~~~~~-DICHLOROBIS(PIPERIDISE)PLATIXJM(~~)
Aug. 20, 1964
is not as great. It should be noted that the delocalization energy is 30% greater for h = 0.2 to 0.4 and k = 1.6 than for the parameters used by West and Powell ( h = 1.0, k = 0.8).5 The delocalization energy increases almost linearly with k . For smaller values of h, the slope decreases, indicating a decreased dependence on k . T h e value of h that gives the prescribed electron densities (in the HMO and OXOMO calculations) is between 0.2 and 0.4. The increased delocalization energy as h decreases also lends weight to the choice of this range for h. This range for h is close to the value 0.15 which was determined by Brown and HeffermanQ2 by reproducing with H M O calculations the electron densities predicted by "self-consistent electronegativity'] M O calculations. T h e value of k t h a t gives the observed bond orders is between 1.4 and 1.6; these values can be used in either the H M O or OAOMO calculations. The choice of the value of k in this range is made more plausible b y the increased delocalization energy calculated for the higher values of k . The range above includes the value (1.56) used by Vincow and Fraenkel to give (22) R. D. Brown and M. L. Hefferman, T r a n s . Faraday S a c , 64, 757 (1958).
[CONTRIBUTIOK FROM
THE ISTITUTO
3257
the spin density distribution observed in the e.s.r. spectra of semiquinone ions.2 3 Table IV shows the difference between the calculated quantities using the H M O and OAOMO methods. I n all cases the differences between the bond orders and electron densities are small. These differences are not large enough to make the extra calculations necessary in the O h O M 0 method worthwhile. Calculations using SCOvalues of 0.1, 0.2, 0.3, and 0.4 resulted in little change for the bond orders and electron densities. As a result, the most appropriate parameters for use in H M O and OAONO calculations of the oxocarbons are k = 0.2 to 0.4 and k = 1.4 to 1.6. The appropriateness of these parameters in the M O calculations of the squarate (C4O4c2) and rhodizonate (c606-') ions will be known when the molecular geometry of these ions has been determined. Acknowledgment.-Part of this work was supported by Grant No. DA-ARO(D)-31-124-G26from the Army Research Office, Durham. The authors wish t o acknowledge Mr. Jerry Farlow for adapting Given's method to the IBLI-7070, and Mr. Tom Hegenbarth for the independent intensity estimations. (23) G. Vincow and G. K . Fraenkel, J . Chem. P h y s . , 34, 1333 (1961).
CHIMICA GENERALE, USIVERSITA
DI
PADOVA,
PADOVA, I T A L Y ]
Rates of Substitution Reactions of Square-Planar Platinum(I1) Complexes. Reactions of trans-Dichlorobis(piperidine)platinum(II)
11.'
BY UMBERTO BELLUCO, LUCIOCATTALINI, AND ALDOTURCO RECEIVED DECEMBER 9, 1963 T h e rates of the reactions of trans-[Pt(CjHllS)2C12]with 36C1-, S O * - , and Sa- and of trans-[Pt(CSHII?;)2N02C1] (CsHllN = piperidine) with S O 2 - are reported. These are compared with the rates obtained previously for similar reactions of trans-[Pt(PEt3)2C12]. The relative increase in reactivity of NOt- in the reaction with [Pt(CjHllN)2C12] is discussed. I t is observed that the ligands in the cis position influence the trans effect of the trans ligand
The first paper of this series reports' the nucleophilic substitution reactions of W l - and NOz- with trans[Pt(PEta)zCl~]in methanol. I n this system nitrite ion is a poorer reagent than chloride ion. This is believed to result from the n-bonding of phosphine with platinum(I1) which in turn makes less important the electrophilic contribution of the entering reagent, e.g., NOZ-.
