This work has been digitalized and published in 2013 by Verlag Zeitschrift für Naturforschung in cooperation with the Max Planck Society for the Advancement of Science under a Creative Commons Attribution4.0 International License.

Dieses Werk wurde im Jahr 2013 vom Verlag Zeitschrift für Naturforschungin Zusammenarbeit mit der Max-Planck-Gesellschaft zur Förderung derWissenschaften e.V. digitalisiert und unter folgender Lizenz veröffentlicht:Creative Commons Namensnennung 4.0 Lizenz.

Photolysis (A 185 nm) of Liquid Tetrahydropyran* Heinz-Peter Schuchmann, Peter Naderwitz, and Clemens von Sonntag Institut für Strahlenchemie im Max-Planck-Institut für Kohlenforschung, Stiftstraße 34-36, D-4330 Mülheim a. d. Ruhr Z. Naturforsch. 33 b, 942-945 (1978); received April 28, 1978 Photolysis, Tetrahydropyran, Quantum Yields, Biradical Intermediates, Photodimer

The main products of the 185 nm photolysis of neat liquid tetrahydropyran are pent-4-en-l-ol (7) (0 = 0.40), 2-(5'-hydroxypentyl)tetrahydropyran (10) (0.21), valeraldehyde (6) (0.13), and pentan-l-ol (8) (0.08). These products are thought to be formed via the bi-radical '0(0112)40112, and/or through intramolecular (6, 7, 8) and intermolecular (10) photoreactions. In the case of 10, the probability of a specific intermolecular photoreaction involving two tetrahydropyran molecules is suggested by the fact that in the photolysis of solutions of tetrahydropyran in cyclohexane and formaldehyde dimethyl acetal products analogous to 10 made up of a tetrahydropyran and a solvent molecule moiety, are not formed.

Introduction The photolysis of saturated cyclic ethers seems to

be more complex than that of open-chain ethers [1]. In the latter the most important primary process at X 185 is the homolytic scission of a C - 0 bond. In the former this reaction would produce a biradical. In the smaller rings (n < 5) the biradical may break up yielding fragments or undergo reclosure, as observed in the photolysis of tetrahydrofurans [2], The results of the present study indicate that inter-molecular processes [3] can become equally, or even more, important, and that great caution must be exercised in the attempt to extend the mechanistic conclusions reached from the photolysis of one compound to a similar molecule, even if structurally closely related.

Results and Discussion Tetrahydropyran is transparent in the near UV

but strongly absorbs the A 185 line [4] of the low pressure Hg arc used in these experiments. X 254 is not photoactive in this system. Products and their quantum yields are listed in the Table. There is no pronounced temperature effect on the quantum yields which indicates that chain reactions [5] do not play a part in this photolysis.

* Part I X of the series: Radiation Chemistry of Ethers. For Part VIII see R. Bausch, H.-P. Schuch-mann, C. von Sonntag, R. Benn, and H. Dreeskamp, J. Chem. Soc. Chem. Commun. 1976, 418-419.

Requests for reprints should be sent to Prof. Dr. C. von Sonntag, Institut für Strahlenchemie im Max-Planck-Institut für Kohlenforschung, Stiftstraße 34-36, D-4330 Mülheim/Ruhr.

Table. 185 nm photolysis of liquid deaerated tetra-hydropyran. Product quantum yields at 20 and 70 °C.

20 °C 70 °C 1 Hydrogen 0.04 2 Ethylene 0.01 3 Cyclobutane 0.002 4 Formaldehyde <0.001 5 2, 3-Dihydropyrana 0.04 0.04 6 Valeraldehyde 0.13 0.13 7 Pent-4-en-l-ol 0.40 0.35 8 Pentanol 0.08 0.08 9 a, a'-Bis(tetrahydropyranyl)b meso 0.015 0.013

d, 1 0.015 0.013 10 2-(5'-Hydroxypentyl)tetra- 0.21 0.25

hydropyranc

11 l,10-Decanediola absent ( < 0.001)

a Reference material from Aldrich, b reference material obtained from the Hg-sensitized

photolysis of tetrahydropyran, c identification on the basis of spectroscopic data

(see text).

