Environmentally Friendly Chemistry with Organophosphorus Syntheses in Focus

The use of the microwave (MW) technique has many advan­ tages in organophosphorus chemistry. MW irradiation may replace a catalyst, or simplify a catalytic system. New reactions may also become possible under MW irradiation. In most cases, microwaves just make the reactions more efficient in respect of rate, selectivity and yield. The benefits are shown via representative examples. In another field, our methods developed for the resolution of cyclic phosphine oxides and phosphinates are summarized. After deoxygenation, the racemic or optically active P­heterocycles were used as P­ligands in platinum complexes that together with other derivatives are potential catalysts. The third topic embraces the optimization of the synthesis of dronic acids/dronates. Starting from the cor­ responding carboxylic acids and using methanesulfonic acid as the solvent, 3 equivalents of phosphorus trichloride is enough as the P­reactant. Applying sulfolane as the solvent, the opti­ mum set of reactants comprises 2 equivalents of phosphorus trichloride and phosphorous acid.


Introduction
In this minireview, our recent results in environmentallyfriendly ("green") chemistry have been summarized. In the first sub-chapter the usefulness of the microwave technique is shown in organophosphorus chemistry. The second sub-chapter summarizes our results on the resolution of cyclic phosphine oxides, as precursors for P-ligands and on the synthesis of platinum complexes as potential catalysts. Catalysts are of utmost importance from the point of view of green chemistry. The last sub-chapter comprises our experiences on the optimization of the synthesis of dronic acid derivatives that are important drugs against bone diseases.

The Potential of Microwave Irradiation in Synthetic Organic Chemistry
The use of the microwave (MW) technique in organic syntheses spread fast in research laboratories, and after almost three decades it is close to gaining industrial applications [1]. These days, sophisticated MW equipment is available and utilized on a wide scale of syntheses, such as substitutions, acylations, esterifications, alkylations, C-C coupling reactions, additions, eliminations, condensations, cycloadditions, rearrangements and the formation of heterocycles [2].
It is possible that a catalyst can be omitted or simplified un der MW conditions. The most valuable benefit is when a reaction can be performed that is impossible on conventional heat ing. This may be the consequence of a so-called special MW effect [3]. The most common benefits from MW irradiation are the considerable shortening of the reaction times and the increase in the selectivity and yield. In this article, the advantages of the use of the MW technique are demonstrated within the discipline of organophosphorus chemistry. Organophosphorus compounds including P-heterocycles find applications in synthetic organic chemistry as reactants, solvents (ionic liquids), catalysts and P-ligands in catalysts, and, due to their biological activity, also as components of medicines and plant protecting agents [4][5][6]. The application of the MW technique in organo phosphorus chemistry is a relatively new field [7][8][9][10][11][12][13].
In this subchapter, the advantages of the application of the MW irradiation are shown via selected examples.

Reactions in which the Catalysts are Replaced or Simplified by MW Irradiation
We found that simple CH-acidic compounds underwent a liquid-solid phase C-alkylation by reaction with alkyl halides in the presence of K 2 CO 3 under solvent-free MW-assisted conditions. It means that the phase transfer catalyst could be substituted by MW irradiation [14,15]. This method was then extended to the alkylation of tetraethyl methylenebisphosphonate (1a), di ethyl cyanomethylphosphonate (1b) and triethyl phosphonoacetate (1c) to afford the corresponding monoalkylated products (2a-c) in variable yields (Scheme 1) [16-18]. The dialkylation of CH-acidic compounds were also elaborated under similar, but more forcing conditions [19].
Double "phospha-Mannich" condensations were also developed applying two equivalents of the formaldehyde and the >P(O)H species to one equivalent of the primary amine (Scheme 3) [32][33][34].

Scheme 3
The bis(phosphinoxidomethyl)amines (4, Y = Ph) were useful precursors of bidentate P-ligands after double deoxygenation that could be used for the synthesis of ring platinum complexes [32][33][34]. (More details in Subchapter 2. 2) The Hirao reaction is the coupling of aryl halides and dialkyl phosphites in the presence of Pd(PPh 3 ) 4 and triethylamine. Many variations were elaborated in the last two decades [35,36]. It is noteworthy that the Pd-catalyzed reaction took place efficiently in the presence of Pd(OAc) 2 without any ligand on MW irradiation (Scheme 4) [37]. The corresponding products (5) were formed in high yields (73-95%). This is the first case for the P-ligand-free accomplishment of the Hirao reaction. Arylphosphonates (5, Y 1 = Y 2 = alkoxy) may also be obtained by the MW-assisted Arbuzov reaction of aryl bromides and trialkyl phosphites [38].
We also found that in the O-alkylation of phenols, the MW irradiation and the phase transfer catalyst synergized each other [39,40]. Quaternary ammonium salts served as special alkylating agents [41]. In the MW-assisted O-alkylation of phosphinic acid, the use of a phase transfer catalyst was useful, when alkyl halides of normal or decreased reactivity were used [42,43].

