Abstract

A new method to synthesize chiral α-amino phosphine oxides is reported. The reaction combines N-substituted imines and diphenylphosphine oxide, and is catalyzed by a chiral magnesium phosphate salt. A wide variety of aliphatic and aromatic aldimines substituted by electron neutral benzhydryl or dibenzocycloheptene groups were excellent substrates for the addition reaction. The dibenzocycloheptene protected imines afforded improved enantioselectivity in the resulting products. Substituted diphenylphosphine oxide nucleophiles also showed good reactivity.
Chiral phosphorus containing compounds such as α-amino phosphonic acids and their derivatives have attracted considerable attention due to their promising biological properties.1 They have found applications as antibacterials,2 anti-viral agents,3 and enzyme inhibitors.4 The absolute configuration of phosphonyl compounds strongly influences their biological properties,5 therefore numerous methods for enantioselective synthesis of chiral α-amino phosphoric acids by hydrophosphonylation of imines (aza–Pudovik reaction) are reported.6,7 However, other α-amino phosphorus derivatives such as α-amino phosphine oxides have had much less attention for their biological properties due to the lack of a direct enantioselective synthetic route.
Recently, organocatalyzed hydrophosphination using secondary phosphines8 has also been demonstrated. Additionally, in 2007, Melchiorre9 and Córdova10 reported enantioselective organocatalytic hydrophosphination of α,β–unsaturated aldehydes with diphenylphosphine using a chiral secondary amine catalyst to provide enantioenriched aldehydes with β–phosphine substitution. However, very few examples exist with phosphine oxides as the nuclophile. In 1999, Shibasaki reported the first catalytic asymmetric addition of diphenylphosphine oxide to cyclic imines, catalyzed by a heterobimetallic lanthanoid complex.11 The corresponding chiral amino phosphine oxide products were obtained in good yield and enantioselectivity, but the substrate scope of this reaction was limited to cyclic imines. In 2009, a chiral guanidinium salt-catalyzed phospha-Mannich reaction of N-tosyl imines with di-1-naphthyl phosphine oxide was reported.12 The α-amino phosphine oxide products were obtained in high yield and high ee. However, only a single example of a diphenyl phosphine oxide as a nucleophile was reported, and with moderate ee. Lastly the addition of dialkyl phosphine oxides to N-acyl pyrroles was reported recently, obtaining products with excellent yield and enantioselectivity.13
The use of chiral phosphoric acids to activate imine substrates for nuclophilic addition is well studied.14 Recently, Ishihara and co-workers reported that purification of phosphoric acids on silica gel can be contaminated with alkali or alkaline earth metals, as their phosphate salts.15 These phosphate salts were found to be active catalysts for new transformations.16
Herein we report a highly enantioselective addition of diphenylphosphine oxide to N-substituted imines catalyzed by chiral BINOL magnesium phosphate.17 During the screening process we soon realized an electron-neutral, unactivated protecting group provided considerable enhancement in the yield and enantioselectivity, similar to that observed by Wulff and coworkers in an aziridination reaction.18
We began by examining the catalytic asymmetric addition of diphenylphosphine oxide 3a to benzhydryl imine 2a as a model reaction (Table 1). During optim izatoin we found that acetonitrile and ‘H(1a) purified on silica gel’ were found to be the best solvent and catalyst for this reaction (entry 4). In the wake of the new results published by Ishihara,15 we screened alkali, and alkaline earth metal phosphate salts. Na(1a) allowed for only a 55% ee (entry 5), while Ca(1a)2 resulted in a 91% ee (entry 6). Mg(1a)2 was found to be the best catalyst, providing 93% asymmetric induction (entry 7). However, lowering the catalyst loading of Mg(1a)2 to 2.5 mol% resulted in a decrease in ee (entry 8). It was interesting that ‘H(1a) washed with hydrochloric acid’ achieved only a 80% ee for the resulting product (entry 9).
Table 1.
Catalytic reaction condition optimization for the enantioselective addition of diphenylphosphine oxide to N-benzhydryl imine.