This paper reports investigations of the reactions of
trans- [Pt(C5H11N)2NOzC1] with NOz- and of trans[Pt(CSH~1N)2C1~] (C5H11N = piperidine) with T - , NOz-, and N3- in methanol. This complex was chosen because Stuart models show t h a t the steric hindrance caused by piperidine is approximately the same as that of triethylphosphine in the previously studied trans- [Pt(PEt3)2C12].Moreover cis-trans isomerization is known not to occur2 for the piperidine compound under the conditions of these experiments. Hence replacement of triethylphospliine with piperidine is expected to affect the kinetic process solely through changes of the electronic features caused by (1) P a r t I : U. Belluco, L. Cattalini, and A . Turco, J . A m . Chem. S O L ,8 6 ,
the different bonding properties of the two ligands. In the course of this investigation i t was also felt necessary to study the reaction of trans- [Pt(PEtJZCl,] with N3-. Experimental Preparation of M a t e r i a l ~ . - - t r a n s - [ P t ( C ~ H ~ ~ S )was ~ C l ~pre] pared from [Pt(SMe&C12]and piperidine in n-butyl ether.2 T h e compound was recrystallized from ethanol, m . p . 252-253'. Anal. Calcd. for CloHg2SsC12Pt: N, 6.4. C1, 16.5. Found: h-,6.3; C1, 16.7. The compound is a nonelectrolyte in methanol ( A M < 1 ohm-' cm.-a mole-'). [Pt(SMel)zCl*]was prepared by the reaction of K2PtCI4 with S(CH3)? in water solution.3 The S(CH3)2 used was obtained from the reaction between CHBI and Ka?S in a water-ethanol m i ~ t u r e . ~ trans-[Pt(CjHI1S)2( SO2>2] was obtained by the reaction of trnns-[Pt(CjHI1S)2C1P1 M) with S a S O ? (lo-' J f ) a t 30" in methanol, containing 10-2 JI toluenesulfonic acid, for 4 days. The compound was recrystallized from methanol, m . p . 258-260'. Anal. Calcd. for C 1 ~ H s 2 S a 0 4 PSt :, 12.25. Found: S ,12.1. trans-[Pt(CjH11S)2( S\;3)2] was obtained by the reaction of trans[Pt(CjHllS)2C12] ( l o w 3Jf) with SaSB ( 5 X 10-2 JI) in methanol a t 30" for 4 days. The product was recrystallized from methanol, m.p. 172-173". Anal. Calcd. for ClaHnS&: S , 2 4 . 9 . Found: S , 24.4.
226 (1964). (2) J. C h a t t , L. A. Ducanson, and L. M . Venanzi, J. Chem. Sac., 4461 (1955).
(3) C. Enebuske, J . prakl. Chem., 3 8 , 358 (1888). (4) D. S . Tarbell and C. Weaver, J . A m . Chem. Soc., 63, 2941 (1941)
UMBERTOBELLUCO, LUCIOCATTALIXI, ASD ALDOTURCO
3258
trans-[Pt( C ~ H I I N ) ~ C I N O was ~ ] obtained b y the reaction of trans-[Pt(C5HI~N)zC12] M) with N a N 0 2 (10-3 M ) in methanol in the presence of toluenesulfonic acid (10-4 ,If) a t 30" for 1 week. T h e product was recrystallized from methanol, m.p. 241-243', A, 2930 A. ( t l n n x 1420). Anal. Calcd. for C,OHT2S302C1Pt:N , 9 . 4 ; CI, 7.9. Found: S , 9 . 6 ; C1,8.0. trans-[Pt(PEts)2C1?]was prepared by the method of Jenseiis as described earlier trans-[Pt(PEt3)?(N ~ ) zwas ] prepared by the reaction of trans[Pt(PEt3)2C12] M ) with excess of S a N s ( 5 X Af) in methanol for 2 days. The product was recrystallized from methanol, m.p. 141-142". A n d . Calcd. for ClzHmNd'aPt: N , 16.3. Found: N, 16.2. Anhydrous C H 3 0 H was prepared by distillation over (CH,O),Mg. The L i W 1 was obtained from LiCl and H 3 T I by isotopic exchange. I t was recrystallized from acetone. Kinetics. trans-[Pt( C5HIIN)2CI2]-NO2-Reactions.