8/19/2019 JOC 1980,45,1035

http://slidepdf.com/reader/full/joc-1980451035 1/7

J . Org. Chem. 1980, 45 , 1035-1041

1035

of ketone 11. Why e nolization is greater for the cyclohexoxy

derivative compared to the other hydrides is not clear,

especially in its comparison to LiMgH,OPh. One expla-

nation is that the amount of axial alcohol produced in the

reaction of every ketone with L iM gH 20 Ph is very nearly

the same as tha t observed for MgH,. Th us

if

MgH, is the

reducing agent for LiMg HzO Ph and MgH, gives 100%

yield in each case,

i t

is not surprising t ha t those hydride

reagents that disproportionate

to

MgH, give no enolization

as is observed in the case of MgH,.

It

was desired to determine if the degree of stereose-

lectivity q - ,

as

higher in the initial stages of the reaction than

after equilibrium had a chance to take place. In this

connection,

4-tert-butylcyclohexanone

as allowed

to

react

with a 10090 excess of lith ium

(2,2,6,6-tetrabenzylcyclo-

hex0xy)dihydridomagnesiate. The results are given in

Table VI. As can be seen from the data , the lower the

temperature, the greater the observed enolization to re-

ductio n ratio. A t -25 C, for example, the major product

(64%) af ter quenching is the s tar ting ketone , 4 - te r t -b ~ -

tylcyclohexanone. It should also be noted tha t the initial

reaction is very fast and t hat af ter 30 s l i t t le more tha n

5% change was observed in the yield of produc t. Also

between

30

and

18000

s,

less than 5% change was observed

in the ratio of alcohols. Th e therm odyn am ic produ ct is

the equatorial alcohol, and if

4-tert-butylcyclohexanone

is allowed to react under the equilibrium conditions in-

herent for

Meerwein-Ponndorf-Verley

or Birch reductions,

the equ atorial alcohol is produced in 98-99% yield. For

lithium alkoxymagnesium hydride reductions, never less

than 86% of the axial alcohol is observed; therefore,

if

equilibration is taking place,

it

is taking place very slowly.

Tabl e VI1 (expt 64-68) lists the results of the reactions

of ketones I, 11, 111, and IV with lithium dialkoxy-

magnesium hydrides prepared according to eq

5 .

The m ost

selective reagents were the bis tetramethylcyc1ohexoxy)

and bis tetrabenzy1cyclohexoxy) ydrides (expt 67 and 68)

which reduced ketone I to provide 89 and 85% axial al-

cohol, respectively. However, a large am oun t of enolization

accompanied th e reactions

(70

and 63 %, respectively). On

the oth er hand, exp t 65 shows that the bis di- ter t-butyl

derivative enolized only 20% of the ketone w hile reducing

the ke tone to

81%

of the axial alcohol. Th e sodium

reagent (expt 66) not only produced a 5 54 5 axial to

equa torial alcohol ratio bu t also enolized 90% of the ke-

tone. When L iH and LiOR were allowed to react under

similar conditions (expt 64), a 74:26 ratio of axial to

equatorial alcohol was observed, but 72 of the ketone

was enolized.

Wh en ketones 11,111, and IV were allowed to react w ith

these reagents, lesser amounts

of

enolization were observed

with very stereoselective results. All the reage nts stud ied

produced 99 and

100%

axial alcohol when allowed to react

with ketones I1 and 111,respectively. Th e reactions with

campho r (ketone IV) produced greater tha n

90

exo al-

cohol with little enolization except for exp t 64 and 66 which

produced 70 and 64%, espectively, of th e sta rtin g ketone.

These reagents represent a me thod of using l i thium a nd

sodium hydrid e for reduction which has n ot been previ-

ously reported.

Acknowledgment. We wish to thank the National

Science Foundation (Grant No. MPS 7504127), Union

Camp, and Alcoa for financial support of this work.

Registry No LiMgHz(O CH3), 72749-25-8; LiMg Hz(O-i-P r),

72749-26-9; LiMg Hz(O-t-B u), 72749-27-0; LiM gHz(O CHz-t-B u),

72749-28-1; LiMgH2(OCHPhz),72749-12-3; LiMgHZ(O-c-CGHll),

72749-13-4; LiM gHz (OP h), 72749-14-5; lithium [(2-methylcyclo-

hexy1)oxylmagnesium hydride, 72749-15-6; lithium (2,6-diiso-

propy1phenoxy)magnesium hydride, 72749-16-7; lithium (2,6-di-

tert-buty1phenoxy)magnesium hydride, 72749-17-8; l i thium

[(2,2,6,6-tetramethylcyclohexyl)oxy]magnesium

ydride , 72749-18-9;

lithium [(2,2,6,6-tetrabenzylcyclohexyl)oxy]magnesium ydride,

72749-19-0; lithium methoxide, 865-34-9; lithium isopropoxide,

2388-10-5; lithium tert-butoxid e, 1907-33-1; lithium 2,2-dimethyl-

propoxide, 3710-27-8; ithium diph enylm ethox ide, 2036-66-0; ithium

cyclohexyl oxide, 4111-51-7; lithium phenoxide, 555-24-8; lithium

2-methylcyclohexyl oxide, 72727-48-1; lithium 2,6-diisopropylphen-

oxide, 72727-49-2; lithium 2,6-di-tert-butylphenoxide, 5894-67-2;

lithium

2,2,6,6-tetramethylcyclohexyl

xide, 72727-50-5; lithium

2,2,6,6-tetrabenzylcyclohexyl

xide, 72727-51 6; magnesium hydri de,

7693-27-8; cis-4-tert-butylcyclohexanol, 37-05-3; trans-4 -tert-bu -

tylcyclohexanol, 21862-63-5; cis-3,3,5-trimethylcyclohexanol,33-

48-2; trans-3,3,5-trimethylcyclohexanol, 67-54-4; 2,2,6,6-tetra-

benzylcyclohexanol, 3849-12-5;

