Abstract
The singlet oxygenation of three polycyclic hydrocarbons, triquinacene, barrelene and homobarrelene was studied. Triquinacene reacted by way of a perepoxide intermediate, transferring an oxygen atom to another triquinacene molecule to give exclusively the mono epoxide. Barrelene, on the other hand, underwent a rare homo-Diels-Alder reaction with 1O2 to give the decomposition product from the initial tetracyclic 1,2-dioxolane leading to benzofuran. The latter reacted with 1O2 in a [2+2] cycloaddition to give an unstable 1,2-dioxetane which collapsed to 2-formylphenyl formate. The latter was independently synthesize via singlet oxygenation of authentic benzofuran. Homobarrelene reacted in a similar fashion to give a homoDiels product, decomposition of which led to a keto aldehyde which was characterized spectroscopically. Computational work confirms the barrelene and homobarrelene reactions with 1O2 as concerted [π2s+π2s+π2s] cycloadditions.
Keywords: singlet oxygen, barrelene, homobarrelene, triquinacene
Graphical Abstract
1. Introduction
Singlet oxygen (1O2) reacts with unsaturated systems mainly by way of three modes: a) with strained and/or electron-rich alkenes, [2+2] cycloadditions are observed leading to unstable 1,2-dioxetanes; b) alkenes and cycloalkenes with allylic protons react via perepoxide intermediates to give allylic hydroperoxides,1 a pathway often referred to as the “ene” reaction; c) with conjugated dienes, mainly [4+2] cycloadditions (i.e., Diels-Alder reactions) are observed resulting in the formation of relatively stable 1,4-endoperoxides. A number of excellent review articles extensively describe the varied reactivities of singlet oxygen toward a plethora of substrates.2 In many cases, alkenes or cycloalkenes that are not particularly electron-rich and lack allylic hydrogens (i.e., those not capable of “ene” reactions) remain unreactive toward 1O2. In addition to the electron-density arguments, the strain factor may also play an important role in the reactivity of a hydrocarbon toward 1O2. In general, cyclic or bicyclic alkenes with greater inherent strain exhibit greater reactivity. Whereas alkynes are inert toward 1O2, allenes and cyclic allenes react readily by polar mechanisms.3 Singlet oxygenations of two systems with unique electronic properties, bicyclopropylidene (1)4 and heptafulvalene (4)5 have been reported. Bicyclopropylidene (1) gives rise to two products, 2 and 3, but the 1,2-dioxetane from a possible 2+2 cycloaddition was not observed. Obviously, the mechanism involved a polar, stepwise process which allowed a cyclopropylcarbinyl-cyclobutyl rearrangement. The formation of the epoxide was ascribed to a perepoxide intermediate which presumably acts as an oxygen-atom donor to another bicyclopropylidene molecule. On the other hand, heptafulvalene, in spite of the presence of multiple π-systems in the molecule undergoes a 2+2 cycloaddition with 1O2 at the most electron-rich central double bond6 to give a 1,2-dioxetane which collapses to two tropone (6) molecules, one of which is formed in its electronically excited state (6*) and partially undergoes rearrangement to benzaldehyde (7) or loses a C=O to benzene (8) by way of a cyclopropanone intermediate. These two examples indicate that reactivity of olefins toward 1O2 highly depends on electron densities at the reactive centers and may react in a concerted fashion or via polar intermediates.
In this study, we report on our findings from tetraphenylporphyrin (TTP) sensitized photooxidations of two prominent members of the C8H8 and C10H10 family of hydrocarbons, respectively, namely, triquinacene (9)7 and barrelene (10),8 as well as homobarrelene (11).9
The latter was included in this study owing to the π-like properties of the annelated cyclopropane ring.
Considering the very low reactivity of bridged bicyclic alkenes and cycloalkenes lacking allylic hydrogens toward singlet oxygen, it was of interest whether the unique array of three π-systems in 1 and 2 would have an effect on the reactivity as well reaction pathway(s) in sensitized photooxygenation reactions.
