Arylation of potassium 2,4-pentanedionate via SRN1 on diazosulfides

$3 oo+ 00 oo4o-4020/91 0 1991Pergamon Pressp

Tc~mhedron Vol 47.No 2, pp 333342.1991 Rmtcdm Gnu Bntam

ARVLATlON OF POTASSIUM 2,4_PENTANEDIONATE

VIA SRNl ON DIAZOSULFIDES

Carlo Dell’Erba, Marlno Novl,* Glovannl Petrlllo, Clnzla Tavanl, and Pa010 Sellandl I.stUo d/ Chrmlca Orgamca,

C N R Centro d! Studfo WI Dw//ordr e lore App/~caz~onr, Corso Europa 26, 16132 Genova, /ta/y (Recewcd In UK 18September

1990)

Summarv; Potassium 2,4_pentanedionate reacts with dlazosulfides (E)-1 and (Z)-2 in DMSO to give 3-aryl-2,4-pentanedlones 3 via an SRNl process. Advantages and drawbacks of such new access to 3 are reported together with relevant mechanlstlc implications.

The large number of lnvestlgatlons concerning the arylation of active methylene compounds mirror the importance that the relevant arylated products have as synthetic targets. Consistently, a survey of current arylatlon methods offers a wide panorama lncludlng: a) SNAr or aryne substitutions respectively on activated' and unactivated2 haloarenes; b) Pd-catalyzed

substltutlons

on

halobenzenes;3

c)

copper-catalyzed

substltutlons

of the halogen of o-halogenoarenecarboxylic acids; 4 d) employing either diaryliodonlum saltssr6 or organoleadTg8 and organoblsmuth7flg reagents; e) reactlons via cyclopentadienyllron complexes of chloroarenes; lo f) homolytic substitution of aromatic hydrocarbons arYlatiOnS

induced

by

either

Mn(III),ll

Ce(Iv),12

or

anodic

oxldatlon13

of

p-

dlcarbonyl compounds; g) free-radical chain arylatlon of 2,4-pentanedlone by action of reducing metal salts on arenedlazonlum tetrafluoroborates;14 h) SRN 1 reactions of enolates with haloarenes.15-18 As regards their general appllcablllty, however, each of the above methods appears to Suffer from

some

practical disadvantages and/or llmltatlons.

As an alternative to the exlstlng methodology it was envisaged, on the basis of our previous studies,1g-23 that dlazosulfldes (Ar-N=N-SR), readily accessible from the corresponding arylamines, might serve as convenient substrates for the of active methylene 'RN' arylatlon compounds. In this line, we report herein on the synthesis of some 3-aryl2,4pentanedlones 3 by reaction of (E)-arylazo phenyl sulfides (E)-1 and (Z)-arylazo &&-butyl sulfides (Z)-2 with potassium 2,4-pentanedionate In DMS0.24 333

334

C DELL’ERBA et al

SPh + Ac2CIi-

+ Ac2CH-

/ N=N Ar

W

Ar-CHAc2

N=N

l

- N2,PhS-

/

- N2,ButS' SBut

AT' (E)-la-c,g,k

(Z)-lb-n

3a-n

l-3

Ar

Ar

a

4-N02C6H4

3-CNC6H4

b

4-CNC6H4

3-PhCOC6H4

C

4-PhCOC6H4

2-N02C6H4

d

4-MeCOC6H4

2-CNC6H4

e

4-BrC6H4

f

4-MeOC6H4

'sH5 2-naphthyl

g

3-N02C6H4

3-pyr1dyl

Results and Discussion The data reported ln the Table show that the reaction of a tenfold excess of potassium 2,4_pentanedionate (generated ln situ by addition of the diketone to an equimolar amount of commercial potassium mbutoxide) with diazosulfides (E)-1 and (Z)-2 In DMSO gives variable yields of the arylation products 3. The

formation of products

process30 the

3 1s belreved1g-23

involving an initial single-electron

pentanedionate

radical anion

anion

to

the

substrate

to occur via

transfer

to

form

an SRNl

[step (l)] from

the

corresponding

(1' or 2') which enters a typical SRNl propagation

cycle

It should be stressed that, though on the whole the [steps w-(4)1. RS-N=N- group behaves as leaving group, the most likely primary event of Step

(2) is the

cleavage

of

the

N-S

bond

to give

a dlazenyl

radical

precursor of the aryl radical. Ar-N=N-SR + (E)-1 or (Z)-2

Are

3;

+

Ac2CH-

W

1' (or 2')

+

Ac2CH*

(1)

1' (or 2s)

-

Are

+

RS-

(2)

Ac2CH-

_

Ar-CHAc2z 3s

Ar-N=N-SR

+

3

+

+

+

N2

1

(3) (or 2;)

(4)

335 Arylauonofpotassium 2,4-pentanedlonate

Table.

