Synthesis 2020; 52(22): 3446-3451
DOI: 10.1055/s-0040-1707219
paper
© Georg Thieme Verlag Stuttgart · New York

Neodymium-Promoted Highly Selective Carbon–Carbon Double Bond Formation of Ketones with Allyl Halides in the Presence of Diethyl Phosphite

Dengbing Xie
,
Yiqiong Wang
,
Bo Yang
,
Songlin Zhang
Key Laboratory of Organic Synthesis of Jiangsu Province, College of Chemistry, Chemical Engineering and Materials Science, Soochow University, Suzhou 215123, P. R. of China   Email: zhangsl@suda.edu.cn
› Author Affiliations

We gratefully acknowledge the project funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions and the National Natural Science Foundation of China (No. 21072143) for financial support.
Further Information

Publication History

Received: 25 April 2020

Accepted after revision: 27 June 2020

Publication Date:
12 August 2020 (online)

 


Abstract

The carbon–carbon double bond formation via neodymium-mediated Barbier-type reaction of ketones and allyl halides in the presence of diethyl phosphite is reported for the first time. The reaction is highly α-regioselective and was conveniently carried out under mild conditions in a one-pot fashion. From a synthetic point of view, a series of conjugated alkenes were obtained in moderate to good yields in this one-pot reaction with feasible reaction conditions.


#

After the first introduction of samarium diiodide in organic synthesis by Kagan,[1] a lot of important individual transformations have been made using it as a mild, neutral, selective and versatile single-electron transfer (SET) reducing and coupling reagent.[2] The successful usage of samarium diiodide promoted other divalent lanthanide reagents LnX2 (Ln = Nd,[3] Tm,[4] Dy,[5] X = I, Br, Cl) in the organic synthesis. In 2003, Evans et al.[3a] used neodymium(II) diiodide as a reductant in the reaction of alkyl halides with ketones or aldehydes. The elegant result indicated that NdI2 can be a single electron transfer reagent similar to SmI2. However, NdI2 is moisture- and air-sensitive, it is also instable in solvent at room temperature. The preparation is much more difficult than that of SmI2. For the above reasons, the application of NdI2 in organic chemistry is still scarcely reported.[6]

Zerovalent lanthanoid metals have strong reduction potential[7] and these metals show great advantages not only as nontoxic, cheap and stable in the air, but also electron-transfer-efficiency as reductant. That is why more and more lanthanoid metals such as Sm,[8] Dy,[9] Ce,[10] Yb,[11] La,[12] and Pr[13] have been employed as useful reagents or catalysts in organic synthesis. However, the application of reduction of zerovalent neodymium metal has been ignored for over one decade until the first report from our group appeared.[14]

Alkenes especially, the dienes not only play an important role in the synthesis of many ligands, but also are found in the structure of numerous natural products and pharmaceutical agents.[15] Therefore, the research on the synthesis of compounds containing carbon–carbon double bonds is of great significance in organic synthesis. For alkene synthesis, one can use the condensation-type reaction of carbonyl compounds such as the Wittig reaction,[16] or related reactions.[17] However, the condensation-type reaction is not effective for the synthesis of dienes. And there are many other methods for the preparation of dialkenes[18] [19] that have some downsides such as difficult synthesis of substrates, harsh reaction conditions, multiple steps, and so on.

Our attempts have been focused on the research of carbon–carbon double bond formation in recent years. We have developed several concise and elegant olefination reactions of carbonyl compounds involving allylsamarium bromide,[20] Grignard reagents,[21] or organozinc reagents[22] (Scheme [1]).