![]() | ||||
|---|---|---|---|---|
| entrya | catalyst (mol %)b | solvent | yield (%)c | ee (%)d |
| 1 | H(1b) Purified on silica gel (10) | toluene | 68 | 0 |
| 2 | H(1c) Purified on silica gel (10) | toluene | 61 | 38 |
| 3 | H(1a) Purified on silica gel (10) | toluene | 72 | 82 |
| 4 | H(1a) Purified on silica gel (10) | CH3CN | 97 | 91 |
| 5 | Na(1a) (10) | CH3CN | 56 | 55 |
| 6 | Ca(1a)2 (5) | CH3CN | 96 | 91 |
| 7 | Mg(1a)2 (5) | CH3CN | 95 | 93 |
| 8 | Mg(1a)2 (2.5) | CH3CN | 92 | 80 |
| 9 | H(1a) washed with HCl (10) | CH3CN | 87 | 80 |
General conditions: 1.2 equiv of imine and 1.0 equiv of diphenylphosphine oxide.
See Supporting Information for details on the synthesis of the catalysts and for the optimization table.
Isolated yields.
Ee determined by chiral HPLC analysis.
A series of substitued imines were then investigated for the asymmetric hydrophosphination reaction in the presence of catalyst Mg(1a)2 under the optimized reaction conditions (Table 2). To our delight, all the imines underwent addition of diphenylphosphine oxide to afford α-amino phosphine oxide products in excellent yield and enantioselectivity. Electron donating substituents in the para- (entry 2), ortho- (entry 3), and meta- (entry 4) positions were all shown to be excellent substrates for the phospha-Mannich reaction. Likewise, the use of electron withdrawing substituents (entries 5–7) in the para-position also provided for high yield and excellent asymmetric induction. A sterically hindered 1-napthyl substituted imine was also an excellent substrate, allowing preparation of 4h in 95% yield with 96% ee (entry 8). A heteroaromatic (2-furyl) substitution also proved to be a good substrate with a high yield and ee for product 4i being observed (entry 9). We also examined the aliphatic substrate scope; cyclohexyl and isopropyl imines gave good yield and enantioselectivity (entries 10 and 11). However, it was found that n-propyl, n-butyl and 3-phenyl propyl substituted imines afforded moderate yields and enantioselectivities when chosen as a substrate for the reaction (entries 12–14).
Table 2.
The catalytic asymmetric addition of diphenylphosphine oxide to N-benzhydryl imines.
![]() | ||||
|---|---|---|---|---|
| entrya | R | product | yield (%)b | ee (%)c |
| 1 | Ph (2a) | 4a | 95 | 93 |
| 2 | 4-MeOC6H4 (2b) | 4b | 95 | 90d |
| 3 | 2-MeC6H4 (2c) | 4c | 96 | 89 |
| 4 | 3-MeC6H4 (2d) | 4d | 95 | 92 |
| 5 | 4-FC6H4 (2e) | 4e | 95 | 90 |
| 6 | 4-BrC6H4 (2f) | 4f | 97 | 92 |
| 7 | 4-NO2C6H4 (2g) | 4g | 97 | 95 |
| 8 | 1-Naph (2h) | 4h | 95 | 96 |
| 9 | 2-Furyl (2i) | 4i | 93 | 91 |
| 10 | c-C6H11 (2j) | 4j | 96 | 91 |
| 11e | iso-propyl | 4k | 92 | 86 |
| 12e | n-Propyl | 4l | 88 | 62 |
| 13e | n-Butyl | 4m | 65 | 52 |
| 14e | Ph-(CH2CH2) | 4n | 73 | 48 |
General conditions: 1.2 equiv of imine and 1.0 equiv of the phopshine oxide with 5 mol % Mg(1a)2.
Isolated yields.
Ee determined by chiral HPLC analysis.
Using the optimized conditions, with Ca(1a)2 as the catalyst a 96% yield and 81% ee of 4 could be obtained.
Imine was generated in situ (see Supporting Information).
