The wide synthetic versatility of five membered rings containing phosphorus

REVISÃO

The wide synthetic versatility of five membered rings containing phosphorus

Vinicius Caliman

Departamento de Química - ICEx - UFMG - 31270-901 - Belo Horizonte - MG

Recebido em 28/4/99; aceito em 21/12/99

The chemistry of cyclopentadiene rings has been widely studied. This review article deals with a similar chemistry of new compounds containing from 1 to 5 phosphorus atoms on the ring substituting the carbon atoms. The neutral rings containing one, two and three phosphorus atoms can be used as building blocks for the synthesis of new organic compounds containing phosphorus. These rings plus the anionic ones also show great potential as ligands in coordination chemistry. The aim of this article is to show how important this new area is and how diverse the chemistry related to a single type of ring can be.

Keywords: cyclopentadiene containing phosphorus; organophosphorus; coordination compounds.

INTRODUCTION

Five membered rings containing phosphorus atoms have been synthesized since 1959. However, nearly 75% of all known rings of this kind were prepared in the last ten years, showing the importance that chemists are giving to this type of compounds. The two areas favored with the development of these rings are organic and coordination chemistry. In a general way, we intend to give the reader a complete view of the chemistry concerning these rings, presenting the synthesis, structural data, reactivity and coordination chemistry of this important class of compounds. The work described in this review is mainly concerned with five-membered organic ring systems containing one, two or three phosphorus atoms. These rings can be neutral (phospholes) or anionic species (phospholides). Mono- di- and triphospholes have already been reported in the literature, and their structure is presented in Figure 1 (A, B and C, respectively). Phospholide anions (also called phosphacyclo-pentadienyls) containing from 1 to 5 phosphorus atoms (D-H) have also been reported.

SYNTHESIS OF PHOSPHOLES, DIPHOSPHOLES AND TRIPHOSPHOLES

Phospholes

Phospholes of the type PC₄R₄R^' (R = R^' = Ph (1)) were first described by Leavitt et al.¹ in 1959 (Figure 2), who reacted the dilithium derivative 1,4-dilithio-1,2,3,4-tetraphenylbutadiene (2) with dichlorophenylphosphine PhPCl₂ to give the 1,2,3,4,5-pentaphenylphosphole (1).

Many routes to prepare phospholes have been described in the literature^2-7. The simplest and most useful method was initially proposed by Mathey⁸ and after a few improvements⁹ the method was subsequently patented¹⁰. It involves a direct dehydrohalogenation by tertiary amines of 1-halophospholenium halides (3), which can be prepared by different methods^11-14 (Figure 3).

When R is phenyl, the yields are often quite high and this allowed 1-phenyl-3,4-dimethylphosphole to be produced on a 30 kg scale. A comprehensive review by Mathey¹⁵ includes several sections dealing with the organic chemistry of these types of ring systems, where different synthetic routes to prepare phospholes can be found.

Diphospholes

Diphospholes with general formula P₂C₃R₃R^' (R = H, Bu^t, Ph; R^' = SiMe₃, P₂C₃Bu^t₃ ) are less well-known and little data have been reported in the literature to date. Mathey and co-workers¹⁶ described the first diphosphole in 1991, via silylation of the bis-(pentacarbonyl)tungsten complex anion (4) (Figure 4) to give the 1,3-diphosphole (5) containing two [W(CO)₅] fragments.

Later, Nixon and co-workers¹⁷ reported the synthesis of a dimeric compound P₄C₆Bu^t₆(6) containing two 1,3-diphosphole units linked by a P-P bond (Figure 5). Compound (6) was obtained by oxidative coupling of two (P₂C₃Bu^t₃ )^- ring anions (7) using a ruthenium complex at low temperature.

Triphospholes

Although triphosphole rings have been synthesized only recently, their organic and coordination chemistry was developed much faster than the chemistry related to diphosphole rings. Figure 6 shows the four possible structures (I-L) for triphosphole rings, however the only two triphospholes reported to date are of type (I).

The triphosphole rings reported so far have been synthesized by alkylation of a triphospholide anion (Figure 7). The anion ring (8 and 9) is treated with a bulky alkyl group BrCH(SiMe₃)₂, at room temperature, to give the triphosphole. As expected, the alkylation of the 1,2,4-triphospholide anions (8 and 9) takes place exclusively at one of the two directly bonded phosphorus atoms due to their less sterically hindered nature. The compounds are obtained in about 50% yield as orange crystals, and both triphospholes (10 and 11) have been fully characterized by several spectroscopic techniques and their structure determined by X-ray diffraction studies^18,19. This simple and reproducible synthetic route allowed the chemistry of this ring to be studied in detail.

To the best of our knowledge there is no example of tetra-and pentaphosphole. Baudler et al.²⁰ tried without success to obtain pentaphospholes P₅R (R = Me, Et) by alkylation of the P₅ anion (12) (Figure 8). It appears that the expected pentaphospholes are so unstable that they give spontaneously a mixture of polyalkylphosphanes.

STRUCTURAL DATA AND AROMATICITY OF PHOSPHOLES AND TRIPHOSPHOLES

Since the synthesis of the first phosphole, the key issue concerning these heterocyclic ring systems was their potential aromaticity. It is important to mention that the concept of aromaticity is dependent upon the criteria employed to measure it, i.e. molecular structure, chemical reactivity, UV spectroscopy. Nevertheless, aromatic phospholes require a flat ring with the electron lone pair on the phosphorus atom highly delocalised.

