The Synthesis of Conjugated Bis-Aryl Vinyl Substrates and Their Photoelectrocyclization Reactions towards Phenanthrene Derivatives

Abstract

The photoelectrocyclization of conjugated vinyl biaryls has proven to be a valuable and efficient strategy for generating phenanthrene derivatives. Contained in this review is an overview of the mechanism for the transformation and a discussion of the reaction scope with a focus on the electrocyclization itself, rearomatization, and the application of the reaction in natural product synthesis.

Graphical Abstract

1. Introduction

Dihydrophenanthrenes are present in a wide array of interesting and important molecules including natural products and materials.² Among the various methods that have been developed to generate them, the photoelectrocyclization of conjugated vinyl biaryl substrates stands among the more powerful. This is mostly because the reaction can lead to the generation of complex systems from relatively simple precursors but it is also a result of the reaction conditions being both sustainable and amenable to the incorporation of a wide array of functionality.³ In spite of these advantages and the first report appearing in 1971, until recently neither the scope of the photoelectrocyclization reaction nor its use in the synthesis of more complex substrates had been studied in any significant detail.⁴ This review outlines what is known about the reaction, including the impact of substitution and reaction conditions on the efficiency, stereoselectivity, and excited state of vinyl biaryl photoelectrocyclization reactions.

2. The Synthesis of Conjugated Vinyl Biaryls

The majority of the approaches to conjugated vinyl biaryls have involved the oxidative decomposition of phenanthrenes, aryl Grignard additions to aldehydes or ketones, and cross couplings. Outlined in this section is an overview of each of these approaches.

2A. Grignard Based Couplings.

Representative of the Grignard approach to these systems was Horgan and Morgan’s generation of vinyl diphenyl analogues from the addition of the biphenyl Grignard from 1 to aldehydes or ketones (Scheme 1).⁵ Dehydration of the resulting alcohols provided the requisite cyclization precursors 2.

Scheme 1.

Vinyl biaryls from Grignard additions

2B. Phenanthrene Oxidations.

As an example of the use of the oxidation of phenanthrenes to access cyclization precursors, Padwa carried out the ozonolysis of phenanthrene derivatives 3 to synthesize bis-2,2′-dialdehyde or ketones 4.⁶ A subsequent Wittig reaction was used to deliver divinyl biaryl cyclization precursors 5 (Scheme 2).

Scheme 2.

Vinyl biaryls from phenanthrene oxidations

2C. Ullmann Couplings.

To our knowledge Kende and Curran were the first to employ a cross-coupling strategy to vinyl biaryl cyclization precursors.⁷ In their synthesis of the natural product juncusol they used an Ullmann coupling reaction between 6 and 7 to give symmetrical bis-aryl 8 (Scheme 3). A subsequent desymmetrizing Wittig reaction generated the desired vinyl biaryl photoelectrocyclization precursor 9.

Scheme 3.

Vinyl biaryls from symmetrical Ullmann couplings

In their synthesis of the natural product santiagonamine, Kelly and co-workers also applied an Ullmann coupling to the synthesis of their photoelectrocyclization precursor (Scheme 4).⁸ They generated unsymmetrical biaryl 12 from the Pd(0)-catalyzed coupling of iodides 10 and 11. A subsequent Wittig reaction gave the unsymmetrical cyclization precursor 13.

Scheme 4.

Vinyl biaryls from unsymmetrical Ullmann couplings

2D. Ene-Yne Cycloadditions.

In addition to employing Grignard based approaches to their cyclization substrates, Lewis and Gevorgyan used a Pd(0) catalyzed ene-yne cycloaddition to generate terphenyl photoelectrocyclization precursor 15 (Scheme 5).⁹

Scheme 5.

Ene-yne cycloadditions to vinyl biaryls

2E. Phenolic Couplings.

Porco and co-workers used a phenolic coupling reaction between quinone ketal 16 and phenol 17 to generate the precursor to the kibdelone family of natural products.¹⁰ They found that a number of Lewis acids were capable of carrying out the desired transformation but during their optimization studies they identified InCl₃ as being the most effective, giving 18 in 70% yield (Scheme 6).

Scheme 6.

Phenol-quinone couplings to vinyl biaryls

2F. Suzuki–Miyaura Couplings.

