2. Results and Discussion
The first issue to address was which catalyst to apply. Our interest in isomerization was indeed triggered by Trost and Kazmaier, who reported facile conversions when triphenylphosphine in warm toluene (80–110 °C) was applied [
4], but a study of the literature revealed that other alternatives appeared to be better. Particularly attractive were polymer-supported phosphines, which in most cases are fairly stable, easy to recover, and could be reused several times without a significant loss of catalytic capacity [
8,
9]. Based on availability, performance, and stability, we decided to apply polystyrene-supported triphenylphosphine (PS-TPP), which is commercially available and easier to recover than PPh
3.
1,1-Diethoxyoct-3-yn-2-one (
1a) was available from another study by our group [
10], and this compound was used to see if the anticipated isomerization took place. When this turned out to be the case,
1a was used to perform exploratory experiments to optimize the reaction conditions. The first reaction was performed essentially as described by Jiang and coworkers, viz. with a catalyst load of 20%, a reaction temperature of 80 °C, and a reaction time of 20 h [
8], and to our satisfaction, we obtained predominantly one product, the expected fully conjugated dienone 1,1-diethoxyocta-3,5-dien-2-one, which appeared to be the 3
E,5
E-stereoisomer
2a (vide infra). The yield was only 29% according to
1H-NMR analyses of the reaction mixture (
Table 1, entry 1), but when the catalyst loading was increased to 30 mol%, the conversion almost doubled even though the reaction temperature had been lowered to 60 °C (entry 2). In order to improve the outcome even more, both the reaction time and the reaction temperature were increased, but these measures combined improved the yield less than increasing the catalytic loading (
Table 1, entry 3). A final aspect to consider was the impact of the addition of an acid as a co-catalyst. Shortly after Trost, Lu, and co-workers published their findings [
4,
5], Rychnovsky and Kim released a study of the influence of a series of weak acids on the PPh
3-catalyzed conversion of conjugated yne esters and ketones to conjugated diene carbonyl compounds [
6]. They reported that acetic acid (used by Trost and Kazmaier) slowed down the reaction and triggered substrate decomposition, but when phenol was applied, these issues appeared to be resolved. To our satisfaction, this was also the case for the isomerization of
1a to
2a under the same conditions, and when the reaction reported in entry 2 was repeated after the addition of phenol, the conversion increased to 65% (
Table 1, entry 4).
In their search for a suitable co-catalyst for PPh
3, Rychnovsky and Kim investigated the catalytic capability of acetic acid, triflic acid, and five phenolic compounds and discovered that when p
Ka was below 7, the co-catalytic impact dropped significantly [
6]. Their best catalyst appeared to be phenol, which has become sort of the standard co-catalyst for PPh
3 in the isomerization under consideration here [
11]. On reflection, it occurred to us that only organic oxyacids had been tested in the search for a co-catalyst even though there are C-H acids with comparable acid strength to that of phenol [
12]. One well-known example of such an acid is pentane-2,4-dione, and when the reaction in entry 5 in
Table 1 was repeated with this 1,3-diketone instead of phenol, the yield of
2a according to
1H-NMR analysis increased to 95% (
Table 1, entry 5). When this was repeated on a preparative scale, diene
2a was isolated in 87% yield by flash chromatography after having recovered PS-TPP by filtration (
Scheme 1).
In order to examine if other 1,1-diethoxyalk-3-yn-2-ones react in the same way, two analogs to
1a,
viz. 1,1-diethoxy-7-methyloct-3-yn-2-one (
1b) and 1,1-diethoxynon-3-yn-2-one (
1c), were prepared and submitted to the reaction conditions summarized in
Scheme 1. The former was synthesized from 5-methylhex-1-yne using the Dulou method for chain elongation (
Scheme 2) [
13], but when the same method was applied to prepare
1c from hept-1-yne, the synthesis failed. We therefore turned our attention to an alternative and much older procedure developed by Wohl and Lange, and this furnished
1c in 38% yield along with a small amount of 8-(diethoxymethyl)pentadeca-6,9-diyn-8-ol (
Scheme 3) [
14].