cis-2-methylcyclohexanol,

443-70-1;

trans-2-methylcyclohexanol,443-52-9; endo-1,7,7-trimethylbicyclo-

[2.2.l]heptan-2-01, 507-70-0; eno-1,7,7-trimethylbicyclo[2.2.l]hep-

tan-2-01, 124-76-5; [(2,2,6,6-tetramethylcyclohexyl)oxy]magnesium

hydride, 72727-52-7; lithium bis[ 2,2,6,6-tetramethylcyclohexyl)-

oxylmagnesium hydride, 72749-24-7; 2,2,6,6-tetramethylcyclo-

hexanone, 1195-93-3; 2,2,6,6-tetrabenzylcyclohexanone, 382-13-0;

2,2,6,6-tetramethylcyclohexanol,

948-41-0.

Hydrometalation. 5. Hydroalumination of Alkenes and Alkynes with

Complex Metal Hydrides of Aluminum in the Presence of Cp,TiCl,

Eugene C. Ashby* and Steph en A. Noding

School of Chemistry,

Georgia

Institute

of

Technology,

A t lan ta , Georgia 30332

Received July 2, 1979

Ter min al alkenes and inter nal alkynes are reduced rapidly a nd in high yield by reaction with LiAlH4, NaAlH,,

LiA1Me3H, NaAlMe3H, LiAlH 2(NR z)z, aA lH2(N Rz)2,

r

Vitr ide [NaA1Hz(0CHzCHz0CH3)z]n the presence

of

a catalytic amo unt

of

CpzTiClz n TH F at room temper ature. When these reactions were quenched with DzO

or

I*,

quantitative yields of the corresponding deuterium or iodine compounds were obtained in most cases. Thi s

method provides a convenient and high-yield route t o alkyl- and vinylaluminum comp ounds as intermediates

in organic synthesis.

Considerable interes t in recent years has been directed

toward the development of carbometalation and hy-

drometalation reactions involving alkenes and alkynes.

Th e reasons for this inte rest are clear: first, alkenes and

0022-3263/80/1945-1035 01.00/0

alkynes are very fundamental and economic building

blocks for more complex organic compounds, an d second,

carbometalation and hydrometalation provide routes to

form carbon-metal bonds which can then be functionalized

980 A me r ic a n C he mic a l Soc ie ty

8/19/2019 JOC 1980,45,1035

http://slidepdf.com/reader/full/joc-1980451035 2/7

1036 J . Org. Clzem.,

(eq

1

and

2)

to form

Vol.

45 , N o . 6 , 1980

numerous classes of compounds.

-

( H i

Carbo metala tion reactions have evolved significantly in

just the last few years due to the efforts of numerous

workers; however, hydrometalation reactions have received

muc h less atte ntio n. Th e work of Brown’ in the devel-

opment of hydroboration as a synthetic tool in organic

chemistry stands alone in terms of the degree of develop-

me nt of the m ethod a nd in terms of its utility. However,

the produc t of hydroboration (R3B)cont ains C-B bonds

which are not easily functionalized com pared t o C-Mg or

C-A1 bonds. In additio n, diborane is quite expensive, thu s

providing sufficient impetus to develop other hydrome-

talatio n reactions, particularly those tha t can form more

active C-M bonds an d whose metal hydride reagent is not

so

expensive.

Recently, Schwartz and co-workers have developed a

hydrometalation reaction involving the addition of

Cp2Zr(H)C1 o alkenes an d alkynesS2 Th e reaction pro-

ceeds well for terminal alkenes and uses stoichiometric

am oun ts of catalyst. In addition , Sat o and co-workers3

have effected th e reduction of alkenes and alkynes with

LiA1H4 in th e presence of transitio n-me tal halides. Al-

thou gh one might assume tha t hydrometalation products

are produced as intermediates, deuterolysis of the reaction

mixture shows that only TiC1, and TiC1, catalysts produce

a deuterium-incorporated product.

Some time ago we reported the direct synthesis of

NaAlH, (eq 3) an d a co nvenie nt synth esis for LiA1H4 (eq

4).4 It is clear th a t these complex metal hydrides, espe-

140 “C

N a

+

A1

4-

2Hz aAlH,

(3)

EbO

NaAlH, LiCl LiA1H4 NaC l (4)

cially NaAlH,, should be a most inexpensive source of

soluble metal-hydrogen compound . In addition, if all four

hydrogens can be mad e to rea ct w ith olefin, NaA1H4 should

inde ed represe nt th e least expensive source of soluble or

insoluble metal hydride available for hydrometalation

reactions (eq

5 ) .

It has been known for some time that

NaAlH4 f 4RHC=CH2

-

aA1(CH2CHzR)4 ( 5 )

NaAlR, compounds react m uch like R3A1 compounds and

are easily functionalized. However, it should be noted that

the ease of functionalization for the alkyl groups is best

for the first group an d progressively more difficult for the

remaining ones.

Bis(dialky1amino)alanes are produced in nearly quan -

titativ e yield and in a high state

of

purity by th e reaction

(1)

H. C. Brown, “Hydroboration”, W.

A.

Benjamin,

New

York, 1962.

(2 ) D.

W. Ha r t a nd

J.

Schwartz,

J .

A m .

C h e m . SOC. 6,

8115 (1974).

(3)

F.

Sato,

S.

Sa to , a nd

M .

Sa to ,

J . O r g a n o m e t . C h e m . , 131,

C26

(1977);

ibid.,

122, C25 (1976);

F.

Sato. S. Sato,

H.

Kodema, and M. Sa to ,

ibid.

142, 71 (1977).