In an effort to examine the scope of the photoelectrocyclization reaction, we sought a general approach to vinyl biaryl cyclization precursors and became intrigued with the use of sequential Suzuki–Miyaura cross-coupling reactions.¹¹ Our plan involved the use of an alkene 20, containing both a leaving group and a boronic acid derivative, in a cross-coupling reaction with vinyl halide or triflate 19 to give 21 (Scheme 7). To be successful, this reaction clearly required that the oxidative addition of X proceed preferentially to Y in the coupling partners 19 and 20, respectively. A second cross-coupling reaction with 22 would then convert 21 into cyclization precursor 23. We envisioned that this approach might be flexible, given that 23 could also be synthesized from the direct coupling of 19 with an appropriately substituted conjugated biaryl compound.

Scheme 7.

Sequential Suzuki–Miyaura couplings to vinyl biaryls

With the aforementioned plan in mind, we focused our initial efforts on the synthesis of bis-aryldihydropyridones.¹² Toward this goal, bromide 26 was synthesized, as illustrated in Scheme 8, by taking advantage of the chemoselective activation of the dihydropyridone iodide 24 in the presence of the bromide in 25. A second Suzuki–Miyaura coupling reaction with boronic acid 27 led to the installation of the second arene to give 28. We were pleased to find that this modular route to biarylalkenes resulted in the synthesis of a diverse array of cyclization precursors.

Scheme 8.

Sequential Suzuki–Miyaura couplings to biaryl dihydropyridones

A similar approach was used to synthesize biaryl cyclopentenones and cyclohexenones, except that vinyl nosylates were used as the coupling partners. The advantage of this was that vinyl nosylates enabled us to utilize readily available 1,3-cyclopentanediones and 1,3-cyclohexanediones as precursors.¹³ In a similar fashion to the iodides described above, the nosylates underwent chemoselective oxidative addition leading to bromophenylcycloalkenone 30 (Scheme 9). An ensuing Suzuki–Miyaura coupling between 30 and a series of arylboronic acids 27 gave the corresponding bis-aryl cyclopentenones and cyclohexenones 31.

Scheme 9.

Biaryl cyclopentenones and cyclohexenones from Suzuki– Miyaura coupling reactions

With some heteroaromatic containing substrates we found it more efficient to carry out the coupling reaction between biaryl halides and the corresponding alkenyl boronic acids or esters.¹⁴ For example, the halogen at the 2-position of dibromopyridine 32 underwent a selective oxidative addition reaction to provide 3-bromo-pyridine 33 after coupling with PhB(OH)₂ (Scheme 10). Biaryl bromide 33 was then subjected to a second Suzuki–Miyaura coupling with 34 to give pyridine cyclization precursor 35.

Scheme 10.

Pyridine photoelectrocyclization precursors

Similarly, symmetrical pyrazines and quinoxalines were employed in coupling reactions with arylboronic acids or esters to give the corresponding 2-aryl-3-halo-arenes 37 (Scheme 11).¹⁵ Cyclization precursors synthesized by using this approach included pyrazine 39 and quinoxaline 40.

Scheme 11.

Pyrazine and quinoxaline cyclization precursors

For some systems the coupling partners came from the halogenation of bis-aryl or heteroaryl compounds via direct electrophilic halogenation or deprotonation and halogenation (Scheme 12).¹⁶ A subsequent Suzuki–Miyaura coupling reaction provides cyclization precursor 43. Examples of products that were synthesized by using these approaches include 4-phenyl oxazole 44, 4-phenylimidazole 45, 2-phenyl indole 46, 3-phenyl thiophene 47, pyrimidoimidazole 48, and pyrazole 49.

Scheme 12.

Sequential biaryl halogenation and Suzuki–Miyaura couplings

In addition to the aforementioned methods, we turned to a Minisci-type reaction and 2-chloronicotinic acid 50 to synthesize pyridine 51, having the second arene at the 3-position of the pyridine ring (Scheme 13).¹⁷ A subsequent Suzuki–Miyaura coupling with 34 gave cyclization precursor 52.

Scheme 13.

Pyridine-containing photoelectrocyclization precursors from Minisci reactions

2G. [2+2]-Cycloadditions to Electrocyclization Precursors.