-The reaction between trans-[Pt(CSHl~hT)2CI2] and NO2- was studied spectrophotometrically in niethanol a t 30" a t a constant ionic strength (LiN03). Preliminary spectrophoton~etrictests had shown that the nitrate ion does not react with trans-[Pt(CSH~1N)2CI2].The initial complex concentration was held constant and t h a t of L-aNO2 varied. I t was also found t h a t t r u n s - [ P t ( C j H ~ ~ N ) ~and Cl~l trans-[Pt(CsH1lN)~(NO2)2] obey Beer's law in methanol in the 3600-2400 A . region for the range of concentration used. A11 spectrophotometric measurements were made on a Beckman DK2A recording spectrophotometer, with a therrriostated cell compartment. The absorption spectra of the piperidine complexes are shown in Fig. 1. In order t o investigate the effect of
Vol. 86
coefficient of the nitro complexes, x = total concentration of nitro complexes, and a = initial concentration of dichloro complex. Th? second step of the reaction with SO.- was followed at 2800 A. by using solutions of truns-[Pt(C5H11Nj~ClSO~]. Isotopic Exchange.-The isotopic exchange experiments with 36C1- were made following procedures described in part I.' trans-[Pt(C5HllN)2C12]-N~- Reactions.-Rates of reaction between trans-[Pt(CjHllN)2Cla] and S3- have been studied in the presence of added LiSOJ t o maintain a constant ionic strength. T h e complex concentration was not varied while t h a t of N a S a covered a 15-fold range. Some runs were made in the presence of various amounts of p-toluenesulfonic acid. In every case there was no change in the observed rate constant. The absorption spectrum of tvans-[Pt(CjHIISjz(S~)z] in methanol is given in Fig. 1. T h e reaction was followed by measuring the optical density a t 2800, 3040, and 3220 A,, where the absorption of the product is considerably larger than t h a t of the starting complex. The absorption was measured against a blank of NaSJ (and p-toluenesulfonic acid when necessary) in the same concentrations as in the solution under investigation.. The reaction product is trans[Pt(C5HIIN)2( X3)?]; however, the rate-determining step must be the substitution of the first chlorine atom, leading t o the formation of the more reactive t r a n s - [ P t ( C j H ~ 1 ~ ) 2 C I N ~ ] . trans-[Pt(PEts)2Cla]-N3Reactions.-The rates of these reactions were studied a t 55" in methanol. T h e procedure followed was the same as t h a t described above.
Results and Discussion T h e values of k&sd for all the reactions studied are given in Table I. The observed rate constants kobsd were obtained by treating the kinetics as a pseudo-
t
f
Wavelength,
1