Ynones
1b and
1c were then treated with PS-TPP in toluene at 60 °C in the presence of pentane-2,4-dione (see
Scheme 1), and after filtration and purification by flash chromatography, the corresponding dienones were obtained in 72% and 79% yield, respectively. The olefinic regions in the
1H- and
13C-NMR spectra of both dienones, 7.41–6.15 ppm and 153.7–122.5 ppm, respectively, are very similar to those of
2a, and this indicates that the diene moiety of all the compounds has the same stereochemistry,
viz.
E,
E (vide infra). Thus,
1b and
1c furnished (3
E,5
E)-1,1-diethoxy-7-methylocta-3,5-dien-2-one (
2b) and (3
E,5
E)-1,1-diethoxynona-3,5-dien-2-one (
2c), respectively.
For 1,4-disubstituted 1,3-butadienes, the
E,
E isomers are the most favorable substrates in Diels–Alder reactions, but to be really useful, they should be electron-rich and not electron-poor like dienones
2 [
15,
16,
17]. We therefore aimed to make electron-rich dienes by converting the electron-withdrawing diethoxyacetyl group in
2 into an electron-donating substituent. This can be achieved in several ways, and we decided to proceed via the corresponding secondary alcohols, which can be converted to the corresponding triflates and halides that are envisaged to undergo reduction by various reagents [
18,
19]. In order to be successful, the reduction must be a regioselective, high-yield reaction, and to test if this would be the case, exploratory experiments with
2a and sodium borohydride were performed. The best results were obtained when the reaction was performed in the presence of cerium(III) chloride (Luche conditions) [
20], and when a preparative reaction was carried out, pure 1,1-diethoxyocta-3,5-dien-2-ol was isolated in 92% yield (
Scheme 4). The compound appeared stable at room temperature, but with an allylic alcohol next to an acetal, subsequent transformations are expected to be challenging.
Unlike ketone
2a, 1,1-diethoxyocta-3,5-dien-2-ol exhibits an
1H-NMR spectrum where the olefinic region is of the first order so that the coupling constants can be accurately determined. The results, summarized in
Figure 2, show that the vicinal coupling constants across the C=C bonds are 15.4 Hz and 15.2 Hz, which are significantly larger than the coupling between the hydrogens in (
Z)-1,2-disubstituted alkenes [
21,
22]. This unequivocally supports the conclusion that the reduction of
2a gives (3
E,5
E)-1,1-diethoxyocta-3,5-dien-2-ol (
3) and therefore, as a corollary, that the PS-TPP-catalyzed isomerization of
1a indeed furnished (3
E,5
E)-1,1-diethoxyocta-3,5-dien-2-one (
2a). And since the olefinic regions of the
1H-NMR spectra of the dienones formed by the isomerization of ynones
1b and
1c are very similar to that for
2a (vide supra), it is concluded that
1b and
1c are converted to (3
E,5
E)-1,1-diethoxy-7-methylocta-3,5-dien-2-one (
2b) and (3
E,5
E)-1,1-diethoxynona-3,5-dien-2-one (
2c), respectively.
4. Materials and Methods
4.1. General Considerations
The chemicals were obtained from commercial suppliers and used without further purification. Anhydrous THF was prepared by using an MB-SPS-800 solvent-purification system. All reactions carried out under anhydrous conditions were performed in oven-dried (150 °C) glassware under nitrogen using the syringe–septum cap technique. Pre-coated aluminum TLC plates (Alugram, 0.20 mm Silica Gel 60 F254) were used for thin-layer chromatography (TLC), and mixtures of hexanes and ethyl acetate were employed as eluents. Visualization of the chromatograms was conducted with phosphomolybdic acid (NH4)4MoO4·4H2O) in ethanol. Flash-column chromatography (FC) was performed manually using silica gel from Fluka Analytical (230–400 mesh) and by eluting it with a mixture of hexanes and ethyl acetate. NMR spectra were recorded on a Bruker Biospin AV500 instrument (500 MHz for 1H, 125 MHz for 13C) in CDCl3 as solvent, using the solvent peaks as references in both 1H- and 13C-NMR spectra (7.26 and 77.16 ppm, respectively). The chemical shifts are reported in ppm, the coupling constants (J) are reported in Hz, and the multiplicity is given as s (singlet), d (doublet), t (triplet), q (quartet), and m (multiplet). Infrared (IR) spectra were recorded on a Nicolet Protege 460 FT-IR spectrophotometer with an attenuated total reflectance (ATR) unit attached. Samples were analyzed neat on a ZnSe crystal, and absorption peaks are reported in wavenumbers (cm−1). High-resolution mass spectra (HRMS) were obtained on a JEOL AccuTOF T100GC mass spectrometer operated in ESI or APCI in positive mode.