Biphenyl cyclobutenone 54 was generated using a [2+2] cycloaddition between 53 and in situ generated dichloroketene followed by reductive removal of the chlorines (Scheme 14).¹⁸ Alkyne precursor 53 was generated either from a Sonagashira reaction of the corresponding ortho-bromo biphenyl or from a Seyfurth–Gilbert or Corey–Fuchs homologation of the corresponding aldehyde.¹⁹

Scheme 14.

Biaryl cyclobutenone photoelectrocyclization precursors

3. Discovery of Vinyl Biphenyl Photoelectrocyclizations and Mechanistic Studies

3A. Discovery.

In 1971, R. J. Hayward and C. C. Leznoff reported that the irradiation of o-phenylstilbene 55 resulted in the formation of a 1:1 mixture of phenylphenanthrenes 56 and 57 in 36% yield (Scheme 15).²⁰ The products came from an expected stilbene photoelectrocyclization to give 56 and an unexpected biphenyl alkene cyclization to give 57.

Scheme 15.

ortho-Phenylstilbene photoelectrocyclizations

Hayward’s results preceded Morgan and Horgan’s more detailed and directed study of 2-vinylbiphenyl substrate 58.⁵ Exposure of 58 to UV light resulted in the generation of dihydrophenanthrene 59 in quantitative yield (Scheme 16). Subsequent to Morgan and Horgan’s discovery of the reaction, a significant amount of effort has been dedicated to determining the definitive mechanism for this transformation and, as a consequence, important insights have been uncovered.³ These are summarized below.

Scheme 16.

Vinyl biphenyl photoelectrocyclizations

3B. Mechanism – Singlet Excited States.

The photo- cyclization of 2-vinylbiphenyls to 9,10-dihydrophenanthrenes involves a two-step sequence after photon absorption (Scheme 17).²¹ The first step involves a 6π conrotatory photoelectrocyclization reaction to give [8,9a]-dihydrophenanthrene 62. While this has been shown to generally occur through singlet excited states, there are several examples of the reaction involving a triplet excited state. Subsequent to the formation of 62, the reintroduction of aromaticity to give 63 involves a thermally allowed, supra- facial [1,5]-hydride shift. Thus, with substituted alkenes, the stereochemistry of the final product was established in the initial cyclization event.

Scheme 17.

Vinyl biaryl singlet mechanism

3C. Mechanism – Triplet Excited States.

The radicallike behavior of the intermediate from triplet excited states means that their cyclizations do not necessarily follow orbital symmetry rules. Mechanistically, following excitation and intersystem crossing, electrocyclization and aromatization via [1,5]-hydride shift gives the corresponding dihydrophenthrene 67 (Scheme 18). Thus, dihydrophenanthrenes having both trans- and cis- stereochemistry can be formed from triplet derived electrocyclizations.

Scheme 18.

Vinyl biaryl triplet mechanism

Lapouyade et al. investigated the photocyclization of vinyl biaryl 68 by using both direct irradiation and sensitization.²² Direct irradiation led to a singlet excited state reaction that was stereospecific during its early stages (Scheme 19, entries 1 and 3). When the reactions were allowed to proceed for longer periods of time the stereoselectivity decreased; this decrease was attributed to a competitive isomerization of the starting material. In contrast to these results, when the photocyclization of 68 was initiated by the triplet sensitizer xanthone, it led to the formation of cis- 70 as the major product regardless of whether (E)-68 or (Z)- 68 was used or how long the reaction was allowed to proceed.

Scheme 19.

Singlet- and triplet-sensitized photoelectrocyclizations

Lapouyade and co-workers surmised that the products from the cyclization of 68 under sensitized conditions were coming from the equilibration of the triplet conformers. They proposed that sensitization of (E)-68 and (Z)-68 initially resulted in the formation of the two planar triplets A and B, respectively (Scheme 20). A and B underwent rapid equilibration to the thermodynamically more stable perpendicular triplet C and subsequently conformers D and E prior to cyclization. The cyclization of D and E resulted in the dearomatized intermediates 71 and 72 in a 9:1 ratio, respectively. A [1,5]-hydride shift then gave the products 69 and 70.

Scheme 20.