In 1970, Egan et al.^21,22 demonstrated by variable temperature NMR studies that the barrier to pyramidal inversion at phosphorus in 1-isopropyl-2-methyl-5-phenylphosphole has the extraordinarily low value of 16 kcal/mol. This value is ca. 20 kcal/mol lower than that for saturated 5-membered cyclic phosphines, indicating (3p-2p)p delocalisation and aromaticity in phosphole systems. Almost at the same time McPhail and co-workers^2,23, who reported the first X-ray molecular structure of a phosphole (Figure 9), presented a convincing argument against phosphole aromaticity. The 1-benzylphosphole (13) has a non-planar geometry with the phosphorus and C(6) being displaced by 0.208 Å and 1.357 Å respectively, to opposite sides of the plane through C(2)C(3)C(4)C(5).

The three following features suggested that phospholes are only weakly aromatic: (i) the pyramidal phosphorus atom (S_angles = 302.7^o), (ii) the rather long intracyclic P-C bonds (1.780 and 1.786 Å), and (iii) the C=C bond distances, C(2)-C(3) and C(4)-C(5) (1.343 Å, each) which lie close to the ethylene value (1.334 Å)²⁴. These conclusions were supported by the results of photoelectron spectroscopic studies, using a series of phospholes (14 and 16) and phospholanes (15 and 17) (Figure 10) reported later by Mathey and co-workers²⁵. These results showed that the electron lone pairs in phospholes (14) and (16) take no part in a cyclic five-membered ring conjugation since saturation of the cis-butadiene system in the five-membered ring has no noticeable influence on the energy of the lone pair. Accordingly, phospholes in their ground state conformations consist of localised diene systems and lone pairs of electrons at the phosphorus atoms. In this sense phospholes are not aromatic.

After some controversy, general agreement was reached and it is widely accepted that phospholes are only weakly aromatic. In a series of five-membered rings containing a heteroatom and a diene fragment, the following order by decreasing aromaticity was obtained, based on structural criteria¹⁵: thiophenes > pyrroles >> furans > phospholes > cyclopentadienes.

In a recent article, Quinn et al.^26,27 synthesized a phosphole having pronounced characteristics of aromaticity. It was reported the synthesis of a sterically crowded 1-(2,4-di-tert-butyl-6-methylphenyl)-3-methylphosphole which contains a noticeably flatter P atom than 1-phenylphosphole. Photoelectron spectroscopy has been explored as a tool to measure the flattening of the phosphorus pyramid in the phosphole.

When the first triphosphole was prepared in 1995, its potential aromaticity was discussed bringing back the discussion concerning the aromaticity involving five membered rings containing heteroatoms.

The chemical behavior and the molecular structure of the triphosphole ring strongly suggest a high aromatic character. Some structural data concerning mono-, di- and triphospholes are given in Table 1. The tricoordinated phosphorus atoms in this type of ring systems (attached to the -R group) are pyramidal in all cases (a + b + g < 360^o). Despite that, when going from mono- to triphospholes, there is a significant increase in the planarity of this phosphorus center (S_angles = 302.7, 320.0 and 342.3^o, respectively) which indicates that some kind of electronic delocalisation (or aromaticity) is increasing in the ring. To support the fact that the aromaticity is increasing from mono- to triphospholes, a comparison is made of the bond lengths of compounds (6) and (10), which have the same (Bu^t) substituent on the carbon. The two P-C bond lengths (a and c, see Table 1) are significantly shorter in compound (10) (1.734(4) and 1.763(4) Å, respectively) than in compound (6) (1.784(5) and 1.816(5) Å, respectively) and the P-C bond length (b, see Table 1) is longer in compound (10) (1.720(3) Å) than in compound (6) (1.673(5) Å). These observations clearly suggest that the electronic delocalisation is much stronger in triphospholes than in mono-phosphole rings. For a completely delocalised diene system the P-C bond lengths would be expected to be as equal as possible and, using the c-b (c minus b) differences as a measure of delocalisation, the triphosphole is more aromatic (0.043 Å) than the diphosphole (0.143 Å).

The 1-[bis(trimethylsilyl)methyl]-3,5-bis(trimethylsilyl)-1,2,4 -triphosphole (11), which was recently synthesised¹⁹, has a planar tricoordinate phosphorus atom. The molecular structure obtained for a single-crystal X-ray diffraction study (Figure 11) reveals the planarity of the five-membered 1,2,4-triphosphole ring system with a sum of angles at the tricoordinate phosphorus atom of 358.7^o. It is noteworthy that the shortest P-C bond length, namely P1-C1 (1.684(7) Å), involves the tricoordinate phosphorus atom P1 and is significantly shorter than P2-C2 (1.745(7) Å), P3-C2 (1.715(7) Å), or P3-C1 (1.743(7) Å), where the phosphorus atoms are dicoordinate.

The above structure is not only planar, but also aromatic, in agreement with previous conclusions drawn for planar phospholes from theoretical studies^28-33.

REACTIVITY OF MONO-, DI- AND TRIPHOSPHOLES

Phospholes

Numerous reviews describe the chemistry of phospho-les^15,34-36. The objective of this account is to summarize the important points concerning the chemistry of this type of compound. The phosphole ring undergoes a wide variety of chemical reactions, for example both electrophilic and nucleophilic reactions occur at the phosphorus atom. Diels-Alder reactions at the diene system, thermal and light induced transformations, expansion or opening of the ring framework, have also been established.

One of the most specific and synthetically useful reactions of phospholes is their cleavage by alkali metals at the exocyclic phosphorus-carbon bond³⁷ (Figure 12). The resulting planar and highly aromatic phospholide anions readily react at the phosphorus atom in a variety of ways to afford new phospholes. This synthetic route allows the replacement of any R-substituent at phosphorus, in phospholes, by a very wide range of other substituents.