(4)

E.

C. Ashby,

G.

Brendel , and

H.

E.

Re dma n,

Inorg.

Chem. ,

2 , 4 9 9

(1963).

(5)

E.

C. Ashby and

S.

Noding.

T e t r a h e d r o n Le t t . ,

4579 (1977).

Ashby and Noding

of Al, Hz , and Rz NH compounds a nd a re also readily

available, inexpensive, and reactive metal hydrides. We

reported earlier th at bis(dialky1amino)alanes add to olefins

and alkynes in the presence of catalytic amounts of

Cp,TiCl, in benzene solution in high yield (eq

6).

We have

CpzTiCIz

HAl(NR2)Z

+

RHC=CHz CHZCH2Al(NR2)2

(6)

just reported the details of HA1(NRzl2 ydroalumination

of alkenes6 and now we wish to repo rt our latest results

in hydroalumination of alkenes and alkynes with NaAlH4,

LiAlH,, and some of their derivatives.

Experimental Section

Apparatus. All reactions were performed under nitrogen or

argon at the b ench by using Schlenk- tube techniques or in a

glovebox equip ped with a recirculating system to remove oxygen

and solvent vapors.’ Calibrated syringes equipped with stain-

less-steel needles were used to transf er reage nts. All glassware

an d syringes were heated in an oven an d cooled under a flow of

nitrogen or argon. All inorganic and organic compounds, including

internal standa rds for GLC, were prepared by weighing the reagent

in a tared volumetric flask and diluting with the appropriate

solvent.

Proton NM R spectra were obtained by using a Varian Model

A-60 60-MHz spectrometer or a

JEOL

Model PFT-100 100-MHz

Fourier transform spectrom eter. All chemical shift values are

expressed in part s per million

6

values) relative

to

Me,Si a s the

internal standard . All mass spectra were obtained by using a

Hitachi RMU -7 mass spectrometer. GLC analyses were obtained

by using an F&M Model 720 gas chromatograph.

I R

spectra were

obtained by using a Perkin- Elmer Model 621 IR spectrometer.

High-pressure reactions were carried o ut by using an au toclave

rated to 15000 psi obtained from th e Superp ressure Division of

the American Ins tru ment Co.

Analytical Methods. Gas analyses were carried ou t by hy-

drolyzing samples w ith 0.1 M HC1 on a s tand ard vacuum l ine

equipped with a Toepler pump.‘ Aluminum was determined by

adding excess standard EDT A solution to hydrolyzed samples

and then back-titrating with standa rd zinc acetate solution at pH

4 with dithizone as an indicator. Amines were analyzed by in-

jecting hydrolyzed samples with an internal standard on the gas

chromatograph. Carbon and hydrogen analyses were carried out

by Atlantic Microlab, Inc.

All produ cts arising from the q uenching of reactions

of

hydrides

and alkynes with HzO,

DzO, Iz,

or

COz

were identified by GLC

and compared to authentic samples obtained commercially or

synthesized by known methods. All NM R spec tra were obtained

in CDC13 or b enzene- d,.

Lithiu m and sodium trimethylaluminohydrides were prepared

by the equimolar addition of a benzene, diethyl ether, or TH F

solution of trimethylaluminum (obtained from Ethyl Corp. and

distilled under v acuum in a drybox) to a lithium or sodium hydride

slurry in the app ropriate solvent. Th e addition was carried out

in a one-necked roun d-bo ttome d Cask equippe d with a magnetic

stirring bar a nd a pressure-equalizing additio n funnel while being

cooled with an ice-water bath . Th e addition funnel was fitted

with a rubber ser um cap which was attached to an argon-filled

manifold connected to a mineral oil filled bubbler by a syringe

needle. After th e addition and stirring (usually 10 min) the

reaction mix ture became a clear, pale brown solution which was

analyzed for aluminum by EDT A titration an d for lithium and

sodium by flame-ionization ph otometr y.

Sodium

bis(2-methoxyethoxy)aluminohydride

Vitride

T)

was

obtained as a 70% toluene solution from Matheson Coleman and

Bell.

Lithium and sodium bis(diethy1amino)- and bis(diisopropy1-

amin0)aluminohydride w ere prepa red by adding, with stirring at

0

“C, a stoichiometric amou nt of diethylamine or diisopropylamine

to a THF solution of lithium or sodium aluminum hydride.

(6)

E.

C. Ashby and

S.

Noding,

J . Org. Chem. .

44,

4364 (1979).

(7)

F.

W. Walker and

E.

C. Ashby,

J . C h e m . Educ . .

45, 654 (1968).

8/19/2019 JOC 1980,45,1035

http://slidepdf.com/reader/full/joc-1980451035 3/7

H y d r o m e t a l a t i o n

Additionally these compounds were prepared by adding a stoi-

chiometric a mo unt of bis(diethy1amino)- or bis(diisopropy1-

amino)alane to activated lithium or sodium hydride. Th e clear,

pale yellow-brown T H F solutions were analyzed for l i thiu m,

aluminum, and hydrogen by the standard m ethods described. If

benzene solutions were desired, the T H F was removed under

vacuum an d replaced by freshly distilled benzene. Th is procedure

was repeated three times. Th e amount of T H F which remained

a c co rd ing to G LC w a s 55 % .

Act ivated I i H was prepared by the hydrogenat ion of te r t -

butylli thium using Me,Si as the internal standard.

Materials. Solvents. Fisher reagent-grade benzene and

hexane were stirred over concentrated H zS04, washed with

NaZCO3 nd then disti lled water, dried over anhydrous MgS 04,

and th en disti l led from NaA1H4 under nitrogen. Th e catalyst

CpzT iClz was obtained from Alfa Inorganics.