Conformational equilibria during triplet photoelectrocyclizations

In a subsequent study, Lazare, Lapouyade, and Bonneau investigated the photoelectrocyclizations of a series of 2-vinylbiphenyls in the presence of triplet sensitizers.²³ In addition to showing that the reaction was successful with a number of vinyl biphenyl substrates, they demonstrated that the triplet excited state of the starting material was higher in energy than the triplet excited state of the electrocyclization intermediate in the substrates that successfully underwent the reaction.

3D. Mechanism – Conformation and Quantum Yields.

While the isolated product yields for the singlet derived photoelectrocyclizations of vinyl biphenyls have generally been high, Lewis and co-workers showed that the quantum yields were typically low (<1%).^9,24 This phenomenon has been attributed to the NEER (Non-Equilibration of Excited Rotamers) principle and that the dominant ground-state conformation of the vinyl biphenyl cyclization precursors has the alkene and the biphenyl anti to one another, as indicated by 58 (see Scheme 21).²⁵

Scheme 21.

Conformational effects on the quantum yields for vinyl biphenyl photoelectrocyclization reactions

Because a larger fraction of their conformers can form the electrocyclization product, symmetrical triphenyl vinyl substrate 74 eliminates the conformational problems seen with 58. Lewis and co-workers showed that the photo- cyclization of 74 to give 75 was photoefficient by virtue of its 87% quantum yield (Scheme 22). The take-home message from these experiments was that, in the absence of conformational issues, the photoelectrocyclization was efficient.

Scheme 22.

Terphenyl photoelectrocyclizations

3E. Mechanism – Rearomatization.

There are several methods of reintroducing aromaticity after photoelectrocyclization. These are discussed below.

3E.1. Rearomatization from Oxidation.

One of the more common methods of inducing the vinyl biaryl photoelectrocyclizations to completion has involved the oxidation of the cyclization intermediate. Typically, exposing the reaction to O₂ or I₂ results in oxidation to the phenanthrene (see Schemes 38, 39, and 42 for examples of this).

Scheme 38.

Kelly and co-workers’ synthesis of santiagonamine

Scheme 39.

Kelly’s second-generation synthesis of santiagonamine

Scheme 42.

Porco and co-workers’ completion of kibdelone A

3E.2. Rearomatization from [1,5]-Hydride Shifts.

Under non-oxidative conditions, several experiments have been important in determining that a [1,5]-hydride shift from an electrocyclization intermediate was important. Horgan et al. showed that Ph₂Se₂ did not affect the reaction of 58, suggesting that a radical intermediate was unlikely (Scheme 23).⁵

Scheme 23.

Radicals are not involved in the rearomatization

A more definitive experiment was carried out by op het Veld and Laarhoven. They showed that the irradiation of pentadeuterated 2-(β-phenylvinyl)biphenyl 76 resulted in the complete incorporation of deuterium at C-9 (Scheme 24).²⁶

Scheme 24.

Rearomatization involving [1,5]-deuteride shift

3E.3. Rearomatization from 1,5-Acyl Shifts.

Padwa and co-workers showed that a [1,5]-acyl shift could be used to rearomatize the electrocyclization intermediate (Scheme 25). The cyclization of vinyl biaryl 78 resulted in dihydrophenanthrenes 79 and 80 as a 3:1 mixture, respectively.⁶ It was interesting that the [1,5]-acyl shift was faster than the [1,5]-hydride shift in this system.

Scheme 25.

Photoelectrocyclizations and [1,5]-acyl shift

3E.4. Rearomatization from Protonation.

As is described in Section 4.E, we have shown that the protonation of the electrocyclization intermediate for some compounds can be an effective means of inducing rearomatization.

4. Vinyl Biaryl Photoelectrocyclization Substrate Scope

We have examined the scope of substrates capable of undergoing the vinyl biaryl photoelectrocyclization reaction. Included here is a summary of the alkenes and the arenes that undergo these reactions.

4A. Biphenyl Pyridone Photoelectrocyclizations.

From a general interest in the chemistry of pyridones,^12,27 we decided to examine the photoelectrocyclization of biphenyl pyridone 81 (Scheme 26). We were delighted to find that the exposure of 81 to 350 nm light resulted in the generation of trans-fused 82 in 90% yield.

Scheme 26.

Pyridone photoelectrocyclizations

4A.1. Regioselectivity of Substituted Pyridone Electrocyclizations.