Zoom Image
Scheme 1 The carbon–carbon double bond formation

As shown in Scheme [1], organometallic reagents were generated by a stepwise procedure under oxygen- or water-free conditions, then the organometallic reagents underwent nucleophilic addition to ketones to generate alkoxides 1, finally elimination of alkoxide mediated by metallic bromide diethoxy(oxo)phosphate 2 afforded the carbon–carbon double bond compounds. More recently, we reported the Barbier reaction using neodymium to generate the intermediate organometallic and its addition to ketones of the benzophenone class, providing tertiary alcohols.[14] Our objective was to merge these concepts and to find out if diethyl­ phosphite is introduced as an additive to the neodymium-mediated Barbier reaction, whether the Barbier reaction could afford the carbon–carbon double bond compounds rather than the tertiary alcohols? Considering the step-efficiency of Barbier reaction compared to Grignard reaction (no need to prepare the organometallic reagents by a stepwise procedure under oxygen or water-free conditions) and the development of new applications of neodymium in organic synthesis, we envisioned that this attempt is meaningful and merited.

Herein we report the utility of neodymium as a mediating metal in the Barbier reaction of carbonyl compounds with allyl halides to produce conjugated alkenes in the presence of diethyl phosphite (Scheme [2]).

Zoom Image
Scheme 2 Carbonyl olefination promoted by neodymium in the presence of diethyl phosphite

Initially, the reaction of benzophenone and allylic bromide were selected as a model reaction (Table [1]).

Table 1 Optimization of Reaction Conditionsa

Entry

Additive (equiv)

Additive:Nd:Bromide (equiv)

Time (h)

Yield (%) b

 1

Ph2P(O)H (1.2)

1.2:3:4

12

 2

(EtO)3P (1.2)

1.2:3:4

12

 3

(EtO)3PO (1.2)

1.2:3:4

12

 4

Ph3P (1.2)

1.2:3:4

12

 5

(EtO)2P(O)H (1.2)

1.2:3:4

 6

54

 6

(EtO)2P(O)H (1.2)

1.2:3:4

12

80

 7

(PhO)2P(O)H (1.2)

1.2:3:4

12

82

 8

(EtO)2P(O)H (0.6)

0.6:3:4

12

27

 9

(EtO)2P(O)H (2.0)

2.0:3:4

12

61

10

(EtO)2P(O)H (1.2)

1.2:2:2.4

12

59

11

(EtO)2P(O)H (1.2)

1.2:1.2:1.6

12

50

a To a suspension of benzophenone (0.5 mmol) and Nd in THF (3 mL) was added bromide and catalytic amounts of I2 (0.025 mmol) under a N2 atmosphere at r.t., and the mixture was stirred for 2 h, then the additive was added to this mixture and stirred at 60 °C.

b Isolated yield based on benzophenone after silica gel chromatography.

As shown in Table [1], no product was obtained by using diphenylphosphine oxide, triethyl phosphite, triethyl phosphate, or triphenylphosphine as additives (Table [1], entries 1–4). Only diethyl phosphite and diphenyl phosphite were the effective additive to promote the olefination reaction (entries 5–7). In terms of cheap and readily available considerations, diethyl phosphite is a better choice. Subsequently, the molar ratio of ketone/bromide/Nd/phosphite was also investigated (entries 8–11). Taken together, the above experimental facts indicate that the optimum conditions for the reaction are the following: diethyl phosphite as an additive at 60 °C with a 1:4:3:1.2 molar ratio of ketone/bromide/Nd/phosphite (entry 6).

Encouraged by these efficient experimental results, we examined the scope of carbonyl compounds and bromides. The representative examples are presented in Table [2]. As shown in Table [2], benzophenone and substituted benzophenones are all workable substrates to react with allylic bromide or allylic iodide in the presence of neodymium and diethyl phosphite giving a range of conjugated dienes in good to excellent yields (Table [2], 2aa, 2ba, 2ca, 2bb). Unfortunately, when acetophenone and 4-bromobenzaldehyde were employed as substrates, no corresponding products were obtained (Table [2], 2da, 2ea).

