Utilizing 2,4,6-tri-tert-butylphenyl as a sterically demanding group (abbreviated as Mes*), we have been successful in isolation, characterization, physical and chemical property investigation of low-coordinate organophosphorus compounds such as those carrying P=P and P=C bonds. Diphosphene, 1-phosphaethene, 1,3-diphosphaallene, 1-phosphaallene, and 3,4-diphosphinidenecyclobutenes are discussed in terms of stereochemistry including E/Z isomerization around P=P or P=C bond and R/S resolution of rotational isomers around crowded P–C bonds with chiral HPLC column. (π-Allyl)palladium complexes ligated with 3,4-diphosphinidenecyclobutene are useful in such catalytic organic reactions as the Tsuji-Trost reaction by which from a chiral s-alcohol, the corresponding chiral amine is obtained with retention of chirality suggesting double inversion process.
The first true “phosphobenzene,” bis(2,4,6-tri-tert-butylphenyl)diphosphene (1), was prepared as the phosphorus analog of azobenzene in 1981, utilizing a sterically demanding 2,4,6-tri-tert-butylphenyl group (abbreviated as Mes*) for protection of unstable P=P bonding system . The steric protection techniques have been widely used to isolate various kinds of unusual organophosphorus compounds with such uncommon structures , . Scheme 1 lists some of the examples with the Mes* group such as phosphaalkene 2 , 1,3-diphosphaallene 3 , 1-phosphaallene 4 , and 3,4-bis(phosphinidene)cyclobutene 5 .
Results and discussion
Preparation of low-coordinate organophosphorus compounds
Diphosphene 1 is prepared by the Grignard-type reaction from 2,4,6-tri-tert-butylphenylphosphonous dichloride (6) with magnesium  (Scheme 2). Mes* and mesityl (2,4,6-trimethyphenyl or Mes) substituted diphosphene 7  can be prepared from a stable 2,4,6-tri-tert-butylphenylphosphine (8) and mesitylphosphonous dichloride (9) by elimination reaction with an organic base like DBU (1,8-diazabicyclo[5.4.0]undec-7-ene).
Phosphaalkenes are conveniently prepared by the phospha-Peterson reaction from silylphosphides and ketones or aldehydes (Scheme 3). 2-Phenyl-1-(2,4,6-tri-tert-butylphenyl)-1-phosphaethene (2) is prepared from lithium (tert-butyldimethylsilyl)(2,4,6-tri-tert-butylphenyl)phosphide (10) and benzaldehyde . The reaction can be applied for the preparation of 3,3-diphenyl-1-(2,4,6-tri-tert-butylphenyl)-1-phosphaallene (12) with diphenylketene . Alternatively, 12 can be prepared from (2,4,6-tri-tert-butylphenylphosphanylidene)(trimethylsilyl)methyllithium (11) and benzophenone by the Peterson reaction . The reaction of 11 with benzaldehyde gives 3-phenyl-1-(2,4,6-tri-tert-butylphenyl)-1-phosphaallene (4) . Interestingly, 4 can also be prepared from phosphaalkene 2 by one-carbon homologation between P=C double bond . The process is achieved by the Doering-Moore-Skattebol reaction using dichlorocarbene as a carbon source to form 2,2-dichloro-1-(2,4,6-tri-tert-butylphenyl)-3-phenyl-1-phosphacyclopropane (13), which can be transferred to a cumulene 4 by dichloride elimination with methyllithium. This one-carbon homologation reaction is useful to prepare 1,3-diphosphaallene , 1-phosphabuta-1,2,3-triene, or 1,4-diphosphabuta-1,2,3-triene .
