Indi an Jo urnal of Che mi stry Vol. 42A, September 2003, pp. 2376-238 1

Ligand driven assembly of a monoorganooxotin cage from the reactions of organotin precursors with phosphonate ligands

Vadapalli Chandrasekhar* & Viswanathan Baskar

Department of Chemistry, Indian Institute of Technology, Kanpur, Kanpur 208 016, India

Received 27 March 2003

The reaction of n-BuSn(OH hCI wi th I-BuP(O)(OH}z in a 1:2 ratio in refiu xi ng tol uene o r stirring in acetoni trile at room temperature form s the monoorganooxotin cage, I (n-BuSn}zO{ 0 2P(OH)-I-Bu} 4 h 1 in good yie lds. The reaction conditions have been varied to optimi ze the y ield of the oxo cluster. An intermediate lead ing to the forma·j ·~.n of J has been identified by :l Ip NM R spectroscopy. Compound 1 can be used as a sing le-source precursor for the formalion of SnP20 7 by a low tem­perature decomposi tion process at 7S(J'C.

Cl usters, cages, rings and coordi nation polymers contai ning organotin units are a topic of considerable interest in view of the large structural diversity that ex ists in these classes of compounds 1,2. In the assem­bly of these compou nds, in most instances, the orga­notin precursor assumes the structure-directing role and steers the reaction towards an eventual product. Thus, the reactivity differences between mono-, di­and tri-organotin precursors towards a common reac­tant such as a carboxy lic acid lead to the formation of varied products whose molecular structures are con­siderably different '. It is surprising to note that such reactiv ity di ffere nces persist even among closely re­lated organotin prec ursors such as the monoorganotin compounds, Il-BuSn(O)OH and n-B uSn(OH hCL Thus, for example, while n-BuSn(O)OH reacts with FcCOOH (Fc=CsH.j-Fe-CsH4) to afford a hexameric drum3 [n-BuSn(0)(02CFc)]6, an analogous reaction of II -BuSn(OH h Cl affords a tri-nuclear tin derivative4

[1I-BuSnCl(02CFc)h(0 )(OH ). We have been recently interested in the use of phosphonate li gands in the assembly of transition and main-group metal clusters and have successfully assembled a number of Cu(II) , and Zn(IT) phosphonate cagesS

.6, where the nuclearity

varied from twelve to three. We have chosen to ex­tend these studies to organotin systems because of the rich cluster chemistry involving Sn-O motifs' . Previ­ously the reac tion of II.-BuSn(O)OH has been investi ­gated with f-BuP(O)(OH)z and the isolation and structural characterization of a cage compound [(11-BuSn)20{02P(OH)-f-Bu} 4h 1 was reported7

. In view of the afore-mentioned reactivity differences between n-BuSn(OHhCl and n-BuSn(O)OH we anticipated the

formation of varied products in the reactions of these tin substrates with phosphonate ligands such as t­

BuP(O)(OHh, t-BuP(0)(OH)(OSiMe3) or (­BuP(0)(OSiMe3h. Contrary to this expectation in all the reactions we isolated the same product viz., the tetranuclear organooxotin cage [(11-BuSnhO{02P(OH)-t- Bu}4h 1 whose formation is evidently steered by the phosphonate ligand and not by the organotin precursor. Further, we have carried out a thermolysis of the cage [(n-BuSnhO{ 0 2P(OH)­I-Bu }4h and have been able to show that the latter is an excellent single source precursor for the formation of SnP20 l These aspects are discussed in the fol­lowing account.

Materials and Methods n-BuSnO(OH), n-BuSn(OHhCI and t-butyl-

phosphonic acid (Aldrich) were used without any further purification. Mono- and d is ilylated phos­phonic acid derivatives were prepared according to literature procedures9 and were characterized spectro­scopically. Melting points were measured using a JSGW melting point apparatus and are uncorrected. Elemental analyses were carried out using a Thermo­quest CE instruments model ENtIa CHNS-O ele­mental analyzer. ' H, 3'p and 119Sn NMR spectra were obtained on a JEOL-JNM LAMBDA 400 model spectrometer using CDCl3 solutions with the corre­sponding chemical shifts re ferenced to tetramethyl si­lane, 85% H3P04 and tetramethyltin respectively. 3'p and " 9Sn NMR were recorded under broad-band de­coupled conditions. IR spectra were recorded as KBr pellets on a Bruker Vector 22 Ff IR spectropho-

CHANDRASEKHAR el al.: LIGAND DRIVEN ASSEMBLY OF A MONOORGANOOXOTIN CAGE 2377

to meter operating from 400-4000 cm-I. TGA was re­corded on a Perkin-Elmer Pyris 6 Thermo Gra­vimetric Analyzer in nitrogen at a heating rate of 20DC/min.