Reactions Otherwise Very Slow or Impossible Under Thermal Conditions
The use of the MW technique was useful in the inverse Wittig-type reaction of 2,4,6-triisopropylphenyl-3-phospholene oxides, -phospholane oxides and -1,2-dihydrophosphinine oxides (all marked by 6) and dimethyl acetylenedicarboxylate to provide β-oxophosporanes 7. Completion under thermal conditions required a ca. 2 week's heating at 150 °C. At the same time, on MW irradiation, the reaction was complete already after 3 h at the same temperature. No solvent was used in either case (Scheme 5) [44,45]. With 2,4,6-triethylphenyl substituent, the reaction took place only on MW irradiation [44].
It is well-known that phosphinic acids (8) do not undergo direct esterification with alcohols to give phosphinates (9) (Scheme 6/A). For this, the esters of phosphinic acids (9) are usually prepared by the reaction of phosphinic chlorides (10) with alcohols in the presence of a base (Scheme 6/B) [4]. Another possibility is the synthesis by the Arbuzov reaction (Scheme 6/C) [4].
The esterification method in use (Scheme 6/B) has the drawback of using relatively expensive P-chlorides (e.g. 10). Besides this, the hydrogen chloride formed as the by-product has to be bound by a base. Consequently, the method is not atomic efficient and is not environmentally-friendly.

Scheme 8
The outcome of the reaction confirmed that indeed the alcohol is phosphinoylated and that the phosphinic acid is not alkylated. The conversion was, however, moderate which was in accord with the results of the quantum chemical calculations. It was found that the thioesterification is rather endothermic (48.5 kJ mol -1 ), and the enthalpy of activation is higher (145 kJ mol -1 ), than that for the reaction with butanol (102 kJ mol -1 ) [49]. Hence, the incomplete conversion even under MW conditions is justified.

Reactions that Became More Efficient Under MW Conditions
There are a lot of reactions within organophosphorus chemistry that became more efficient on MW irradiation [7]. The advantages include faster and more selective reactions. Besides this, in most of the cases, there is no need for solvents. Such reactions include, for example, Diels-Alder cycloadditions [52], fragmentation-related phosphorylations [53] and Arbuzov reactions as shown above in Subchapter 1.2 [38].
The hydroxy-methylenebisphosphonates are analogues of dronic acids/dronates which are used widely in the treatment of bone diseases [57]. See Subchapter 3.
In summary, the MW technique was shown to have an increasing potential in organophosphorus chemistry. In certain instances, MW irradiation may substitute catalysts. In other cases, it may make possible otherwise impossible transformations, or, as in most cases, simply promotes the reactions.

Resolution of the Optical Isomers of Cyclic Phosphine Oxides
The preparation of P-chiral phosphines in enantiopure form is of great interest, as the transition metal complexes of these P(III)ligands may be applied as enantioselective catalysts in various homogeneous catalytic reactions. A well-established route for the preparation of the optically active P-chiral phosphines is the resolution of the racemic phosphine oxides followed by deoxygenation taking place, in most of the cases, by retention [58,59].
The absolute P-configurations of the enantiomers of the 3-phospholene oxides (17) were determined by X-ray crystallography and CD spectroscopy [60,61,67,69,[74][75][76]. Crystals suitable for X-ray were obtained in the presence of the corresponding resolving agents (22)(23)(24)(25) that gave insights into the interactions between the host and guest molecules, and made possible a better understanding of the phenomenon of the mo lecular recognition.

Complexation Reactions of Cyclic Phosphines and Bisphosphines
P-heterocyclic derivatives are of interest among the P(III)compounds that are used as ligands in transition metal complexes applied as catalysts in homogeneous catalytic reactions. In the last years, the synthesis and catalytic activity of a variety of transition metal complexes incorporating mono-or bidentate 5-, 6-or 7-membered P-heterocyles were described [77]. Our research group also studied the synthesis of the platinum complexes with P-heterocyclic and bisphosphine ligands [78,79]. Many of them were tested as catalysts in the hydroformylation of styrene. In many instances, the borane complexes of the corresponding P-heterocycles that may be regarded as the pre cursors of phosphines, were also synthesized. The phosphines may be liberated from the borane complexes by reaction with secondary amines (e.g. diethylamine) [80].
Several bis(phosphinoxidomethyl)amines (39) obtained by the double Kabachnik-Fields reaction (see Subchapter 1.2) of the corresponding amine, two equivalents of paraformaldehyde and two equivalents of Ph 2 P(O)H were subjected to double deoxygenation using phenylsilane. The reaction of bisphosphines (40) so obtained with dichlorodibenzonitrile platinum(II) af forded the corresponding cis-platinum complexes (41) (Scheme 17). The structure of the platinum complexes 41a, 41c and 41f, as well as the intermolecular interactions were investigated by quantum chemical calculations, and X-ray crystallography [33,34,89].
In most of the cases, an unusual preference for the branched aldehyde was observed that makes our Pt-complexes valuable.

