.4(7)
34
5
8.4(3)
0.0(1.3)
0.45(5)
410
3
35.3
26
14.7
510
4
15.6
12.2
6.9

 
 
 

Table III. Distortion Parameters and Bond Lengths of Bridgehead Amides. 4

Bridgehead
Tether
cN
cC
t
Lactam
Length
(deg)
(deg)
(deg)
14
3
54.9(1)
1.4(2)
16.7(1)
24
4
46.4(2)
1.2(3)
7.5(2)
34
5
38.2(2)
-0.2(3)
0.9(2)

Figure 15 demonstrates the high rate of hydration of 4 compared with an unstrained non-bridgehead alkene.

Figure 15. Relative rates of hydration for 4 and 2-methylbut-2-ene.

Figure 16 shows the relative rates of the 1,3-dipolar cycloaddition of picryl azide with 4, 5, and cyclohexene.

Figure 16. Relative rates of 1,3-dipolar cycloadditions of picryl azide to cyclohexene, 4 and 5. 10

Discussion

Compounds 1-5 are anti-Bredt compounds because they contain bridgehead olefins. In addition, 1-2 and 4-5 have values of S that are less than 9, demonstrating the limitations of Fawcett‚s rules. However, by applying Wiseman‚s postulate, one sees that 1-5 each contain a bridgehead double bond that is trans in the larger of the two rings in which it is endocyclic (Figure 17), displaying the breadth and applicability of Wiseman‚s findings.

Figure 17. Trans structures of 1-5.

Though 1-5 exist and can be isolated, they are nevertheless strained species. The x-ray crystallography data in Table II quantifies this strain in the form of torsional distortion (t) of the p bond and the rehybridization (c) of the carbon atomic orbitals. The conformational mobility, or rather, the ability of the tethered ring to assume various readily accessible conformations determines the degree of strain present in the bridgehead olefin.4 The observed decrease in strain (decrease in t and c) when progressing from 1 to 3 and from 4 to 5 results from the increase in conformational mobility associated with adding an additional carbon to the tether. Figure 18 displays two views (A and B) of the strained bridgehead olefins in products 1-3. Perspective A shows the projection of the olefin when viewed along the bridgehead double bond axis, clearly depicting that the olefin is distorted from the ideal planar geometry. Perspective B provides a clear view of the rehybridization at both olefinic carbons.4 These trends are in accord with the principles outlined in the introduction: as the size of the tethered ring increases, the strain present at the bridgehead olefin decreases.

Figure 18. Depiction of rehybridization and distortion of olefins in 1, 2, and 3.9

Products 1-3 have an amide functionality at the bridgehead position. Figure 19 displays two views (A and B) of three representatives of this group, where A and B both correspond to the analogous A and B in Figure 12. The rehybridization trends at the amide nitrogen (cN) and the exocyclic amide carbon (cC) are similar to those observed for the olefinic carbons; that is, the nitrogen and carbon atomic orbitals involved in p bonding show increasingly more s-character when progressing from 3 to 1 (Table III). Likewise, the torsional strain present in the

Figure 19. Depiction of rehybridization and distortion of amides in 1, 2, and 3.

bridgehead amide also increases from product 3 to 1. One can apply the same rationale to explain the geometric trends in the amide as were applied in the olefinic bond.

A simplified application of this analysis can be used to describe the trends of 4-5, which contain only a bridgehead olefin. The same trends apply to 4 and 5: the torsional strain and rehybridization values also decrease with an increase in tether length (Table II).

While the rehybridization and torsional strain trends are similar for both the bridgehead amide and olefin, the amide more easily accommodates the strain resulting from the short tether lengths. To understand this, one should consider the resonance structures of the two functionalities in Figure 19. The amide is not as strained as the olefin because its predominant resonance form (A) is unaffected by torsional strain whereas the prevailing olefin resonance form (C) is susceptible to larger torsional strain.. It is interesting to note that NMR studies have shown the bridgehead carbon becomes more electron deficient as the olefin becomes more strained, while the exocyclic carbon gains electron density, thus polarizing the bond.