Synthesis of the tetranuclear organotin cage {(n-BuSnhO(02P(OH)-t-Buj4h

(f) Reactions of n-BuSn(OHhCI with t-BuP(O)(OHh - Various reactions were carried out in toluene or acetonitrile in varying SniP stoichi­ometric ratios (Table 1). lllustrati ve procedures are given below.

(a) n-BuSn(OHhCI and t-BuP(O)(OHh were sus­pended in 100 mL of toluene. The suspension was stirred for 12 h at room temperature by which time the solution became almost clear. Subsequently, the solu­tion was filtered through a G-4 frit to remove small amounts of unreacted n-BuSn(OHhCI. The clear fil­trate was evaporated to dryness under reduced pres­sure to obtain a white solid. This was dissolved in 20 mL of CHCI) and was left undisturbed for 48 h at room temperature . Colorless block-like crystals iden­tified as 1 were isolated . Mp. 275"C (d). Lit: 280"C (d).7 Anal. Calcd. for C48H 116026PSSn4: C, 31.40 ; H, 6.60. Found: C 31.50; H, 6.80. 'H NMR (CDCI), ppm) 0.87 (t) (-C HTC HTC HTCH,l, 12H), 1.10 (d), 1.14 (d) (C-(CH3h 72 H), 1.26 (m), 1.36 (m) 1.64 (m) ((-CHr CHr CHr CH 3, 24 H) ; 31p NMR (CDCh, ppm) 31.1 ("J("711 19Sn_O_3Ip :229, 239 Hz), 22.4 eJ( 11 7/ 119Sn_O_3I p : 274, 286 Hz); II<JS n NMR

(CDCl), ppm) -630.4 (tt, 2J(S n-O-P) : 239,286 Hz).

(b) The reactions were carried out in a 250 mL round-bottom flask fitted with a Dean-Stark apparatus attached to a condenser for removal of water formed during the reaction as an azeotropic mixture. For each experiment, Il -B uSn(OHhCI and t-BuP(O)(OHh were suspended in 100 mL of toluene and were heated un­der reflux for 6 h. Subsequently, the solution was cooled to room temperature and filtered through a G-4 frit to remove small amounts of unreacted n­BuSn(OH)2CI. The clear filtrate was evaporated to dryness under reduced pressure to obtain a white solid . The solid was dissolved in 20 mL of CHCl3 and was left undi sturbed for 48 h at room temperature. Colorless block-like crystals of 1 were isolated .

(c) 50 mL of acetonitrile was taken in a 100 mL round-bottom flask to which was charged 1/ ­

BuSn(OH)2Cl and t-BuP(O)(OHh The resulting sus­pension was stirred for 12h at room temperature by which time a white solid got precipitated out of aceto-

nitri le. This was filtered, dried, redissolved (CHCI3 30 mL) and filtered again. The clear solution obtained on fi ltration was reduced to half its original volume and was left undisturbed for 48 h to isolate the prod­uct as colorless block-like crystals as indicated by its analytical and spectroscopic data which matched with that obtained in the first preparation (la).

(2) Reaction of n-BuSnO(OH) or n-BuSn(OHh Cl with t-BuP(O)(OH)(OSiMe3)-(a) This reaction was carried out in a 250 mL round-bottom flask fitted with a Dean-Stark apparatus attached to a condenser to remove the water formed during the reaction as an azeotropic mixture. n-BuSnO(OH) and t­

BuP(O)(OH)(OSiMe3) were taken in 100 mL of tolu­ene and heated under reflux for 6 h. Subsequently, the reaction mixture was cooled to room temperature and filtered through a G-4 frit to remove the insoluble im­purities. The clear filtrate was evaporated to dryness under reduced pressure to get a sticky white solid. This was washed twice with 10 mL portions of n­hexane and vacuum-dried to get a powdery white solid . This was dissolved in 20 mL of CHCl) and left undisturbed for 48 h at room temperature. Colorless block-like crystals were isolated.