Reactions dionate

of

diazosulfides

1

and

2

with

potassium

2,4-pentane-

in DMSO a

Expt. no.

Diazosulfide

1

(E)-la

2 3

Conditions

b

3a-n yield (%) crd

LL, 5 h

72

(E)-lb

LL, 0.5 h

60

(E)-lb

LL, 0.5 he

46

LL, 0.5 hf

39

4

(E)-lb

5

(Z)-2b

6

(Z)-2b

CPE,

0.43 F/mol

69 h

7

(E)-2b

CPE,

0.56 F/m01

73 h

8

(E)-lc

I, 1.5 h

48

9

(Z)-2c

I, 28 h

45 bd

I, 26 h

74 ad

10

(Z)-2d

I, 27 h

64

11

(Z)-2e

I, 2.5 h

43 Ad

12

(Z)-2f

1, 15 h

9

13

(E)-lg

3.5 h + I, 1 h

71

14

(Z)-29 (Z)-2h

I, 23 h

51 kd

I, 5 h

84

16

(Z)-21

1, 2.5 h

58 '

17

I, 3 h

74

18

(Z)-2j (E)-lk

I, 0.75 h

67

19

(Z)-2k

I, 22 h

77

20

(Z)-21

I, 6 h

7

15

LL,

21

(Z)-2m

I, 8 h

26

22

(Z)-2n

I, 17 h

60

’ [Dlazosulflde] = 0 065 M, [Ac2CH-K+]

= 0 650 M, unless otherwlse specrfied b Expenments were carried out ether

at the laboratory light (LL), or under irradtation (I) by a sunlamp, or by constant potential electrolysis (CPE)

’ Yieldsrefer

to products isolated by column chromatography, unless differently stated d In expts 5, 9, 1 I, 14. and 16 the unreacted substrate was found by ‘H NMR to be partially lsomenzed into the (E)-form e (Ac,CH-K+] = 0 33 M f [Ac2CH-K+] = 0 13 M g 9% of unreacted substrate recovered h Determined by ‘H NMR ’ 25% of unreacted substrate recovered ’ 23% of unreacted substrate recovered k 21% of unreacted substrate recovered ’ 12% of unreacted substrate recovered

336

C DEU%RJ.BA et al

The results obtained show that while the reactions on diazosulfides (E)-1 are mostly spontaneous processes (although irradiation does increase the overall reaction rate), photostimulation carry out experiments on diazosulfides

by a sunlamp

is needed to

(Z)-2. The differential reactivity

of the two kinds of substrates is well evidenced when comparing expts 2 and 5 or, more closely, expts 8 and 9 which were carried out under similar (irradiation) conditions: Completion substrate

the photo-induced

in 1.5 hr, while with is

recovered,

which

reaction

on

(E)-lc goes

to

(Z)-2c, after 28 hr 25% of unreacted

was

partially

lsomerlzed

Into

the

more

stable (E)-isomer.31 In the light of the involvement of an SZNl mechanism, the lower reactivity of S-tert-butyl dlazosulfides

(Z)-2, which has been

already observed in the reactions with aryloxides,23 most likely is the result of a concomitance of factors: i) a more negative reduction potential of (Z)-2 with respect to the corresponding S-phenyl derivative (E)-1,23 which negatively influences both the lnltiation step (1) and the propagation step (4); ii) a slower fragmentation of radical anions 2; with respect to the corresponding 1; in consequence of a presumable higher energy of the N-S bond. Actually, S-tert-butyl diazosulfides (Z)-2 generally show greater the corresponding thermal stability than (E)-1.28,32 On a practical point of view, the greater stability of diazosulfides they

are

more

chromatography definition

(Z)-2 represents an advantage because, in the first place, easily

and,

in

purifiable the

second

of the applicability

e.g.