Another specific reaction of the phosphole ring takes place when a nucleophilic attack at the phosphorus atom is achieved by treatment with alkyllithium compounds (Figure 13). Thus, 1-phenyl-3,4-dimethylphosphole (19) easily reacts with tert-butyllithium to give 1-tert-butyl-3,4-dimethylphosphole (20) via an intermediate phosphoranyl anion (21)³⁸. This reaction is a good example of the enhanced reactivity of phosphorus in phospholes toward nucleophilic reagents. The corresponding reaction does not appear to work with LiMe, which is less nucleophilic than LiPh.

It has also been possible to force phospholes to undergo Diels-Alder reactions at the diene system but, as a general rule, it can be stated that the dienic system of phospholes is poorly reactive. This observation is obviously correlated with the cyclic delocalisation within the ring, which in spite of its relative weakness reduces the availability of the 4 p-electrons. Nevertheless, compound (19) cleanly reacts with N-phenylmaleimide at moderate temperatures to give the expected [4 + 2] adduct (22) in 50% yield³⁹ (Figure 14).

As expected, upon oxidation, sulphurisation, quaternisation or complexation of the phosphole lone pair, the reactivity of the dienic system sharply increases.

Diphosphole

The only reported reaction concerning the diphosphole ring (6) is the cleavage of the exocyclic phosphorus-phosphorus bond by an excess of Na metal in monoglyme¹⁷ (Figure 15). As the diphosphole used in this reaction is a dimer containing two P₂C₃Bu^t₃ units, connected by the P-P bond, the sole product from this reaction is the (P₂C₃Bu^t₃ )^-(7) anion. This route has been described as the exclusive way to a pure sample of compound (7).

Triphosphole

The chemistry of triphosphole rings is just starting to develop due to its relatively recent synthesis¹⁸ and it appears to be very promising. Like phospholes, the triphospholes also suffer a nucleophilic attack when treated with alkyllithium. The reaction of 1,2,4-triphosphole (10) with alkyllithium led to addition of the alkylgroup to the P atom of the ring, adjacent to the sp³-hybridized P to give rise to the triphosphacyclopentenyl-lithium complex P₃C₂Bu^t₂ CH(SiMe₃)₂RLi(DME) (R = Me (23), Buⁿ(24)) ⁴⁰ (Figure 16).

This type of unsaturated ring anion system is known also for the compound (P₃C₂(SiMe₃)₄ ) recently described by Thelen et al.⁴¹.

The triphosphole (10) also undergoes an electrocyclisation reaction when irradiated with visible light to give a triphosphabicyclo compound (25)⁴² (Figure 17). This compound is highly fluxional in solution and undergoes rapid [1,3] suprafacial sigmatropic shifts of PX from PA to PB. This process is facilitated by the very weak P-P bond and by the low inversion barrier of the PX atom. Compound (25) has been characterized as a complex containing a [W(CO)₅] fragment (26) and by a full structurally characterized insertion product (27) formed by treatment with [PtCl₂(PEt₃)]₂ ⁴³.

The last two examples about the reactivity of the triphosphole (10) deal with sulfur and selenium elements. The ring reacts with these elements in different ways. A complex reaction mixture results from the treatment of (10) with S₈ in benzene in the presence of Et₃N and the unusual pentasulfide P₃C₂Bu^t₂ CH(SiMe₃)₂S₅(28) (Figure 18) was isolated in 28% yield and structurally characterized by NMR, mass spectrometry and by a single crystal X-ray diffraction study.

The analogous reaction with selenium unexpectedly gave a low yield of the 1,3,5-ditertiarybutyl-1,2,4-selenadiphosphole P₂C₂Bu^t₂ Se (29) isolated as yellow oil that was fully characterized by its ¹H, ¹³C, ³¹P and ⁷⁷Se NMR spectra⁴⁰.

COORDINATION CHEMISTRY OF PHOSPHOLES AND TRIPHOSPHOLES

Phospholes exhibit a great versatility as ligands in coordination chemistry. In a review article Mathey wrote⁴⁴, "For a given phosphole, no less than ten different types of complexes can be obtained".

Literature reports concerning the coordination chemistry of phospholes and polyphospholes essentially concentrate on the former ring system.

Phospholes, which can be synthesized up to 30 kg, have an extensive coordination chemistry⁴⁴. They can be considering as two, four and six electron donors. Two electrons can be donated from the phosphorus lone pair, the diene system behaves as a four electron donor, and when both coordination modes operate simultaneously the ring acts as a six electron system.

Several metal fragments have been attached to a phosphole via the phosphorus lone pair. The 1,3,5-triphenylphosphole (TPP) (30) for example reacted with 1 equivalent of [M(CO)₆] (M = Cr, Mo, W) in sealed tubes at high temperatures to give the monosubstituted compound [M(CO)₅L] (L = TPP) (31)⁴⁵ (Figure 19).

The disubstituted compound [M(CO)₄L₂] (M = Cr, Mo; L = DMPP) was also obtained when 1-phenyl-3,4-dimethylphosphole (DMPP) reacts with [M(CO)₄(piperidine)₂] in dichloromethane⁴⁶. Under relative mild conditions [Fe₂(CO)₉] also reacted with DMPP to afford [Fe(CO)₄L] (L = DMPP)⁴⁷. A series of platinum (II) complexes of the type cis-[PtX₂L₂] (X = Cl, Br, I; L = 1-R-3,4-dimethylphosphole; R = CH₃, n-C₄H₉, t-C₄H₉, C₆H₅, CH₂C₆H₅) have been prepared via the reaction of phospholes either with PtCl₂ or K₂PtCl₄ in ethanol⁴⁸. The tris-phosphole complexes of formula [PtXL₃]⁺X^-(L = phosphole ligand) were also obtained when phospholes were used in excess.