Alkenes. 1-Oc tene (b p 122-123 C), 1-methyl-1-cyclohexene

(bp 110-111 C), styrene (bp 145-146 C), cis-2-hexene (bp 6 7 4 8

C), trans-2-hexene (bp 107-103 C), 2-ethyl-1-hexene (bp 11S 120

C) , cyclohexene (b p 82-83 C), 3,3-dim ethyl-l-buten e (neo-

hexen e) (b p 40-41

C),

and methylenecyclohexane (bp 102-103

C) were obtained from Chemical Samples Co. or Aldrich

Chem ical Co. and disti l led and stored over 4A molecular sieves.

Alkynes. 2-Hexyne (b p 83-84 C), 4-octyne (bp 132-133 C),

1-phenyl-1-propyne (b p 185-186 C), 1-octyne (bp 124-125 C),

and phenyle thyne (b p 170 C a t 19 mm ) were obta ined from

Chemical Sam ples Co. or Aldrich C hemical Co. and distilled a nd

stored over 4A molecular sieves.

1-(Trimethylsily1)-1-octyne as prepare d as discussed in ref

6. Anal. Calcd for Cl1Hs#i: C, 72.44; H , 12.16. Found: C, 72.23;

H , 12.13.

Preparation of Com plex Aluminohydrides. LiA1H4 (Alfa

Inorganics) solutions in T H F or diethyl ether and NaAlH4 (Alfa

Inorganics) solutions in T H F were prepa red by refluxing LiAlH,

an d NaAlH, in the app ropr iate solvent for at least 24 h followed

by fil tration through a frit ted-glass fil ter funnel in the drybox.

The resulting clear solutions were standardized for aluminum

conten t by EDT A and for hydrogen by stan dard vaccum-l ine

techniques.* Activated LiH was prepar ed by the hydrogenation

of tert-butylli thium or n-butylli thiu m a t 4000 psi of hydrogen

for 12 h a t rm m tem perature in hexane. Th e resulting L iH slurry

was removed via syringe under an argon atmosp here or in a

drybox.

Sodium hydride

as

a 50% oil dispersion was obtained from Alfa

Inorganics. Th e oil was removed by repeated washing and de-

cantation using freshly disti l led hexane.

General Reactions of Alkenes and Alkynes w ith Complex

Metal H ydrides. A 10

18

mm test tube with a Teflon-coated

magn etic stirring bar w as flamed an d cooled under a flow of argon

or nitrogen.

A

saturated solution of CpzTiClz n TH F (0.125 M )

was prepared ( th e solutions had to be made fresh each day). One

or two milliliters of th e Cp,TiCl,-THF solution was introd uced

into the vessel and the n th e alkene or alkyne added. Immediately

afte r additi on, a violet color developed but diminished in intensity

with t ime. The n the reac t ion mixture was st ir red a t room tem-

peratur e or at higher temperatures, depending upon the reactants,

for up to 40 h in some cases. In general, the reactio ns involving

internal alkynes were complete in 20 min. Th e reactions were

quenched by various ineans (see General Quenching Techniques

section) an d worked up by the regular m ethod (additio n of water,

extraction with dieth yl ether or hexane, and drying over MgS0,).

Most products were separated by GLC using a 6 ft, 10 Apiezon

L 60-80 column with a helium flow rate of 45 mL /mi n: 1-octene

(110 C, oven tem pera ture) , 1-methyl-1-cyclohexene

(50

C),

2-ethyl-1-hexene

(50

C), and cyclohexene (50 C).

A

20-ft, 10%

T C E P column with a he l ium f low ra te of 45 mL/min was used

for I- hexen e, cgs-2-hexene, rans-2-hexene, neohexene (45 C, flow

ra te 25 mL/ni in) , 2-hexyne (70 C), 1-octyne

(70

C), 1-( t r i -

methylsily1)-1-octyne (100 C), an d diphenyl-1-propyne (125 C).

A 10-ft,

5

Carbowax 20M column was used for diphenylethyne

(200 C, flow rate 60 mL/m in). Th e yield was calculated by using

a su it a bl e hydroc arbon in te rna l s t a nda rd for ea ch c ase ( I Z - C ~ ~ H ~ ,

n-C14H30,or T Z - C I ~ H ~ ~ ) ,nd the products were identified by

J . Org. Chem., Val. 45, N o. 6, 1980 1037

comparing t he retention times of authentic sam ples with those

of the produc ts under similar conditions and /or by coinjection

of products and authentic samples obtained commercially or

synthesized by known methods.

General Quenching Techniques. Quenching with HzO.

After the desired reaction tim e for the catalytic hydrom etalation

reaction described above, th e reaction was quen ched with wa ter

or a sa tu ra ted so lu t ion of ammonium chlor ide to produce the

proton ated species. Th e am oun ts of recovered startin g mate rial

and products were determined by the m ethods described above.

The same procedure used for

quenching with HzO was followed. Th e amou nts of recovered

starting material and products were determined by the GLC

methods and conditions described above. Each produc t was

collected from the gas chromatograph and submitted for mass

spectral analysis. Th e corrected percent of deuterium incorpo-

ration for the product was calculated by comparing the mass

spectrum of the proto nated species with the m ass spectrum of

the de uterated species and by sub tracting the con tributions of

naturally occurring isotopic components from each molecular ion

peak. Thi s procedure was followed for all unsaturated substrates

observed und er these reaction conditions.