4.2. Preparation of 1,1-Diethoxyalk-3-yn-2-ones 1
The preparation of 1,1-diethoxyoct-3-yn-2-one (
1a) has been described elsewhere [
10].
4.2.1. 1,1-Diethoxy-7-methyloct-3-yn-2-one (1b)
A two-necked, round-bottomed flask, equipped with a magnetic stirrer, a condenser, and a septum, was flushed with N
2 and charged with 5-methyl-1-hexyne (1.09 g, 11.3 mmol) and anhydrous diethyl ether (50 mL). The resulting solution was cooled to 0 °C (water/ice), and BuLi (2.5 M solution in hexane) (4.5 mL, 11.3 mmol) was added over 5 min. The mixture was then stirred for 15 min before diethoxyacetonitrile (DAN) (1.46 g, 11.3 mmol) was added in one batch. The stirring continued at bath temperature for 1 h, water (20 mL) was added dropwise, and the hydrolysate was then stirred for 1 h. The product was isolated by extraction with diethyl ether (2 × 50 mL) in a separatory funnel, and the combined organic extracts were dried (MgSO
4) and then concentrated in vacuo after filtration. From the resulting orange crude product, 0.80 g (31%) of slightly impure
1b was isolated by FC (9:1 hexane/ethyl acetate) as a yellowish oil. The presence of impurities is evident from the NMR spectra in the
Supplementary Materials, but the dominant peaks are due to
1b and support structure elucidation.
1H NMR (CDCl3, 500 MHz): δ 4.71 (s, 1H), 3.74-3.58 (m, 4H), 2.42 (t, J = 7.4 Hz, 2H), 1.72 (hept, J = 6.6 Hz, 1H), 1.50 (m, 2H), 1.27 (t, J = 7.1 Hz, 6H), 0.91 (d, J = 6.6 Hz, 6H). 13C NMR (CDCl3, 125 MHz): δ 183.2, 101.8, 99.0, 79.3, 63.1 (2C), 36.5, 27.4, 22.2 (2C), 17.5, 15.3 (2C). IR (neat) νmax: 2957, 2928, 2209, 1683, 1467, 1386, 1368, 1317, 1257, 1116, 1062, 911, 840, 816, 755. HRMS (APCI+/TOF): calcd for C13H22O3Na [M + Na+] 249.1467, found 249.1464.
4.2.2. 1,1-Diethoxynon-3-yn-2-one (1c)
An oven-dried, two-neck, round-bottomed flask, equipped with a magnetic stirrer, a condenser, and a septum, was flushed with N
2 and charged with anhydrous THF (12 mL) and 1-heptyne (0.38 g, 4.0 mmol). A 2.5 M hexane solution of EtMgBr (1.6 mL, 4.0 mmol) was then added dropwise, and the resulting mixture was heated to 58 °C and left stirring for 1.5 h. 1-(Diethoxyacetyl)piperidine (DAP) (0.87 g, 4.0 mmol) was then added neat, and the mixture was stirred for 20 h. The reaction was quenched by the dropwise addition of 0.5 M H
2SO
4 (10 mL) and stirred for 1 hr before being worked up by extraction with diethyl ether (2 × 10 mL). The combined organic extracts were dried (MgSO
4), filtered, and concentrated in vacuo on a rotary evaporator. TLC analysis of the residue revealed two products, which were isolated by FC (9:1 hexane/ethyl acetate) and gave 0.34 g (38%) of
1c and 26 mg (4%) of 8-(diethoxymethyl)pentadeca-6,9-diyn-8-ol as clear, yellowish oils. The presence of impurities in 8-(diethoxymethyl)pentadeca-6,9-diyn-8-ol is evident from the NMR spectra in the
Supplementary Materials, but the dominant peaks are due to this alcohol and support the elucidation of its structure.