crystallization

by

their

place,

use

allows

and/or

a

better

range of the studied reactions without

complications brought about by some lnstablllty of the substrate, which would affect the significance of the outcome. Besides the above practical advantages, the fact that, at variance of the S-phenyl

analogues,

diazosulfides

(Z)-2 do not react

spontaneously

with potassium 2,4-pentanedlonate allows to attain further evidences for the mechanism effect

of

confirmed

of the studied reactions.

light, by

the

SRNl

electrochemical

character

In fact, besides of

experiments

the

system

carried

out

the catalytic

herein

on

both

could

be

(Z)- and

(E)-2b [this last kind of dlazosulflde was studled because a more or less advanced (Z)-2 to (E)-2 lsomerization was always observed in the course of the studied reactions]. The cyclovoltammetric analysis of the two substrates above showed a sharp decrease in the height of their irreversible reduction peak by addltlon of increasing amounts of potassium 2,4_pentanedionate. Such behaviour is indicatlve33 of an electrocatalytic process

well

in

agreement

with

the

expectation

of

a

chain

process

consecutive to the formation of the substrate radical anion at the cathode

337

Arylauon ofpotassmm2,4-pentandonatz

surface.

On

electrolyses

a more quantitative basis the results of preparative (expts 6 and 7), carried out at the potential of the first

reduction wave of the substrate

on either

presence of a tenfold excess of potassium

(Z)-2b or

(E)-2b and in the

2,4-pentanedionate,

showed the

formation of the arylation product 3b in yields comparable to those obtained (expt. 5) in the photostimulated reaction on (Z)-2b. This last outcome together with the values of 0.43 and 0.56 electrons per molecule, obtained by chronocoulometry of the same experiments, leaves little doubts about the intervention of a chain process In the system herein. Further analysis of the data reported ln the Table shows that, in agreement with previous results on the SRNl reactivity of the same dlazosulfides with carbon nucleophiles20t22R23 and of halobenzenes with monoanions of /3-dicarbonyl compounds, 15t17r34-38 the yields of arylation products are satisfactory provided that an electronwlthdrawing substituent is present In the Ar of

(E)-1 and

(Z)-2. Thus dlazosulfldes

(Z)-2f and

(Z)-21 gave only trace amounts of the corresponding a-arylated-6dlcarbonyl derivatives, whose yields slightly improve In the case of the 2-naphthyl derivative

(Z)-2m and become

again

satisfactory

with the 3-

pyridyl analogue (Z)-2n. As a matter of fact, the lnltlal use of enolates of P-dlcarbonyl compounds in SRNl reactions was dlsappointlng as the and monoanions from malonlc esters, 34-37 acetoacetic esters, 34,36,37 /3-dlketones34s36-38 all failed to react with phenyl radicals and even34 with the more electrophllic 2-qulnolyl radicals. Later on, the influence of either haloarenes methylene

a cyano15t17r18 In

facilitating

or a benzoyl17 SRNl

group

reactions

compounds has been evidenced;

In the

with

aryl moiety

monoanions

of

of

active

such favourable effect has been

attributed15 a) to a faster aryl radical/nucleophile

coupling due to an

increased

favoured electron-

transfer

electrophlllclty from

the

radical

of Are anion

and b) to a more of

the

substltutron

substrate. In the system hereln, as the propagation thermodinamlcally

and

kinetically

favoured

In

all

product

step

to

the

(4) should be

cases,3g

the

meagre

yield of arylatlon product observed with (Z)-2f, (z)-21, and (Z)-Pm has to be

attributed

to

the

relatively

lower

electrophilicity

of

the

aryl

radicals Involved In the relevant SRNl propagation cycles. Most likely, in such cases the key step (3) 1s so slow that competltlve pathways for the aryl radicals can overcome the substltutlon-product-formlng

coupling with

the nucleophile. In previous papers,20-23 we have often underlined that rn SRNl reactions on diazosulfides main competing paths for aryl radicals are: a) hydrogen atom transfer from the medium or electron transfer from whichever reducing species present (followed by protonatlon of the ensulng