Reference compounds were available to allow assignment of all products but product 10. The identity of the latter as 2-(5'-hydroxypentyl)-tetra-hydropyran was established on the basis of spectro-scopic data (see Experimental).

The possible primary processes and some of the subsequent reactions are shown in the Scheme. We believe that C - 0 bond homolytic cleavage is an important primary process in the present photo-lysis, just as in the case of other ethers [1] (reaction (2)). This gives rise to the biradical whose self-disproportionation leads to 4-penten-l-ol (7) (reaction (10)) or valeraldehyde (6) (reaction (11). (6 and 7 might, however, be formed to a considerable extent in the true molecular processes 3 and 4 which cannot be separated from the other routes: reactions

H.-P. Schuchmann et al. • Photolysis of Tetrahydropyran 943

0 a

o

ICH2)5OH

O CH, 8 Q* + H O

v

H - O

e- * o

a • o

CH? — CHO CH2O + | I

5 CH2 CH2

15 2 CH2 — CH2

CHJ—CHO

I F * 1 1 1 6 C H 2 - C H 2

o H

o Scheme. Primary processes in the 185 nm photolysis of liquid tetrahydropyran.

(10), (11), and (13)). Moreover, the biradical with its oxy l end could abstract a hydrogen atom from the solvent cage: reaction (8). This would lead to a caged radical pair as shown in the Scheme. The radical HO-(CH2)5 can form %-pentanol (8) through disproportionation (reaction (14)) of this pair in the cage, dihydropyran (5) being the other product. Some also will come free and abstract a hydrogen atom from the substrate in the bulk of the solution.

Product 10 could in principle be formed in two ways (reactions (1) and (9)). I f the disproportiona-tion/combination ratio of the caged radical pair were known, an upper limit could be estimated from 99 (5) for the contribution of reaction (9) to 99 (10). Failing this, we cannot at present definitely settle the question whether reaction (1) or leaction (9) is the major source of product 10 (see below).

Process (5) is of little importance as demonstrated b y the low quantum yields of formaldehyde, ethylene, and cyclobutane, probably via a tetra-methylene intermediate. The fragmentation of the tetramethylene species into two molecules of ethylene is a familiar process (c/. [6]).

Hydrogen atoms (from reaction (6)) readily abstract from the substrate, and give molecular hydrogen and substrate radicals (preferentially those in a-position to the ring oxygen). The alter-native route to molecular hydrogen (reaction (7)) may also take place. Their sum 9? (H2) equals 0.04, showing that processes (6) and (7) are not impor-tant.

The sum of the quantum yields of all primary processes approaches unity. This indicates that little reclosure of the biradical, in so far as it is being formed, occurs. In some tetrahydrofurans [2] this process plays a major role whereas a reaction of the biradical with the substrate and the formation of the "cage product " equivalent to 10 is - although observed - comparatively small, with a quantum yield of the order of IO - 2 .

The formation of product 10 is probably the most interesting feature of this photolysis. In the photo-lysis of the neat tetrahydropyran, 10 may be formulated as the combination product of a caged radical pair. Evidence for the latter assertion comes from the absence (99 < IO - 3 ) o f 1,10-decanediol (11), the expected combination product of two HO(CH2)s radicals. I f all combination products were formed without the cage, then on statistical grounds using 9? (9) = 0.03 and 99 (10) = 0.21 which are proportional to the encounter probabilities, one would calculate 0.36 for the decanediol quantum yield which is far off the experimental value of less than 0.001. On the basis of the cage hypothesis one might expect 99 (pentanol) to increase with temperature. Surpris-ingly, no measurable temperature effect on the quantum yields was found. However, if an out-of-cage contribution of the HO(CH2)5 radical to the pentanol yield is only small (99 »0 .05 ) the expected change in 99 (pentanol) might be within the experi-mental error.