Similarly, 1,3-bis(2,4,6-tri-tert-butylphenyl)-1,3-diphosphaallene (3) is prepared in various ways as shown in Scheme 4. Lithium silylphosphide 10 is allowed to react with 0.5 equiv. carbon dioxide to give 2,2-phosphinosiloxy-1-phosphaethene 14 as a synthetic intermediate by phospha-Peterson reaction . Elimination of siloxide with butyllithium from 14 gives 3. Starting from bis(dichlorophosphino)methane (15), another synthetic intermediate 17 is prepared by the reaction with 2,4,6-tri-tert-butylphenyllithium to bis[chloro(2,4,6-tri-tert-butylphenyl) phosphino]methane (16), followed by HCl elimination with DBU. Elimination of hydrogen chloride with potassium tert-butoxide from 17 leads to diphosphaallene 3 . As mentioned above, 3 can be prepared from diphosphene 1 by Doering-Moore-Skattebol reaction. The reaction of 1 with dichlorocarbene forms 3,3-dichloro-1,2-bis(2,4,6-tri-tert-butylphenyl)-1,2-diphosphacyclopropane (18), which can be transferred to 3 by dichloride elimination with methyllithium . This process is useful to prepare unsymmetrical 1,3-diphosphaallene .
3,4-Bis(2,4,6-tri-tert-butylphenylphosphanylidene)-1,2-bis(trimethylsilyl)cyclobutene (5) is prepared by the phospha-Cope rearrangement of 1,2-bis(2,4,6-tri-tert-butylphenyl)-1,2-bis(trimethylsilylethynyl)diphosphane (21) to 1,6-bis(2,4,6-tri-tert-butylphenyl)-3,4-bis(trimethylsilyl)-1,6-diphosphahexatetra-1,2,4,5-ene (22) at room temperature as shown in Scheme 5 , . Diphosphahexatetraene 22 is further converted to diphosphanylidenecycobutene 5 on heating in a refluxing benzene or toluene as an air-stable compound. Diphosphane 21 can be prepared from (2,4,6-tri-tert-butylphenyl)(trimethylsilylethynyl)phosphine (19) after deprotonation to lithium phosphide 20, followed by oxidative coupling with 1,2-dibromoethane at low temperature. Compound 5 is a phosphorus congener of interesting dimethylenecyclobutene in terms of a 6 π-electron skeleton and is expected as a good bidentate ligand for transition metal complex formations. The application for catalytic organic syntheses is highly expected and hereafter this fundamental framework of diphosphanylidenecyclobutene (or phosphinidenecyclobutene) is abbreviated as DPCB , .
As shown in Scheme 5, the DBCB with phenyl in place of Tms (trimethylsilyl), 1,2-diphenyl-3,4-bis(2,4,6-tri-tert-butylphenylphosphanylidene)cyclobutene (23) is prepared in a similar way employed for 5 , . Alternatively, the reaction of (phenylethynyl)(2,4,6-tri-tert-butylphenyl)phosphinous chloride (24) gives 23 by the Grignard coupling with magnesium to 1,2-bis(phenylethynyl)-1,2-bis(2,4,6-tri-tert-butylphenyl)diphosphane (25), followed by phospha-Peterson and thermolysis.
Photo-isomerization reactions around the P=P and P=C bond
Photoreaction concerning the P=P of diphosphenes is to be discussed (Scheme 6). The reaction is wavelength and temperature dependent , . With a Pyrex filter, which cuts off higher energy (λ>275 nm), E-1 shows photo-isomerization at −78°C, indicating a photo-equilibrium with Z-1 around the P=P bond, whereas at room temperature, the Z-1 goes back to E-1 most probably the ground state is high enough to exceed the energy barrier of the equilibrium, due to the large steric repulsion within the Z-1. On the other hand, irradiation without a filter at −78°C is strong enough to cleave a double bond to give 2,4,6-tri-tert-butylphenylphosphinidene (26) as an intermediate, which affords 5,7-di-tert-butyl-3,3-dimethyl-2,3-dihydro-1H-phosphindole (27) through C–H insertion into a neighboring methyl group almost quantitatively.