The regioselectivity of reactions was monitored by the use

of

NM R for 1-phenyl-1-propyne and

1-(trimethylsily1)-1-octyne

fter

the produc ts were isolated by preparative GLC. For l-pheny l-

1-propyne, 95% of the m ajor product, cis-1-phenyl-1-propene,

was deuterated, with 90% of the deu terium located on C -1 as

indicated by a quar tet of triplets at 5.78 ppm J = 7 Hz, 1 H )

and a double t a t 1 .89 ppm ( J

=

7 Hz, 3 H). The o ther i somer

(deuterium located on C -2) showed a singlet at 1.89 ppm and a

multiplet a t 6.36-6.54 ppm. No trans-1-p henyl-1-p ropene was

observed. However the two major products, 1-phenylprope ne

(15%)

and 3-phenyl-1-propene

( 1 5 )

contained 55 and

85

deuterium, respectively. Th e deuterium in the 3-phenyl-1-propene

was located on C- 3 as indicated by a multiplet at 4.4-5.8 pp m

( 3 H ) .

T he products from t he reduction of I- trimethylsily1)-1-octyne

w ere p re p ar ed i n d e ~ e n d e n t l y . ~is-1-(Trimethylsily1)-1-octene

was prepared by the hyd rogenation of 1-(trimethylsily1)-1-octyne

with 5 Pd /C in 95% ethanol and the reaction monitored until

the desired am oun t of hydrogen was absorbed. Th e cis isomer

was collected an d purified via GLC under the aforem entioned

conditions. Th e tran s isomer was also detected by GLC . Th e

cis isomer, when coinjected under GLC conditions with the

product of the hy droalumination reaction, showed a trace char-

acteristic of only the cis isomer. T he tran s isomer obtained from

the hyd rogenation reaction had a reten tion time identical with

th at of the minor product from the hydroalumination reaction.

Th e following data were obtained for cis-1- trimethylsily1)-1-

octene: IR (neat,

film

2960

(s),

2940

(s),

2860 (m ), 1600 (m). 1470

(m ), 1260 [for octene: IR (neat, fi lm) 2960 (s) , 2860 (m ), 1600

(m ), 1470 (m), 1260 (s), 850 (br,

s)];

NMR (CCl,, Me&) 6 0.14

s,

9 H), 0.74-2.66 (m, 11H), 2.15 (9,

2 H ,

J

= 8

Hz), 5.49 (d, 1

H , J

=

13.0 Hz) , 6.33 (d t, 1 H ,

J

= 14 and 7 Hz); mass spectrum,

m / e

(relative intensity) 184 (M+,2), 170 (13), 169 (70), 141 (4) ,

125 (4), 114 (23), 109 (13), 99 (2 6), 85 (15), 73 ( loo ) , 67 (9), 59

(91),44 (21), 41 (14). An al. Calcd for CllHXS i: C, 71.65; H , 13.12.

Foun d: C, 71.54; H , 13.14.

Th e NMR spectrum of the trans isomer matched t he spectrum

found in the 1iterature:'O NM R (CC14, Me&) 6 0.16

(s,

9 H ) ,

0.6-1.6 (m, 11H ) , 2.1 ( m , 2 H) , 5.6 (d , 1 H ,

J =

18 Hz), 6.0 (d t ,

1

H, J

= 18

and 6 Hz) .

Quenching with Is.

A

known concentration of iodine in

benzene was prepared, an d a stoichiometric amou nt was added

to the catalytic hydroalumination reaction after th e desired time.

Th is mixture was the n allowed to stir at room tem perature for

1h. Afterward, water was added followed by a satur ated sodium

thiosulfate solution. T he organic layer was the n separated an d

dried over sodium sulfate and analyzed by GLC or NMR. Thi s

proced ure was followed for 2-hexyne. Th e iodoalken es were

identified by NM R in the following way. When th e catalytic

hydro alum ination reaction of 2-hexyne was quenched with

DzO,

Quenching with D20.

(8) D. F. Shriver , The Manipula t ion

of Air

Sensit ive Compo unds ,

McGraw-Hill,

New

York,

1969.

(9)

R.

A . Walker and

R.

A . Hickner ,

J .

Am .

Chem.

SOC.

0, 5298

(10) F.

A.

Carey and J. R. Toler ,

J . Org.

Chem., 41, 1966 (1976).

(1958).

8/19/2019 JOC 1980,45,1035

http://slidepdf.com/reader/full/joc-1980451035 4/7

1038 J .

Org. Chem.,

Vol. 45, No. 6 , 1980

the produc ts were the cis alkenes which were confirmed by co-

injection of authentic samples in the gas chromatograph. From

Zweifel's work, it is known that hydroaluminated com pounds

quen ched with iodine main tain their regiochemistry. Ther efore,

the iodoalkene obtained from th e quenching with I of th e reaction

involving 2-hexyne was 2- and 3-iodo-cis-2-hexene. N M R was

used t o distinguish between th e two isomers:

Ashby

a n d N o d i ng

S c h e m e I.

Proposed M echanism of Hydrom eta la t ion o f

Olefins by LiAIH, in the Presence of Cp,TiC12

L I A I H ~+ C p 2 T 1 C 1 2 C P ~ T I ( H ) C

+

L I A I H ~ C I ( 8 )

\

/ I

/c=c

H

BU \Me

methyl s ing le t a t 2 . 3 6 ppm

I \

c=c / H

BU

\ M e

m ethyl doublet at 1 . 6 3 p p m ,

J =

7 Hz

Anal. Calcd for C6Hl1I: C, 34.30; H , 5.28. Found: C, 34.51;

H. 5.25.

R e s u l t s

and

Discus s ion

Sat o and co-workers3 reported th at ti ta nium te tra-

chloride and zirconium tetrachloride catalyze the addition

of LiA1H4 to olefinic double bonds to produce the corre-

sponding lithium organoaluminate. They reported, for

example, that 1-hexene produced 99% n-hexane when

allowed to react with LiA1H4 n T H F at room temperature

for 30 min in t he presence of

2

mol

7'0

of T iC14 followed

by hydrolysis (eq

7).