Next we chose to investigate the imine protecting group on the nucleophilic addition. We began this study by preparing 5a from benzaldehyde and 5H–dibenzo[a,d]cyclohepten-5-amine.19 Optimization of the reaction conditions revealed the best selectivity when 2 equivalents of imine 5a were treated with 1 equivalent of diphenylphosphine oxide in CH2Cl2 at room temperature (Table 3). In order to determine the utility of the new protecting group, we synthesized four new imines (Table 3, 5b–e). We opted to synthesize the imines which had induced relatively lower enantioselectivity with the benzhydryl protecting group in our previous study (see Table 2). When we subjected 5a to the optimized reaction conditions, we found a 90% yield and 99% ee for the resulting product 6a. For the benzhydryl protecting group, product 4a was obtained with 96% yield and 93% asymmetric induction. For imine 5b, 92% ee was obtained for product 6b, while in case of the benzhydryl protected imine a 90% ee was observed in product 4b. Imines 5c and 5d both underwent diphenylphosphine oxide addition with excellent yield and enantioselectivity. For the products 6c and 6d, an increased asymmetric induction was observed when compared to the benzhydryl protected counterpart 4c and 4e respectively. However, for the hetero-aromatic (2-furyl) imine, the product 6e was obtained in lower enantiomeric excess (87%) than product 4i (91%). In the case of aliphatic aldimines, imines derived from isobutyl aldehydes gave comparable results in terms of yield and enantioselectivity (4k, and 6f). The products 6g, 6h, showed better asymmetric induction than products 4l, 4m. The dibenzocycloheptene protecting group was critical for asymmetric induction in 6i, the product was obtained in 93% ee, while product 4n gave only 48% ee. Overall 5H–dibenzo[a,d]cyclohepten-5-amine derived imines gave better asymmetric induction in the resulting products than benhydryl imines, while yields for both protecting groups were comparable. We hypothesized that the dibenzocycloheptene protected imines could have enhanced CH-π20 and π-π21 stacking interactions with catalyst Mg(1a)2 in transition state, which lead to better asymmetric induction in the resulting products.
Table 3.
The catalytic asymmetric addition of diphenylphosphine oxide to imines 5
![]() | ||||
|---|---|---|---|---|
| entrya | R | product | yield (%)b | ee (%)c |
| 1 | Ph (5a) | 6a | 90 | 99d |
| 2 | 4-MeOC6H4 (5b) | 6b | 92 | 92 |
| 3 | 2-MeC6H4 (5c) | 6c | 92 | 96 |
| 4 | 4-FC6H4 (5d) | 6d | 95 | 96 |
| 5 | 2-Furyl (5e) | 6e | 98 | 87 |
| 6e,f | iso-Propyl | 6f | 72 | 85 |
| 7d | n-Propyl | 6g | 73 | 80 |
| 8d | n-Butyl | 6h | 84 | 74 |
| 9d | Ph-(CH2CH2) | 6i | 86 | 93 |
General conditions: 2.0 equiv of imine and 1.0 equiv of the phosphine oxide with 5 mol % Mg(1a)2.
Isolated yield.
Determined by chiral HPLC analysis.
Using the optimized conditions, with Ca(1a)2 as the catalyst a 95% yield and 90% ee of 6 was obtained, and with ‘H(1a) washed with HCl’ a 90% yield and 79% ee of 6 was found.
Imine was generated in situ.
(S)-H(1a) was used to synthesize Mg(1a)2.
We have also shown that the benzhydryl group can be removed with good yield to provide the α-amino phosphine oxide 7a, with preserved enantioselectivity (Scheme 1). Protection of the amine with a Boc group allowed for product 8a, which was used to deterimine the absolute stereochemistry of the chiral phosphine oxide through X-ray crystallographic analysis (see Supporting Information).
Scheme 1.
Deprotection of 4a and determination of absolute configuration
The reactivity of various aliphatic/aromatic phosphine oxides22 was evaluated for hydrophosphination reaction (table 4). To our satification, aromatic phosphine oxides bearing either electron withdrawing (entries 1 and 2) or electron donating substituent (entries 3 and 4) underwent nucleophilc addition to N-benzhydryl imine affording products in good yield and excellent enantioselectivity. However, aliphatic phosphine oxides such as benzyl, di t-butyl and diethyl phosphine oxide as well as sterically demanding aromatic phosphine oxide (2,6-dimethyl phosphine oxide) did not undergo phospha-Mannich reaction under optimized conditions.
Table 4.
The catalytic asymmetric addition of aromatic phosphine oxides (3b–e) to N-benzhydryl imine.
![]() | ||||
|---|---|---|---|---|
| entrya | R | product | yield (%)b | ee (%)c |
| 1 | 4-ClC6H4 (3b) | 9b | 93 | 84d |
| 2 | 4-FC6H4 (3c) | 9c | 88 | 84 |
| 3 | 4-MeOC6H4 (3d) | 9d | 79 | 87 |
| 4 | 4-MeC6H4 (3e) | 9e | 63 | 89 |
General conditions: 1.2 equiv of imine and 1.0 equiv of oxide 3b–e with 5 mol % Mg(1a)2.