Optimization of the Synthesis of Dronic Acids/Dronates
1-Substituted-1-hydroxy-1,1-methylenebisphosphonic acid derivatives (dronic acids or dronates) can be used against various bone diseases. Fig. 2 shows geminal-bisphosphonates (42) and the pyrophosphate analogues (43). The methylene-function of bisphosphonates presents a basis for tailoring the biological and physiological properties through the appendage of R 1 and R 2 side-chains. In dronic acids and dronates, the R 1 group represents a hydroxy substituent. The resulting 1-hydroxy-1,1bisphosphonate arrangement provides a tridentate functionality towards binding Ca 2+ ions, and hence promotes the affinity for species responsible for the accumulation of phosphates in the bone tissues [57]. The practical synthesis of dronic acids involves the three-component reaction of a substituted carboxylic acid, phosphorous acid and a P-chloride, which is usually phosphorus trichloride, and rarely phosphorus oxychloride. The molar ratio of the components applied varied on a broad scale and the solvents applied embrace a wide spectrum including aromatics (chlorobenzene and toluene), 1,4-dioxane, 1,2-dimethoxyethane, PEG-400, n-octane, sulfolane, methanesulfonic acid (MSA) [57].
Regarding the synthesis of zoledronic acid (44), the ratio of imidazolylacetic acid, phosphorous acid and phosphorus trichloride varies significantly as reflected by the 1:1-5:2-4.6 combinations reported. The yields provided are rather variable and move on a broad scale of 24-86%. With lack of purity criteria, the yields may refer to crude mixtures consisting mainly of the dronic acid and its monosodium salt. In the literature, the ratio and the excess of the P-reactants has never been explained or commented on. In most cases, inadequate ratios and unnecessary excesses were used in the unoptimized reactions. This resulted in a 'black-box' in respect of the synthesis of dronic acids/dronates. Moreover, a large quantity of hydrochloric acid is formed by hydrolysis of the excess phosphorus trichloride. We were the first who studied, the synthesis of zoledronic and risedronic acids in detail to establish the optimum choice and ratio of the reactants. In addition, we wished to obtain insights into the reaction sequence for the formation of hydroxy-methylenebisphosphonic acids [91].
First, the reaction of imidazolylacetic acid (44), phosphorous acid and phosphorus trichloride leading to zoledronic acid (45a) was studied in MSA at 80 °C (Scheme 19) [91]. The optimal set of parameters comprised the use of only phosphorus trichloride as the P(III) species in a quantity of 3.2 equivalents. After completion of the reaction (3 h), the mixture was hydrolyzed, neutralized and the crude product containing the mixture of zoledronic acid (45a) and its monosodium salt (45a-Na) was recrystallized from aqueous hydrochloric acid to afford pure zoledronic acid (45a) in a yield of 49% in a purity of 98%.

Scheme 19
Similar experiments were carried out for the synthesis of risedronic acid (45b) applying pyridylacetic acid (46) and phosphorus trichloride (Scheme 20.) [91]. In this case, there was no need for a recrystallization at the end. The use of 3.1 equivalents of phosphorus trichloride (again in the absence of phosphorous acid) led to a yield of 74% of risedronic acid 45b. been applied in vain in earlier dronic acid syntheses, and quantities of phosphorus trichloride beyond 3.2 equivalents were unnecessary. The unjustified use of phosphorous acid and the unnecessary excess of phosphorus trichloride results in extra costs and environmental burdens (after the excess phosphorus trichloride was hydrolyzed) [91,92]. On the other hand, when phosphorus trichloride was used in a quantity less than 3 equivalents, the conversions were incomplete causing again extra costs in the production of dronic acids/dronates. Based on the above reactions, we also elaborated the synthesis of ibandronate (45c) and alendronate (45d) [93]. Etidronate (45e) [94], fenidronate (45f) [95] and pamidronate (45g) [96] were also prepared in this way. The experimental data are summarized in Table 1. It can be seen, that in certain causes the acid, while in the other instances the monoNa salt, or the diNa salt were formed.

Mechanistic Considerations on the Formation of Dronic Acids/Dronates
A possible route for the formation of dronic acids in the reaction of a carboxylic acid with phosphorus trichloride in MSA is shown in Scheme 21. Our idea is that 1 equivalent of the phosphorus trichloride is needed to convert the acid to the corresponding acid chloride (47). In MSA as the solvent, a mixed anhydride RC(O)O(O) 2 SMe (48) may also be formed from the acid chloride and a molecule of solvent, or in another way. In the second step, the intermediates (acid chloride 47 and anhydride 48) formed are reacted further with the remaining part of the phosphorus trichloride, a step that is followed by hydrolysis and pH adjustment (Scheme 21) [97].
When sulfolane was used as the solvent in the synthesis of pamidronic acid from b-alanine (49) instead of MSA, no

Scheme 22
The use of sulfolane as the solvent in the synthesis of other dronic acid/dronates and the mechanism of these reactions are under study. We also investigated the possibility of applying other P-reactants.
In summary our minireview may have provided an overview on at least three aspects of "green" chemistry performed in our laboratory, on MW-assisted organophosphorus chemistry, on the synthesis of P-ligands and Pt-complexes incorporating them to form potential catalysts, and in the field of optimizing the synthesis of dronic derivatives.