Complexes containing phospholes as four electron ligands are not very common. This type of complex seems to be very unstable due to the high reactivity of the lone pair of the phosphorus atom. In order to obtain them it is necessary to deactivate the phosphorus atom by attaching electron withdrawing substituents to it and to prevent complexation through the lone pair by steric hindrance. Working with TPP (30) and [Fe(CO)₅], Cookson et al.⁴⁵ obtained the TPP ring h⁴-bonded to an [Fe(CO)₃] fragment (32) (Figure 20).

The first complex in which the phosphole ring acted as a six-electron donor via the phosphorus lone pair and the dienic system was discovered by Braye et al.⁴⁹. Thus, the reaction of 1,2,3,4,5-pentaphenylphosphole (PPP) (1) with [Fe₃(CO)₁₂] affords complex (33) in 36% yield, among other products (Figure 21).

Later, Mathey and co-workers^50,51 showed that the main products of the reaction of [Mn₂(CO)₁₀] with various phospholes were the tridentate complexes [Mn₂(CO)₇L] (L = phospholes) analogous to (33). Using 1-phenyl-3,4-dimethylphosphole (19) and [Fe₃(CO)₁₂] in boiling toluene Mathey and Muller⁴⁷ reported the synthesis of compound (34) (Figure 22) which unlike (33) does not contain an iron-iron bond.

Mathey and co-workers⁵² also discovered that complexes of type (34) gave sandwich complexes such as (35) when heated at 150 ^oC with an excess of the phosphole (Figure 23). Compound (35) which showed a head-to-tail configuration, was fully characterized by X-ray crystal structure analysis. Depending on the starting materials, a number of mixed heterobimetallic sandwich complexes could be obtained.

The coordination chemistry of triphospholes is developing rapidly. As mentioned earlier, the first triphosphole was synthesized only recently. However the results reported so far show the great potential as ligands for this new class of compounds.

Like phospholes, triphospholes can also behave as two, four and six electron donors. The major difference are the two extra phosphorus atoms in the triphosphole rings. Each phosphorus atom has a lonely electron pair, which can be shared with an electrophilic metal center.

Figure 24 shows the reaction of triphosphole (10) with a half equivalent of [PtCl₂(PEt₃)]₂. The product h¹-trans-[PtCl₂(PEt ₃)(P₃C₂Bu^t ₂CH(SiMe₃)₂)] (36) was the first example of a triphosphole containing a metal complex. In this complex the triphosphole ring behaves as a 2e donor bonded to the metal via the sp²-hybridized phosphorus atom lying between the two carbon atoms.

Compound (36) was characterized by several spectroscopic techniques and its molecular structure was determined by an X-ray diffraction study⁵³.

Compound (10) can also behave as a 4e donor. When compound (36) reacts with another half equivalent of [PtCl₂(PEt₃)]₂ the product h¹-h¹-trans-[{PtCl ₂(PEt₃)}₂-(P₃ C₂Bu^t₂CH(SiMe ₃)₂)] (37) is formed and the second coordination of a Pt-fragment occurs unexpectedly to the only sp³-hybridized phosphorus atom (Figure 24), which is the most sterically hindered⁵⁴.

Unlike phospholes, which use the diene system, the triphosphole ligand in compound (37) donates 4e from the two phosphorus lone pairs. However, triphospholes can also use the diene system to donate 4e together with another 2e from the sp³-hybridized phosphorus atom, acting as a 6e donor.

Treatment of compound (10) with [M(CO)₄(h⁴-NBD)] (M = Cr, Mo; NBD = norbornadiene) in tetrahydrofuran, afforded the red complexes [M(CO)₃{h⁵-P₃C ₂Bu^t₂CH(SiMe₃ )₂}] (M = Cr (38), Mo (39)) in 20% yield. Likewise reacting (10) with {W(CO)₅(THF)] gives [W(CO)₃{h⁵-P₃C ₂Bu^t₂CH(SiMe₃ )₂}] (40). All three compounds were characterized by several spectroscopic techniques. Compounds (38) and (39) were also fully structurally characterized by a single-crystal X-ray diffraction study⁵⁵. Especially noteworthy is the observation that in both (38) and (39) the tricarbonil fragment is coordinated to all five atoms of the 1,2,4-triphosphole ring system, which is planar. This ring in complexes (38)-(40) can therefore be considered as donating 6e to the metal center (Figure 25).

The same ligating behavior was found in the zerovalent ruthenium complex [Ru{h⁵-P₃C₂Bu ^t₂CH(SiMe₃)₂ }(h⁴-C₈H₁₂)] (41) obtained from the displacement of naphthalene by the 1,2,4-triphosphole (10)⁵⁶.

A different type of ligation for the triphosphole ring system occurs in compound [Fe₂(CO)₆{m-P₃C ₂Bu^t₂CH(SiMe₃ )₂}] (42) where the ring system acts as a bridging ligand to the two Fe(CO)₃ units which are linked together by a Fe-Fe bond¹⁸. The 1,2,4-triphosphole is also considered to act as a 6e donor, and its geometry is considerably distorted from that of the parent compound (10).

PHOSPHOLIDE AND POLIPHOSPHOLIDE ANIONS

Phospholide and polyphospholide anions P_nC_5-nR_5-n (n = 1-5) (R = H, aryl, alkyl) are aromatic five-membered rings resulting from the successive replacements of CR units of C₅R^-₅ with phosphorus atoms. Figure 26 shows the cyclopentadienyl anion C₅R^-₅(M) and all the possible analogous compounds containing phosphorus (N-T).