Next, in order to investigate the generality of the bromide and test the regioselectivity of the Barbier type reaction, 3,3-dimethylallyl bromide was employed as the substrate (Table [2], 2ac, 2bc, 2cc, 2fc). As expected, no γ-adducts were observed, only the α-adducts were obtained in good yields. To further broaden the reaction scope, 3-bromo-2-methylpropene and cinnamyl bromide were employed to react with a series of benzophenones (2ae, 2ge, 2fe, 2he, 2cf, 2hf), and the α-adducts were obtained exclusively in moderate to good yields. The results indicated that the neodymium-mediated Barbier-type reaction was highly α-regioselective, which is consistent with the results of our former work.[14] Finally, propargyl bromide was tested and the reaction was not successful (2ag).

Table 2 Olefination of Carbonyl Compoundsa

2aa

2ba

2ca

2ba

2da

2ea

2ac

2bc

2cc

2fc

2ae

2ge

2fe

2he

2cf

2hf

2ag

a To a suspension of ketone (0.5 mmol) and Nd (1.5 mmol) in THF (3 mL) was added bromide (2 mmol) and catalytic amounts of I2 (0.025 mmol) under a N2 atmosphere at r.t., and the mixture was stirred for 2 h, then diethyl phosphite (0.6 mmol) was added to this mixture and stirred at 60 °C for 12 h.

b Isolated yields based on ketone after silica gel chromatography.

According to our previous work,[20a] [21] a similar mechanism was proposed (Scheme [3]). The on-site neodymium reagent 4 undergoes nucleophilic addition to carbonyl compound 3 to generate alkoxide 5. The diethyl phosphite 6 reacts with the on-site neodymium reagent 4 to form metallic bromide diethoxy(oxo)phosphate 7. Then alkoxide 5 reacts with metallic bromide diethoxy(oxo)phosphate 7 to form a six-centered transition state 8, and the final elimination reaction affords the carbon–carbon double bond compound 9.

Zoom Image
Scheme 3 Possible reaction mechanism

In summary, we have developed a highly α-selective Barbier reaction of ketones with allyl halides promoted by neodymium in the presence of diethyl phosphite to afford carbon–carbon double bond compounds for the first time. As a complement to the traditional carbonyl olefination, the carbon–carbon double bond formation was highly α-regio­selective and step-efficient. From a synthesis point of views, a straightforward strategy to synthesize conjugated alkenes in moderate to good yields was developed.

THF was distilled from Na/benzophenone under N2. All reactions were conducted under a N2 atmosphere. Metallic Nd and all solvents were purchased from commercial source, and used without further purification. Flash column chromatography was carried out on Merck silica gel (300–400 mesh). 1H and 13C NMR spectra were recorded on a Varian-Inova 400 spectrometer. Solvent for NMR is CDCl3, unless the otherwise noted. Chemical shifts are reported in δ units (ppm) relative to the singlet (0 ppm) for TMS. Data are reported as follows: chemical shift, multiplicity (standard abbreviations), coupling constants (Hz), and integration. 13C NMR spectra were recorded at 100 MHz. Chemical shifts are reported in parts per million relative to the central line of the multiplet at 77.5 ppm for CDCl3. High-resolution mass spectra were obtained with a GCT-TOF instrument. All chemicals were purchased from Aldrich, Alfa, or Acros chemical company and used without further purification. Petroleum ether (PE) used refers to the 60–90 °C boiling point fraction of petroleum.


#

Dienes 2; General Procedure

To a suspension of the respective carbonyl compound (0.5 mmol) and neodymium (216 mg, 1.5 mmol) in THF (3 mL) was added the appropriate allyl halide (2 mmol) and catalytic amounts of I2 (6 mg, 0.025 mmol) under a N2 atmosphere at r.t. The mixture was stirred for 2 h. Then diethyl phosphite (83 mg, 0.6 mmol) was added to this mixture and stirred at 65 °C for 12 h, and quenched with dil. HCl. The resulting mixture was extracted with Et2O (3 × 10 mL), and dried (anhyd Na2SO4). The solvent was removed by evaporation under reduced pressure. Purification by column chromatography on silica gel (300–400 mesh, PE as eluent) afforded the corresponding olefin (Table [2]).


#

1,1-Diphenylbuta-1,3-diene (2aa)

Colorless oil; yield: 82.4 mg (80%).