C DEIL%RBA~~ al

15 cm from the reaction ftask (Pyrex), an appropnatefy posnroned fan served to matntarn the reaction temperature around 25 ‘C The end of reackon was fudged by ceasing of nrtrogen evolution and/or TLC anafysls. Usual work-up involved pounng of the reaction moctureinto

1ce/3%

HCI followed by extraction wtth ether and washing of the combined

extracts wtth bnne In the case of the expenment on (Z)-2n

the reaction mrxture was diluted Hnth water and the pH

adjusted to neutrakty wRh 3% HCI After drying of the ether extracts (Na2S04) the sofvent and the excess 2,4pentanedione was dlstrlfedoff n a rotary evaporator (60 ‘C, 5-10 mmHg) Column chromatography on s~kcagel of the residue (petroleum ether and petroleum ether-dichloromethane gave pure 3-aryl-2,4-pentanediones

3-ANI-2.4-Dent~

msrtures of graduatfy increasmg pofanty as ekJatIt)

3 in vanabfe yields (see Table)

3 ‘H NMR

analysis

revealed that in CDCl, sotution all the isolated compounds 3 emst

only as enol tautomers 41 3-(4-Nrtrophenyf)-2,4-pentanedlone

3a m p 119 8-121

3-(4-Cyanophenyf)-2,4-pentanedione3b

0 “C (EtCH) (lrt ,14 m p 119 “C)

m p 146 7-147

3-(4-Benzoylphenyf)-2,4-pentanedrone

X

6 “C (benzne) (In ,15 m p 145-150

m p 126 9-127 2 “C (benme)

3-(4-Acetylphenyl)-2,4-pentanedrone

3d m p 97 4-98

3-(4-Sromqphenyl-2,4--pentenedrone

36 m p 112 9-113

‘C)

(V ,17 m p 131 ‘C)

1 “C (petroleum ether) (In ,14 98 “C) 3 “C (petrofeum ether), ‘H NMR, 6 1 89 (6H, s), 2 64

(3H, S), 7 31 and 8 00 (2H each, AA’BB’, J 8 2 Hz), and 16 71 (lH, s) 3-(4-Methoxyphenyf)-2,4-pentanedlone 3-(3-N~~enyl)-2,4-pentenedrone

31 m p 66 5-69 5 ‘C (petroleum ether) (W ,14 m p 70 ‘C) 3~ m p 77 9-78

1 “C (benzme), ‘H NMR, 6 1 90 (6H, s), 7 53-7 65 (2H m),

8 05-8 32 (2H, m), and 16 76 (iH,s) 3-(3-Cyanophenyf)-2,4-pentanedrone

3h m p 108 O-109 3 “C (benzme) (N ,15 m p 108 ‘C)

3-(3-Se~/vrphe~~-2,4-~n~ione

31 m p 90 3-90

8 “C (benzme) (V ,I4 104 “C), ‘H NMR, 6 1 92 (6H, s),

7 33-7 68 (6H, m), 7 68-7 90 (3H, m), and 16 70 (1 H, s) 3-(2-Nrtrophenyl)-2,4-pentanedlone

31 m p 68 l-69

3-(2-Cyanophenyl)-2,4-pentanedrone 3-Phenyl-2,4-pentanededlone 3-(2-N8ph~~-2,4-pentenadrone

1 “C (petroleum ether) (kt ,14 m p 67 “C)

3k m p 92 8-94 0 ‘C (benzme) (W ,15 m p 92-94

‘C)

31 m p 55 O-56 0 “C (petroleum ether) (kt ,14 m p 57 ‘C) 3m

m p 88 9-89 9 “C (petroleum ether), ‘H NMR, 6 1 90 (6H, s), 7 20-7 95

(7H, m), and 16 73 (1 H, s) 3-~3-PylldL~-2,4-pentenedrone

3n oil which easily solIdfires on cookng in an Ice-bath,

‘H NMR, 6 1 90 (6H, s),

7 30-7 60 (2H, m), 8 45-8 60 (2H, m), and 16 79 (1 H, s)

E/ecfmcbem#x/ defermrnafrons Cyclic voftammetry, constant potential electroksis, and chronocoulometry were carned out usmg the same instrumentation and condkions described elsewhere 23