All of these anions have been isolated, some only very recently, and a recent review covers extensively both the organic and coordination chemistry of some of these interesting species⁵⁷.

SYNTHESIS OF PHOSPHOLIDE AND POLYPHOS-PHOLIDE ANIONS

The best route to synthesize a phospholide ion involves the cleavage of the P-C exocyclic bond, in an appropriate phosphole system, by an alkali metal³⁷. This reaction has already been shown in Figure 12. However, in order to get pure anions without by-product contamination, this synthetic route requires the use of some special phospholes^58-61, and two examples are given in Figure 27.

It is interesting to note that other electropositive metals such as samarium and ytterbium can also cleave the P-P bond⁶². Some non-conventional phospholide anions have also been made through alkali metal induced P-P bond cleavage^58,63, as shown in Figure 28.

The 1,3-diphospholide anion (7) (see Figure 5) was first detected by Bartsch et al.⁶⁴. It was formed as a by-product in the synthesis of the 1,3,4-triphospholide anion (8) (see Figure 7) by the reaction of PºCBu^t with LiP(SiMe₃)₂, which was first described by Becker et al.⁶⁵. Later, Cowley⁶⁶ and Nixon⁶⁷ demonstrated that reduction of the phosphaalkyne, PºCBu^t, by a sodium amalgam leads to a 1:1 mixture of the di- and triphospholide anions mentioned above (Figure 29).

Recently, Mathey and co-workers⁶⁸ published the first report describing the synthesis of 1,2-diphospholide anions (43). The initial step involved the formation of the 1,2-diphosphacyclo-pentene (44) by insertion of a phosphinidene into 1,2-dihydrophosphetes (Figure 30). Compound (44) was reduced by lithium in tetrahydrofuran to give the expected compound (43) via a delocalised phosphaallylic anion.

Baudler et al.^69,70 have reported that 1,2,3-triphospholide (45), 1,2,3,4-tetraphospholide (46) and 1,2,3,4,5-pentaphospholide (47) anions result from treatment of white phosphorus (P₄) with sodium in dimethoxyethane (Figure 31), where the solvent clearly participates in the reaction mechanism.

Following the early report no more work has appeared on the tetraphosphorus species, however a specific synthetic route for a 1,2,3-triphospholide anion has been published⁷¹ as shown in Figure 32. Compound (48), which was synthesized by a specific route, reacted with potassium in dimethoxyethane by breaking the P-P bond. The delocalisation of the negative charge over the P-P-P unit in compound (49) prevents the collapse of the 5-membered ring during the subsequent cleavage of the P-Ph bonds to give (50).

A better synthetic route for the pentaphospholide anion (47) has also been reported^72,73. Sodium dihydrogenphosphide reacts with white phosphorus, in tetrahydrofuran, in the presence of 18-crown-6 to give a pure tetrahydrofuran solution of NaP₅, because the by-products tend to precipitate from the mixture (Figure 33).

STRUCTURAL DATA AND AROMATICITY OF PHOSPHOLIDE AND POLYPHOSPHOLIDE ANIONS

Structural data on phospholide and polyphospholide anions are still scarce. The only reported X-ray crystal structure analysis of a phospholide anion was published by Douglas and Theopold in 1989⁷⁴. The molecular structure of Li(TMEDA)PC₄Me₄(51) is shown in Figure 34 and features a planar ring that is h⁵-coordinated to lithium. The coordination sphere of the alkali metal is completed by a chelating tetramethylethylenediamine (TMEDA) ligand. The carbon-carbon bond distances within the heterocycle (C(1)-C(2) = 1.396, C(2)-C(3) = 1.424, C(3)-C(4) = 1.396 Å) are very similar, indicating a high degree of delocalisation. Similarly, the phosphorus-carbon bonds (P-C(1) = 1.750, P-C(4) = 1.752 Å) are significantly shorter than P-C single bonds (ca. 1.83 Å), consistent with partial double-bond character. These structural features are a direct evidence of the highly aromatic nature of the phospholyl anions.

Delocalisation is also obvious in the tetraphenyltetraphospha-fulvalene dianion (52), which is shown in Figure 35⁷⁵. The molecular structure of this compound shows that the two rings are strictly coplanar and the C-C bridge bond is rather long (1.482(5) Å), indicating only a weak conjugative interaction between the two halves of the molecule. All four P-C bonds within each ring are very similar in length, lying between 1.748(3) Å and 1.761(3) Å. The intracyclic C-C bond distance (1.401(4) Å) suggests some partial double-bond character. This data support the description of this type of ring system as aromatic.

The only other available data concerning the 1,2,4-triphospholide anion (8) is shown in Figure 36. The molecular structure of the lithium salt of the 1,2,4-(P₃C₂Bu^t ₂)^- anion was determined by Becker et al.⁶⁵ and Caliman et al.⁷⁶, however only some partial structural information has been published.

The P-C bond lengths are very similar to those recorded for the anion (PC₄Me₄)^- (51) and the dianion (P₂C₃Ph₂)₂ ^-2(52) being 1.75 Å, and the P-P bond is substantially shortened with respect to a normal single bond (2.11 Å versus 2.21 Å).

These three X-ray crystal structure analyses convincingly establish that phospholide and polyphospholide anions are highly aromatic species. A recent theoretical study by Malar⁷⁷ fully supports this conclusion. The following aromaticity order for this type of ring system has been computed (100% of relative aromaticity of (C₅H₅)^-): (P₅)^- 100.6 > 1,2,4-(P₃C₂H₂)- 95.9 > (P₄CH)^- 94.8 > (PC₄H₄)^- 90.7 > 1,3-(P₂C₃H₃)^- 88.4 > 1,2-(P₂C₃H₃)^- = 1,2,3-(P₃C₂H₂) ^- 86.0%.