1H NMR (CDCl3/TMS, 400 MHz): δ = 7.38–7.19 (m, 10 H), 6.71 (d, J = 8.0 Hz, 1 H), 6.47–6.43 (m, 1 H), 5.40–5.36 (m, 1 H), 5.13–5.10 (m, 1 H).

13C NMR (CDCl3/TMS, 101 MHz,): δ = 143.10, 142.04, 139.61, 134.92, 130.37, 128.48, 128.15, 128.12, 127.54, 127.46, 127.35, 118.57.

HRMS (EI+): m/z calcd for C16H14 (M+): 206.1096; found: 206.1087.


#

1,1-Bis(4-methoxyphenyl)buta-1,3-diene (2ba)

Colorless oil; yield: 100 mg (77%, from allyl bromide), 118.3 mg (91%, from allyl iodide).

1H NMR (CDCl3/TMS, 400 MHz): δ = 7.13–7.09 (m, 2 H), 7.05–7.02 (m, 2 H), 6.82–6.78 (m, 2 H), 6.73–6.69 (m, 2 H), 6.50 (d, J = 12.0 Hz, 1 H), 6.41–6.31 (m, 1 H), 5.23 (dd, J = 16.0, 4.0 Hz, 1 H), 4.96 (dd, J = 12.0, 4.0 Hz, 1 H), 3.72 (s, 3 H), 3.67 (s, 3 H).

13C NMR (CDCl3/TMS, 101 MHz): δ = 158.85, 142.34, 136.70, 135.21, 131.55, 128.79, 126.64, 117.21, 52.28.

HRMS (EI+): m/z calcd for C18H18O2 (M+): 266.1307; found: 266.1324.


#

1,1-Bis(4-chlorophenyl)buta-1,3-diene (2ca)

Colorless oil; yield: 103.2 mg (75%).

1H NMR (CDCl3/TMS, 400 MHz): δ = 7.36 (d, J = 8.0 Hz, 2 H), 7.24–7.11 (m, 6 H), 6.67 (d, J = 12.0 Hz, 1 H), 6.43–6.33 (m, 1 H), 5.44–5.40 (m, 1 H), 5.18 (d, J = 8.0 Hz, 1 H).

13C NMR (CDCl3/TMS, 101 MHz): δ = 140.60, 140.11, 137.54, 134.27, 133.56, 133.54, 131.67, 129.27, 128.72, 128.55, 128.43, 119.82.

HRMS (EI+): m/z calcd for C16H12Cl2 (M+): 274.0316; found: 274.0341.


#

4-Methyl-1,1-diphenylpenta-1,3-diene (2ac)

Colorless oil; yield: 91.3 mg (78%).

1H NMR (CDCl3/TMS, 400 MHz): δ = 7.38–7.34 (m, 2 H), 7.32–7.28 (m, 1 H), 7.27–7.24 (m, 4 H), 7.23–7.19 (m, 3 H), 6.87 (d, J = 11.4 Hz, 1 H), 5.93 (dp, J = 11.3, 1.4 Hz, 1 H), 1.88 (s, 3 H), 1.75 (s, 3 H).

13C NMR (CDCl3/TMS, 101 MHz): δ = 143.08, 140.17, 139.69, 137.65, 130.55, 128.09, 128.07, 127.40, 126.98, 126.87, 124.47, 123.17, 26.47, 18.60.

HRMS (EI+): m/z calcd for C18H18 (M+): 234.1409; found: 234.1392.


#

1,1-Bis(4-methoxyphenyl)-4-methylpenta-1,3-diene (2bc)

Colorless oil; yield: 108.7 mg (74%).

1H NMR (CDCl3/TMS, 400 MHz): δ = 7.11–7.03 (m, 4 H), 6.81–6.62 (m, 4 H), 6.63 (d, J = 11.6 Hz, 1 H), 5.84 (d, J = 11.2 Hz, 1 H), 3.71 (s, 3 H), 3.66 (s, 3 H), 1.76 (s, 3 H), 1.65 (s, 3 H).