Fig 1.-The absorption spectra of tlle piperidine complexes : a, trans-[Pt(C5H~1S)~C12] 7 X M; b, trans-[Pt(CsH~iN)z(xo2)47 x 10-1 M; c, t y a n ~ - ~ ~ t ~ ~ 57 x ~ ~10-4 l l J sI ; ~ 2 ~ ~ ~ 0 2 ~ d , trans-[Pt(CjHl~S)2(N8)11 .If. CHsO-, some runs were made in the presence of sodium methoxide. It was found t h a t addition of CH30Na (10-3 Jf) has no effect on the reaction rate. In other experiments p-toluenesulfonic acid was added t o the reaction mixture. m'ith the addition of small amounts (3.35 x M)of p-toluenesulfonic acid. the results changed dramatically. The rate increased and the spectral changes accompanying the reaction were quite different from t h a t of a reaction containing no added acid. T h e effect of acid on this reaction is being studied and the results of this investigation will be reported later. Applying these observations t o the preparative scale it was possible to isolate the intermediate trans-[Pt( C5HllL7)2C1X02] by the method described above. The spectrum of this compound is given in Fig. 1. The consumption of trans-[Pt( C5H11N)2C12] to give trans-( Pt(C5HllhT)~C1NO~] was followed by measuriug the rate of change of optical density a t the wave length of the isosbestic point a t 2665 A . (Fig. 1). T h e simple relation
was used t o calculate the concentration of trans- [Pt(CsHIIS)SClnl during the course of the reaction, where el = molar extinction coefficient of thc dichloro compound, e> = molar extinction ( 5 ) K. A . Jensen, Z. nnorg. niircriz. Chefis., 229, 225 (1936)
time, minutes Fig 2 -Typical first-order plots for the experimental data for trans-[Pt(PF;t3)~C12]-Ss-reactlons with [complex] = 1 58 X lo-' M , solvent = CH30H, temperature = 30" a , 0 10 31 NaNa, b, 0 O i 5 111 N a S a , c, 0 05 J1 S a S d , d , 0 025 .lIX a S ? , e, 0 0125 M liaS8
first-order process. The kobsd values of Table I were obtained graphically according to the first-order rate law (Fig. 2 ) log
a ~
a -x
-
-
kt ~
2.303
The variation of the pseudo-first-order rate constants, k o b s d , with initial concentration of reagent was con-
~ ~ U ~ S - D I C H L O(PIPERIDINE) ROBIS
Aug. 20,1964
PLATISUM
(11)
3259
TABLE I
TABLEI1
RATESFOR REACTIONS OF PLATXNUM( 11) COMPLEXES WITH 36Cl-, Nos-, ASD Na- IN C H 3 0 H AT 30"
KATE CONSTAXTS FOR THE REACTIONS O F PLATINUM(I1) COMPLEXES WITH 36Cl-, NOz-, ASD N3- IN CH,OH
Complex concn
,M
trans- [ P ~ ( C S H ~ I N ~ C I Z ] 0 005 005 005
005 005 0007 0007 0007
0007
Reactant concn., M
0.010 033 ,050 ,075 ,152 XOz- (NaNOz) .01 .02 .03 ,05 Xz- (NaNOa) ,0175 ,0375 'T1- (LiCI)
00094 00094 050 00094 050 00094 00070 050 00094 075 00094 .10 trans- [ P ~ ( C S H I I N ) L X N O Z ] 0 000625 NOZ- (NaNOz) 0,005 000600 ,010
000639 000632 000710 000651 trans- [Pt(PEts)zCliIb 0 000138 000168 000158 000158 000158 a kobsd =
,02 ,015
0.265 ,242 225 ,200 ,123
,045 ,035 025 ,005 ,0925
,0725 060 ,060 ,060
Reactant
T e m p , kl X 105. k? X 104, 'C sec -1 M-1 sec - 1
X 108, sec. -1
30 30 30 30 30 55 55 55
0.24F
0 442 610 ,850 1.505 0.32 52 .83 1.185 0.98 2.09 2.60 2.65
2.61
,035
4.29
.01
5.48
0.05 ,045
0.96 1.42
,050
1.83 1.56 2.62 3.61
0.0126 ,025 ,050 075 .10
0.0973 ,085 ,060 ,035 .01
0.30 0.50 0.90 1 33 1.73
,035
Complex
kobd
,040 ,020 ,005
,035
Xa- (NaNs)
Added LiXOa concn., M
a
15 15 15
7 7 4 4 47 4 5 10
20 9 53 57 2 0 4 16
4 25 3 52 97 2
3
U. Belluco, L. Cattalini, and A. Orio, Gazz. chim. ital., in b Data taken from ref. 1.
press.