REACTIVITY OF PHOSPHOLIDE AND POLYPHOSPHOLIDE ANIONS

All known reactions of phospholide and polyphospholide anions take place at phosphorus atoms. Thus, protonation by acids gives unstable 1H-phospholes and 1H-polyphospholes that instantly give dimers of various structures^78-82. Two representative cases are exemplified in Figures 37 and 38.

More generally, all the classical electrophilic species (RX, O₂, S₈, X₂, etc.) preferentially react at phosphorus to give the expected 1R-phosphole (R = alkyl, aryl, S, O, etc.), either as stable or unstable species⁵⁷. The alkylation at phosphorus by alkyl halides is by far the most useful reaction of phospholide anions (Figure 39)^37,60,83.

It is noteworthy that alkylation of polyphospholide anions usually gives unstable polyphospholes (see Figure 8) and the only two successful examples of this type of reaction are shown in Figures 4 and 7.

COORDINATION CHEMISTRY OF PHOSPHOLIDE AND POLYPHOSPHOLIDE ANIONS

The coordination chemistry of cyclopentadienyl rings containing phosphorus is concentrated on four anionic rings, specifically (i) phospholide, (ii) 1,3-diphospholide, (iii) 1,2,4-triphospholide and (iv) pentaphospholide anions. The other anion ring systems contribute with only a few examples to the literature and will not be shown here. Many reviews^44,57,84-86 discuss in detail this type of chemistry where many more examples can be found.

All these anionic ring systems have been h⁵-bonded to metal centers and due to the phosphorus electron lone pairs available, several other coordination possibilities exist.

Some rare earth metals such as samarium (II) and ytterbium (II) have recently been used to prepare bent metallocenes containing the phospholyl anion. These complexes were obtained either by reaction of the appropriate phospholide anion with the corresponding metal diiodide, or directly by cleavage of the P-P bond of 1,1-biphospholyls with the activated metallic powders (Figure 40)⁶². Single crystal X-ray structure analysis of the 2,5-diphenylphospholylytterbium derivative shows a centroid-Yb-centroid bond angle of 129^o and indicates that the geometry of the h⁵-ligated phospholyl rings is similar to that of the free ion.

Interestingly, the 1,3-diphospholide ring (7) was the only one that gave a h³-complex. Nixon and co-workers⁸⁷ showed that [NiBr₂] reacts with a mixture of di- (7) and triphospholide (8) anions to give the 18 electron nickelocene complex (53) (Figure 41). This was the first example of a complex containing one normal and one "slipped" ring in a simple metal sandwich complex containing no other attendant ligands.

An attempt to make a 19 electron metallocene by treatment of [CoCl₂] with the 1,2,4-triphospholide anion (8) led to the synthesis of compound (54) (Figure 42)⁸⁸. The complex may arise from the expected intermediate paramagnetic compound [Co(h⁵-P₃C₂Bu ^t₂)₂] by H abstraction from the ether solvent used in the reaction.

It is noteworthy that the co-condensation of metal atoms, generated by the metal vapor synthesis (MVS) techniques, with the phosphaalkyne P¾CBu^t, leads to a variety of novel organometallic sandwich compounds^89-93. Interestingly, the 1,3-di- (7) and 1,2,4-triphospholide (8) rings are often obtained as ligands in the complexes synthesized by this route (Figure 43).

Pioneering work by Scherer and co-workers⁹⁴ has established that the pentaphospholide anion (P₅)^- also shows the basic h⁵- and h⁵-h¹- coordination mode and up to four metals have been connected to the phosphorus lone pairs of a central (h⁵-P₅) unit. In addition the P₅ ring anion behaves as a bridging ligand between two iron centers in a triple-decker structure (Figure 44).

This behavior was specific for the (P₅)^- anion until recently when Hitchcock et al.⁹⁵ reported a similar triple-decker structure using the 1,2,4-P₃C₂Bu^t ₂ ring system bridging two ruthenium centers.

ACKNOWLEDGEMENTS

I thank the Brazilian Government via the CNPq and the UFMG via Pró-Reitoria de Pesquisa-PrPq, for the financial support and Professor John F. Nixon for his suggestion.

REFERENCES

1. Leavitt, F. C.; Manuel, T. A.; Johnson, F.; J. Am. Chem. Soc. 1959, 81, 3163.

2. Coggon, P.; McPhail, A. T.; J. Chem. Soc., Dalton Trans. 1973, 1888.

3. Quin, L. D.; Bryson, J. G.; J. Am. Chem. Soc. 1967, 89, 5984.

4. Campbell, I. G. M.; Cookson, R. C.; Hocking, M. B.; Hughes, A. N.; J. Chem. Soc. 1965, 2184.

5. Fongers, K. S.; Hogeveen, H.; Kingma, R. F.; Tetrahedron Lett. 1983, 24, 1423.

6. Märkl, G.; Potthast, R.; Angew. Chem. Int. Ed. Engl. 1967, 6, 86.

7. Ferao, A. E.; Deschamps, B.; Mathey, F.; Bull. Soc. Chim. Fr. 1993, 130, 695.

8. Mathey, F.; C. R. Seances Acad. Sci., Ser C 1969, 269, 1066.

9. Brèque, A.; Mathey, F.; Savignac, P.; Synthesis 1981, 983.

10. Brèque, A.; Muller, G.; Bonnard, H.; Mathey, F.; Savignac, P.; Eur. Patent. 41447, 1981; Chem. Abstr. 1982, 96, 143076x.