13C NMR (CDCl3/TMS, 101 MHz): δ = 159.06, 158.87, 139.27, 136.46, 136.42, 132.94, 131.98, 128.91, 123.70, 123.02, 113.77, 113.71, 55.48, 55.43, 26.75, 18.87.

HRMS (EI+): m/z calcd for C20H12O2 (M+): 294.1620; found: 294.1631.


#

1,1-Bis(4-chlorophenyl)-4-methylpenta-1,3-diene (2cc)

Colorless oil; yield: 107.6 mg (71%).

1H NMR (CDCl3/TMS, 400 MHz): δ = 7.36–7.22 (m, 4 H), 7.17–7.12 (m, 4 H), 6.84 (d, J = 12.0 Hz, 1 H), 5.86 (d, J = 12.0 Hz, 1 H), 1.88 (s, 3 H), 1.77 (s, 3 H).

13C NMR (CDCl3/TMS, 101 MHz): δ = 141.10, 139.10, 138.13, 137.14, 131.84, 128.32, 125.26, 122.69, 26.51, 18.67.

HRMS (EI+): m/z calcd for C18H16Cl2 (M+): 302.0629; found: 302.0617.


#

1,1-Bis(4-fluorophenyl)-4-methylpenta-1,3-diene (2fc)

Colorless oil; yield: 87.8 mg (65%).

1H NMR (CDCl3/TMS, 400 MHz): δ = 7.14–7.06 (m, 4 H), 6.99–6.84 (m, 4 H), 6.70 (d, J = 11.6 Hz, 1 H), 5.78 (d, J = 12.8 Hz, 1 H), 1.79 (s, 3 H), 1.68 (s, 3 H).

13C NMR (CDCl3/TMS, 101 MHz): δ = 163.55, 163.44, 139.35, 138.43, 137.77, 136.12, 132.37, 132.29, 129.54, 129.16, 129.09, 124.86, 123.06, 115.49, 115.44, 115.37, 115.28, 115.23, 115.16, 26.73, 18.86.

19F NMR (376 MHz, CDCl3): δ = –114.91, –115.48.

HRMS (EI+): m/z calcd for C18H16F2 (M+): 270.1220; found: 270.1985.


#

(3-Methylbuta-1,3-diene-1,1-diyl)dibenzene (2ae)

Colorless oil; yield: 58.3 mg (53%).

1H NMR (CDCl3/TMS, 400 MHz): δ = 7.27–7.23 (m, 3 H), 7.21–7.10 (m, 7 H), 6.58 (s, 1 H), 4.91 (d, J = 12.0 Hz, 2 H), 1.40 (s, 3 H).

13C NMR (CDCl3/TMS, 101 MHz): δ = 143.27, 142.40, 141.49, 140.62, 130.71, 130.30, 128.07, 127.91, 127.45, 127.19, 119.08, 22.04.

HRMS (EI+): m/z calcd for C17H16 (M+): 220.1252; found: 220.1239.


#

(E)-1-Methoxy-4-(3-methyl-1-phenylbuta-1,3-dien-1-yl)benzene (2ge)

Colorless oil; yield: 68.8 mg (55%).

1H NMR (CDCl3/TMS, 400 MHz): δ = 7.26–7.22 (m, 5 H), 7.13–7.11 (m, 2 H), 6.88–6.86 (m, 2 H), 6.62 (s, 1 H), 5.01–4.97 (m, 2 H), 3.83 (s, 3 H), 1.51 (s, 3 H).

13C NMR (CDCl3/TMS, 101 MHz): δ = 158.86, 143.66, 142.61, 141.22, 132.85, 131.46, 130.56, 128.02, 127.55, 127.16, 118.80, 113.30, 55.18, 22.07.

HRMS (EI+): m/z calcd for C18H18O (M+): 250.1358; found: 250.1370.


#

4,4′-(3-Methylbuta-1,3-diene-1,1-diyl)bis(fluorobenzene) (2fe)

Colorless oil; yield: 73 mg (57%).