The photoelectrocyclization of non-symmetrical stilbene derivatives have been repeatedly demonstrated to deliver a nearly equal mixture of regioisomeric products.^3,28 In contrast to this, we identified an interesting relationship between the electronic nature of the substituents on the terminal arene and the regioselectivity of the cyclization reaction of dihydropyridones (Scheme 27). When the terminal arene contained an electron-withdrawing substituent at the meta-position, as with 83, the reaction was completely regioselective; favoring outside isomer 84 instead of the corresponding inside isomer (cf. 85, Scheme 28).

Scheme 27.

Regioselective pyridone photoelectrocyclizations

Scheme 28.

Non-regioselective pyridone photoelectrocyclizations

In contrast to the results with electron-withdrawing groups, the presence of an electron-donating group on the terminal arene resulted in a 1:1 mixture of outside and inside regioisomers 84 and 85, respectively (Scheme 28). We proposed that the selectivity was due to the relative rate of the [1,5]-hydride shift. The [1,5]-hydride shift with the substrates having electron-withdrawing substituents was slower and reversible, allowing equilibration between the isomers.

4B. Cyclopentenone and Cyclohexenone Photoelectrocyclizations.

The photochemistry of 81 and 83 encouraged us to examine the scope of alkenes capable of undergoing the photoelectrocyclization. We turned to cycloalkenones and found these reactions to also be successful, giving dihydrophenanthrenes in moderate to high yields (Scheme 29). Among the successful substrates were tetrasubstituted enones 88 and 89, giving quaternary substituted dihydrophenanthrenes 92 and 93 in 65% and 75% yield, respectively.

Scheme 29.

Cyclohexenone and cyclopentenone photocyclizations

4B.1. Regioselectivity in Cyclopentenone Photoelectrocyclizations.

In light of the results obtained with the dihydropyridones, it was interesting that the photoelectrocyclizations of meta-substituted cyclopentenone substrates 94 and 95 gave the corresponding outside isomer 96A exclusively, regardless of the electronegativity of the substituent (Scheme 30). Given that we had previously found that dihydrophenanthrenes lacking quaternary substitution underwent slow oxidation upon exposure to air, we chose to oxidize the initial dihydrophenanthrene products immediately after the reaction. In keeping with our reversibility argument, we propose that the [1,5]-hydride shift is slow in the cyclopentenone electrocyclization intermediates regardless of the electronic nature of the arene substituents.

Scheme 30.

Regioselective cyclopentenone photoelectrocyclizations

4C. Cyclobutenone Photoelectrocyclization Reactions.

In an effort to expand the scope of the reaction even further, we also explored the photochemistry of biphenyl cyclobutenone 54. When 54 was exposed to 350 nm light we only isolated cyclobutenone ring-opened products 98 and 99 (Scheme 31).

Scheme 31.

Cyclobutenones in photoelectrocyclization reactions

Given that the photoelectrocyclization of 54 could lead to what we presumed would be a strained trans-fused fused [4.2.0]cyclooctadiene, it was not surprising that it was not successful. In an effort to determine whether the photocyclization of 54 might be reversible and whether there might be other means of inducing its rearomatization, we proposed to examine the reactivity of dichlorocyclobutenone 100 (Scheme 32). If the electrocyclization of 100 were reversible, aromaticity could be reintroduced through the elimination of HCl. In the event, we isolated 104 in 75% yield after 100 was exposed to 350 nm light. The isolation of 104 was consistent with the notion that these reactions are reversible. Mechanistically, we believe that this result was a consequence of the elimination of HCl from 101 and a subsequent ring opening to give ketene 103. Methanol addition gives 104.

Scheme 32.

Reversible cyclobutenone photoelectrocyclizations

4D. Five-Membered-Ring Heteroarenes in Photoelec- trocyclizations.

Having successfully demonstrated that a number of cycloalkenones underwent the photoelectrocyclization, we next decided to examine the scope of arenes capable of undergoing the reaction. To this end, we explored the photocyclization of bis-aryl cyclohexenones in which one of the arenes was a five-membered-ring heteroaromatic. Thiophenes, imidazoles, pyrazoles, and pyrimidoimidazoles all underwent an efficient photoelectrocyclization reaction to give 105, 106, 107, and 108, respectively (Scheme 33). Among the substrates that did not undergo the reaction was oxazole 44. Interestingly, when the corresponding isomeric oxazole precursor was subjected to 350 nm light, we isolated 109 as the corresponding 3° alcohol in low yield.