The anomalous value for kl in the reaction between trans- [Pt(PEts)ZCl2] and K3- is possibly outside the limits of experimental error. There is no apparent reason for this discrepancy and a careful investigation a t very low azide concentration will be necessary before any theoretical significance can be attached to the observation.
t
R/ [complex]
Reaction in CHIOH a t 55"
sistent with the expression
since the plots of kobsd v s . [ Y ] give straight lines with nonzero intercepts, as in Fig. 3. Equation 1, where kl is a first-order rate constant for solvent-controlled reaction, conforms t o the general law found for the substitution reactions of platinum(11)square-planar complexes.6 T h e over-all reaction for the kinetics discussed here can be summarized as
+Y
J. kz
trans- [PtLZCIY]
+Y
4
fast
trans- [PtL2ClY]
The calculated rate constants k1 and k2 are presented in Table 11. I t was necessary to study the kinetics of the two complexes, trans- [Pt(PEt&Xn] and trans[Pt(C6HllN)2X2], a t different temperatures; however, this fact does not complicate the discussion. The data in Table I1 show that all the reactions of trans[Pt(C5H1lN)2Cl2]proceed a t a higher rate than those Of trans- [Pt(PEts)zCl2]. (6) F . Basolo and R. G. Pearson, "Progress in Inorganic Chemistry," Vol. 4, F. A . Cotton, E d . , Interscience Publishers, I n c . , Xew York, N. Y., 1962, p. 392.
[Y -].lo2 Fig. 3 -The variation of kobsd with reagent concentration for trans-[Pt(C5H1,N)~C1~] X- reactions: a, X - = N1-; b,
x-
=
so2-;c , x-
+
=
36c1-.
I n the case of the reaction with trans-[Pt(PEta)zClz], C1- should act as a better nucleophile than in the reaction with trans- [Pt(C5Hl1N)2Cl2]since the acceptor properties of platinum are expected to be enhanced by the presence of the phosphine molecule in the complex. The fact that the opposite occurs tells us t h a t rearrangement of the square configuration to give the assumed trigonal bipyramidal transition state apparently involves a larger free energy change in the case of the phosphine complex, and the most obvious explanation is t h a t bond making is more developed in this case. Additional reasoning is necessary in order to explain the larger increase of the reaction rate with NOz- in going from trans- [Pt(PEtr)2Clz] to trans- [Pt(C5HIIN)2C12]. The most important effect caused by the replacement of the two phosphine molecules by two piperidine molecules is probably the enhancement of
3260
UMBERTO BELLUCO. LUCIOCATTALINI, A X D XLDOTURCO yC5Hll
Cl NC5H11
products
Fig. 4.-The
transition state for [Pt(CsHllS)2ClaN02].
the ability of platinum to donate its a-electrons to the incoming reagent, thus favoring the electrophilic attack of the nitrite ion. In other words N O a - can make relatively stronger bonds in the activated complex of the reaction with the piperidine compound. I t has already been shown t h a t NOz- reacts rapidly with negatively charged platinum(I1) complexes but slowly with positively charged ~ o m p l e x e s . ~Thus its effectiveness is apparently due mainly to its electrophilic ability. T h e previous paper’ in this series gives a short explanation of the biphilic nature of this reagent. It is interesting to note t h a t the results reported here were obtained with a neutral complex, where the formal charge on the central atom has been changed by varying the nature of two neutral ligands. Nevertheless the effect on the reaction rates is of the same type as t h a t obtained by Gray’ by replacing neutral ligands with charged ligands, for example, going from trans- [Pt(NH&C12] to [Pt(r\”a)aCl]+. It is tempting to conclude that on the one hand