2. Results and Discussion
The first system we studied was triquinacene (9). The photooxidation was affected with singlet oxygen, generated photochemically by means of a 250 W sodium lamp, using tetraphenylporphyrin (TPP) as sensitizer in a CDCl3 solution at room temperature. The progress of the photooxidation was monitored by 1H NMR. After completion of the reaction, the product was isolated by preparative TLC and identified as 12, the monoepoxide of triquinacene. The epoxide formation is consistent with the previous results on alkenes not capable of ‘ene’ reactions. However, in the case of norbornene which reacts exceptionally slowly with 1O2, the result was considerably different from that of 9 in that norbornene gave primarily the dialdehyde stemming from the 1,2-dioxetane cleavage, and traces of the corresponding epoxide.10 We believe that 9 reacts with 1O2 in a step-wise manner, and the epoxide is formed from a perepoxide intermediate via loss of an oxygen atom, possibly to another triquinacene molecule.11 We could not detect any other products stemming from a [2+2] cycloaddition, i.e., a 1,2-dioxetane or a dialdehyde.
The electronic structure of barrelene (10) with three π-systems in a bicyclic framework has been subject to several studies. Computational work and photoelectron spectroscopy indicate that barrelene is a strained molecule. This conclusion is supported by the fact that the enthalphy of hydrogenation for the first double bond in barrelene is ΔHh = −37.6 kcal/mol, which seems to be the largest ΔHh value ever observed for the hydrogenation of a single bond.12 Whereas the dihydro and tetrahydro derivatives of barrelene (bicyclo[2.2.2]octa-2,5-diene and bicyclo[2.2.2]oct-2-ene, respectively) proved to be inert toward 1O2, compound 10 reacted at 0 °C with singlet oxygen rather rapidly. After the formation of the first photooxidation product, which proved to be unstable and whose structure could be assigned in the crude 1H NMR spectrum. The structure of the postulated intermediate, 13, points to an initial homo-Diels-Alder cycloaddition reaction between barrelene and 1O2. The 1,2-dioxolane moiety in 13 undergoes an O-O fission to give the oxygen-centered diradical 14.13
The allylic radical undergoes a 1,2-hydrogen shift to give the ketoaldehyde 15.14
A formal homo-Diels-Alder reaction between barrelene and 1O2 would be a thermally allowed [π2s+π2s+π2s] process by the Woodward-Hoffmann rules,15 and barrelene and a close relative, Nenitzescu’s hydrocarbon (tricyclo[4.2.2.02,5]deca3,7,9-triene) are known to undergo homo-Diels-Alder reactions,16,17, respectively.
The pathway to 13 was explored computationally. Two paths were identified. Along with a concerted homo-Diels-Alder path, a stepwise path going through an intermediate perepoxide. At the M06/6–311+G** level,18,19 the two paths have very similar barriers (10.8 kcal/mol for the concerted path and 11.2 kcal/mol for the stepwise). However, at the MP2/6–311+G** level, the concerted path is favored by a considerable margin and has a barrier of only 5.8 kcal/mol (the perepoxide intermediate is 22 kcal/mol endothermic at this computational level).
Compound 15 proved to be quite labile thermally and during the course of the photooxygenation decomposed to a new product which upon further photooxygenation gave dialdehyde 19 as the sole product. Its formation is rationalized in terms of acid-catalyzed rearrangement of 15 to 2-(2-hydroxyphenyl)acetaldehyde (16) which spontaneously (and presumably also under acid catalysis) gives 17. Benzofuran (17) then undergoes a [2+2] cycloaddition with 1O2 to give the corresponding 1,2-dioxetane intermediate 18 which is thermally labile and collapses to the observed product 19.
Since the photooxygenation was conducted in CDCl3, and the latter contains traces of HCl, it is quite reasonable to postulate the benzofuran formation from 16. To test the hypothesis whether the final product, 2-formylphenyl formate (19) is formed from 17, an authentic sample of benzofuran (17) was subjected to singlet oxygenation20 under the same conditions as barrelene, Indeed, 17 reacted with 1O2 in a slow reaction to give 19 exclusively. It was hydrolytically sensitive and partially hydrolyzed to salicylaldehyde on the column. A pure sample of 19 was fully characterized by NMR and HRMS.