Scheme 33.

Five-membered heteroaryl photoelectrocyclization reactions

4D.1. Indoles Undergo Triplet Photoelectrocyclizations.

In contrast to the other heteroaromatic substrates, the photoelectrocyclization of indole 110 resulted in the exclusive generation of cis-fused dihydrophenanthrene analogue 112 (Scheme 34). To determine the mechanism through which 112 was forming, we synthesized and examined the photochemistry of labeled precursor 111 and found that it underwent a [1,5]-deuteride shift to generate 113. This experiment definitively demonstrates that 112 and 113 result from 114 and 115, which come from triplet excited states.

Scheme 34.

Indole-containing biaryl cyclohexenone photoelectrocyclizations proceed via triplet excited states

4E. Rearomatization via Protonation of the Electrocyclization Intermediate.

We reported the stereodivergent photocyclization of cyclohexenone biaryls in which one of the arenes was a pyrazine, a quinoxaline, or a pyridine.^4b With the pyrazine and quinoxaline substrates, we found that the photocyclization of the precursors under neutral conditions proceeded stereoselectively to give the expected trans-fused products (Scheme 35). In contrast, when trifluoroacetic acid (TFA) was included in the reaction mixture, the reaction gave mostly or exclusively the corresponding cis-fused dihydrophenanthrene. To account for this outcome, we hypothesized that the electron-deficient heteroarene slowed down the [1,5]-hydride shift, making protonation of the electrocyclization intermediate competitive.

Scheme 35.

Heteroaromatics in photoelectrocyclization reactions

Interestingly, with pyridine substrates, the position of the nitrogen turned out to be important in the protonation reaction. When the nitrogen was positioned proximal to the phenyl ring the reaction was not affected by a proton source. That is, the reaction gave trans-product 128 regardless of conditions. When the nitrogen was proximal to the cyclohexenone, cis-product 127 was the major product when TFA was added to the reaction, but these reactions were not as selective as the pyrazine or quinoxaline reactions.

Oxazoles proved themselves to be interesting targets in this chemistry. As was mentioned above, 44 did not undergo a photoelectrocyclization while the cyclization of 129 was inefficient, giving 3° alcohol 109 in 24% yield (Scheme 33). We proposed that these reactions were reversible and that the equilibrium favored the cyclization precursor. Consistent with this hypothesis, the addition of TFA to the reaction of 129 led to the generation of cis-fused dihydrophenanthrene 130 in 84% yield (Scheme 36). Neutralized 130 proved to be somewhat unstable, decomposing to alcohol 109 when exposed to SiO₂.^4c

Scheme 36.

Oxazoles in photoelectrocyclizations

4E.1. Protonation of Cyclobutenones.

We have examined whether the trapping of the electrocyclization intermediate with a proton could be applied to cyclobutenones. As was outlined above, we had previously shown that the photoelectrocyclization of cyclobutenone 54 only gave products from cyclobutenone ring-opening, namely 98 and 99 (Scheme 31) but that the related dichlorocyclobutenone 100 gave phenanthrene 104 from cyclization and elimination of HCl (Scheme 32). We proposed that we might be able to generate cis-fused bicyclo[4.2.0]octadienes by including TFA in the cyclobutenone reactions. In the event, whereas TFA by itself was not enough, a mixture of TFA and TFAA gave cis-fused cyclobutanol 133 in 37% yield after reduction with NaBH₄ (Scheme 37).^4b Interestingly, when the reaction was run in the presence of ethanol we only isolated ethyl esters 135. These reactions are consistent with the idea that the electrocyclic ring opening of the cyclobutenone and the bis-aryl cyclobutenone electrocyclization reaction are reversible.

Scheme 37.

Cyclobutenone photoelectrocyclization reactions

5. The Application of Photoelectrocyclizations in Natural Product Synthesis

The majority of the applications of the photocyclization of conjugated vinylbiaryl in total synthesis has involved the construction of either phenanthrenes or [9,10]-dihydrophenanthrenes.