However, the reactions were not

TiCI,

H 2 0

C4H9CH=CHz LiAlH,

-

6H14

( 7 )

quenched with

DzO

which in effect monitors the produc-

tion of the hydrometalated intermed iate. When these

reactions were quench ed with bromine, only 70 of the

corresponding bromide was obtained in most cases.

Therefore, the prod uct obtained from the quenching of the

reacti on with Hz O could be misleading in term s of the

formation of the intermed iate hydrom etalated product.

We have recently reported th at th e reaction of LiAlH4 with

olefins in the presence of NiClz results in th e formation

of the reduction product in quantitative yield in almost

every case; however, addition of

DzO

to the reaction

mixtu re resulted resulted in very low deute rium incorpo-

ration in the product.12

Since previous work h as no t definitely shown the pres-

ence of a hyd rometalation intermediate in the reduction

of olefins with complex metal hydrides, we investigated

the hydrometalation reaction of unsatu rated molecules in

more detail, particularly looking for a different hydride

and/or catalyst that would give a higher percentage of

hydrometalated intermediate as evidenced by deuterium

incorporation in the product. Lithium and sodium alu-

minum hydride, li thium and sodium trimethylalumino-

hydride, li thium and sodium

bis(dialky1amino)alumino-

hydride, a nd Vitride [sodium bis(2-methoxyethoxy)-

aluminohydride] were allowed t o react with a series of

alkenes and alkynes in a 1:l ratio in th e presence of 5 mol

of CpzTiC lz in TH F. Th e results are presented in

Tables

I

and

11.

Titanocene dichloride was chosen as the

catalyst because it worked

so

well in the previously studied

hydrometalation reaction with

bis(dialkylamino)alanes,6

an d since th en Sato3 has investigated other catalysts.

T he results

of

Tables

I

and I1 show that all of the hy-

drid es behaved similarly. Wh en allowed to react with

termina l alkenes, the reactions were over in 10 min a t room

11) G.

Zweifel

and C. C.

Whitney,

J . A m . Chem.SOC.,

9, 2753 (1967);

G . Zweifel and

R.

B. Steel, ibid.. 89, 2754 (1967): G. Zweifel. J.

T.

Snow.

a n d

C.

C. Whitney, ibid.,

90,

7139 (1968).

( 12 )

E.

C.

Ashby and

J. J.

Lin, Tetrahedron Lett., 51, 4481 (1977) .

L

R C H ~ C H ~ A I H ~ L IC P ~ T I I O )

'H

R C H ~ C H ~ A I H ~ L I D 2 O --.- RCH2CH2D 11)

temperature. After quenching with

DzO,

each product was

analyzed via mass spectrom etry and the m ass spectrum

of the product compared to the mass spectrum of the

deuterated species obtained from quenching the corre-

sponding Grignard reagent with

DzO.

The percent deu-

terium incorporation in the product was considered an

indication of the intermediate formation of hydrometalated

species. All ter min al alkenes, i.e., 1-octen e, 1-hexe ne,

styrene, methylenecyclohexane, and 2-ethyl-1-hexene

(except neohexene), provided high yields of deuterium

incorporation pro duct (95-100% ). A mechanism for this

reaction is proposed in Schem e

I.

Th e low yield of deu -

terium incorporation product

( 5 5 )

in the case of neo-

hexene is undoubtedly due to the steric influence of a

tert-buty l group attach ed to the C=C double bond

(I) .

c1

\

/ c p

\CP

CH31

,'

JH3 ,

i \ b / H

C H 3

I

Another observation in this study which parallels the

bis(dialky1amino)alane alumination reaction is that the

overall yield of the alkane decreased in the following

manner: octane (98% ) hexane (99% ) > ethylbenzene

(85 )

> methylcyclohexane

(70%)=

2-ethylhexane

(70%)

>

neohexane (60%).

The internal alkenes (cis-2-octene, cis-2-hexene, and

trans-2-hexene) were no t

as

readily hydrometalated

as

he

terminal alkenes. In all cases only low yields ( 1 4 % ) of

the corresponding alkanes were observed. Th e reactions

were carried out as described for the terminal alkenes

except the reaction time was increased

to

36 h.

In

addition,

the reaction tem per atu re was increased to 60 OC, bu t to

no avail. Sato 3 found th at in terna l alkenes could be hy-

droaluminated if the reactions were carried out for 120 h

a t 60 C in th e presence of TiC14 or ZrC1,. We were hoping

that CpzTiClzwould allow th e reaction to ta ke place under

milder conditions, but unfortunately this did not happen.

For term inal alkynes (1-octyne and phenylethyne ), re-

sults similar to those observed for the hydroalumination

reactions with

[

(i-Pr)zN]zAIHwere realized, namely, de-

protonation

of

the acetylenic proton followed by addition

to the unsaturate d linkage. Only small

amounts

of alkenes

(3-1070)

nd alkanes (4-13

7 0)

were observed. T he starting

alkyne showed a large amount (85 )

of

deuterium in-

corporation a t the terminal carbon atom when th e reaction

8/19/2019 JOC 1980,45,1035

http://slidepdf.com/reader/full/joc-1980451035 5/7

8/19/2019 JOC 1980,45,1035

http://slidepdf.com/reader/full/joc-1980451035 6/7

1040 J . Org.

Chem. , Vol .