The results from the singlet oxygenation of barrelene are remarkable since during the photooxygenation a series of reactions are taking place in a cascade fashion, namely, a homo-Diels-Alder reaction, thermal isomerization of the strained peroxide intermediate, subsequent rearrangement of the labile product, intramolecular cyclization-dehydration, followed by 1O2 addition to the benzofuran intermediate, finally a 1,2-dioxetane cleavage to give the final product 19, the net transformation being C8H8 (9) + 2 1O2→C8H6O3 (19) + H2O!
The third hydrocarbon, homobarrelene (11) chosen for this study is a close relative of barrelene electronically, owing to the π-like properties of the cyclopropane Walsh orbitals, and indeed this consideration was born out by photoelectron spectroscopy as well theoretical work by Gleiter and de Meijere.21 Moreover, the strain in the three-membered-ring should also render this hydrocarbon a much more reactive target toward 1O2 than the unreactive dihydro and tetrahydro analogs of 9. Singlet oxygenation of homobarrelene (11) was significantly slower than both barrelene and triquinacene. The sole product proved to be somewhat unstable and the photooxidation was interrupted after 50% conversion since prolonged photooxidation was accompanied with partial decomposition of the product; chromatography of the mixture was also detrimental to the isolation of a pure product. However, the mixture spectrum was clean and devoid of any other impurities or byproducts to permit full spectroscopic characterization. Based on all NMR data (1H and 13C), the structure of 22 was assigned to the sole product.
Thus, annelation of a cyclopropane ring onto the barrelene framework rendered the bicyclic diene considerably less reactive toward 1O2, however, the mode of reaction (homo-Diels-Alder) and the decomposition pattern of the 1,2-dioxolane intermediate 20 was identical to that shown in Scheme 2 for barrelene. It is interesting to note that of the two C-C bonds to the bridgehead carbon in the five-membered peroxide unit in 20, only the one distal to the cyclopropane was broken, leading to the trans (in reference to the relative stereochemistry of the two cyclopropane rings) product 22. Modeling at the M06/6311+G** and MP2/6–311+G** levels predict a smaller barrier for breaking the distal bond (by 2.5 and 3.3 kcal/mol, respectively). The greater strain in the distal bond is evidenced by the longer C-C bond in 20, 1.540Å vs. 1.532Å at the M06/6–311+G** level.
Scheme 2.
Singlet oxygenation of triquinacene (9)
If indeed 1O2 reacts with 11 in a homo-Diels-Alder reaction, it would be worthwhile to test whether other dienophiles would also react with 11 in this mode. To the best of our knowledge, no cycloadditions of 11 have previously been reported. We chose two reactive dienophiles, 4-methyl-1,2,4-triazolin-3,5-dione (23, MTAD) and tetracyanoethylene (25, TCNE). Both dienophiles reacted with 11 at room temperature in CH2Cl2 solutions (TCNE proved to be more reactive than MTAD) to give the homo-Diels-Alder cycloadducts 24, and 26, respectively. They were isolated by flash chromatography and fully characterized by HRMS and NMR spectroscopy.
