bpo c chapter 31

7/29/2019 Bpo c Chapter 31

1/25

31N-METALCOMPOUNDS

n the 85 years following Kekulk's brilliant proposal for the structure of ben-zene, organic chemistry underwent a tremendous expansion, and in the processa wide variety of paradigms or working hypotheses were developed about whatkinds of compounds could "exist" and what kinds of reactions could occur. Inmany cases, acceptance of these hypotheses appeared to stifle many possiblelines of investigation and caused contrary evidence to be pigeonholed as"interesting but not conclusive." As one example, the paradigm of angle strainwas believed to wholly preclude substances that we know now are either stableor important reaction intermediates, such as cubane (Section 12- 1O), cyclo-propanone (Section 17-1 I), and benzyne (Sections 14-6C and 23-8).No paradigm did more to retard the development of organic chemistrythan the notion that, with a "few" exceptions, compounds with bonds betweencarbon and transition metals (Fe, Co, Ni, Ti, and so on) are inherently un-stable. This idea was swept away in 1951 with the discovery of ferrocene,(C,H,),Fe, by P. L. Pauson. Ferrocene has unheard of properties for anorganoiron compound, stable to more than 500" and able to be dissolved in,and recovered from, concentrated sulfuric acid! Pauson's work started anavalanche of research on transition metals in the general area between organicand inorganic chemistry, which has flourished ever since and has led to animproved understanding of important biochemical processes.


2/25

31-1 Metallocenes31-1 METALLOCENESThe discovery of ferrocene was one of those fortuitous accidents that waswholly unforeseeable- the kind of discovery which, over and over again, haschanged the course of science. Pauson was trying to synthesize fulvalene,I , by first coupling two molecules of cyclopentadienylmagnesium bromidewith FeCl, and then dehydrogenating the product:

20gJ3-2H- fulvalene, 1

The rationale for the coupling reaction was that phenylmagnesium bromidewith FeCI, gives high yields of biphenyl, presumably by way of an unstablephenyliron compound:

The reaction product was a beautifully crystalline, highly stable, orangesubstance, CIoH,,Fe, which Pauson formulated as a simple combination oftwo cyclopentadienide anions and fevro~ts on with two C-Fe bonds, 2. How-ever, the product soon was shown by a variety of physical methods to have the"sandwich" structure, 3:

dicyclopentadienyl iron, 2 ferrocene, 3The bonding between the metal and the cyclopentadiene rings involves then- electrons of the two rings, all carbons being equally bonded to the centralferrous ion. The latter, in accepting a share of 12 n- electrons from two cyclo-pentadienyl anions, achieves the 18 outer-shell electron configuration' ofthe inert gas, krypton. Analysis of the structure of crystalline ferrocene shows

lFigure 6-4 (p . 154) shows that iron(0) has 8 electrons in the 4s and 3d orbitals. Fer-rous ion (Fern) then will have 6 outer-shell electrons. This 6 plus the 12 n- electronsof the two cyclopentadienide rings makes the 18-electron total and the krypton elec-tronic configuration.


3/25

31 Transit ion-Metal Organic Compoundsthat when you look down on the molecule along the ring-iron-ring axis thecyclopentadiene rings are seen to be staggered with respect to one another,as show n in 4. Ferrocene has mp 173"and, although stable to sulfuric acid, it isreadily oxidized by nitric acid t o th e less sta ble ferricinium ion:

ferr ic inium ion

Like benzene, ferrocene does not react easily by addition but doesundergo electrophilic substitution. F o r example, Friedel-Crafts acylation(Section 2 2-4F ) with C H 3C O C l gives both a mono ethanoylferrocene and adiethanoylferrocene. The two acyl groups become attached to two differentrings and, because only one diethanoylferrocene can be isolated, the cyclo-pentadienyl groups appear to be free to rotate about the axis of the carbon-iron bonds:

C O C H ,CH3COC1 CH3COC1AICI, >

C O C H ,

Ferrocene is only one of a large number of compounds of transitionmetals with the cyclopentadienyl anion. Other metals that form sandwich-typ e stru ctures similar to ferrocen e include nickel, titanium, cobalt, ruthen ium ,zirconium, and osmium. The stability of metallocenes varies greatly with themetal an d its oxidation state; ferrocene, ruthenocene, and osmocene ar e par-ticularly stable because in each t he m etal achieves the electron ic configurationof an inert gas. Alm ost the ultim ate in resistance to oxidative attac k is reache din (C5 H5 ),Co@ , obalticinium ion, which can be recov ered from boiling aquaregia (a mixture of concentrated nitric and hydrochloric acids named for itsability to dissolve platinum and gold). I n cobalticinium ion, the m etal has the18 outer-shell electrons characteristic of krypton.Man y oth er unsaturated organic com pounds can form n- complexes w ithtransition metals. A substance tha t is in som e ways analogous to ferrocene isthe complex of two benzen e mo lecules with chromium metal, called dibenzene-chromium. T he bonding involves zerovalent chromium and the n- electron s of


4/25

31-1 Metal locenes 1587

the two benzene rings. In dibenzenechromium, the electronic configurationof the chromium atom is similar to that of krypton:

dibenzenechromium

Although dibenzenechromium is thermally quite stable, it is less sothan ferrocene and melts with decomposition at 285" to give benzene andmetallic chromium. Furthermore, it appears to lack the aromatic character ofeither benzene or ferrocene as judged by the fact that it is destroyed by re-agents used for electrophilic substitution reactions.Several transition-metal complexes of cyclobutadiene have been pre-pared, and this is all the more remarkable because of the instability of theparent hydrocarbon. Reactions that logically should lead to, cyclobutadienegive dirneric products instead. Thus, 3,4-dichlorocyclobutene has been de-chlorinated with lithium amalgam in ether, and the hydrocarbon product isa dimer of cyclobutadiene, 5. However, 3,4-dichlorocyclobutene reacts withdiiron nonacarbonyl, Fe2(CO)S, o give a stable iron tricarbonyl complexof cyclobutadiene, 6, whose structure has been established by x-ray analysis.The T-electron system of cyclobutadiene is considerably stabilized by com-plex formation with iron, which again attains the electronic configuration ofkrypton.

Oxidation of 6 with ceric iron, Ce(IV), releases cyclobutadiene which quicklydimerizes, but can be trapped by good dienophiles such as ethyl propynoateto give a cycloadduct.


5/25

1508 31 Transi t ion-Metal Organic CompoundsMany metallocene derivatives are known of other conjugated cyclicpolyenes. Exam ples are bis(cyc1ooctatetraene)uranium (uranocene, 7) and bis-(pentalenylnickel), 8 (see Section 22- 12B) :

Many of the metallocene compounds display unusual reactivities andreactions, of which none is more startling than the discovery by the Russianchemist, M. E. Vol'pin, of absorption of dinitrogen, N,, by titanocene,(C,H,),Ti, to form a complex o r com plexes that can be reduced easily toform ammonia. Th e nature of these complexes is in doub t, but very clearevidence has been obtained by J. E . Bercaw for the structure of the com plex9 formed from decamethylzirconocene and dinitrogen:

9 (for clarity, the 20 methylgrou ps on the r ings have beenomit ted)

This complex treated with acids gives NH 2-NH, and some NH ,.

Exercise 31 -1 If the ferrocene rings in 3 were not free to rotate, how many differentdichloroferrocene isomers would be expected (including chiral forms)? How couldthe substi tut ion method (Section 1-IF) be used to determine which of the isomerswas which?Exercise 31-2 The cyclobutadiene iron complex, 10, has been prepared optica l lyactive, and when oxidized with Ce(lV) in the presence of tetracyanoethene gives amixture of cyclobutadiene cyclo add ucts, all of which are optical ly inactive.

10 (one chiral form)

a. Draw the other chiral form of 10.


6/25

31-2A T-Type Alkene a nd Cycloalkene Complexes 1509b. Write structures for the cycloadducts that would be expected to be formed if 10were oxidized with Ce(lV) in the presence of tetracyanoethene.c. How does formation of opt ical ly inact ive products ind icate that the cycloadductsare formed from the cyclobutadiene corresponding to 10 ?d. Is cycloaddit ion of an alkene to cyclobutadiene best regarded as a [2+ 21 or a[4 + 21 react ion?Exercise 31-3* Assuming the molecular formula of 9 is established as C4,H6,N6Zr2,explain how the propo sed structure is consistent with I5 N nmr spectra as follows.Made with 15N=14N, 9 shows three widely separated resonance lines of equal in-tensity. Ho wever, when 9 is made with I5N=l5N, two of the peaks become doubletswith a spacing of 6 Hz.