E/Z-Diphosphene photo-isomerization on the carbonylmetal complexes (M=Cr, Mo, W) are studied. (2,4,6-Tri-tert-butylphenyl)(2,4,6-trimethylphenyl)diphosphene (E-7) forms pentacarbonylchromium(0) complex E-28 and on exposure to room light Z-28 is obtained as a stable isomer at room temperature. The structure of Z-28 is confirmed by X-ray analysis .
Theoretical calculation for the transition states of E/Z isomerization of HP=PH (E-29 and Z-29) is examined in terms of rotation or inversion mechanism . Rotation energy (including P=P partial bond cleavage followed by rotation) is lower than inversion mechanism (movement on the H–P=P–H plane) by about 10 kcal/mol. The energy diagram is shown in Fig. 1 together with the estimated structures 30 and 31 calculated for transition states for inversion and rotation, respectively.
On the other hand, phosphaethene E-2 is isomerized to Z-2 on exposure to light and the two can be separated through a silica-gel column chromatography as shown in Scheme 7 , . Each isomer is stable at room temperature and is characterized by X-ray analysis, which shows that the Mes* group is almost perpendicular to the P=C system whereas the Ph group is on the P=C plane in both E and Z cases .
1,3-Diphosphaallene 3 and 1-phosphaallene 4 are isomerized on irradiation of light. Karsch reported the X-ray analysis for 1,3-diphosphaallene 3 and some important geometries are shown in Fig. 2, demonstrating that the two Mes* groups are almost perpendicular attached on the straight P=C=P shaft. Compounds 3 and 4 are of axial dissymmetry similar to allenes like penta-2,3-diene. The structure of 3 has been characterized by X-ray analysis as shown in Fig. 2 , which demonstrates that the two Mes* groups are almost perpendicular attached on the straight P=C=P shaft.
As expected, both racemic 3  and 4  can be separated by chiral HPLC column such as Cellulose Carbamate or (–)-PTrMA to give (–) and (+) enantiomers. Each separated fraction can be analyzed by CD spectrum. Interestingly, if the separated fractions are exposed to light, the optical activity fades away almost instantaneously, while in the dark, their activity stays unchanged even warmed up to 50°C for 16 h (for (–)-4). The rate of activity loss with light (λ>370 nm) is estimated, k=2.2×10−3 s−1 .
As an example of photo-isomerization of DPCBs, E,E-5 is submitted for photo-reaction (Scheme 7). E,Z-5 is formed on irradiation of light, but no Z,Z-5 is observed even on a long irradiation time, most probably due to large congestion within the DPCB 5 .
Iodine-induced isomerization of DPCB and phosphaalkene
Contrary to the case of the photo-isomerization of DPCB, it seems likely that the iodine-induced isomerization of E,E-5 to E,Z-5 is irreversible or the equilibrium balance is largely in favor of the E,Z-form . Thus, attempted isomerization of E,Z-5 to E,E-5 fails. Similar results are obtained for E,E-23, although the reaction is slower than that of E,E-5. A probable mechanism of the isomerization is shown in Scheme 8. Initially, a carbonium ion is formed assisted by iodine, where the bond order of the –P=C< bond is lowered to make the rotation around the phosphorus-carbon bond easier in 32. The predominant formation of the E,Z derivatives is ascribable to less steric repulsion between the Mes* group and the substituent at the cyclobutene-ring carbon in the E,Z-derivatives than that in the E,E-derivatives. A similar iodine-induced isomerization is observed in a reversed way for a simple phosphaalkene 2 (Scheme 8). The E/Z ratio in the equilibrium state is ca. 95 (E): 5 (Z) in the presence of iodine, while the photo-equilibrium ratio is 3 (E): 7 (Z) . Thus, the iodine-induced E/Z isomerism seems to be a general property of phosphaalkenes affording a thermodynamically more stable isomer as a major product.