The data for 1c are as follows:
1H NMR (CDCl3, 500 MHz): δ 4.71 (s, 1H), 3.74–3.59 (m, 4H), 2.40 (t, J = 7.2 Hz, 2H), 1.60 (m, 2H), 1.43–1.37 (m, 2H), 1.36–1.29 (m, 2H), 1.26 (t, J = 7.0 Hz, 6H), 0.90 (t, J = 7.3 Hz, 3H). 13C NMR (CDCl3, 125 MHz): δ 183.2, 101.8, 98.9, 79.3, 63.1 (2C), 31.8, 27.4, 22.2, 19.4, 15.3 (2C), 14.0. IR (neat) νmax: 2976, 2958, 2931, 2872, 2209, 1680, 1457, 1371, 1321, 1253, 1120, 1062, 908, 838, 730, 703. HRMS (ESI+/TOF): calcd for C13H22O3Na [M + Na+] 249.1467, found 249.1470.
The data for 8-(diethoxymethyl)pentadeca-6,9-diyn-8-ol are as follows:
1H NMR (CDCl3, 500 MHz): δ 4.43 (s, 1H), 3.92–3.86 (m, 2H), 3.78–3.72 (m, 2H), 2.91 (s, 1H), 2.23 (t, J = 7.2 Hz, 4H), 1.53 (m, 4H), 1.39–1.30 (m, 8H), 1.27 (t, J = 7.1 Hz, 6H), 0.89 (t, J = 7.2 Hz, 6H). 13C NMR (CDCl3, 125 MHz): δ 105.6, 85.2, 78.2, 66.3, 65.8, 31.2, 27.3, 22.5, 19.0, 15.4, 14.1. IR (neat) νmax: 3456, 2968, 2927, 2859, 2238, 1458, 1376, 1328, 1117, 1067, 1019, 909, 729, 650, 623, 552. HRMS (ESI+): calcd for C20H34O3Na [M + Na+] 345.2406, found 345.2409.
4.3. Isomerization of 1 Using PS-TPP
4.3.1. General Procedure
A 10 mL, one-necked, round-bottomed flask was charged with 30 mol% PS-TPP (0.1 g, 3 mmol/g), alkynone 1 (1.0 mmol), toluene (2 mL), and acetylacetone (0.15 g, 1.5 mmol). The solution was then heated to 60 °C and stirred for 20 hr. Diethyl ether (5 mL) was added, and after approximately 30 min, the filtration and concentration of the filtrate on a rotary evaporator followed. From the resulting residue, the corresponding (3E,5E)-1,1-diethoxyalka-3,5-dien-2-one (2) was isolated by FC (30:1 petroleum ether/ethyl acetate).
4.3.2. (3E,5E)-1,1-Diethoxyocta-3,5-dien-2-one (2a)
When 1a (0.21 g, 1.0 mmol) was reacted following the general procedure, 0.18 g (87%) of 2a was furnished as a yellowish oil.
1H NMR (CDCl3, 500 MHz): δ 7.39 (dd, J = 10.4 Hz, 15.4 Hz, 1H), 6.44 (d, J = 15.4 Hz, 1H), 6.31–6.19 (m, 2H), 4.74 (s, 1H), 3.72–3.55 (m, 4H), 2.21 (m, 2H), 1.24 (t, J = 7.0 Hz, 6H), 1.05 (t, J = 7.4 Hz, 3H). 13C NMR (CDCl3, 125 MHz): δ 194.7, 148.3, 145.5, 128.3, 122.5, 102.3, 62.9 (2C), 26.2, 15.2 (2C), 12.8. IR (neat) νmax: 2974, 2933, 2876, 1633, 1630, 1593, 1457, 1371, 1319, 1300, 1259, 1099, 1056, 1001, 951, 914, 841, 823, 687, 568. HRMS (APCI+/TOF): calcd for C12H20O3Na [M + Na+] 235.1310, found 235.1313.
4.3.3. (3E,5E)-1,1-Diethoxy-7-methylocta-3,5-dien-2-one (2b)
When
1b (0.23 g, 1.0 mmol) was reacted following the general procedure, 0.17 g (72%) of
2b was obtained as a yellowish oil. The presence of impurities is evident from the NMR spectra in the
Supplementary Materials, but the dominant peaks are due to
2b and support structure elucidation.