31 2 OTHER ORGANOMETALLIC COMPOUNDSOF TRANSITION METALS31 2A P-Type A1 kene an d Cy cloal kene Co mplexesNot all organometallic compounds of transition metals date from the discoveryof ferrocene. Many have been known for a long time but their structures werenot understood. A conspicuous example is the anion of a substance known asZeise's salt, which is formed from the reaction of ethanol with chloroplatinicacid, H,PtCl,. The anion has the formula Pt (C,H,) Cl,? Although knownsince 1830, it finally was shown to have a structure with both ethene carbonsbonded to platinum. The carbon-to-metal bonding usually is formulated as an complex, I I a , or charge-transfer complex (Section 22-42>).Alternatively,we can think of the bonding in I l a as between platinum as a Lewis acid(electron-pair acceptor) and ethene as a Lewis base (electron-pair donor):

I l a I l b I I cThe arrow in I I a symbolizes donation of n electrons. However, be-

cause the stability of the ion is much greater than would be expected foreither a simple acid-base or charge-transfer complex, it is postulated thatunshared d electrons from the metal participate in the bonding. This is sym-bolized by the dashed arrow in l Ib, which stands for donation of d electronsinto the n* orbital of the double bond or, as it is often called, "back bonding."Perhaps most simple is I1C, where the C-Pt bonding is formulated as a three-


7/25

31 Transition-Metal Orga nic Com poundsmem bered ring with essen tially C-Pt o- onds. In this form ulation , full partici-pation of a platinum electron pair is assumed.Many complexes of alkenes, cycloalkenes, alkynes, and cycloalkyneswith t ransi tion metals are now known. So m e examples are:

M any substance s of this type have potent ia l synthetic usefulness as catalystso r a s reagents.

Exercise 31-4 The d ipheny lethyne co mp lex with Pt(O), analogous to 12, as beenshown by x-ray diffraction analysis to have C-C=C bond a ngles of about 140" anda central C-C bo nd distanc e of 1.32A. Exp lain wh ich of the formulations, lla, 1b, orI1c , seems most reasonable to account for the x-ray data for this complex.

31-28 Al kyl-Transition-Metal BondsOrganometallic compounds of transition metals with alkyl-to-metal bonds formany years were regarded as highly unstable substances and pron e to dissoci-ate into radicals that would couple or disproportionate, as i l lustrated by thefollowing sequence:

The fact is that the stabi l i ty depends on the character of the a t tached alkylgroups. Transition-m etal com pou nds with CH,-, (CH ,),CCH 2-, andC6H 5CH 2- groups are in many cases very much more stable than those withCH,CH2-, (CH,),CH-, and (CH,),C- groups, even though the eas e offormation of radicals by dissociation would be expected to be especiallyfavorable with the C6 H5C H2- group. Furthermo re, decompositions that givealkene and metal hydride o r disproportionat ion prod ucts (a lkene and alkane)may fail to give coupling products altogether. These facts and many others


8/25

31- 26 Alkyl-Transition-Metal Bond s 1511indicate that an important mode of decomposition of a variety of alkyl-substituted transition-metal compounds does not proceed by a radical mechan-ism. Instead, there is transfer of a P hydrogen from the alkyl group to the metalto form a n-type alkene complex and a metal-hydride bond. Decompositionof this complex produces the alkene and the metal hydride. These changesare reversible.

cr complex -rr complexFormation of the alkane then can result from cleavage of an alkyl-metal bondby the metal hydride:

This mechanism but not the sequence of Equation 31-1 makes clearwhy CH3-, (CH,),CCH,--, and C,H,CH,- are stable attached to transi-tion metals, because each of these substituents lacks a P hydrogen that wouldpermit formation of a T-bonded complex. The hydride-shift reaction is es-pecially important in hydrogenation and carbonylation reactions, as will beshown in Section 31-3. (Also see Exercise 31-14.) In fact, the reversiblerearrangement of n to CT complexes has wide generality and includes alkylshifts as well as hydride shifts. An important example of the alkyl-shift mech-anism is the polymerization of ethene by Ziegler-Natta catalysts (R,Al andTiCl,) ; the chain-building sequence is given in Section 29-6A.

In understanding these reactions, it is helpful to view the metal-alkenen complex as an incipient carbocation (just as n complexes of halogens areincipient carbocations). Alkyl and hydride shifts then bear analogy to carbo-cation rearrangements. This may be an oversimplification but it makes thechemistry easier to follow.

In other reactions, we will see that the metal can act as a nucleophilic reagent.