DPCB model compounds 33–35 and the related species 36 and 37 (C4H4P2) are theoretically calculated by ab initio method using GAUSSIAN 88 Program at the HF/6-31G* level . Calculated relative energies (MP2/6-31G*//HF/6-31G*) are shown in Fig. 3, indicating that DPCB framework is the most stable among other isomers. The difference between E and Z in the model case is not large as calculated for diphosphene HP=PH (Fig. 1) in contrast to the real DPCBs stabilized with bulky Mes*.
Rotational isomers around the P–C bond
In place of Mes*, using Dbt (3,5-di-tert-butyltolyl or 2,4-di-tert-butyl-6-methylphenyl) as a protecting group, 3,4-bis(2,4-di-tert-butyl-6-methylphenylphosphanylidene)-1,2-diphenylcyclobutene (38) is prepared from (2,4-di-tert-butyl-6-methylphenyl)phenylphosphinous chloride (39) and tert-butyllithium followed by the phospha-Cope and thermolysis  via 1,2-bis(2,4-di-tert-butyl-6-methylphenyl)-1,2-bis(phenylethynyl)diphosphane (40) as shown in Scheme 9. DPCB with Dbt 38 consists of two isomers, anti-38 and syn-38, in a 2:1 ratio according to the NMR analysis at room temperature. The transition metal complexes 41 with six-group metals such as tungsten carbonyl are prepared to show that anti-41 and syn-41 are obtained in the same ratio of 2:1. The isomer syn-41 can be analyzed by X-ray crystallography, while the other racemate of anti-41 can be successfully separated by HPLC using Chiralcel OD column to indicate a 1:1 mixture of (–) and (+) rotational isomers monitored by CD spectrum . Most likely rotation around the bond between phosphorus and ipso-Dbt carbon is restricted at room temperature. Even in 5 and 23, rotation around P-Mes* bond is supposed to be highly hindered because Mes* is more bulky than Dbt.
Stereochemistry in DPCB-catalyzed organic reactions
π-Allylpalladium complexes possessing the DPCB ligand 42 are formed from E,E-23 and [Pd(π-allyl)Cl]2 followed by silver triflate (trifluoromethanesulfonate)  (Scheme 10). The triflate 42 thus obtained is stable and can be investigated by X-ray analysis, showing that the π-allyl system is perpendicular to the DPCB-plane that controls the stereochemistry.
Among many DPCB-catalyzed organic reactions , the Tsuji-Trost reaction is particular interesting in terms of stereochemistry . The reaction is useful and enables allylic OH can be replaced directly by a nucleophile such as alkyl or amino group. Particularly, in this article the following two reactions are to be mentioned as shown in Scheme 10.
From E/Z mixture of 2-buten-1-ol (7:1, mainly E), the reaction gives the same E/Z mixture of the products, indicating the stereochemistry is maintained. More interestingly, from 98.5% ee of (R,E)-4-phenylbut-3-en-2-ol the reaction gives anilino derivative of the same stereo configuration (R). The product indicates that double inversion process is involved during the reaction as suggested in Scheme 10. The fact that the allyl system is perpendicular to the DPCB-plane is playing an important role to control the reaction. DPCB complexes are stable and easy to handle . Moreover, the phenyl groups on the cyclobutene ring can be replaced by substituted phenyls such as with p-MeO and p-CF3. Various kinds of functionalized alkyl or aryl groups can be placed on the cyclobutene ring to construct oligomers or polymers. In addition, protecting groups on the phosphorus can be replaced by such as Dbt, Tip (2,4,6-triisopropylphenyl), and some others with chiral auxiliaries, although less hindered groups might reduce the stability of the DPCB , .