1H NMR (CDCl3, 500 MHz): δ 7.41-7.36 (m, 1H), 6.45 (d, J = 15.4 Hz, 1H), 6.24–6.15 (m, 2H), 4.74 (s, 1H), 3.72–3.56 (m, 4H), 2.47–2.41 (m, 1H), 1.24 (t, J = 7.1 Hz, 6H), 1.05 (d, J = 6.8 Hz, 6H). 13C NMR (CDCl3, 125 MHz): δ 194.8, 153.7, 145.8, 126.5, 122.8, 102.4, 63.0 (2C), 29.9, 21.9 (2C), 15.3 (2C). IR (neat) νmax: 2958, 2925, 2870, 1695, 1631, 1595, 1465, 1369, 1279, 1242, 1100, 1060, 1002, 950, 904, 825, 692, 567. HRMS (ESI+/TOF): calcd for C13H22O3Na [M + Na+] 249.1467, found 249.1468.
4.3.4. (3E,5E)-1,1-Diethoxynona-3,5-dien-2-one (2c)
When 1c (0.23 g, 1.0 mmol) was reacted following the general procedure, 0.18 g (79%) of 2c was afforded as a yellowish oil.
1H NMR (CDCl3, 500 MHz): δ 7.41–7.35 (m, 1H), 6.43 (d, J = 15.4 Hz, 1H), 6.27–6.18 (m, 2H), 4.74 (s, 1H), 3.72–3.56 (m, 4H), 2.18–2.14 (m, 2H), 1.46 (m, 2H), 1.24 (t, J = 7.1 Hz, 6H), 0.92 (t, J = 7.3 Hz, 3H). 13C NMR (CDCl3, 125 MHz): δ 194.8, 147.0, 145.6, 129.5, 122.6, 102.4, 63.0 (2C), 35.4, 22.0, 15.3 (2C), 13.8. IR (neat) νmax: 2961, 2929, 2873, 1694, 1631, 1593, 1457, 1371, 1313, 1269, 1099, 1057, 1002, 902, 857, 801, 686. HRMS (ESI+/TOF): calcd for C13H22O3Na [M + Na+] 249.1467, found 249.1468.
4.4. Preparation of (3E,5E)-1,1-Diethoxyocta-3,5-dien-2-ol (3)
A 10 mL flask was charged with 2a (0.22 g, 1.04 mmol), EtOH (10 mL), and CeCl3 × 7H2O (0.46 g, 1.23 mmol). The mixture was stirred at rt during the addition of NaBH4 (0.040 g, 1.04 mmol), and the stirring was continued until all the hydride had dissolved (10 min). Water (20 mL) was then added, and the hydrolysate was extracted with DCM (2 × 20 mL). Some NaCl was added to break the emulsion formed during the extraction. The combined organic extracts were dried (MgSO4), then filtered, and finally concentrated in vacuo on a rotary evaporator. From the residue obtained, 0.21 g (92%) of 3 was isolated as a clear oil by FC (9:1 hexane/ethyl acetate).
1H NMR (CDCl3, 500 MHz): δ 6.33 (dd, J = 10.5 Hz, 15.4 Hz, 1H), 6.06 (dd, J = 10.5 Hz, 15.2 Hz, 1H), 5.75 (dt, J = 6.5 Hz, 15.2 Hz, 1H), 5.62 (dd, J = 6.2 Hz, 15.4 Hz, 1H), 4.28 (d, J = 6.0 Hz, 1H), 4.10 (m, 1H), 3.75 (m, 2H) 3.58 (m, 2H), 2.28 (d, J = 3.7 Hz, 1H), 2.10 (m, 2H) 1.24 (t, J = 7.0 Hz, 3H), 1.21 (t, J = 7.0 Hz, 3H), 1.00 (t, J = 7.3 Hz, 3H). 13C NMR (CDCl3, 125 MHz): δ 137.2, 132.9, 128.8, 127.9, 105.0, 72.7, 63.9, 63.7, 25.8, 15.5, 15.5, 13.6. IR (neat) νmax: 3449, 2958, 2929, 2873, 1457, 1372, 1120, 1062, 1024, 903, 737. HRMS (ESI+/TOF): calcd for C12H22O3Na [M + Na+] 237.1467, found 237.1469.