Some doubt is thrown on the hypothesis that much 10 is formed via reaction (9), by the results of the photolysis of tetrahydropyran in solution where there is no indication that the biradical leads to a radical pair with the participation of a cage molecule other than tetrahydropyran. Products such as (5 ' -hydroxy-pentyl)cyclohexane are absent when the solvent is cyclohexane. Thus it would follow that the formation of 10, at least in solutions of tetra-hydropyran, is probably the consequence of a very specific photochemical reaction between two tetra-hydropyran molecules. Examples of other inter-

944 H.-P. Schuchmann et al. • Photolysis of Tetrahydropyran 944

molecular photochemical processes which occur at }. 185 have been described [3].

The behaviour of 99 (10) and <p (7) as a function of tetrahydropyran concentration in its solutions is complex. In cyclohexane, 99 (10) goes through a broad maximum centered at around v : v 1:1, then tends steeply toward zero as the tetrahydropyran concen-tration falls further (9910 (Tetrahydropyran in cyclo-hexane, vol .%) : 0.21 (100), 0.25 (80), 0.29 (60), 0.29 (50), 0.29 (40), 0.29 (25), 0.19 (10), 0.06 (5), 0.02 (1)). In formaldehyde dimethyl acetal, 99 (10) also tends to zero with falling tetrahydropyran concentrations (the absorption of the solvent at 185 nm [7] being taken into account), and there is no maximum (9910 (Tetrahydropyran in formalde-hyde dimethyl acetal), vo l .%) : 0.21 (100), 0.14 (80), 0.13 (70), 0.12 (25), 0.05 (5)). In contrast, y (7) shows a definite increase with falling tetrahydro-pyran concentration in cyclohexane, (997 (Tetra-hydropyran in cyclohexane, vo l .%) : 0.40 (100), 0.49 (80), 0.60 (60), 0.47 (50), 0.59 (40), 0.76 (25), 0.70 (10), 0.80 (1)), and in formaldehyde dimethyl acetal shows no definite change (997 (Tetrahydro-pyran in formaldehyde dimethyl acetal, vo l .%) : 0.40 (100), 0.29 (80), 0.24 (70), 0.29 (25), 0.30 (5)). It is apparent that in the photolysis of these mixtures concentration dependent solvent effects [8] come into play which are at present not understood.

Experimental

Tetrahydropyran (Merck) was fractionated to a GC purity >99 .92%. Sample preparation, irradia-tion and analysis was largely as described [9]. Irradiations were carried out at 20 and 70 °C. The dose rate as determined by the ethanol actinometer [1] was 4.3 X 1017 quanta min - 1 per 2 ml sample. Doses ranged from 2.2 x 1018 to 26 X 1018 quanta in the kinetic experiments, corresponding to conver-sions between about 0.02 and 0.25%. Mixtures of tetrahydropyran with cyclohexane and formal-dehyde dimethyl acetal at various concentrations were also irradiated. The products contained in the

irradiated liquid were determined on a 40 m Carbo-wax 20 M, and/or on a 100 m PPG coated glass capillary column. w-Cgtho and W-C18H38 were used as internal standards; the GC molecular response factors of the products were determined using authentic material (sources see Table), or estimated on an incremental basis (c/. [10]). The quantum yields of the larger products are considered accurate within ± 10%. The gaseous products were purged from the sample and introduced into the gas Chromatograph by the appropriate carrier gas as described [11]. Formaldehyde was determined photometrically by the acetylacetone-ammonium acetate method [12]. The other products were identified by means of authentic reference material and/or GC-MS. Mass spectra were done on an Atlas MAT CH4 instrument.