Sterically protected low-coordinate P=P and P=C bonds undergo E/Z isomerization on irradiation of light or in the presence of iodine. Sterically demanding groups such as 2,4,6-tri-tert-butylphenyl (Mes*) and 2,4-di-tert-butyl-6-methylphenyl (Dbt) suffer from hindered rotation around the bulky phenyl–P bond. Diphosphinidenecyclobutene (DPCB)-transition metal complexes promote catalytic organic reactions such as the Tsuji-Trost reaction, which involves double inversion process. Sterically protected low-coordinate compounds are of interest in terms both of physicochemical features and of chemical reactivity. By integrated theoretical consideration on these compounds might help in attaining a variety of goals in pure and applied chemistry.
A collection of invited papers based on presentations at the 22nd International Conference on Phosphorous Chemistry (ICPC-22) held in Budapest, Hungary, 8–13 July 2018.
The author thanks Prof. A. J. Arduengo, III at the University of Alabama, Prof. Shigekazu Ito at Tokyo Institute of Technology, Prof. Kozo Toyota and Dr. Shigeru Sasaki at Tohoku University, Prof. Fumiyuki Ozawa at Kyoto University, Prof. Shigeru Nagase at Institute for Molecular Science (Okazaki), and Prof. Yoshio Okamoto at Nagoya University for supporting the research mentioned in this article.
 M. Yoshifuji, I. Shima, N. Inamoto, K. Hirotsu, T. Higuchi. J. Am. Chem. Soc.103, 4587 (1981); 104, 6167 (1982).10.1021/ja00405a054Search in Google Scholar
 M. Yoshifuji. J. Organomet. Chem.611, 210 (2000).10.1016/S0022-328X(00)00458-7Search in Google Scholar
 M. Yoshifuji. presented in part at the joint symposium of the 20th International Symposium “Advances in the Chemistry of Heteroorganic Compounds” and the “XVII International Symposium on Selected Problems of Chemistry of Acyclic and Cyclic Heteroorganic Compounds,” November 23, 2017, Lodz, Poland; presented at the 22nd International Conference on Phosphorus Chemistry, Budapest, July 11, 2018; submitted, Phosphorus, Sulfur, Silicon, Relat. Elem. Proceedings of the 22nd ICPC.Search in Google Scholar
 M. Yoshifuji, K. Toyota, N. Inamoto. Tetrahedron Lett.26, 1727 (1985).10.1016/S0040-4039(00)98323-0Search in Google Scholar
 M. Yoshifuji, K. Toyota, N. Inamoto. J. Chem. Soc. Chem. Commun. 689 (1984).10.1039/C39840000689Search in Google Scholar
 M. Yoshifuji, S. Sasaki, N. Inamoto. Tetrahedron Lett.30, 839 (1989).10.1016/S0040-4039(01)80630-4Search in Google Scholar
 M. Yoshifuji, K. Toyota, M. Murayama, H. Yoshimura, A. Okamoto, K. Hirotsu, S. Nagase. Chem. Lett. 2195 (1990).10.1246/cl.1990.2195Search in Google Scholar
 M. Yoshifuji, K. Shibayama, N. Inamoto, T. Matsushita, K. Nishimoto. J. Am. Chem. Soc.105, 2495 (1983).10.1021/ja00346a077Search in Google Scholar
 M. Yoshifuji, K. Toyota, K. Shibayama, N. Inamoto. Tetrahedron Lett.25, 1809 (1984).10.1016/S0040-4039(01)90047-4Search in Google Scholar
 M. Yoshifuji, H. Yoshimura, K. Toyota. Chem. Lett. 827 (1990).10.1246/cl.1990.827Search in Google Scholar