11. Isaacs, N. S.; El-Din, G. N.; Synthesis 1989, 967.

12. Quin, L. D.; Borleske, S. G., Engel, J. F.; J. Org. Chem. 1973, 38, 1858.

13. Mercier, F.; Holand, S.; Mathey, F.; J. Organomet. Chem. 1986, 316, 271.

14. Mathey, F.; Lampin, J. P.; Thavard, D.; Can. J. Chem. 1976, 54, 2402.

15. Mathey, F.; Chem. Rev. 1988, 88, 429.

16. Sierra, M. L.; Maigrot, N.; Charrier, C.; Ricard, L.; Mathey, F.; Organometallics 1991, 10, 2835.

17. Al-Juaid, S. S.; Hitchcock, P. B.; Matos, R. M.; Nixon, J. F.; J. Chem. Soc. Chem. Commun. 1993, 267.

18. Caliman, V.; Hitchcock, P. B.; Nixon, J. F.; J. Chem. Soc. Chem. Commun. 1995, 1661.

19. Cloke, F. G. N.; Hitchcock, P. B.; Hunnable, P.; Nixon, J. F.; Nyulászi, L.; Niecke, E.; Thelen, V.; Angew. Chem. Int. Ed. Engl. 1998, 37, 1083.

20. Baudler, M.; Akpapoglou, S.; Ouzounis, D.; Wasgestian, F.; Meinigke, B.; Budzikiewicz, H.; Münster, H.; Angew. Chem. Int. Ed. Engl. 1988, 27, 280.