9/25

1512 31 Transi tion-Metal Organic Com poun ds31-2C Carbene-Metal ComplexesBonding between a transition-metal atom and one sp2-hybridized carbon canbe represented by the following valence-bond structures:

Stable transition-metal complexes of this type are known and others have beenrecognized as likely intermediates in a number of reactions. Rightly or wrongly,they are called carbene-metal complexes, although they also can be regardedeither as metal-stabilized carbocations or as metal-stabilized ylides (Section16-4A).Perhaps the most common examples of this type of carbon-metalbonding are the metal carbonyls, in which the carbon monoxide ligand func-tions as the "carbene":

When represented in this way the chemistry of carbonyl complexes of transi-tion metals becomes easier to understand. Hydroformylation reactions andother carbonylations that are catalyzed by transition-metal complexes fre-quently involve hydride or alkyl transfers from the metal atom to the "positive"carbonyl carbon (Sections 16-9G, 31-3, and 3 1-4):

We now proceed to describe some selected reactions that can beunderstood within this framework of cr-, 7 ~ - , nd carbene-type bonding be-tween the metal and carbon.

31-3 TRANSITION-METAL COMPOUNDS AS REAGENTSFOR ORGANIC SYNTHESES31 3A Reactions of Zirconoce ne Ch lorohydrideSome transition-metal hydrides show promise as synthetic reagents of thesame general applicability as the boron hydrides (Section 11-6). An excellentillustration is provided by the work of J . Schwartz with zirconocene chloro-


10/25

31-3A Reactions of Zirconocene Chlorohydride 1513hydride, 13, which is available by reduction of zirconocene dichloride:

C1 0 0Na AIM, (OCH2CH20CM3),C1

(The cyclopentadienide rings in 13 are shown as being nonparallel and this isin accord with x-ray diffraction studies of metallocenes that have extra sub-stituents on the metal.) Henceforth we will abbreviate the structure 13 by(Cp),ZrClH.

Alkenes react with (Cp),ZrClW to form alkyl-Zr bonds with zirconiumbecoming attached to the least-hindered primary carbon:CH2=CHCH2CH2CM2CH3CM3CH=CHCH2CH2CN[, I (Cp) ZrClH /"'(Cp),Zr\

The initial step in this kind of reaction is formation of the n-alkene complexfollowed by hydride transfer:

These reactions must be reversible for an alkene with an internal double bondto form an adduct with the metal atom at the end of the chain. The process isseen as a series of interconversions between 7~ and cr complexes, which permitsthe metal atom to move to the least-hindered (primary) carbon:


11/25

31 Transition-Metal Organic Compounds

7 ' ROH(Cp),Zr, H30@ RCHOC-R

Figure 31-1 Synthetic reactions of alkylzirconocenes. In general, thereactions parallel those of the boranes (Sections 11-6 and 16-9G).

The alkyl chlorozirconium compounds undergo a variety of usefulreactions, as can be seen from Figure 31-1. Similar reactions also can becarried out with alkynes by way of complexes such as (Cp),Zr(Cl)CH=CH,.

One of the elegant features of these reactions is the formation of crys-talline [ (Cp) Zr (Cl)],O on treatment of the reaction products with water.This substance can be converted back to zirconocene dichloride with HC1 andthence back to (Cp),ZrClH:

A very important reaction of alkyl transition-metal complexes withcarbon monoxide results in formation of an acyl derivative, as can be seen for

H/(Cp) Zr in Figure 3 1 1. This reaction proceeds by way of a carbon-\


12/25

31-3A Re actio ns of Zircon ocen e Chlorohy dride 3515monoxide complex of the metal, which then rearranges by an alkyl shift:

This sequence of steps is an important part of the mechanism of the hydro-formylation of alkenes (0x0 reaction), to be discussed in Section 3 1-4B, andalso is related to the carbonylation reactions of boranes discussed in Sec-tion 16-96.

Exercise 31-5 Write the sequence of steps whereby (Cp),ZrClH reacts with 2-methyl-2-pentene to form (Cp),Zr(CI)CH2CH,CH2CH(CH3),. Why is there no appreciableamount of (Cp),Zr(CI)CHzCH(CH3)CH2CH2CH3 n the prod uct?Exercise 31-6 Show how (Cp),ZrCIH could be used to achieve the fol lowing con-versions:

Exercise 31-7* The stereochemistry of reactions in which Zr-C bond s are formedand cleaved can be deduced from the results of the following reactions, where Dis hydrogen of mass 2.