3. Conclusions
We have studied the singlet oxygenations of three popular hydrocarbons, namely, triquinacene (9), barrelene (10) and homobarrelene (11). These three compounds were chosen owing to their unique array of π-electrons in the bridged polycyclic frameworks, sometimes associated with homoconjugative interactions of the π-systems. Also, due to lack of allylic hydrogens (abstraction of the bridgehead hydrogens was not feasible due to possible formation of bridgehead alkenes), these hydrocarbons are not capable of a common pathway encountered in alkene singlet oxygenations, namely the ‘ene’ raction. Moreover, the lack of electron-donating groups (e.g., methoxy) on the double bonds in 9-11 might not only render them unreactive toward 1O2, also the 2+2 cycloaddition pathway leading to 1,2-dioxetanes would be absent. In these cases studied, the reactivity would be dictated by π-electron density as well as strain factors. Triquinacene (9) proved to be quite reactive toward 1O2 and reacted presumably via a perepoxide intermediate which -by contrast to norbornene- gave the corresponding epoxide rather than the 1,2-dioxetane. Barrelene was also quite reactive under the same conditions, however, an unusual homo-Diels-Alder reaction occurred in this case leading to a tricyclic peroxide which was unstable at room temperature and underwent homolytic 1,2-dioxolane cleavage. The resulting ketoaldehyde could be observed by 1H NMR during the reaction, however, it gradually rearranged to 2-(2-hydroxyphenyl)acetaldehyde (16) which spontaneously cyclized to benzofuran under loss of water. Benzofuran then underwent 2+2 cycloaddition with 1O2 to give the corresponding 1,2-dioxetane which gave the final product, 2-formylphenyl formate (19) via a 1,2-dioxetane cleavage. Compound 19 was independently prepared by singlet oxygenation of a commercial sample of benzofuran under the same conditions supporting the mechanism in Scheme 3 postulating benzofuran as precursor of the final product.
Scheme 3.
Further photooxidative transformations of barrelene by way of benzofuran
Homobarrelene did not prove to be as reactive as triquinacene or barrelene. It reacted quite slowly with 1O2 and gave one single product which stemmed from the homo-Diels-Alder product. The peroxide intermediate underwent selective C-C cleavage to give 22 exclusively. This product was characterized by NMR spectroscopy, however, it also proved unstable on silica gel.
The homo-Diels-Alder reactions of barrelene and some of its analogs have been reported. The fact that barrelene and homobarrelene react with 1O2 by a homo-Diels-Alder reaction is unprecedented in reactions of singlet oxygen with dienes. In one other case a homo Diels-Alder reaction was reported, though the substrate was 2,3,5,6-tetramethoxybenzobarrelene which gave a small amount of the homo-Diels-Alder product alongside mainly the expected 1,2-dioxetanes.22 The authors ascribe the homo Diels-Alder product to a stepwise mechanism via a perepoxide intermediate, whereas our computational studies clearly support the concerted, symmetry-allowed [π2s+π2s+π2s] cycloaddition. The lack of reactivity of bicyclo[2.2.2]octa-2,5-diene toward 1O2 is simply a consequence of greater flexibility and significantly less strain in the latter system. Finally, homobarrelene (11) was shown to undergo homo-Diels-Alder reactions with MTAD and TCNE as well, to give stable analogs of the peroxide 20 postulated in Scheme 4.
Scheme 4.
Singlet oxygenation of homobarrelene (11)
In summary, this work nicely demonstrates the sensitivity of singlet oxygen to electronic and strain factors in its reactions with bridged polycyclic olefins. Whereas the three isolated double bonds in triquinacene do not permit a possible homoDiels-Alder reaction (in this case a 2+2+2+2 cycloaddition which would have been disallowed from the ground state by the Woodward-Hoffmann orbital symmetry rules15), barrelene and homobarrelene appear to be ideally suited to observe a homoDiels-Alder reaction with 1O2. We are currently investigating the reactivities of other interesting bridged bicyclic compounds toward singlet oxygen, capable of exhibiting diverse reactivities, in hopes of gaining a better understanding of the factors that affect the various modes of reaction of 1O2.
Supplementary Material
Figure 1.
Hydrocarbons whose singlet oxygenations were studied in this work
Scheme 1.
Singlet oxygenations of bicyclopropylidene (1) and heptafulvalene (4)
Scheme 2.
Singlet oxygenation of 10: first intermediate (15)
Scheme 5.
Homo-Diels-Alder reactions of 11 with MTAD (23) and TCNE (25)
HIGHLIGHTS.