21. Egan, W.; Tang, R.; Zon, G.; Mislow, K.; J. Am. Chem. Soc. 1970, 92, 1442.

22. Egan, W.; Tang, R.; Zon, G.; Mislow, K.; J. Am. Chem. Soc. 1971, 93, 6205.

23. Coggon, P.; Engel, J. F.; McPhail, A. T.; Quin, L. D.; J. Am. Chem. Soc. 1970, 92, 5779.

24. Bartell, L. S.; Bonham, R. A.; J. Chem. Phys. 1957, 27, 1414.

25. Schäfer, W.; Schweig, A.; Märkl, G.; Hauptmann, H.; Mathey, F.; Angew. Chem. Int. Ed. Engl. 1973, 12, 145.

26. Quin, L. D.; Keglevich, G.; Ionkin, A. S.; Kalgutkar, R.; Szalontai, G.; J. Org. Chem. 1996, 61, 7801.

27. Nyulaszi, L.; Keglevich, G.; Quin, L. D.; J. Org. Chem. 1996, 61, 7808.

28. Chesnut, D. B.; Quin, L. D.; J. Am. Chem. Soc. 1994, 116, 5779.

29. Nyulászi, L.; J. Phys. Chem. 1995, 99, 586.

30. Nyulászi, L.; J. Phys. Chem. 1996, 100, 6194.

31. Dransfeld, A.; Nyulászi, L.; Schleyer, P. von R.; Inorg. Chem. 1998, 37, 4413.

32. Nyulászi, L.; Inorg. Chem. 1996, 35, 4690.

33. Glukhotsev, M. N.; Dransfeld, A.; Schleyer, P. von R.; J. Phys. Chem. 1996, 100, 13447.

34. Mathey, F.; Top. Phosphorus Chem. 1980, 10, 1.

35. Quin, L. D.; The Heterocyclic Chemistry of Phosphorus, Wiley Interscience, New York, 1981.

36. Mathey, F.; Phosphorus & Sulfur 1994, 87, 139.

37. Braye, E. H.; Caplier, I.; Saussez, R.; Tetrahedron 1971, 27, 5523.

38. Mathey, F.; Tetrahedron 1972, 28, 4171.

39. Mathey, F.; Mercier, F.; Tetrahedron Lett. 1981, 22, 319.

40. Caliman, V.; Hitchcock, P. B.; Nixon, J. F.; Sakarya, N.; Bull. Soc. Chim. Belg. 1996, 105, 675.

41. Thelen, V.; Schmidt, D.; Nieger, M.; Niecke, E.; Schoeller, W.; Angew. Chem. Int. Ed. Engl. 1996, 35, 313.

42. Bachrach, S. M.; Caliman, V.; Nixon, J. F.; J. Chem. Soc. Chem. Commun. 1995, 2395.

43. Caliman, V.; Hitchcock, P. B.; Nixon, J. F.; Chem. Commun. 1998, 1537.

44. Mathey, F.; Fischer, J.; Nelson, J. H.; Struct. Bonding (Berlin) 1983, 55, 153.

45. Cookson, R. C.; Fowles, G. W. A.; Jenkins, D. K.; J. Chem. Soc. 1965, 6406.

46. Santini, C. C.; Fischer, J.; Mathey, F.; Mitschler, A.; J. Am. Chem. Soc. 1980, 102, 5809.

47. Mathey, F.; Muller, G.; J. Organomet. Chem. 1977, 136, 241.

48. MacDougall, J. J.; Nelson, J. H.; Mathey, F.; Inorg. Chem. 1982, 21, 2145.

49. Braye, E. H.; Hübel, W.; Caplier, I.; J. Am. Chem. Soc. 1961, 83, 4406.

50. Mathey, F.; J. Organomet. Chem. 1975, 93, 377.

51. Rosalky, J. M.; Metz, B.; Mathey, F.; Weiss, R.; Inorg. Chem. 1977, 16, 3307.

52. Santini, C. C.; Fischer, J.; Mathey, F.; Mitschler, A.; Inorg. Chem. 1981, 20, 2848.

53. Caliman, V.; Hitchcock, P. B.; Nixon, J. F.; Heteroatom Chem. 1998, 9, 1.

54. Caliman, V.; D. Phil. Thesis, University of Sussex at Brighton, England 1996.

55. Caliman, V.; Hitchcock, P. B.; Nixon, J. F.; Nyulaszy, L.; Sakarya, N.; Chem. Commun. 1997, 1305.

56. Hitchcock, P. B.; Nixon, J. F.; Sakarya, N.; Chem. Commun. 1996, 2751.

57. Mathey, F.; Coord. Chem. Rev. 1994, 137, 1.

58. Mathey, F.; Mercier, F.; Nief, F.; Fischer, J.; Mitschler, A.; J. Am. Chem. Soc. 1982, 104, 2077.

59. Charrier, C.; Mathey, F.; Tetrahedron Lett. 1987, 28, 5025.

60. Charrier, C.; Maigrot, N.; Mathey, F.; Organometallics 1987, 6, 586.

61. Charrier, C.; Bonnard, H.; Mathey, F.; J. Org. Chem. 1982, 47, 2376.

62. Nief, F.; Mathey, F.; Synlett 1991, 745.

63. Bevierre, M. O.; Mercier, F.; Ricard, L.; Mathey, F.; Angew. Chem. Int. Ed. Engl. 1990, 29, 655.

64. Bartsch, R.; Hitchcock, P. B.; Nixon, J. F.; J. Chem. Soc. Chem. Commun. 1987, 1146.

65. Becker, G.; Becker, W.; Knebl, R.; Schmidt, H.; Weeber, U.; Westerhausen, M.; Nova Acta Leopold. 1985, 59, 55.

66. Cowley, A. H.; Hall, S. W.; Polyhedron 1989, 8, 849.

67. Bartsch, R.; Nixon, J. F.; Polyhedron 1989, 8, 2407.

68. Maigrot, N.; Avarvari, N.; Charrier, C.; Mathey, F.; Angew. Chem. Int. Ed. Engl. 1995, 34, 590.

69. Baudler, M.; Düster, D.; Ouzounis, D.; Z. Anorg. Allg. Chem. 1987, 544, 87.

70. Baudler, M.; Hahn, J.; Z. Naturforsch. B 1990, 45, 1139.

71. Maigrot, N.; Sierra, M.; Charrier, C.; Mathey, F.; Bull. Soc. Chim. Fr. 1994, 131, 397.

72. Baudler, M.; Etzbach, T.; Chem. Ber. 1991, 124, 1159.

73. Baudler, M.; Ouzounis, D.; Z. Naturforsch. B 1989, 44, 381.

74. Douglas, T.; Theopold, K. H.; Angew. Chem. Int. Ed. Engl. 1989, 28, 1367.

75. Maigrot, N.; Ricard, L.; Charrier, C.; Mathey, F.; Angew. Chem. Int. Ed. Engl. 1992, 31, 1031.

76. Caliman, V.; Hitchcock, P. B.; Jones, C.; Nixon, J. F.; Unpublished results.

77. Malar, E. J. P.; J. Org. Chem. 1992, 57, 3694.

78. Bartsch, R.; Hitchcock, P. B.; Nixon, J. F.; J. Organomet. Chem. 1992, 433, C11.

79. Caliman, V.; Hitchcock, P. B.; Nixon, J. F.; J. Organomet. Chem. 1997, 536, 273.

80. Lauzon, G.; Charrier, C.; Bonnard, H.; Mathey, F.; Fischer, J.; Mitschler, A.; J. Chem. Soc. Chem. Commun. 1982, 1272.

81. Charrier, C.; Bonnard, H.; Lauzon, G.; Mathey, F.; J. Am. Chem. Soc. 1983, 105, 6871.

82. Lauzon, G.; Charrier, C.; Bonnard, H.; Mathey, F.; Tetrahedron Lett. 1982, 23, 511.

83. Deschamps, B.; Mathey, F.; J. Chem. Soc. Chem. Commun. 1985, 1010.

84. Nixon, J. F.; Chem. Ind. (London) 1993, 404.

85. Nixon, J. F.; Chem. Soc. Rev. 1995, 319.

86. Nixon, J. F.; Chem. Rev. 1988, 88, 1327.

87. Bartsch, R.; Hitchcock, P. B.; Nixon, J. F.; J. Organomet. Chem. 1989, 373, C17.

88. Bartsch, R.; Hitchcock, P. B.; Nixon, J. F.; J. Chem. Soc. Chem. Commun. 1988, 819.

89. Arnold, P. L.; Cloke, F. G. N.; Nixon, J. F.; Chem. Commun. 1998, 797.

90. Arnold, P. L.; Cloke, F. G. N.; Hitchcock, P. B.; Nixon, J. F.; J. Am. Chem. Soc. 1996, 118, 7630.

91. Callaghan, C; Clentsmith, G. K. B.; Cloke, F. G. N.; Hitchcock, P. B.; Nixon, J. F.; Vickers, D. M.; Organometallics 1999, 18, 793.

92. Cloke, F. G. N.; Flower, K. R.; Hitchcock, P. B.; Nixon, J. F.; J. Chem. Soc. Chem. Commun. 1995, 1659.

93. Avent, A. G.; Cloke, F. G. N.; Flower, K. R.; Hitchcock, P. B.; Nixon, J. F.; Vickers, D.; Angew. Chem. Int. Ed. Engl. 1994, 33, 2330.

94. Scherer, O. J.; Brück, T.; Wolmershäuser, G.; Chem. Ber. 1989, 122, 2049.

95. Hitchcock, P. B.; Johnson, J. A.; Nixon, J. F.; Organometallics 1995, 14, 4382.

Publication Dates

• Publication in this collection
  08 Dec 2000
• Date of issue
  June 2000

History

• Accepted
  21 Dec 1999
• Received
  28 Apr 1999