Isolated yield.
Determined by chiral HPLC analysis.
With CH2Cl2 as a the solvent a 78% yield and 80% ee of 9b was obtained, while using Ca(1a)2 as the catalyst resulted in a 77% yield and 79% ee of 9b.
In summary, we have developed a new method where diphenylphosphine oxides can be added to imines with excellent yield and enantioselectivity catalyzed by a chiral phosphate salt. This method provides simple and direct access to chiral α-amino phosphine oxides. In this study, we have examined addition of diphenylphosphine oxide to two different N-substituted imines. The imines synthesized from 5H–dibenzo[a,d]cyclohepten-5-amine provided better enantioselectivity than those derived from benzhydryl imines, while both imines gave comparable yields. Also, the substituted diphenylphosphine oxides were excellent nucleophiles obtaining chiral α-amino phosphine oxides in good yield and enantioselectivity.
Supplementary Material
Acknowledgments
We thank the National Institutes of Health (NIH GM-082935) and the National Science Foundation (NSF-0847108) for financial support.
Footnotes
Supporting Information Available. Characterization, chiral HPLC conditions, experimental preparation and spectra is available free of charge via the Internet at http://pubs.acs.org.
References
- 1.For reviews on biological activity of α-Amino phosphonic acids and their derivatives, see: Kafarski P, Lejczak B. Phosphorus sulfur silicon Relat Elem. 1991;63:193–215.Romanenko VD, Kukhar VP. Chem Rev. 2006;106:3868–3965. doi: 10.1021/cr051000q.
- 2.(a) Atherton FR, Hassall RW, Lambert RW. J Med Chem. 1986;29:29–40. doi: 10.1021/jm00151a005. [DOI] [PubMed] [Google Scholar]; (b) Huber JW, Gilmore WF, Robertson LW. J Med Chem. 1975;18:106–108. doi: 10.1021/jm00235a024. [DOI] [PubMed] [Google Scholar]; (c) Pratt RF. Science. 1989;246:917–919. doi: 10.1126/science.2814513. [DOI] [PubMed] [Google Scholar]
- 3.Huang J, Chen R. Heteroat Chem. 2000;11:480–492. [Google Scholar]
- 4.(a) Allen MC, Fuhrer W, Tuck B, Wade R, Wood JM. J Med Chem. 1989;32:1652–1661. doi: 10.1021/jm00127a041. [DOI] [PubMed] [Google Scholar]; (b) Bird J, De Mello RE, Miles-Williams AJ, Rahman SS, Ward RW. J Med Chem. 1994;37:158–169. doi: 10.1021/jm00027a020. [DOI] [PubMed] [Google Scholar]; (c) Smith WW, Bartlett PA. J Am Chem Soc. 1998;120:4622–4628. [Google Scholar]
- 5.Patel DV, Rielly-Gauvin K, Ryono DE, Charles AF, Rogers WL, Smith SA, Deforrest JM, Oehl RS, Petrillo EW., Jr J Med Chem. 1995;38:4557–4569. doi: 10.1021/jm00022a022. [DOI] [PubMed] [Google Scholar]
- 6.For review, see: Merino P, Marqués-López E, Herrera RP. Adv Synth Catal. 2008;350:1195–1208.Ordóńez M, Rojascabrera H, Cativiela C. Tetrahedron. 2009;65:17–49. doi: 10.1016/j.tet.2008.09.083.