our results are in agreement with the biphilic character of K O p - and on the other hand they are in agreement with the general picture which looks at the phosphine molecule as a u-donatinga-accepting ligand. The results of the reactions of trans- [Pt(C5H11N)2C1N02] with C1- show t h a t the replacement of one chloride by nitrite as a ligand in trans- [Pt(CSH11N)2Cla] causes a significant decrease of the reaction rate with C1-. However, the opposite is true for the reaction with NOz-; the chloronitro complex reacts faster than the dichloro. The presence of N O a - in the position trans to the leaving chloride ion might have been expected to increase the reaction rate in both cases, owing to the larger trans effect of nitrite groups as compared to chloride. The composition of the transition state for the reaction between trans- [Pt(C5H11N)2C1z] and NOz- is the same as that of the reaction between trans-[Pt(C5H11N)zClNOa] and C1-. In both cases, there are two chlorines and a nitro group in the trigonal plane and, if all bonds were well developed, the transition state would be identical (Fig. 3 ) . Thus the slower reaction in the second case is in agreement with the observation that the nitro group stabilizes the square-planar complex more effectively than does CI-.* Since C1- reacts faster with trans[ Pt(C5HllX)3C12 ] than with trans- [ Pt ( C5HllN)3ClNO?],the nitrite could not have stabilized the transition state as much as i t did the ground state. The reactivity order toward NOp-, trans- [Pt(C5HllN)2NOsC1] > trans- [Pt(CiHl,N)zC12],also suggests t h a t the introduction of a second nitrite in the transition state has a greater effect than introduction of the first nitrite group i n the substrate. In other words, the substitution of a chloride by a nitrite group in the penta( 7 ) H. B. G i a y and I NOz-, showing that nitrite destabilizes the pentacoordinate transition state relative to chloride. This behavior is opposite to t h a t observed for the case of the piperidine complex trans- [Pt(CSH11N)2C12]and perhaps indicates that relevant platinum orbitals are no longer available to NO2- in the bipyramid when phosphine molecules are bound in the complex. The reactions with N3- indicate t h a t this ion behaves as a nucleophilic reagent. The azide ion has an empty a-antibonding orbital which could confer electrophilic properties to this reagent by receiving electrons from the metal. However, this a-orbital is mainly localized on the central n i t r ~ g e nand, , ~ when the attack occurs through one of the terminal nitrogens, i t will shift toward the opposite end of the triatomic group,I0 thus causing the potential electrophilic character to decrease further. The observed rate constants for the substitution by N3- show that this ion behaves similarly to C1-. The increase of the reactivity of these ions on going from the reactions with trans- [Pt(PEt3)2C12]to those with trans- [Pt(CsH1lN)?C19]is approximately the same. T h e ratio k2’’’-’’ kZc’- for the reactions of the two ions with the single ( 9 ) We are indebted t o referee I1 (IO) W. D Closson and H B G r a y , J A m . Chem. S O L 86, , 290 (1R68)
Aug. 20, 1964
ACETYLENE TUNGSTEN CARBONYL COMPLEXES
complexes is slightly in favor of the azide ion, showing that N3-is a better nucleophilic reagent than is C1-. Acknowledgments.-We are grateful for financial support from the North Atlantic Treaty Organization
[ COKTRIBUTION F R O M
THE
32G1
(N.A.T.O.) and the Italian "Consiglio Nazionale delle Ricerche." Lye thank Professors F. Basolo and R . G. Pearson and Dr. 51. L. Tobe for helpful correspondence and stimulating discussions.