References and Notes

1 Joshi, S S , Gambhir, I R J Am Chem Sot 1956, 78,2222 263,&l

Bourdas, J , Mahmu, C CR Aced SC!, Ser C lQ66,

Bourdais, J , Germsun,C Tetrahedron Letf 1970,195

2 Leake,WW,Levute,R

JAm

Chem Sot 1959,81,1627

3 Seto, K, Takahashi, S , Tahara, T J Chem Sot , Chem Commun lQ85122

341

Arylahon of potassmm 2,4_pentanwhonate

4 Hardy Hurttey, W R J Chem Sot

1929,18?0

Nast, R , Mohr, R , Schuttze, C Chem Ber 1969,96,2127

KA , Rttchre, E , Taylor, WC Ausfraf J Chem 1974,27,2209 Ames, DE,

Dodds, WD

Perktn Trans 11975,1267

J Chem Sot,

Bruggmk, A,

Per&In Trans 7 1975,705

Cmgotbs,

McKillop,A Tetrahedron 1975,31,2607

Bacon, RGR,

Murray, JCF

Ames, D E , Rbelro, 0 J Chem Sac , Per&m Trans 11975,l

J Chem Sot,

S90 McKlllop, A , Rao, D P

Synthesis 1977,759 5 Neilands, 0, Vanags, 0, Gudnntece, F J Gen Chem USSR 1959,28,1201

Bennger, F M , Forgione, P S , Yudrs,

M D Tetrahedron 1959, 8, 49 Gnedanus, J W , Rebel, W J , Sandrn, R B J Am Chem Sot Bennger, F M , Galton, S A , Huang, S J lbrd 1962,84,2819 714 Bennger, F M

1962, 84, 1504

Bennger, F M , Forglone, P S J Org Chem 1963,28,

, Forgrone, P S Tetrahedron 1963, 19, 739 Hampton, KG , Hams, T M , Hauser, C R J Org

Chem 1964, 29, 3511 Bennger, F M , Daniel, W J , Galton, S A, Rubm. G /b/d 1955, 31, 4315 Hampton, KG, Hams, TM 1989%

, Hauser, C R Org .Synfh 1971,57,126

2135 Chen, Z , Jn,

Toman, K., Machrya, K., Ichimoto, I , Veda, H Agr 6101 Chem

Y ,Stang, P J J Org Chem 1987,52,4115

6 For reviews see VarvogLs,A. 8ynfhesr.s1994,709 7 Abramovrtch, R A, Barton, D H R , Fir-ret,J -P

Monarty, R M , Vard, R K. /b/d 1999,431

Tetrahedron 1999,44,3039

8 Pnhey, J T , Rowe B A. Austral J Chem 1979,32,1561, Chem 1994,37,1245 9 Barton, D H R

IbId 1989,33,113,

Tetrahedron Lett 1999,985, Austral J

Kozyrod, R P , Prnhey, J T , Org Synth 1984,82.24

, Lester, D J , Motherwell, W B , Barros Papoula, M T J Chem Sot , Chem Commun 1999, 246

Barton, D H R , Blaze~ewsk~,J -C , Charplot, B , Lester, D J , Motherwell, W B , Barns Papoula, M T /b/d 1999,827 10 Abd El AZIZ,A S , Lee, C C , Prorko, A, Sutherland, R G Synfh Commun 1958,18,2Ql 11 Crtterio, A , Fancelk, D , Santi, R , Paganr, A, Bonsrgnore, S Gaze Chum /fa/ 1958, 7 78,405

Crtteno, A , Santa, R

,

Fiorani, T , Strologo, S J Org Chem 1999,64,2703 12 Baciocchi, E , Dell’Alra, D , Ruuconl, R Tetrahedron Left 1966,27,2763 13 Ismarl, M T Bull Chem Sot Fr 1997,488 14 CMeno, A, Ferrano, F J Chem Res (M) 1993,2656,

(S) 1993,308

15 Beugelmans, R , Bars-Choussy, M , Boudet, B Tetrahedron, 1952,38,3479 16 Beugelmans, R , Bars-Choussy, M furop 17 Gturan, M A, Pmson, J ,Saveant, J -M

J Med Chem 1959,23,539

, Threbault, A Tetrahedron Lea 1999,30,1373

18 Combellas, C , Lequan, M , Lequan, R M , Simon, J , Threbauft, A J Chem Sot , Chem Commun 1999,542 19 PetrillO, G

, NOVI, M , Garbanno, G , Dell’Erba, C Tefrahedron Leff 1985, 26, 6365, Tetrahedron 1996, 42, 4007