Figure 20. Resonance forms of amide and olefin functionality.4

The strain in anti-Bredt compounds 1-5 renders them more reactive species than compounds with non-bridgehead olefins. This increased reactivity results from the natural tendency to relieve strain. Figures 15 displays the relative rates of hydration of bridgehead alkene 4 compared to 2-methylbut-2-ene. Understandably, 4 is more reactive than the acyclic alkene, because the former is more strained. Based on the data in Table II, 4 is more strained than 5, because 5 has one extra carbon in its tethered ring. Since 4 is more strained, it is expected that it should also be more reactive. Indeed, the data in Figure 16, which compares the relative rates of 1,3-dipolar cycloaddition of picryl azide to 4 and 5, validates this expectation: 4 is four orders of magnitude more reactive than 5. These examples of facile additions across bridgehead double bonds testify to the reactivity of bridgehead alkenes. Given the fast reaction rates and the number of different possible reactions, we see that bicyclic systems with a bridgehead alkene are a nice template from which to generate highly functionalized bicyclic systems.

Bridgehead amides like those found in 1-3 also exhibit high reactivity resulting from their inherent strain. One would expect the amide bond in A to be more susceptible to solvolysis than a common amide bond. This idea is supported in Figure 21 by the observation of Pracejus that A reacts 104 times faster than B in basic methanol solution.

Compound Relative rates of hydrolysis A 104

B 1

Figure 21. Comparison of rates of hydrolysis.

Conclusion

The inclusion of a p bond at the bridgehead position of a bicyclic ring system results in a geometrical distortion of the double bond. This leads to an increase in energy which is the cause for the difficulty of such systems to cyclize. A review of the reactivity of some strained p systems showed an increase of the reactivity in comparison to unstrained double bonds. It was shown that both an olefinic double bond and a lactam are distorted when placed at the bridgehead. The ability of the lactam to sustain the distortion more easily was explained with a resonance structure.

The authors achieved their goal in defining the basis of bridgehead p bond strain, Bredt's rule, and Wiseman's postulate. It is believed that the large number of explained examples provide for a better understanding of the effects of strain upon structure and reactivity in bicyclic cycloalkenes.

TEST QUESTIONS

QUESTION 1

Enol ethers are known to undergo facile hydrolysis in dilute acids to yield aldehydes or ketones. The rate-limiting step in this process is usually the electrophilic addition of H+ to the C=C bond. The analogous olefins require much stronger acids for H+ addition. For example, 1-cyclohexeneyl methyl ether (2) undergoes acid-catalyzed hydrolysis 1.4 x 105 times faster than 1-methylcyclohexene (10).4

However, it has been observed that glacial acetic acid adds to the C=C bond in bicyclo[3.3.1]non-1-ene (3) at room temperature quantitatively in under 2 minutes. Under the same conditions however, 9-oxabicyclo[3.3.1]non-1-ene (2) requires 68 hours to reach 50% conversion2. Explain the reactivity trends of olefins and enol ethers towards electrophilic addition of H+, then explain why the trends are reversed when the p bond is in a bridgehead position. In addition, explain why 3 is more reactive than 4 towards the addition of acetic acid.
 
 

 ANSWER:

The addition of H+ to olefins and enol ethers yields a carbocation intermediates (5 and 6). However, enol ethers yield more stable cations than olefins due to the large resonance stabilization imparted by the p orbitals on the adjacent oxygen (7) Since formation of the carbocation is the rate limiting step in most additions of this type, the reactivity of enol ethers towards electrophiles is expected to be much larger when compared to olefins.

  The situation in 3 and 4 is very different. Structure 3 shows high reactivity to acetic acid due to the torsional and pyramidal distortion of the p bond which polarizes the olefin as shown in 8a. This polarization resembles the transition state 9a, which leads to the formation of stabilized cation 10a and greatly facilitates the reaction.

In enol ether 4, one might expect an even greater reactivity due to the oxygen adjacent to the bridgehead carbon. However, the constrained geometry prevents the oxygen p orbitals from overlap with the low lying LUMO of the bridgehead carbocation, thus preventing resonance stabilization of the cation. The bridgehead carbocation is further destabilized by the inductive effect of oxygen, which draws electron density away from the electropositive bridgehead carbon. The end result is a high energy intermediate 10b, which greatly slows the reaction.






EXAM QUESTION #2
 
 

Structures 1-3 in Figure 1 contain bridgehead olefins. Judge which structures, if any, will be stable enough to be isolated at room temperature. Explain how your answer is justified by the rules and observations of Bredt and Wiseman.




Figure 1. Three hypothetical structures containing a bridgehead olefin.

ANSWER 1:

Structure 1 is not directly observed, but it has been trapped as a reactive intermediate with furan as the Diels-Alder cycloadduct. It is forbidden by Bredt‚s rule. According to Wiseman‚s treatment of bridgehead olefins, the strain in 1 would be analogous to trans-cyclohexene, which is an extremely reactive species.
 