RESEARCH DEPARTMENT, THESTAXDARD OIL COMPASY (OHIO), C L E V E L A S D ,
OHIO]
Novel Acetylene Tungsten Carbonyl Complexes BY D. P. TATE,J. AI. AUGL,W. 51. RITCHEY,B. L. Ross,
AND
J . G. GRASSELLI
RECEIVED MARCH30. 1964 Disubstituted alkynes such as 3-hexyne, diphenylacetplene, and methylphenylacetylene react with tris(acetonitri1e)tungsten tricarbonyl or mono(acetonitri1e)tungsten pentacarbonyl to give ( RC=CR)3UT(CO). The structure and a qualitative molecular orbital picture are discussed. Similar reactions with the trisacetonitrile complexes of chromium and molybdenum cause cyclization to give hexaalkplbenzene or tetraphenylcyclopentadienone. Monosubstituted acetylenes do not give identifiable products under comparable reaction conditions. The reaction of stilbene with mono(acetonitri1e)tungsten pentacarbonyl gives a normal olefin complex, mono(sti1bene)tungsten pentacarbonyl
Introduction Metal carbonyls can react with alkynes to give one of three general types of complexes : (1) the alkynes might cyclize with formation of a homocyclic ring system such as cyclobutadiene derivatives, benzene derivatives,;-' cyclic ketones such as cyclopentadienones, quinones, and tropones,2,3.a-14 or with formation of heterocyclic rings13.15-17., (2) the alkyne acts as a dinuclear bridging ligand2,3,18-22; and ( 3 ) the alkyne is bonded to a single transition metal atom without ring formation. Only a very few compounds of the latter type are reported in the l i t e r a t ~ r e . ~ ~ - ~ O (1) L. Malatesta, G. Santarella, L. hl. Vallarino, and F. Zingales, Angew. Chem., 72, 34 (1960). (2) W. Hubel, E. H . Braye, A . Clauss. E. Weiss, U. Kruerke, D. A . Brown, G. S. D . King, and C. Hoogzand, J . Znorg. .VUG!.Chem., 9 , 204 (1959) (3) R. P. Dodge and V . Schomaker, S a t u r e , 186, 798 (1960). ( 4 ) A . T. Blomquist and P. M . Maitlis, J . A m . Chem. SOL.,84, 2329 (1962). ( 5 ) W . Hubel and C. Hoogzand, Ber., 93, 103 (1960). (6) 51. Tsutsui and H . Zeiss, J . A m . Chem. SOC.,82, 6255 (1960). ( 7 ) G. S . Schrauzer, Ber., 94, 1403 (1961). (8) J. R. Let0 and F. A. Cotton, J . A m . Chem. SOL.,81, 2970 (1959). (9) G. N. Schrauzer, Chem. I n d . (London), 1404 (1958). (IO) E . Weiss and W. Hubel. J . Znovg. iVucl. Chem., 11, 42 (1959). (11) E. Weiss, R . G. Merenyi, and W. Hubel, Chem. Znd. (London), 107 (1960). (12) hl. L. H. Green, L. P r a t t , and J G. Wilkinson, J . Chem. Soc., 989 (1960). (13) U'. Hubel and E. Weiss, Chem. Znd. (London), 703 (1959). (14) H. W. Sternberg, R . hlarkby, and I . Wender, J . A m . Chem. SOC.,8 0 , 1009 (1958). (15) J. R . Case, R. Clarkson, E . R . H . Jones, and M. C. Whiting, Proc. Chem. SOL., 150 (1959). (16) 0. S. Mills and G . Robinson, ibid., 156 (1959). (17) G. Albanesi and hl. Tovaglieri, C h i m . I n d . (Milan), 41, 189 (1959). (18) H . Greenfield, H. W . Sternberg, R . A . Friedel. J. Wotiz, R. hfarkby, and I. Wender, J . A m . Chem. SOL.,78, 120 (1956); i b i d . , 76, 1457 ( 1 9 5 4 ) , W. G. Sly, ibid., 8 1 , 18 (1959). (19) H. F. Brain, C . S. Gibson, J . A. J . Jarvis, R . F. Phillips, H . M . Powell, and A. J. Tyabji, J . Chem. S o c . , 3686 (1952). ( 2 0 ) J . F. Tilney-Basset, i b i d . , 577 (1961). (21) J . F. Tilney-Basset and 0. S. Mills, J . A m . Chem. S o c . , 81, 4757
(1'359). (22) hl. J. Ilubeck, ibid., 82, 502 (1960) (23) J. C h a t t , G. A. Kowe, and A. A. Williams, Pvoc. Chem. Soc., 208 (1957). (24) R. Colton, R. Levitus, and S. G. Wilkinson, Nature, 186, 233 (1960). ( 2 5 ) J. C h a t t , L. A. Duncanson, and K . G. G u y , i b i d . , 184, 5 2 6 (1959). (26) G . Wilke and G. Herrmann, A w ~ P wChem. '. Intern. E d . Engl., 1, 549 (1'362). (27) A. L). Gelman, S. V. Bukhovets, and E. Meilakh, Compl. y e i z d . m a d . sci. C R S S , 46, 105 (1945). ( 2 8 ) S. V . Bukhovets, Z z v . Sektovo Plaliny i Drug. Blagovodn. , M e l d Inst. Obschch. i Xeougan. K h i m . A k a d . S a u k S S S R , 29, 55 (1955).