- 7.For Organocatalytic hydrophosphonylation, see: Joly GD, Jacobsen EN. J Am Chem Soc. 2004;126:4102–4103. doi: 10.1021/ja0494398.Akiyama T, Morita H, Itoh J, Fuchibe K. Org Lett. 2005;7:2583–2585. doi: 10.1021/ol050695e.For recent work on hydrophosphonylation, see: Nakamura S, Nakashima H, Yamamura A, Shibata N, Toru T. Adv Synth Catal. 2008;350:1209–1212.Cheng X, Goddard R, Buth G, List B. Angew Chem Int Ed. 2008;47:5079–5081. doi: 10.1002/anie.200801173.Zhao D, Wang Y, Mao L, Wang R. Chem Eur J. 2009;15:10983–10987. doi: 10.1002/chem.200901901.For hydrophosphoynlation of ketimines, see: Nakamura S, Hayashi M, Hiramatsu Y, Shibata N, Funahashi Y, Toru T. J Am Chem Soc. 2009;131:18240–18241. doi: 10.1021/ja908940e.For therotical study on the chiral phosphoric acid catalyzed hydrophosphonylation, see: Yamanaka M, Hirata T. J Org Chem. 2009;74:3266–3271. doi: 10.1021/jo900404b.Akiyama T, Morita H, Bachu P, Mori K, Yamanaka M, Hirata T. Tetrahedron. 2009;65:4950–4956.
- 8.Bartoli G, Bosco M, Carlone A, Locatelli M, Mazzanti A, Sambri L, Melchiorre P. Chem Commun. 2007:722–724. doi: 10.1039/b613477g. [DOI] [PubMed] [Google Scholar]
- 9.Carlone A, Bartoli G, Bosco M, Sambri L, Melchiorre P. Angew Chem Int Ed. 2007;46:4504–4506. doi: 10.1002/anie.200700754. [DOI] [PubMed] [Google Scholar]
- 10.Ibrahem I, Rios R, Vesely J, Hammar P, Eriksson L, Himo F, Córdova A. Angew Chem, Int Ed. 2007;46:4507–4510. doi: 10.1002/anie.200700916. [DOI] [PubMed] [Google Scholar]
- 11.Yamakoshi K, Harwood SJ, Kanai M, Shibasaki M. Tetrahedron Lett. 1999;40:2565–2568. [Google Scholar]
- 12.Fu X, Loh W, Zhang Y, Chen T, Ma T, Liu H, Wang J, Tan C. Angew Chem Int Ed. 2009;48:7387–7390. doi: 10.1002/anie.200903971. [DOI] [PubMed] [Google Scholar]
- 13.Zhao D, Mao L, Wang Y, Yang D, Zhang Q, Wang R. Org Lett. 2010;12:1880–1882. doi: 10.1021/ol100504h. [DOI] [PubMed] [Google Scholar]
- 14.For reviews on chiral phosphoric acid, see: Akiyama T, Itoh J, Fuchibe K. Adv Synth Catal. 2006;348:999–1010.Akiyama T. Chem Rev. 2007;107:5744–5758. doi: 10.1021/cr068374j.Terada M. Chem Commum. 2008:4097–4112. doi: 10.1039/b807577h.
- 15.Hatano M, Moriyama K, Maki T, Ishihara K. Angew Chem Int Ed. 2010;49:3823–3826. doi: 10.1002/anie.201000824. [DOI] [PubMed] [Google Scholar]
- 16.Zhang Z, Zheng W, Antilla JC. Angew Chem Int Ed. 2011;50:1135–1138. doi: 10.1002/anie.201006595.Drouet F, Lalli C, Liu H, Masson G, Zhu J. Org Lett. 2011;13:94–97. doi: 10.1021/ol102625s.For reports on metal and Brønsted acid catalysis, see: Rueping M, Koenigs RM, Atodiresei I. Chem- Eur J. 2010;16:9350–9365. doi: 10.1002/chem.201001140.
- 17.Lv J, Li X, Zhong L, Luo S, Cheng JP. Org Lett. 2010;12:1096–1099. doi: 10.1021/ol1000928. [DOI] [PubMed] [Google Scholar]
- 18.Zhang Y, Lu Z, Desai A, Wulff WD. Org Lett. 2008;10:5429–5432. doi: 10.1021/ol802431v. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Hjelmencrantz A, Friberg A, Berg U. J Chem Soc, Perkin Trans. 2000;2:1293–1300. [Google Scholar]
- 20.For a review, see: Nishio M. Tetrahedron. 2005;61:6923–6950.
- 21.(a) Janiak C. J Chem Soc, Dalton Trans. 2000:3885–3896. [Google Scholar]; (b) Roesky HW, Andruh M. Coord Chem Rev. 2003;236:91–119. [Google Scholar]
- 22.Abell JP, Yamamoto H. J Am Chem Soc. 2008;130:10521–10523. doi: 10.1021/ja803859p. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.