Triquinacene, Barrelene and Homobarrelene
Singlet Oxygen
Homo-Diels-Alder reactions of barrelene and homobarrelene
Oxygen atom transfer from a perepoxide intermediate
Acknowledgements
We thank Professor Armin de Meijere, University of Göttingen, for a gift of barrelene and triquinacene. Dr. Ozer acknowledges a post-doctoral fellowship from the Turkish Scientific and Technological Council (TUBITAK). I. Erden gratefully acknowledges support of this work by funds from the National Institutes of Health (Grant No. 1SC1GM082340), S. Gronert acknowledges support from the National Science Foundation (CHE-156582). The mass spectrometry work at SFSU was in part supported by a grant from the National Science Foundation (CHE-1228656). The SFSU authors also acknowledge the support from National Science Foundation for the purchase of the 500 MHz HMR instrument through grants DBI 0521342 and DBI 1625721.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
This paper is dedicated to the memory of Professor Wolfgang Lüttke, a wonderful teacher, researcher and human being
Declaration of interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References and Notes
- 1.Though the perepoxide intermediate may be advanced as the precursor of an epoxide product or a 1,2-dioxetane, Houk, Foote et al. propose a ‘no intermediate, two-stage’ mechanism for the ‘ene’ reaction of singlet oxygen with alkenes carrying allylic hydrogens based on theoretical calculations:; Singleton DA, Hang C, Szymanski M, Meyer MP, Leach AG, Kuwate KT, Chen JS, Greer A, Foote CS, Houk KN, J. Am. Chem. Soc 2003, 125, 1319–1328. [DOI] [PubMed] [Google Scholar]
- 2.a) Gollnick K, Schenk GO, Oxygen as Dienophile in 1,4-Cycloaddition Reactions, Hamer J, Ed.; Academic Press: New York, 1967, Vol. 8, pp 255–344; [Google Scholar]; b) Singlet Oxygen, Wasserman HH, Murray RW, Eds.; Academic Press: New York, Vol. 40; [Google Scholar]; c) Frimer AA, Chem. Rev 1979, 79, 359–387; [Google Scholar]; d) Adam W, Heil M, Mosandl T, Saha-Möller CR, Dioxetanes and α-Peroxy Lactones, Four-membered Ring Cyclic Peroxides in Organic Peroxides, Ando W, Ed., John Wiley and Sons: Chichester, 1992, Chapter 45, pp 221–254; [Google Scholar]; e) Clennan E, Pace A, Tetrahedron, 2005, 61, 6665–6691; [Google Scholar]; f) Frimer AAin The Chemistry of Peroxides, Patai S, Ed.; John Wiley and Sons: New York, 1983, pp 201–234. [Google Scholar]
- 3.a) Erden I, Martinez TR, Tetrahedron Lett. 1991, 32, 1859–1962; [Google Scholar]; b) Erden I, Song J, Cao W, Org. Lett 2000, 2, 1983–1985. [DOI] [PubMed] [Google Scholar]
- 4.a) Erden I, de Meijere A, Rousseau G, Conia J-M, Tetrahederon Lett. 1980, 21, 2501–2504; [Google Scholar]; b) de Meijere A, Erden I, Weber W, Kaufmann D, J. Org. Chem 1988, 53, 152–161. [Google Scholar]
- 5.Erden I, Kaufmann D, Tetrahedron Lett. 1981, 22, 215–218. [Google Scholar]
- 6.Fleming I, J. Chem. Soc., Perkin Trans 1973, 1, 1019–1021. [Google Scholar]
- 7.Woodward RB, Fukunaga T, Kelly RC, J. Am. Chem. Soc 1964, 86, 3162–3164. [Google Scholar]
- 8.a) Zimmerman HE, Grunewald GL, Paufler RM, Sherwin MA, J. Am. Chem. Soc 1969, 91, 2330–2338; [Google Scholar]; b) de Meijere A, Weitemeier C, Angew. Chem. Int. Ed 1976, 15, 686–687. [Google Scholar]