The CH-CH co upl ing constants in the proton nmr spec tra of 14 and 15 are about13 Hz, Work out the favora ble conformations a nd the like ly co nfiguration s of 14 and 15and the stereochemistry of the addition and cleavage reactions. (Review Section9-1 OH.)


13/25

31 Transition-Metal Organic Compounds31-3B A N ucle oph i ic Transit ion-Metal Reagent.Sodium Tetracarbonylferrate(-11)Sodium reacts with iron pentacarbonyl to produce a salt known as sodiumtetracarbonylferratet-TI)2, Na2Fe CO),, which has been shown by J . P.Collman and co-workers to have considerable potential as a reagent for organicsynthesis.

The tetracarbonylferrate dianion is a good nucleophile and reacts with a1 kylhalides or alkyl sulfonate esters by the SN2mechanism (with inversion) toform C-Fe bonds:

R20 01 /COFe(CO), + RBr --+ OC-Fe, + BroI COcoThe resulting anion undergoes insertion with carbon monoxide or ketoneformation with acyl halides in a manner similar to alkylchlorozirconocenes(Section 3 1-3A):

01 ,COOC -Fe,I COco0IR-C-R'

0IThe product of CO insertion has the potential of transferring R-C:@, and isconverted to RCHO with acids, to RCOX with halogens, or to RC0,H byoxidation:

i,co c12OC-Fe, > RCOClI CO -c1@

2The designation (-11) indicates that the iron in this substance can be regarded asbeing in the -2 oxidation state.


14/25

31 4A Hydrogenation31-4 SOME HOMOGENEOUS CATALYTIC REACTIONSINVOLVING TRANSITION-METAL COMPLEXES31 4A HydrogenationThe mechanisms of hydrogenation of alkenes over finely divided metals suchas nickel, platinum, and so on (Section 11-2) now are u nderstood in a generalway. However, these reactions are extremely difficult to study because theyoccur o n a m etallic surface whose structure is hard to define. In contrast, themechanisms of hydrogenation with hom ogeneous cataly sts are known in con-siderable detail and provide insight into their heterogeneous counterparts.Ho mo geneo us hydrogenation catalyzed by the four-coordinated rhodiumcomplex, Rh [ (C ,H5 ) PI ,Cl, has b een particularly well investigated. W ith thiscatalyst, the first step is formation of the six-coordinated rhodium hydride ofknown configuration, 16 , n which w e abbreviate the ligand, triphenylphosphine,(C,H5),P, as L:

T he next step is coordination of the alkene (here ethen e) with 1 6 with loss ofL to give the n com plex 17 , also of know n configuration:

Stable ethene complexes of R h similar to 1 7 have been isolated and show n tohave the n-complex structure . Form ation of 1 7 mu st be an equilibrium processbecause addition of extra L reduces the rate of hydrogenation by shifting theequilibrium to the left.Hy droge nation proce eds by hydride rearrangement of 1 7 to a five-coordinated ethyl-rhodium complex, 18 . Th is com plex regains a ligand m ole-cule to replace the one lost previously, thereby giving the six-coordinatedcomplex , 19 :


15/25

31 Transit ion-Metal Organic C ompou nds

CH2=CH2CH3I LCH , ~I ,* LI * CH,

H - ~ h H-R~&/

Figure 31-2 Merry-go-round diag ram that sum marizes the steps in thehy dro ge na tion of ethene by Rh[(C,H,),P],CI. The key featu res are theflow of reactants into and the prod ucts out of the cycle , the loss and re-gain ing of the l igand , and the regenerat ion of the catalyst . The l ig andtr iphenylphosphine is represented by L.

Stab le complexes with Rh-CH 2CH , bonds s imi lar to 19 have been wellchara cterize d. T h e final s tep is formation of etha ne from 19 with regenerationof RhL3C1:

Al though we abbrev ia te (CG H,) ,P as L and show l it tle role for it o r forthe Cl a t tached to rhodium in the react io n, the se l igands play a very importan trole in providing the electronic and steric environment around the rhodium,which makes efficient catalysis possible. A useful d iagram of how th e catalystfunction s in the overal l react ion is sh ow n in Figu re 31-2.