1H NMR (CDCl3/TMS, 400 MHz): δ = 7.19–7.13 (m, 4 H), 7.05–6.93 (m, 4 H), 6.58 (s, 1 H), 4.99 (s, 2 H), 1.48 (s, 3 H).

13C NMR (CDCl3/TMS, 101 MHz): δ = 162.25 (1 J C,F = 247.45 Hz), 161.02 (d, J = 4.90 Hz), 141.97, 139.40, 139.29, 136.30 (4 J C,F = 3.03 Hz), 131.84 (3 J C,F = 7.07 Hz), 130.98, 129.07, 128.99, 119.44, 115.01 (2 J C,F = 21.21 Hz), 22.06 (s).

19F NMR (CDCl3, 376 MHz): δ = –114.66, –115.12.

HRMS (EI+: m/z calcd for C17H14F2 (M+): 256.1064; found: 256.1079.


#

(E)-4-(3-Methyl-1-phenylbuta-1,3-dien-1-yl)-1,1′-biphenyl (2he)

Colorless oil; yield: 77 mg (52%).

1H NMR (CDCl3/TMS, 400 MHz): δ = 7.59–7.56 (m, 1 H), 7.50–7.41 (m, 3 H), 7.35–7.31 (m, 2 H), 6.29–7.23 (m, 8 H), 6.73–6.67 (m, 1 H), 5.02–4.98 (m, 2 H), 1.51 (s, 3 H).

13C NMR (CDCl3/TMS, 101 MHz): δ = 142.41, 142.38, 142.19, 141.18, 141.00, 140.66, 140.52, 139.95, 130.95, 130.77, 130.65, 130.32, 128.76, 128.73, 128.10, 127.97, 127.79, 127.57, 127.26, 127.23, 126.96, 126.93, 126.77, 126.52, 119.2, 119.18, 22.18, 22.05.

HRMS (EI+): m/z calcd for C23H20 (M+): 296.1565; found: 296.1534.


#

4,4′-(4-Phenylbuta-1,3-diene-1,1-diyl)bis(chlorobenzene) (2ef)

Colorless oil; yield: 138.6 mg (79%).

1H NMR (CDCl3/TMS, 400 MHz): δ = 7.34–7.18 (m, 2 H), 7.18–7.12 (m, 5 H), 7.11–7.09 (m, 4 H), 6.99–6.97 (m, 2 H), 6.77 (d, J = 8.0 Hz, 1 H), 5.22–5.19 (m, 1 H), 4.99–4.94 (m, 1 H).

13C NMR (CDCl3/TMS, 101 MHz): δ = 140.68, 140.25, 140.03, 139.33, 137.87, 133.34, 132.26, 131.21, 128.30, 127.68, 126.91, 119.32.

HRMS (EI+): m/z calcd for C22H16Cl2 (M+): 350.0629; found: 350.0617.


#

4-(1,4-Diphenylbuta-1,3-dien-1-yl)-1,1′-biphenyl (2hf)

Colorless oil; yield: 152.2 mg (85%).

1H NMR (CDCl3/TMS, 400 MHz): δ = 7.31–7.15 (m, 19 H), 6.43–6.34 (m, 1 H), 6.25–6.10 (m, 1 H), 6.10–5.96 (m, 1 H).

13C NMR (CDCl3/TMS, 101 MHz): δ = 143.63, 141.40, 137.64, 137.59, 131.37, 130.35, 129.88, 128.43, 128.39, 127.62, 126.90, 126.88, 126.28, 125.98, 125.95, 114.58.

HRMS (EI+): m/z calcd for C28H22 (M+): 358.1722; found: 358.1741.


#
#

Supporting Information



Zoom Image
Scheme 1 The carbon–carbon double bond formation
Zoom Image
Scheme 2 Carbonyl olefination promoted by neodymium in the presence of diethyl phosphite
Zoom Image
Scheme 3 Possible reaction mechanism