(b) n-BuSn(OHhCI and t-BuP(O)(OH)(OSiMe3) were taken together in 100 mL of toluene and heated under reflux for 6 h. The work-up of the reaction was performed as detailed above(2a) to obtain 1.

(3) Reaction of n-BuSn(OHhCl with t-BuP(O)(OSiMe3h-n-BuSn(OHhCI and t-BuP(O)(OSiMe3h were reacted with each other by adopting a procedure as described above (2b) to ob­tain 1.

(4) Reaction of a mixture of n-BuSnO(OH) and /1-BuSn(OH)zCl with t-BuP(O)(OHh- n-BuSnO(OH), n-BuSn(OH)2Cl and t-BuP(O)(OH)2 were reacted to­gether by adopting the procedure given above(2a) to afford 1.

Powder X-ray diffractio/1 crystallography

Powder X-ray diffractogram was recorded using a Rich Seifert l sodebyflex X-ray unit (model 2002) with Cu Ka radiation and Ni filter.

Results and Discussion The reactions of I/-BuSn(OHhCI with t-

BuP(O)(OH)2 have been carried out under different stoichiometric ratios in various reaction conditions using toluene or acetonitrile as the solvent (Table 1). These reactions proceed with the elimination of hy-

2378 INDIAN J. CHEM., SEC. A, SEPTEMBER 2003

drogen chloride and/or water. Surpri singly, none of these reactions needed the ass istance of a hydrogen chloride scavenger and cage 1 is almost exclusively formed. The yield of formati on of the cage has been optimized (- 80%) and the maximum yield is ob­tained in a reaction involving a 1:2 stoichiometry of the tin and phosphorus reactants. This also corre­sponds to the actual ratio of these components in the product. Interestingly even in other stoichiometries ( 1:1 ,2:3,3:4) we were able to isolate only 1 although in sl ightly reduced yields. The reaction proceeds equally well both in conditions where the water gen­erated in the reaction is removed dynamically (by the use of a Dean-Stark apparatus using toluene as the solven t under reflux conditions; Table 1, entri es 2,4,6,8) or where it is left in the reaction medium (Ta­ble 1, entries 1,3,5,7,9). Contrary to the formation of the chlorine free oxo-ti n cage 1 in the present in stance

in a sl ightly analogous reaction invol ving BuSn(OH)2CI and FcCOOH the isolated product [n­BuSnCI(02CFc)h(O)(OH) contains chlorine substitu­ents on tin4. Similarly, the reaction of n-BuSn(OHh CI with phosprunic acids also leads to the product which contains chlorine substituents on tin 10. Other instances of the formation of varied products in reactions of n­BuSn(O)OH or n-BuSn(OH)2C1 with a common rea­gent have been recorded (Table 2).

In order to test the reactivity of the silylated phos­phonate li gands, t-BuP(O)(OH)(OSiMe3) and t­BuP(O)(OSiMe3h were reacted with n- BuSn(OHhCI. Hi gh yields of 1 are formed in both of these reactions also by the elimination of trimethyls ilylchloride or hexamethyldisi loxane (Table I, entries] 1,12).

It has been reported previously that the reaction of a mixture of n-BuSn(O)OH and n-BuSn(OHhCl with benzoic acid leads to the formation of a ladder com-

Table I-A summary of the reactions of vari ous phospho nate li gands with II-BuSnO(OH) and fI-BuSn(OHhC I

SI No. Reactants Sn: P molar Experimental conditions Yield

Organotin precursors (g) Phosphonate substrates (g) ratio Solvent Conditio ns Time (h) g %

I II -BuSn(OH hCl ( 1.22) I-BuP(O)(OH h (1.37) 1:2 toluene r. t" 12 1.80 80

2 n-B uSn(O HhCI ( 1.22) 1- BuP(O)(OHh (1.37) 1:2 toluene reflux 6 1.8 1 80

3 II - Bu~n(OH)2C I ( 1.22) I-BuP(O)(OHh (1 .37) 1:2 aceto nit rile r.t 12 1.78 78

4 n-BuSn(OH hCI ( 1.00) I-B uP(0)(OH)2 (0.84) 2:3 toluene reflux 6 !.I 8 63 _ 5 n-B uS n(OH)2CI ( 1.00) I-BuP(O)(OH)2 (0.84) 2:3 acetoni trile r.t 12 1.2 1 65