45, No ,

1980

Ashby and Noding

Scheme I

D / H

P h

'Me

b

p -c

A '

D 2 @

B

lcjc

\ / D

,c=c

P h 'Me

ti

B'

Table 11. Reac t ions of Selected Alkenes and Alkynes

wi th Com plex

(Dialky1amino)aluminohydrides

n t h e

Presence of 5 mol Cp.TiCl ,=

D

com plex a lkene o r y ie ld , incorp ,

a lum inum hydr ide a lkyne p roduc t

L i A I H , ( N E t , ) , 1 - o c t e n e o c t a n e 9 6 9 8

1 - o c t y n e 1 - o c t e n e 10

oc tane 11

4 - o c t y n e 4 - o c t e n e 9 5 9 7

c i s -2 -oc tene oc tane 5

LiAlH,(

N-i -Pr? ') : 1 -oc tene oc tane 92 96

1 - o c t y n e 1 - o c t e n e 1 2

o c t a n e 11

4 - o c t y n e 4 - o c t en e 9 7 9 7

c i s -2 -oc tene oc tane 2

N a A I H , ( N E t , ) , 1 - o c t e n e o c t a n e 15

1 - o c t y n e 1 - o c t e n e

15

o c t a n e 1 3

4 - o c t y n e 4 - o c t en e 9 3

c i s . 2 -oc tyne

octane t race

NaAlH, (N- i -P r , ) , 1 -oc tene oc tane 95 96

1 - o c t y n e 1 - o c t e n e

10

o c t a n e 1 3

4 - o c ty n e 4 - o c te n e 9 1 9 3

c i s -2 -oc tene oc tane 2

he reac t ions were ca rr i ed ou t in THF a t room tem -

pera tu re f o r 2 h in a

1

1 m ola r ra t io

of

com plex a lum ino-

hydr ide to a lkene or

a lkyne .

pheny l ring. Also, th e bulky cyclopentadie nyl groups on

the tita nium can ad apt a staggered arrangement with the

phenyl group if the tit ani um is closer to the number one

carbo n, whereas this kind of arrang eme nt would be more

difficult and would involve the methyl group if the ar-

rangem ent were on the opposite side. Therefore, the least

hindered position would be th e one where the titanium

would be located closer to the number one rather tha n the

number two carbon. Once A and B form , hydrogen m i-

gration can take place, resulting in the formation of an

1VI

CHZ

M

H

H \ I

i

i - C - C = C H z

P

h H t '

intermediate with C being the most stable resonance

contributor. Interm ediate C can then undergo the

transmetalation step (Scheme

I)

with [(i-Pr),NI2A1H fol-

lowing the @-hydride dd ition

to

yield

D

on quenching with

D 2 0

or

abstract hydrogen from the solvent to form

E.

Once

E,

a terminal alkene, is formed,

it

may also undergo

catalytic hydrometalation to yield, after similar steps, the

protonated or de uterated 1-phenylpropane. I n all cases,

it is probably m ost likely t ha t the protonated products

result from a homolytic cleavage

of

the intermediate ti-

tanium species before the transmetalation step takes place.

Therefore, the hydroalumination of internal alkynes shows

great promise for the formatio n of vinylaluminates which

are

known to

undergo further reaction with

C 0 2 ,

yanogen,

and halogens to form a,@-unsaturated arboxylic acids,

nitriles, and vinyl halides.ll

As noted earlier, the terminal oc tyne reactions were no t

clean. Therefore,

1- trimethylsily1)-1-octyne

as prepared

by the reaction of 1-octynyllithium with chlorotri-

methylsilane (eq

13).

Whe n this alkyne was allowed to

C6HI3C=CH n-Bu Li 6HI3C=CLi

e3SiC1

-CIHlO

LiCl

C6Hl3C=CSiMe3 (13)

react under the conditions state d in Tables

I

and 11, the

reaction was slow even at 60 C. Th e best results obtained

showed a 35% yield of cis-1-(trimethybily1)-1-octene

65%

deuter ium incorporat ion) and a

15%

yield

of

1-( t r i -

methylsily1)hexane

(20

deuterium incorporation). Pre-

sumably because of the bulkiness of the trimethylsilyl

group, some homolysis of the interm ediat e alkenyl tita -

nium compound takes place, resulting in hydrogen ab-

stractio n from solvent, causing poor de uterium incorpo-

ration of the product when

D 2 0

s added (eq

14 .

In conclusion, hydroalumination reactions using complex

aluminohydrides work extremely well for terminal alkenes

and internal alkynes but do not work well for internal

alkenes and terminal alkynes. An interesting observation

which is a result of this work is that t,he regiochemistry

8/19/2019 JOC 1980,45,1035

http://slidepdf.com/reader/full/joc-1980451035 7/7

J .

Org.

Chem. 1980,45, 1041-1044

1041

/ H

S H

C

p2T

-

C

H

/ c = ,

- C=C

-

/’ ‘SiMe3

\

H13C6

S i +le3 Hx

14)

involved for both the bis(dialkylamino)alane4 and the

alu min ate reactions is approximately the same. Th is in-

dicates th at t he regiospecific step m ust be th e sam e for

both reactions. We suggest th at the complexation of the

intermediate titanium hydrido compound

or

the formation

of the alkyltitanium compound is the rate-determining

step, after which the subsequent steps of Scheme

I

follow.

Acknowledgment. We are indebted to the Nat ional

Science Foundation (Grant No.

MPS 7504127) ,

Union

Ca mp Corp., and the Alcoa Foundation for financial sup-

\

/ H

/ c = c

H

H1326 ‘SiMe3

port of this work.

Regis t ry

No.