- 9.a) Daub J, Schleyer P. v. R., Angew. Chem. Int. Ed 1967, 7, 468–469; [Google Scholar]; b) de Meijere A, Weitemeier C, Schallner O, Chem. Ber 1977, 110, 1504–1522. [Google Scholar]
- 10.Jefford CW, Chem. Soc. Rev 1993, 22, 59–66. [Google Scholar]
- 11.A similar oxygen atom transfer in the singlet oxygenation of trans-cyclooctene with ‘hard to reach’ allylic hydrogens thus incapable of an ‘ene’ reaction, as in the present case, was reported:; Leach AG, Houk KN, Foote CS, J. Org. Chem 2008, 73, 8511–8519. [DOI] [PubMed] [Google Scholar]
- 12.Turner RB, J. Am. Chem. Soc 1964, 86, 3586–3587. [Google Scholar]
- 13.Whether the 1,2-dioxolane moiety in 13 undergoes O-O bond homolysis to give diradical 14 is a triplet sensitized process or is dominated by a thermolysis process cannot be answered at this time. Toward that end, peroxide 13 would have to be isolated and subjected to decomposition in the absence of a triplet sensitizer. See the following references:; a) Scaiano JC, Wubbels GN, J. Am. Chem. Soc 1981, 103, 640–645; [Google Scholar]; b) Belh SJ, Walalawela N, Lekhtman S, Greer A, ACS Omega, 2019, 4, 22623–22631, [DOI] [PMC free article] [PubMed] [Google Scholar]; c) Zaklika KA, Thayer AL, Schaap AP, J. Am. Chem. Soc 1978, 100, 4916–4918. [Google Scholar]
- 14.The thermal isomerization mode of 13 to give the keto aldehyde 15 closely resembles that of 2,3-dioxabicyclo[2.2.1]heptane:; Coughlin DJ, Salomon RG, J. Am. Chem. Soc 1979, 101, 2761–2763. [Google Scholar]
- 15.Woodward RB, Hoffmann R, Angew. Chem. Int. Ed 1969, 8, 781–932. [Google Scholar]
- 16.Zimmerman HE, Grunewald GL, J. Am. Chem. Soc 1964, 86, 1434–1436. [Google Scholar]
- 17.Erden I, de Meijere A, Tetrahedron Lett. 1980, 21, 1837–1840. [Google Scholar]
- 18.Computational modeling was completed using the Gaussian1615 suite of quantum mechanical programs. All calculations were performed at the M06/6–311+G** or MP2/6–311+G** level.
- 19.Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, Scalmani G, Barone V, Petersson GA, Nakatsuji H, Li X, Caricato M, Marenich AV, Bloino J, Janesko BG, Gomperts R, Mennucci B, Hratchian HP, Ortiz JV, Izmaylov AF, Sonnenberg JL, Williams-Young D, Ding F, Lipparini F, Egidi F, Goings J, Peng B, Petrone A, Henderson T, Ranasinghe D, Zakrzewski VG, Gao J, Rega N, Zheng G, Liang W, Hada M, Ehara M, Toyota K, Fukuda R, Hasegawa J, Ishida M, Nakajima T, Honda Y, Kitao O, Nakai H, Vreven T, Throssell K, Montgomery JA Jr., Peralta JE, Ogliaro F, Bearpark MJ, Heyd JJ, Brothers EN, Kudin KN, Staroverov VN, Keith TA, Kobayashi R, Normand J, Raghavachari K, Rendell AP, Burant JC, Iyengar SS, Tomasi J, Cossi M, Millam JM, Klene M, Adamo C, Cammi R, Ochterski JW, Martin RL, Morokuma K, Farkas O, Foresman JB, and Fox DJ, Gaussian 16, Revision A.03, Gaussian, Inc, Wallingford CT,2016. [Google Scholar]
- 20.Although the singlet oxygenation of benzofuran was mentioned in an article, details and product characterization were not published:; Berdahl Dr. R., Wasserman HH, Isr. J. Chem 1983, 23, 409–414. [Google Scholar]
- 21.Gleiter R, Böhm MC, de Meijere A, Preuss T, J. Org. Chem 1983, 48, 796–802. [Google Scholar]
- 22.Adam W, Balci M, Cakmak O, Peters K, Saha-Moller CR, Schulz M, Tetrahedron, 1994, 50, 9009–9024. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.