31-4B Hydroformylat ion of Al kenes ( 0 x 0 React ion)T h e c onvers ion of a lkenes to a ldehydes wi th ca rbon monoxide and hydrogenin the presence of a cobalt catalyst is an important react ion (Section 16-9F):

C H O


16/25

31 -4 B Hydroformylat ion of Al kenes (0 x 0 R eaction)

Figure 31-3 Catalyt ic cyc le for the hydroformylat ion of a lkenes asdeve loped by G. Wilkinson. The stereochemical configurations of thepart ic ipants in the c ycle are uncer ta in.

A mechanism for this kind of transformation has been established by G.Wilkinson for a rhodium catalyst , RhL,H(CO), in which L is (C,W ,),P, inaccord with the cycle of Figure 31-3. This cycle shows that the reaction isclosely related to the hydrogenation cycle of Figure 3 1-2. T he new steps are

Exercise 31-8 Explain how 2-methylpropanal could be formed in substantial amountin the cy cle of Figure 31-3 with propene as the starting alkene.


17/25

1520 31 Transit ion-Metal Organic Comp ounds31-4C Carbonylation of MethanolA successful commercial synthesis of ethanoic acid starts with methanol andcarbon monoxide in the presence of a rhodium catalyst and hydrogen iodide:

RhC13.3H20, HI , H,C02HCH30H + Co 1754 400 psiThe reaction has some similarity to the hydroformylation reaction de-

scribed in Section 31-4B. The hydrogen iodide is required to transformmethanol to methyl iodide. The rhodium catalyst then reacts with the methyliodide as a nucleophilic reagent:

Complexation with carbon monoxide follows:

A shift of the methyl group from rhodium to carbon and then hydrolysis givesthe acid and regenerates HI and the rhodium catalyst:

31-4D The Al kene Me tathesis R eactionOne of the most curious catalytic reactions of alkenes ever discovered isalkene metathesis or alkene dismutation, in which two alkenes exchange alkyli-dene groups, usually over a tungsten catalyst. The essence of the reaction isillustrated by a commercial process for converting excess propene to a mixtureof ethene and butenes:cq3 ,H cy3 ,H

C + C WC16-A1Cl3I II i CH3CH=CHCH3 + CM2=CH2CH2 CH2 (c is and trans)


18/25

31-4D The Alkene Metathesis Reaction 1524T h e reaction p roducts are those expected if cyclobu tanes were intermediates,but formation and cleavage of cyclobutanes is not the correct mechanismbecause cyclobutanes generally are not converted to alkenes over alkene-metathesis catalysts.

After a g reat deal of research on the m echanism of this reaction, it nowappea rs l ikely that the crucial s tep is the formation of carben e metal complexesand that the products are formed by recombination of the carbenes withalkene in the various possible ways:

Exercise 31-9 Explain how an alkene-metathesis catalyst might convert a cyclo-alkene into (a) a long -chain unsaturated polymer, (b) a mixture of large-rin g polymers,and (c) a catenane (interlocking carbon r ings l ike two l inks in a chain).

31-5 n-PROPENYL COMPLEXES OF NICKELA considerable body of highly useful chemistry based on nickel has been de-veloped, largely by the German chemist, G. Wilke. Many of these reactionsinvolve what are called -rr-propenyl (T-allyl) complexe s and their fo rm ation


19/25

1522 31 Transit ion-Metal Organic C ompou ndshas a close analogy in the formation of fe rrocen e from cyclopentadienyl-magnesium compounds and ferric chloride (Section 31-1). Treatment ofNiBr, with tw o moles of 2-propenylm agnesium bromide gives a stable (albeitoxygen sensitive) substance of composition (C3H5),Ni :

trans (75%) c is (25%)

Unlike C3H 5M gBr, the metal comp ound h as a very complex proton nm rspectrum (see Exe rcise 3 1-10). Ana lysis of the spectrum indicates it arisesfrom a mixture (75 :25) of tw o (C 3H 5),N i isomers with each isomer having itsC,H5 groups in a rigid planar arrang eme nt as follows:

These facts can be accommodated by the trans- and cis-di-n-propenylnickelstructures, 20. Di-n-propenylnickel has many interesting reactions, amongwhich are the following examples:

The n-propenyl-type structures are more stable for nickel than forother metals such as iron. With 1,3-butadiene, Fe(CO), forms a double


20/25

31-5 r-Propen yl Com plexes of Nickel 1523T complex, whe reas N i( C O ), produc es a bis-71.-propenyl-type structure, 21:

With more 1,3-butadiene, 21 is conv erte d first to 22,which after rearrangeme ntreacts with 1,3-butad iene to give back 21 with liberation of trans, m ns , rans-1,5,9-cyclododecatriene:

2 50 C H 2 = C - C H = C H 2 ,+ N i 21 +\trans,trans,trans-l,5,9-cyclododecatriene

The overall sequence thus provides a catalytic route for the cyclic trirneriza-tion of 1,3-butadiene.Ethene and alkynes react with 21 in the presence of excess carbonmonoxide to give ten-mem bered ring compound s, whereas the reaction of 21


21/25

31 Transition-Metal Organic Compoundswith excess carbon monoxide results in formation of a mixture of six- andeight-membered rings:

Exercise 31-10 The nmr spectrum of 2-propenylmagnesium bromide in ether isshown in Figure 31-4. With the ai d of the discu ssion in Sections 9-10C and 9-10E andthe know ledge that the CH, resonance of ethylmagnesium brom ide comes at 38 Hzupfield from tetramethylsilane, sketch the nmr spectrum you would expect forCH,=CHCH,MgBr. Co nsid er po ss ible ways of reco ncilin g your exp ecte d spe ctrumwith the actual sp ectrum shown in Figure 31-4. (Review Section 27-2.)

i nc reas ed

377 148

Figure 31-4 Nmr spectrum of 2-propenylmagnesium bromide in diethylether solution at 60 MHz with reference to tetramethylsilane at 0 Hz. Theoff-scale bands are due to the diethyl ether, and the signals designatedC,H,, are du e to 1,5-hexadiene (co up lin g pro du ct resu lting du rin g forma-tion of the Grignard reagent).


22/25

31-6 Vitamin B,, as an Organometallic Com pound 1525Exercise 31-11 When one m ole of azabenzene (pyridin e), which is a good l igand,is added to a solution of one mole of 20 in diethyl ether, a complex of composition(C3H5),NiNC5H5 s formed in which the very com plex proton spectrum of the C3H5groups of 20 becomes greatly simplified and essentially l ike that of Figure 31-4.Explain how complexation of one mole of azabenzene with nickel in 20 could sogreatly simplify the proton nmr spectrum.Exercise 31-12* n-Propenyl(ethy1)nickel decomposes at -70" to give propene andethene. If the ethyl grou p is lab ele d with deu teriu m as -CH,-CD,, the pro duc ts areC3H 5D an d CD2=CH 2. If it is la be le d as -CD,-CH,, the pro du cts are C,H, +CD2=CH2. Are these the products expecte d of a rad ica l decom position, or of a re-versible hydride-s hift followed by decom position as in the mec hanism of Section31-2B? Su ppos e the hyd ride-shift step were n ot reversible, what pro ducts wo uld youexpect then?

31-6 VITAMIN B,, AS AN ORGANOMETALLIC COMPOUNDThe structure of vitamin B,, shown in Section 30-6B with a cyanide ion coor-dinated with cobalt is not the active form of the vitamin but is a particularlystable form, convenient to isolate and handle. The active form is a coenzymethat is remarkable in having a carbon-cobalt bond to an essentially alkyl-typecarbon. The carbon-cobalt bond is to a 5'-deoxyadenosyl group, and if weabbreviate vitamin B,, coordinated to cyanide as 23, the coenzyme can bewritten, in the same style, as 24. (You will notice that 23 is an abbreviation ofthe formula of Section 30-6B turned 180.)

cyanocobalamine, 23 HO OHdeoxyadenosylcobalamine, 24

Both 23 and the B,, coenzyme, 24, are compounds of Co(I11) and both sub-stances have all electrons paired. B,, can be reduced to a form with Co(I1)which has an unpaired electron and gives an esr signal (Section 27-9). Thecobalt-carbon bond of 24 appears to be formed from 23 by removal of thecyano group and a two-electron reduction to Co(1). The reduced cobalt is


23/25

9526 31 Transition-Metal Organic Compoundspowerfully nucleophilic an d probably is alkylated with adenosine triphosphate(ATP, Section 15-5F) o form 24:

Vitamin B,, coenzyme participates in several biological reactions butnone is m ore unusual, or a s hard t o rationalize, as its role in the interconversionof methylpropanedioyl CoA (methylmalonyl CoA, 25) to butanedioyl CoA(succinyl CoA, 26):3

CO~ 'SCOA C0" 'SCOA

methylmalonyl CoA, 25 succinyl CoA, 26

This rearrangement, which is important in the biochemical utilization of pro-panoic acid, has been show n to involve transfer of a hydrogen from the CH,-group of 25 to the -CH2R group of 24.Then rearrangement and formationof 26 occurs along with reformation of 24:

3For the structure of CoA, see Section 18-8F.