6 n-BuSn(OHhCI (1.25) I-BuP(O)(OHh (0.94) 3:4 toluene reflux 6 1.32 57

7 n-BuSn(OHhCI (1 .25) I-BuP(O)(OH)2 (0.94) 3 :4 acetonitrile r.t 12 1.36 58

8 II -BuSn(OH h CI (1 .00) I-BuP(O)(OHh (0.56) 1:1 to luene refl ux 6 0.80 43

9 n-BuSn(OHhCI (1.00) I-BuP(O)(OH h (0.56) 1: 1 acetolli tri Ie r.t 12 0.76 41

10 II-B uSnO(OH) (0.72) t-B uP(O)(OH)(OSiMe,) ( 1.46) 1:2 tol uene refl ux 6 1. 12 71

II n-B uSn(OH h Cl (0.89) I-B uP(O)(OH)(OS iM eJ) ( 1.54) 1:2 toluene reflux 6 1.2 1 73

12 n-BuSn(OHhCI ( 1.04) I-BuP(O )!GS iMe3)2 (2.40) 1:2 tol ue ne reflux 6 1.42 73

13 I1-BuSn(OHh CI (0.44) & I-BuP(O, \ OH)2 ( 1.00) 0.5:0.5:2 toluene reflux 6 1.37 83

I1-BuSnO(OH) (0.38)

a room temperature Powder X-ray diffraction crystallography Powder X-ray diffractogram was recorded using a Rich Seifert Isodebyflex X-ray unit model 2002 with Cu Ka radiation and Ni filter.

Table 2-Various structural forms obtained in the reac tion of Il-BuSn(O)OH or n-BuSn(OHhCI with carboxylic and phosphinic ac ids

SI No. Ligands Structural forms on reacti on with n-BuSn(O)(OH) II-BuSn(OH)2C1

I C.,H~-Fe-C;l-!4C(0)OH drum3 trinuclear c1uster4

2 (C6H11 h P(0)OH cube & butterfly 1 1,12 O-capped c1uster lO

3 Ph2P(O)OH O-capped c1uster J3 extended tetra- nuclear c1 usier1o

4 {t-BuhP(O)OH cube l2 crown-butterfly c1uster1o

CHANDRASEKHAR el at.: LIGAND DRIVEN ASSEMBLY OF A MONOORGANOOXOTIN CAGE 2379

pound [{n-BuSn(O)02CPh h{n-]3uSn(Cl)(02CPh)2} h containing terminal chlorine atoms, in contrast to the anticipated drum, [11-BuSn(O)02CPh]614. Accordingly it was of interest to test the reaction of I-BuP(O)(OH)c with a mixture of n-BuSn(O)OH and II-BuSn(OH)2CI to explore if other types of products can be isolated.

However, under these reaction conditions also cage 1 is formed almost exclusively (Table 1, entry 13). Thus, it appears that irrespective of whether n­BuSn(OHhCl or n-BuSn(O)OH are used as the start­ing organotin precursors, their reactions with phos­phonate ligands leads to the tetranuclear cage 1 as the sole product (Scheme J) . Such an exclusive formation

of a single cluster type involving varied reactants, stoichiometries and reaction conditions is quite un­common in organotin chemistry and points out to the rapid formation of a common intermediate. Although we were unable to obtain unambiguous evidence for the existence of an intermediate species in the present instance, recently we have shown with benzyltin compounds the existence of a half cage [(PhCH2)2Sn20(02P(OH)-t-Bu)4] that rapidly dimer­izes to a tetranuclear cage [(PhCH2)2Sn20(02P(OH)-t­BU)4h I5

. It appears that with n-butyltin derivatives in the present instance the reactivity of such an interme­diate is even quicker (Scheme 2).

4n ·8uSn(OHhCI +

81·8uP(O)(OH )2

411·8uSn(OH}zCI +

81·BuP(O)(OSiMej)2

211·BuSn(OHhCI + 211·BuSn(O)OH + 81·BuP(O)(OHh

a

[(Il·BuSnhO· {02P(OH)t.Bu}4n

411·8uSn(OHhCI +

81·Bul'(O)(OH)OSiMe3

411·8uSn(O)OH +

81·BuP(O)(OH)OSi:l1e3

Scheme I-(a) so lvenilcondition-toluene, reflux/CH,CN, r.t. eliminated products: 6H20, 4HCl. (b) solvent/condition·toluene, reflux, eliminaled products: 4H 20 , 4Me3SiCl. 2Me3SiOSiMe3. (c) solvent/condition·toluene, ref1ux , eliminated products: 2H20, 4Me3SiOSiMe}. (d) solvent/condition-toluene, refl ux, e liminated products: 6H20 , 2HCl. (e) solvent/condition·toluene, reflux, eliminated products: 4Me3SiCI,6Me3SiOSiMe3'