5.1. Kelly’s Synthesis of Santiagonamine.

For instance, in the total synthesis of santiagonamine by Kelly and co-workers (Scheme 38),⁸ the key step of their synthesis was the photocyclization reaction vinyl biaryl 13. However, upon irradiation in the presence of iodine as an oxidant and propylene oxide to neutralize the HI generated in the reaction,²⁹ 13 was converted into the undesired phenanthrene 136. They reasoned that 13 mostly existed as conformer 138, which would lead to 136 after irradiation and electrocyclization.

To solve this problem, Kelly and co-workers converted 13 into 139 having a lactone linker that locked the structure into the desired conformation (Scheme 39). As a result, the desired product 140 was generated efficiently. To reach santiagonamine, the dimethylethylamino side chain was incorporated in three steps from 140 via an allyl Stille coupling reaction to give 141, a Johnson–Lemieux oxidative cleavage of the alkene to give aldehyde 142, and a reductive amination reaction.

5.2. Kende and Curran’s Juncusol Synthesis.

In Kende and Curran’s total synthesis of Juncusol,⁷ the photocyclization of bisaryl alkene 143 resulted to the formation of 9,10- dihydrophenanthrene 144 in 60–65% yield. Parikh–Doering oxidation of 144 and Wittig olefination of the resulting aldehyde provided Juncusol after removal of the methyl ethers (Scheme 40).

Scheme 40.

Juncusol completion

5.3. Porco’s Synthesis of the Kibdelones.

Following their failed attempts at carrying out Diels–Alder cycloadditions to produce the A-D ring of the kibdelone family of natural products, Porco and co-workers employed a photoelectrocyclization of 18 to generate the core structure as 146 in 61% yield (Scheme 41).^10,30 They also isolated a small amount of the desired oxidized cyclized product 147. The addition of oxygen to the reaction failed to give more of 147.

Scheme 41.

Porco and co-workers synthesis of kibdelones

In contrast to the inability to affect the oxidation of the photoelectrocyclization intermediate using oxygen, Porco and co-workers found that iodine was an effective oxidant to generate the requisite phenanthrene in 70% yield (Scheme 42). Removal of the TBDPS group provided bisphenol 147. oxa-Michael coupling with iodocyclohexenoate 148 enabled them to generate 149. Hydrolysis of the ester and a cyanuric chloride mediated Friedel–Crafts cyclization gave the kibdelone A skeleton as 150. Ketal hydrolysis and oxidation gave kibdelone A. They observed an interesting pH effect on the final oxidation. When the oxidation was carried out under acidic conditions both the B- and D-rings were oxidized but only the B-ring underwent oxidation under the conditions shown.

6. Conclusions

This review has illustrated that the photoelectrocyclizations of vinyl biaryl substrates to give phenanthrenes and dihydrophenanthrenes can be an efficient and interesting process (Scheme 43). These cyclizations proceed mostly through singlet excited states but there are multiple examples of triplet excited state reactions. Many of these reactions have been shown to be reversible, obeying Curtin– Hammett kinetics with products depending upon the reaction conditions. Many issues remain unresolved, including asymmetric variations of these transformations, getting a better grasp of the excited states, and intercepting the intermediates with electrophiles other than protons.

Scheme 43.

Possible vinyl bis-aryl photoelectrocyclization pathways

Biographies

Xuchen Zhao was born in Zhaoyuan, P. R. of China. He obtained his B.S. degree in applied chemistry from Beijing University of Chemical Technology in 2011. After working for a short period of time in industry, he moved to the University of Utah where he received his Ph.D. in chemistry in 2020. His doctoral research with Jon Rainier focused on the 6π-photoelectrocyclization of vinyl-bis-aryls. He is currently working in the research group of Professor Amos B. Smith, III as a postdoctoral fellow on the synthesis of potential HIV inhibitors.

Jon Rainier was born in Anaheim, California (USA). He received his B.S. in chemistry from the University of California, Irvine. After a short stint in the aerospace industry as a quality assurance technician, he received an M.S. in chemistry from California State University, Long Beach where he worked under the guidance of Kensaku Nakayama. He subsequently received his Ph.D. from the University of California, Riverside where he studied quinone methides under the direction of Steven R. Angle. He then moved to the University of Pennsylvania as an NIH postdoctoral fellow under the mentorship of Amos B. Smith, III where he worked on the synthesis of penitrem D. He began his independent career at the University of Arizona and moved to the University of Utah in 2002. His research interests are focused on organic synthesis and the synthesis of bioactive small molecules.