Now, M , Petnllo, G , Sanirana, M L Tettahedron Len 1966,27,6129 20 Now, M , Petnllo, G

, Dell’Erba, C Tetrahedron Leff 1987, 28, 1345 Petnllo, G , Now, M , Garbanno, G , Dell’Erba,

C Tetrahedron 1957,43,4625 21 Now, M , Garbanno, G , Petnllo, G , Dell’Erba, C Tetrahedron 1990,48,2205 22 Petrrllo,G , Non, M , Dell’Erba, C Tetrahedron Lett 1999,30,6Qll 23 PetrillO,G , NOVI,M , Dell’Erba, C , Tavanr, C , Bena, G Tetrahedron, In press 24 It IS worth recalling that in the preparation of drazosulfides, by coupling of arenedrazonrum chlorides with thiols,2526 the formatron of (Z) --Isomers IS kfnetrcalfyfavoured 27-29 In

the caseof the S-phenyl dertvatnms, however, the rate

of @) to (E) isomertzation IS so high that rt occurs almost completely dunng work-up of the reaction mtxtures and only (E) -isomers are practically isolated

2749

On the contrary, due to a much lower rsomenzatron rate, S-ferr-

butyl dlazosuffides can be obtamed as pure (Z)-stereclsomers 27-29

342

C DEU.%RBA et al

25 Sakla, A.6 , Masoud, N K., Saw~ns,2, Ebatd, W S Helv Chum Acta 1@74,57,481 26 Rltchte, C D , Vktanen, P 0 J Am Chem Sot 1972,94,1589 27 van Wet, H , Kooyman, E C Ret Trav Chum Pays-Bas

1967,86,993

28 Brokken-Ztjp, J , van der Bogaen, H Tetrahedron 1973,29,4169

, van Beek, J R G C , Boven, J , Schoot. C J J Org Chem 1971,36,3194 de ROSSI, R H ‘Aromattc Nucleophikc Substnutron by the SRN 1 Mechantsm’, ACS Monograph 178,

29 van Beek, LKH 30 Ross& R A,

Amencan Chemical Society, Washington, D C

,1983

31 Atthough stable as pure solids or oils, the (Z)-diazosulffdes 2 slowfy rsomenze in soltiron to the more stable (E)isomers 27-29 32 van Zwet, H , Rerdng, J , Kooyman, E C Ret Tfav Chum Pays-Bas 33 Saveant, J -M Act Chem ffes 1980,73,3X3

1971,90,21

Andneux, C P , Hapot, P , Savbnt,

J -M Chem Rev 1990,99,723

34 Hay, J V , Wolfe, J F J Am Chem Sac 1975,97,3702 35 Semmelhack, M F , Bargar, T J Am Chem Sot 1999,102,7765 36 Bunnett, J F , Sundberg, J E J Org Chem lB76,41,1702 37 Ocetveen, E A. I van der Plas, H C Ret Trav Chum Pays- Bas 1979,98,441 38 Scamehom, R G , Hardacre, J M , Lukanich, J M , Sharpe, L R J Org C/rem 1984,49,4881 39 As regards the propagation step (4’) of an SRN 1 cycle, the favourable effect of a phenylthioazo group (L = PhSNN-) on an otherwise thermodynamrcalfy dtsfavoured equikbnum (L = halogen) has been already evtdenced In ArNus

+

AIL

e

ArNu

+

ArLs

(4’)

the reactions between diazosulfides (E) - 1 and cyanide ion 2o We confidently believe that the same favourable effect be in play also throughout the system herein for example, the peak reduction potential of 3b (EP -204

V, vs

Ag/AgN03 0 01 M in DMSO) was found, by cyclic voltammetty, to be more negattve than that measured for (E)-2b (Ep - 145 V) In the same condrtions 40 The (E)-2 to 0-2

isomenzatton IS induced by light,27-2g therefore to lrmttsuch possrbilrtymost operations were

carried out in glassware wrapped n aluminium foil 41 Emsley, J , Freeman, N J , Bates, P A, Hursthouse, M B J Chem Sot , Per/w Trans 7 1989,297