I-Octen e, 111-66-0; octane, 111-65-9; 1-hexene,

592-41-6; hexane, 110-54-3; styren e, 100-4 2-5; thylb enzen e, 100-41-4;

methylenecyclohexane, 1192-37-6;methylcyclohexane, 108-87-2;2-

ethyl-1-hexene, 1632-16-2;2-ethylhexane, 589-81-1;neohexene, 558-

37-2; 2,2-dimethylbutane, 75-83-2; cis-2-hexene, 7688-21-3; ran s-2 -

hexene, 4050-45-7;cyclohexene, 110-83-8;cyclohexane, 110-82-7;1-

methyl-I-cyclohexene, 591-49-1; I-octyne , 629-05-0; phenylethyne,

536-74-3; 4-octyne, 1942-45-6; cis-4-octene, 7642-15-1; 2-hexyne,

764-35-2; 1-phenyl-1-propy ne,673-32-5; 1-phenylpropane , 103-65-1;

1-deuterio-cis-1-phenyl-1-propene,

2087-52-6; 2-deuterio-cis-l-

phenyl-1-propene, 72090-05-2;3-phenyl-l-propene, 300-57-2; -( tri -

methylsily1)-1-octyne, 5719-55-8;

1-(trimethylsilyl)octane,

429-76-3;

cis-1- trimethylsily1)-1-octene,

7365-48-7; I-iodoo ctane, 629-27-6;

2-iodo-cis-2-hexene, 72087-50-4; 3-iodo-cis-Z-hexene, 72087-51-5;

Cp2T iClz, 1271-19-8; LiAIH,, 16853-85-3; NaAlH,, 13770-96-2;

LiAlMe3H,62816-22-2; NaA1Me3H, 66484-08-0; NaA I(OC H2C H20 C-

H3)2Hz, 22722-98-1; cis-2-octe ne, 7642-04 -8; LiA 1HZ (NE tz),, 2749-

21-4; LiA1Hz(N-i-Prz)2, 2749-22-5; NaA1H2(NEt&, 62259-84-1;

NaA1 HZ(N-i-Pr2),, 2749-23-6.

Hydrometalation.

6.

Evaluation of Lithium Hydride as a Reducing Agent

and Hydrom etalation Agent

Eugene C. Ashby* and Stephen A. Noding

School

of

Chemistry, Georgia Institute

of

Technology, Atlan ta, Georgia

30332

Received June

11

1979

Th e reactions of activa ted lithium hyd ride with carbonyl compounds (aldehydes, ketones, esters, and enones),

alkenes, and alkynes in the presence of transition -metal halides were investigated. Significant reaction involving

the above substrates w as accomplished only when a n equimolar amo unt of VCI, was used in con junction with

th e lithium hydrid e. Aldehydes were reduced to their corresponding alcohols in high yield (95-97%), and ester s

were also reduced to their corre spond ing alcohols in high yield (93-95%) with a small amou nt (5-7%) of th e

corresponding aldehyde (of the carboxylic acid portion) formed. Reductions of cyclohexanones were highly

stereoselective. In this connection, 4-tert-butylcyclohexanoneas reduced to the axial alcohol in 96% yield and

with 82% stereoselectivity. T he only enone

to

be reduced w as cinnamaldehyd e, which gave the 1,2-reduction

product

in

90% yield. Termin al olefins were reduced to alkanes whereas interna l olefins were completely unreactive.

In thi s connection, 1-octene was reduced to oc tane in 95% yield. In this example,

30%

deuterium incorporation

of the product was observed when hydrolysis was effected with DzO. Such a result indicates formation of th e

intermed iate octyllithium in 30% yield. Since alkynes and inte rnal olefins are not reduced a t all with LiH and

VCl, and te rminal olefins are, reduction of enynes and dienes with LiH a nd VCl, could serve as a selective reduction

metho d for the red uction of a terminal double bond in t he presence of a triple bond a nd also for the selective

reduction of a terminal doub le bond in the presence of an internal double bond.

Lith ium, sodiu m, an d potassium hydrides have been

widely used as bases in sy ntheti c organic chemistry, but

have no t been used

as

reducing agents. Some time ago we

developed a simple metho d for the prep aratio n of a very

active form of lithium hydride.’ Th is method involves the

room-temperature hydrogenolysis of tert-butyllithium (eq

1).

Th e resulting white solid is easily filtered from th e

room

t e m p

t -C4HSLi

+ H2

i -C4HI0 LiH

(1)

reaction mix ture b ut is most conveniently used as a slurry

due to its high degree of pyroph oricity. Th is form of

lithium hydride app ears to be much more reactive tha n

the lithium hyd ride th at is commercially available and

which is prepared at about

450

“C. A preliminary ex-

periment? showed that benzophenone is reduced to

benzhydrol in

6%

yield when stirred w ith activated lithium

1)E. C. Ashby a nd

R.

D. Schwartz ,

Inorg.

Chem., 10, 355 (1971).

(2)

E. C. Ashby and

R.

Boone, unpublished results.

0022-3263/80/1945-1041 01.00/0

hydride for

2 h,

whereas no trace of benzhydrol was found

when commercial lithium hydride was used under th e same

conditions.

Recently Caubere and co-workers showed that the

reagent NaH-R ONa-M X, is capable of reducing

halide^,^^^

ketones, alkenes, and

alkyne^.^^^

Although these reactions

were not catalytic, they d id provide a

75-95

yield of the

corresponding alkane s, alcohols, alkanes, an d alkenes , re-

spectively.

In our continuing search for new stereoselective reducing

agents and new hydride systems to effect hydrometalation

of alkenes and alkynes, we investigated the reactions of

activated LiH with various carbonyl compounds, alkenes,

and alkynes in the absence and presence of catalysts. The

(3) G.

Guillaumet, L. Mordentia , and

P.

Caubere ,

J .

Organomet.

(4)

G.

Guillaumet, L. Morde nt ia, a nd

P.

Caubere ,

J .

Organomet.

( 5 )

J.

J.

Brume t ,

L.

Lordenti , B. Loubinaux, and P. Caubere ,

Tetra-

(6)

J. J.

Brume t a nd

P.

Caubere ,

Tetrahedron Lett., 3947 (1977).

Chem., 92,

43 (1975).

Chem. , 102, 353 (1977).

hedron Lett.,

1069

(1977).

980

Am erican Chem ica l S oc ie ty