HO~ tBU /.B\ / OH

P ....... ><:P /....... \ o 0 ........ 0 0

".Bu"- \/ \ L-".BU Sll ____ ~Sn

/ \, 0 ./~ o ' . 0

~;o 6~/ /p"" /p",-

HO /·Bu /.Bu 01-1

Scheme 2

2380 INDIAN J. CHEM .. SEC. A, SEPTEMBER 2003

45 40 1

35

ppm

(b)

Fig. I- Ia represents the 3I p NMR of a product obtained from the reacti on of Il -BuSn(OH)2Ci and I-B uP(O)(O Hh in a 1:2 sto ich i­ometry in acetonitrile at room temperature.The reaction mixture was worked up after 4 h of the reacti on and the so lven t evaporated to give the half-cage la. The peaks marked with * represents the half-cage la with 11 9Sn satellites. I b shows the 31 P NMR of the cage 1 (after complete conversion) with 11 9Sn sate llites.

Accordingly, the 31p NMR spectrum of a product obtained from the reaction of I1-BuSn(OH)2CI and f­

BuP(O)(O Hh ( I :2, acetonitrile, room temperature, 4h) shows the presence of peaks at 32.8 and 31.2 ppm due to the half-cage 1a (and a small amount of the fully converted cage 1 (Fig. 1a). The phosphorus chemical shi fts obtained for 1a closely correspond to the half-cage [(PhCH2hSn20 (0 2P(OH)-f-Bu)4] iso­lated by us recently' s. Ia rapidly dimerizes to 1 and the 31 P NM R of the latter reveals two resonances at 31. I and 22.4 ppm (Fig. 1b). These are due to the two different types of phosphorus centers (A and B) in-

lL -----.--------'1-------., 1 1

~28. 6 -631 .1 -633. 6 ppm ~. 1

Fig . 2_ " 9Sn NMR of the cage I showing a triplet of triplets centered at - 630. I ppm

involved in bridging the di-tin motifs. The dimeriza­tion of 1a to 1 is quite rapid as indicated by our inabil­ity to obtain a clean 11 9Sn NMR for this intermediate.

The 119Sn NMR of 1 is shown in Fig. 2 and reveals a triplet of triplets centred at -630.4 ppm. The 2J(Sn­O-P) values are 239 and 286 Hz. These parameters are simi lar to those reported previousl/.

11 9Sn NMR chemical sh ifts are extremely useful as diagnostic signatures for the various structural forms of organooxotin cages and clusters. Thus, the chemi­cal shift observed for 1 is quite unique and is among the most upfield shifted resonances observed for or­ganotin compounds l.l

5.

Therlllal decomposition of Cage 1 to SnPZ0 7

Although several organooxo tin cages have been reported thus far, there have been no attempts to use h f h . f . I 16-18 t ese as precursors or t e generation 0 materIa s .

Encouraged by the favourable combination of Sn:P ratio as well as the nature of the organic groups in 1 we investigated its thermal decomposition 19. Thermo­gravimetric analysis of the organooxotin cage 1 was carried out under N2 at a heating rate of 20·C min-I and the curve obtained is shown in Fig. 3.

A three-stage weight loss precedes the final de­composition to afford a stable material. The first two weight losses are gradual while the third is much sharper. The onset of the first weight loss at 270·C is followed by the second loss at 364 ·C. A gradual de­crease in weight continues till 460·C. At this stage the total weight loss is 20.3 %. Heating the sample further results in a large weight loss of about 37.3 % in the temperature regime of 470-S20·C. After this there is

CHANDRASEKHAR el al.: LIGAND DRIVEN ASSEMBLY OF A MONOORGANOOXOTIN CAGE 2381

1110

90

80

lI'! 70 :c CO

~ 60

SO

40

30 200 400 600 800 1000

Temperature (C)

Fig. 3- Thermogravimetric analysis curve of 1.

- z:o:-b ---'- 30-'----".'00---'- 5~O ---'----::6';;-'0 -

28 (dog ..... ) 10

Fig. 4-Powder XRD pattern for the material obtained on ther­molysi s of 1.

no further appreciable weight loss even on continuing the heating to 1000°C. At this stage the final char yield is 42%.