For further confirmation of product 10 (see Table), deaerated tetrahydropyran was irradiated at A 185 on a preparative scale to a total conversion of about 3%. Evaporation of volatile and unreacted material yielded a crude product which was refined by preparative GC on a 10 m stainless steel column (i. d. 10 mm) packed with DC 550-coated Chromo-sorb P (20%), and then subjected to TMS derivatiza-tion and IR, 13C NMR, and mass spectrometry. The TMS derivative was obtained by treating 10 with an excess of bis-trimethylsilyltrifiuoroacetamide for 2 h at 50 °C and was gas chromatographed on a 50 m OS 138-coated glass capillary column. I R (model 621 Perkin Elmer) of the underivatized compound 10 showed the presence of a hydroxyl group (solution: 3610 cm - 1 , 3450 cm - 1 (broad); film: 3400 cm - 1 , very broad). Broad-band decoupled i3C NMR (Bruker W H 270) at 67.89 MHz showed ten signals at 24.06, 25.77, 26.28, 26.64, 32.42, 33.21, 37.08, 62.63, 68.39, and 77.85 ppm (standard TMS). Off-resonance 13C NMR revealed that the signal at 77.85 ppm is a doublet whereas the others are triplets, suggesting one branching point in the molecule. The mass spectrum of 10 showed the presence of the a-tetrahydropyranyl structural unit (m/e 85, 100%), and a probable molecular weight of 172 (m/e 172, 1%). Further peaks are at m/e 41 (30%), 55 (20%), 29 (15%), 57 (15%), 67 (12%), 31 (10%), 43 (10%), 130 (3%). The mass spectrum of the TMS derivative of 10 indicated a molecular weight of 244 (2%). Further peaks are at m/e 85 (100%), 75 (38%), 73 (30%), 81 (27%), 95 (25%), 41 (22%), 67 (20%), 55 (18%), 154 (6%).

[1] C. von Sonntag and H.-P. Schuchmann, Adv. Photochem. 10, 86 (1977).

[2] N. Kizilkilic?, H.-P. Schuchmann, and C. von Sonntag, to be published.

[3] a) D. Sänger and C. von Sonntag, Z. Naturforsch. 25b, 1491 (1970); b) H.-P. Schuchmann, C. von Sonntag, and D. Schulte-Frohlinde, J. Photochem. 3,267 (1974/75).

[4] L. W. Pickett, N. J. Hoeflich, and T.-C. Liu, J. Am. Chem. Soc. 73, 4865 (1951).

[5] E. Qetinkaya, H.-P. Schuchmann, and C. von Sonntag, J. Chem. Soc. Perkin II, in press.

[6] K. J. Laidler and L. F. Loucks, in C. H. Bamford and C. F. H. Tipper (eds.): Comprehensive Chemical Kinetics Vol. 5, p. 17, Elsevier, Amster-dam 1972.

H.-P. Schuchmann et al. • Photolysis of Tetrahydropyran 945

[7] H.-P. Schuchmann and C. von Sonntag, J. Chem. Soc. Perkin Trans. II 1976, 1408.

[8] a) D. Schulte-Frohlinde, D. Sänger, and C. von Sonntag, Z. Naturforsch. 27 b, 205 (1972); b) H.-P. Schuchmann, C. von Sonntag, and D. Schulte-Frohlinde, J. Photochem. 4, 63 (1975). *

[9] H.-P. Schuchmann and C. von Sonntag, Tetra-hedron 29, 1811 (1973).

[10] R. Kaiser, Chromatographie in der Gasphase, Vol. IV, Bibliographisches Institut, Mannheim 1965.

[11] F. Weeke, E. Bastian, and G. Schomburg, Chromatographia 7, 163 (1974).

[12] B. Kakäc and Z. J. Vejdelek, Handbuch der Photometrischen Analyse Organischer Verbin-dungen, p. 257, Verlag Chemie, Weinheim 1974.