We recently reported the preparation of a new acetylenic complex of tungsten that appears to be of the latter type.31 This study has now been extended to other alkynes and the other subgroup VIB metals. Experimental ( C H ~ C H ~ C E C C H ~ C H ~ ) ~ W(I) ( Cwas O ) prepared as described in an earlier ( C G H ~ C ~ C C G H ~ ) ~ (111.-Diphenylacetylene W(CO) (4.5 9 . ) and (CHXK)3W(C0)3(3.3 9.) in 55 ml. of ethanol were refluxed under nitrogen for 18 hr. Carbon monoxide was evolved (about 380 cc.) (calcd. for 2 equiv., 422). The pale yellow crystals (5.2 9 . ) were collected, taken up in benzene, and precipitated with ethanol giving (C&C=CCsHj)3U'(CO), m.p. 193", color change around
175'. Anal. Calcd. for Ca3HbaOW: C , 69.25; H , 1.03; W, 24.67; mol. wt., 746. Found: C, 69.77, 69.96; H , 4.34, 4.21; W , 24.55; mol. wt., 754 (osmometer, benzene). This compound may be made similarly starting with diphenylacetylene and (CH;CN)W(CO)s. I t can also be prepared by simply heating the reactants to 70' under vacuum without solvent for a few hours. (CH3C=CC6Hj)3W(CO) (III).-Methylphenylacetylene i10 ml.) and ( C H L C N ) ~ W ( C O (3.9 ) ~ g.) were heated t o 90" with stirring under reduced pressure (195 m m . ) . Gas evolution ceased within 1 hr. The pressure was then lowered to 5 mm. and the excess methylphenylacetylene distilled off a t 90". The brown viscous residue was taken up in 8 ml. of ether and filtered. This solution was added to 80 ml. of methanol. Dropwise addition of water precipitated light yellow crystals which on crystallization from benzene gave (CH3C=C-C6Hb)3U'(CO), m.p. 96-98". Anal. Calcd. for C28H240U7:C , 60.02; H , 4.32; W', 36.81; mol. wt., 560. Found: C, 60.3; H , 4.3; W, 33.4; mol. wt., 56-1 (osmometer, benzene). Oxidation with excess I:! in refluxing ethanol evolved 0.97 equiv. of CO based on a molecular weight of 560. Methylphenylacetylene could be recovered from a Dry Ice trap after pyrolyzing this compound under vacuum. (C6HjCH=CHC6Hj)W(CO)5.--A mixture of trans-stilbene (0.6 g.) and (CH&2N)W(CO)5( 1 . 0 g.) was heated t o 130' under good vacuum until gas evolution ceased. The black residue was extracted with benzene and the benzene solution evaporated to dryness. The orange-red solid was crystallized from hesanebenzene t o give (stilbene)U'(CO)5,m . p . 135'. Anal. Calcd. far C,,H,yObU': C , 45.25; H , 2.38. Found: C, 45.27, 45.24; H , 2.82, 2.68. Nuclear Magnetic Resonance Spectra.-Proton magnetic resonances were observed with a fully equipped Varian DP-60 (29) S. V Bukhovet? and K. A. lfolodova, Z h . 4rorgan. K h i m . . 2 , 776 (19571, ibid., 3 , 1540 (1958). (30) K. A. hfolodova, ibid., 3 , 2-172 (1958) (31) D. P. T a t e and J . hl Augl, J . A m . Chem. S O L ,86, 2174 ( 1 9 6 3 ) .