24/25

Additional Reading 1527We formulate the intermediate oxidized forms of 25 and 26 with cobalt-to-carbon bonds, but there is no definitive evidence that this is correct. The overallreaction involves attack on the CH,- of 25, not an easy reaction to carry out inthe laboratory, except with reagents such as Cl., because this CH, is not ad-jacent to a double bond or other activating group. Furthermore, there is no verygood analogy for the rearrangement step. At present, although it is known that24 is reduced to give CH,-R, the details of this important biochemical mech-anism remain to be elucidated by further research.

Additional Reading

F. A. Cotton and G. Wilkinson, Advanced Inorganic Chemistry, 3rd ed., Wiley-Interscience, New York, 1972, Chapter 23.C . A. Tolman, "The 16 and 1 8 Electron Rule in Organometallic Chemistry and Homo-geneous Catalysis," Chem. Soc. Rev. 1, 337 (1 972).T. J. Katz a nd N. Acton, "Bis( pentalen ylnickel)," J.Amer. C hem . Soc. 94, 3281 (1972);J. Tsuji, "Carbon-Carbon Bond Formation via Pa lladium Com plexes," Accts. Chem.Res. 2, 144 (1 969).R. Cramer, "Transition-Metal Catalysis Exemplified by Some Rhodium-PromotedReac tions of Olefins," A ccts. Ch em. Res. 1, 186 (1968).M. F. Semm elhack, "Formation of Carbon-Carbon Bon ds via n-A llylnic kel Com-pounds," Organic Reactions 19, 115 (1972).D. E. Bubl i tz and K. L. Rinehart, Jr., "The S ynthesis of Substituted Ferrocenes a ndOther T-Cyclopentadienyl-Transit ion Metal Compounds," Organic Reactions 17,1 (1969).J. M. Swan an d D. St. C. Black, Org anom etall ics in O rganic Synthesis, Chapman andHall, London, 1974. This is an informative but concis e accoun t for the non specialist.

Supplementary Exercises

Exercise 31-13 Palladium has many interesting uses in organic syntheses. Thefollowing sequence of reactions also could be achieved by forming and carbonatinga Grignard reagent, but would not be stereospecific as it is with palladium. Devisemechanistic steps for the reaction that account for the stereochemical result [L is(C,H,),P]. Rev iew Sec tions 31 -2, 31 3, and 31 4.

CH3 CH, L CH, LI I I CO I IR-C,H,CHBr Pd(gkC,H,CH-Pd-Br ---+ C,H,CHCO-Pd-BrI ILoptically active

ICH30HL > S-C,H,CHCO,CH, + Pd(O)L, + HBr


25/25

1528 31 Transit ion-Metal Organic CompoundsExercise 31-14* a. When a metal is complexed with an alkene, there are two pos-sible ways for nu cleophiles to be com e attache d to carbon, as i l lustrated here withpal ladium:

.Nu I, -Pd-CH2-CH, :Nu inversion1 CH,-Pdt--11 II CH2 I CH2 Ia - p d t l l --P d-CH2-CH,:Nu retentionI CH2

NU / I

Show how these mechanisms in combination with others described in this chaptercan explain how PdCI, ca n convert CH2=CH2 to CH,CHO (Wacker process). Yourmechan ism must b e in accord with the fact that, when the reaction is carried out inD,O, there is no deu terium in the ethanal formed .

[This reaction is used for large-sca le production by oxidizing the Pd(0) back to Pd(ll)with Cu(ll). Thus Pd(0) + 2Cu(ll) --+ d(ll) + 2Cu(l), and then the Cu(1) is convertedba ck to Cu(ll) with 0,. The overall result is CH2=CH2 + 120, -+ CH,CHO.]b. The balance between the competi t ive nucleophi l ic reactions described in Parta is a delicate one as judged from the following results:,eutral, H ~ Ras D,L mixture)H\ ,CH3 OCH,C=C + PdCI, + CO + CH,OH

trans

Write m echan istic steps that wil l ac coun t for the difference in stereochemical resultsof these reactions, noting that in one case there is a single carbon ylation reaction andin the other a dicarbonylation reaction.

bpo c chapter 31

Documents