The TGA curve can be explained by the following sequence of decomposition steps. The first step is a dehydration involving the elimination of four water molecules from a total of eight P-OH's present in the cage. The second weight loss is the decomposition of the cage framework leading to the elimination of two molecules of P20 S• The total calculated weight loss corresponding to these two steps is 19.4%, which is very close to the experimental weight loss observed (20.3%). The third and the major event occurs around 470°C and involves the elimination of twelve mole­cules of i-butene whose calculated weight loss (37.3%) matches the experiment (37%).

To analyze the material formed on thermolysis, cage 1 was calcined independently at 750°C for 15 h in flowing N2. The powder X-ray diffraction pattern (Fig. 4) of the calcined material shows the presence of a predominantly crystalline SnP20 7 phase8

• The broadening of the powder XRD peaks indicates a small percentage of other unidentified amorphous

phases of Sn-O-P networks. FT-IR analysis of the calcined material showed a broad peak centered around 1063 cm-' confirming the presence of Sn-O-P network2o.

In conclusion, we have shown a ligand driven as­sembly of the tetranuclear oxotin cage 1 in the reac­tions of phosphonate ligands with either fl­

BuSn(O)OH or n-BuSn(OH)2CI. We have also dem­onstrated the application of 1 as a single source pre­cursor for SnP20 7 by a low-temperature decomposi­tion process.

Acknowledgment We are thankful to the Department of Science and

Technology, New Delhi, India for financial support.

References Chandrasekhar V, Nagendran S & Baskar V, Coord Chem Rev, 235 (2002) I.

2 Holmes R R, Acc chelll Res, 22 (1989) 190. 3 Chandrasekhar V, Nagendran S, Bansal S, Kozee M A &

Powell DR. Angew Chem 1m Ed, 39 (2000) 1833. 4 Kumara Swamy K C, Nagabrahmanandachari S & Raghura­

man K, J organomel Chem, 587 (1999) 132. 5 Chandrasekhar V & Kingsley S, Angew Chem 1m Ed, 39

(2000) 2320. 6 Chandrasekhar V, Kingsley S, Rhatigan B, Lam M K &

Rhei ngold A L, Illorg Chem, 41 (2002) 1030. 7 Kumara Swamy K C, Schmid C G, Day R 0 & Holmes R R,

JAm chem Soc, 112 (1990) 223. 8 JCPDS 29 # 1352, 1.54184. 9 Chakraborty D, Chandrasekhar V, Bhattacharjee M, Kriitzner

R, Roesky H W, Noltemeyer M & Schmidt H G, Inorg Chem, 39 (2000) 23.

10 Kumara Swamy K C, Day R 0 & Holmes R R, Inorg Chem, 3 1 ( 1992) 4184.

II Kumara Swamy K C, Day R 0 & Holmes R R, J Am chem Soc, 109 (1987) 5546.

12 Holmes R R, Kumara Swamy K C, Schmid C G & Day R 0, JAm chem Soc, 110 (1988) 7060.

13 Day R 0, Holmes J M, Chandrasekhar V & Holmes R R, J Am chem Soc, 109 (1987) 940.

14 Chandrasekhar V, Schmid C G, Burton S D, Holmes J M, Day R 0 & Holmes R R, Illorg Chem, 26 ( 1987) 1050.

15 Chandrasekhar V, Baskar V, Steiner A & Zacchini S, Or­ganomelallics, 21 (2002) 4528.

16 (a) Brinker C J & Scherer G W, Sol-gel science (Acadamic Press, Boston), 1990. (b) Corriu R J P & Leclercq D, Angew Chem 1m Ed Engl, 35 (1996) 1421.

17 (a) Che M, Cheng Z X & Louis C, J Am chem Soc, 117 (1995) 208. (b) Mostafa M R & Youssef A M, Maler Lett, 34 (1998) 405.

18 Chandler C D, Roger C & Hampden-Smith M J, Chelll Rev, 93 (1993) 1205.

19 (a) Terry K W & Tilley T D, Chem Maler, 3 (1991) 1001. (b) Terry K W, Lugmair C G & Tilley T D, JAm chem Soc, 119(1997)9745. .

20 Mal N K, IchiKawa S & FujiWara M, Chem Commun (2002) 112.