ANSWER 2:

This structure is stable and isolable at room temperature. For example, the newly discovered natural product Pteridanoside contains such a bridgehead olefin. The bridgehead alkene is trans in an eight membered ring, suggesting reasonable stability.

Figure 2. Pteridanoside

ANSWER 3:

Although long sought, adamantene has never been isolated. It has, however, been trapped as a reactive intermediate by electron-rich dienes. It can be seen as a bicyclo[3.3.1]non-1-ene system, which is marginally stable, with an additional ring formed by an additional methylene group.
 

 References

1Danishefsky,S.J.; Master, J,; Young, Wendy B.; Link, J.T.; Lawence, B.; Magee, T.; Jung, D.; Issacs, R.; Bornmann, W.; Alaimo, C.; Coburn, C.; Di Grandi, M.;. Total Synthesis of Baccatin III and Taxol, J. Am. Chem. Soc. 1996, 118, 2843-2859.

2 Koebrich, G. Bredt Compounds and the Bredt Rule. Angew. Chem. Int. Ed. Engl., 1973, 12, 464-473.

3 Bredt, J.;Thouet, H.; Schmitz. Annalen. 1924, 437,1.

4 Wiseman, J.R.; Pletcher, W.A. Bredt‚s Rule III The Synthesis and Chemistry of Bicyclo[3.3.1]non-1-ene. J. Am. Chem. Soc. 1970, 956-962.

5 March, J., Advanced Organic Chemistry, 4th Edition, John Wiley & Sons, New York, 1994, p. 158-164.

6 Prelog, V. J. Uber Steriche Hinderung in Bruckenringen(Bredtsche Regel) und uber diue meso-trans-Stellung in Konden sierten Ringsystemen des Hexamethylens. Annalen Der Chemie, 437 1924, 1-13.

7 Fawcett, F.S. Chem. Rev. 1950, 47, 219.

8 Wiseman, J.R. Bredt‚s Rule. Bicyclo[3.3.1]non-1-ene. J. Am. Chem. Soc. 1967, 5966-1968.

9 Lease, T.G.; Shea, K.J. The Type 2 Intramolecular Imino Diels-Alder Reaction. Synthesis and Structural Characterization of Bicyclo[n.3.1] Bridgehead Olefin/Bridgehead Lactams. J. Am. Chem. Soc. 1993, 115, 2248-2260.

0 Shea, K.J., Kim, J. Influence of Strain on Chemical Reactivity. Relative Reactivity of Torsionally Strained Double Bonds in 1,3-Dipolar Cycloadditions. J. Am. Chem. Soc. 1992, 114, 4846-4855.

11 Pracejus, H.Neues Zur Sterischen Mesomeriehinderung Bei Lactamen Vom Typ Des a-Chinuclidons. Tetrahedron, 1965, 21, 2257-2270.

12 Fife, T.H.; Vinyl Ether Hydrolysis. The Facile General Acid Catalyzed Conversion of 2-Ethoxy-1-Cyclopentene-1-Carboxylic Acid to Cyclopentanone. J. Am. Chem. Soc. 1965, 1084-1089.

13 Kresge A.J.; Sagatys D.S.; Chen H.L.; Vinyl Ether Hydrolysis. 9. Isotope Effects on Proton Transfer from the Hydronium Ion; J. Am. Chem. Soc., 1977, 99:22, 7228.

14 Guenzet J.; Camps, M.; Catalyze Par Les Acides De Lewis-I: Etude Cinetique De La Reaction D‚Esterfication Des Olefines Catalysee Par SnCl4. Tetrahedron, 1974, 30, 849-856.

15 Keese, R; Krebs, E.P. Concerning the Existence of 1-Norbornene. Angew. Chem. Internat. Edit. 1971, 10, pp. 262-263

16 Castillo F. U.; Sakagami Y.; Alonso-Amelot, M.; Ojika, M.; "Pteridanoside, the First Protoilludane Sesquiterpene Glucoside as a Toxic Component of the Neotropical Bracken fern Pteridium Aquilinum var. Caudatum" Tetrahedron 1999, 55, pp.12295-12300

17 Dauben, W. G.; Robbins D.; "New Applications of the Wittig Reaction to the Preparation of Strained Bridgehead Olefins" Tetrahedron Letters 1975, 2, pp. 151-154