 M. Yoshifuji, S. Sasaki, T. Niitsu, N. Inamoto. Tetrahedron Lett.30, 187 (1989).10.1016/S0040-4039(00)95155-4Search in Google Scholar
 M. Yoshifuji, K. Toyota, H. Yoshimura. Phosphorus Sulfur Silicon Relat. Elem.76, 67 (1993).10.1080/10426509308032360Search in Google Scholar
 H. H. Karsch, F. H. Köhler, H.-U. Reisacher. Tetrahedron Lett.25, 3687 (1984).10.1016/0040-4039(84)80105-7Search in Google Scholar
 M. Yoshifuji, S. Sasaki, N. Inamoto. J. Chem. Soc. Chem. Commun. 1732 (1989).10.1039/c39890001732Search in Google Scholar
 G. Märkl, P. Kreitmeier, H. Nöth, K. Polborn. Angew. Chem. Int. Ed.29, 927 (1990).10.1002/anie.199009271Search in Google Scholar
 M. Yoshifuji. J. Synth. Org. Chem. Jpn.61, 1116 (2003).10.5059/yukigoseikyokaishi.61.1116Search in Google Scholar
 R. Appel, V. Winkhaus, F. Knoch. Chem. Ber.120, 243 (1987).10.1002/cber.19871200218Search in Google Scholar
 A.-M. Caminade, M. Verrier, C. Ades, N. Paillous, M. Koenig. J. Chem. Soc. Chem. Commun. 875 (1984).10.1039/C39840000875Search in Google Scholar
 M. Yoshifuji, T. Sato, N. Inamoto. Chem. Lett. 1735 (1988).10.1246/cl.1988.1735Search in Google Scholar
 M. Yoshifuji, T. Hashida, N. Inamoto, K. Hirotsu, T. Horiuchi, T. Higuchi, K. Ito, S. Nagase. Angew. Chem. Int. Ed. Engl.24, 211 (1985).10.1002/anie.198502111Search in Google Scholar
 M. Yoshifuji, K. Toyota, I. Matsuda, T. Niitsu, N. Inamoto, K. Hirotsu, T. Higuchi. Tetrahedron44, 1363 (1988).10.1016/S0040-4020(01)85915-9Search in Google Scholar
 H. H. Karsch, H.-U. Reisacher, G. Müller. Angew. Chem. Int. Ed. Engl.23, 618 (1984).10.1002/anie.198406181Search in Google Scholar
 M. Yoshifuji, K. Toyota, T. Niitsu, N. Inamoto, Y. Okamoto, R. Aburatani. J. Chem. Soc. Chem. Commun. 1550 (1986).10.1039/c39860001550Search in Google Scholar
 M. Yoshifuji, K. Toyota, Y. Okamoto, T. Asakura. Tetrahedron Lett.31, 2311 (1990).10.1016/0040-4039(90)80215-8Search in Google Scholar
 K. Toyota, K. Tashiro, T. Abe, M. Yoshifuji. Heteroat. Chem.65, 549 (1994).10.1002/hc.520050521Search in Google Scholar
 K. Toyota, K. Tashiro, M. Yoshifuji, S. Nagase. Bull. Chem. Soc. Jpn.65, 2297 (1992).10.1246/bcsj.65.2297Search in Google Scholar
 M. Yoshifuji, Y. Ichikawa, K. Toyota, E. Kasashima, Y. Okamoto. Chem. Lett. 87 (1997).10.1246/cl.1997.87Search in Google Scholar
 T. Minami, H. Okamoto, S. Ikeda, R. Tanaka, F. Ozawa, M. Yoshifuji. Angew. Chem. Int. Ed.40, 4501 (2001).10.1002/1521-3773(20011203)40:23<4501::AID-ANIE4501>3.0.CO;2-KSearch in Google Scholar
 F. Ozawa, M. Yoshifuji. Dalton Trans. 4987 (2006).10.1039/b610327hSearch in Google Scholar
 F. Ozawa, H. Okamoto, S. Kawagishi, S. Yamamoto, T. Minami, M. Yoshifuji. J. Am. Chem. Soc.124, 10968 (2002).10.1021/ja0274406Search in Google Scholar
 N. Yamada, K. Abe, K. Toyota, M. Yoshifuji. Org. Lett.4, 569 (2002).10.1021/ol017169hSearch in Google Scholar
©2018 IUPAC & De Gruyter. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. For more information, please visit: http://creativecommons.org/licenses/by-nc-nd/4.0/