amine'8 on C h r o m o ~ o r b . ~ gT h e retention tixries of some alkylpyridines on this substrate have been determinedlo and it was observed t h a t the values of 2- and 4-isomers differ greatly, t h e 4-isomers having longer retention times. I t is interesting t o note t h a t 2-ethylpyridine has a smaller retention time than 4-picoline. This could be ascribed t o the hindrance t o solvation of the 2-substituted pyridines on the (18) Visco Products Co , Inc , Houston, Texas (IS) Johns-Manville Corporation. (20) A W. Decora and G. V Dinneen, Paper presented before t h e Analytical Division, American Chemical Society Meeting, September, 1958, Chicago, Ill.
polar solvent. .I complete separation of the stdrtitig xnaterial and t h e reaction product was in such cases achieved. The column was kept at 150". Helium was used as a carrier gas. I n the case of 4-alkylpyridines homologs, a shorter column-about 10 f e e t - c o u l d be used, since the separatiori is quite large; for t h e lower 2-alkylpyridine homologs, an additional 10-foot column improved t h e separation. T h e pressure of the carrier gas was changed from 10 t o 30 Ib./sq. inch according to the increase in length of the column.
Acknowledgment.-The authors wish to thank N r . Ed. 11. Lewicki for technical assistance. Evazsros. ILL
[CONTRIBUTION FROM THE DEPARTMENT O F CHEMISTRY O F THE POLYTECHNIC I N b l l r L 1 E U F BROOKL\l.zi]
Diaryliodonium Salts. XIV. Reactions of Organometallic Compounds with Iodosobenzene Dichlorides and with Iodonium Salts1s2 BY F. MARSHALL BERINGER JOSEPH IV. DEHX,JR.,
AND
MURRAY LVINICOL
RECEIVED XOVEMBER 30, 1959 At low temperatures ar>-llithium coinpounds react with aryliodoso dichlorides in ether, tetrahydrofuran or tolucllc t i 1 give 20-50Y0 of diaryliodonium salts. More aryllithium compound or arylmagnesiurn halide converts t h e iodoniurn salt t o a triaryliodine, from which the iodoiiiuni salt can be recovered in good yield by cleavage with acid A triarjliodine or a n alkyldiaryliodine allowed t o warm t o 0" or above in ether gives products formed by decomposition to iodo compounds and free radicals, which may react with each other or with solvent.
Introduction.-Three steps may be considered in the reactions of organometallic compounds (RM) with iodine(II1) chlorides.
benzene and iodobenzene and with phenylmagnesium bromide gave biphenyl, iodobenzene arid a trace of a diphenyliodonium salt. The appearance of ethylbenzene, biphenyl and R M f IClp ----f RIClz + hlCl (A) iodobenzene suggests the further reaction of the R M f R I C L --+ RJC1 MCl Grignard reagent with the iodonium salt formed (B) in the initial reaction. Examples of reaction C R M f RJC1 +RjI + MC1 (C) are the formation a t low temperature of triphenylReaction X is exemplified by the formation of iodineg,l0from diphenyliodonium iodide and phenyltruns-2-chlorovinyliodoso dichloride from the cor- lithium and of phenyl-2,2'-biphenylyliodine1from responding mercuric chloride by treatment with 2,2'-biphenylyliodonium iodide and phenyllithium. iodine t r i ~ h l o r i d e . ~There are few examples of re- Both trisubstituted iodines are yellow. The latter action A alone, as in most cases it is followed by cyclic trisubstituted iodine was relatively stable to reaction B t o give an iodonium salt; e.g., bis- 100", while triphenyliodine decomposed above 0' (trans-2-chlorovinyl)-mercury with iodine tri- to give, as one product, biphenyl. chloride gave bis-(trans-2-chlorovinyl)-iodonium In 1953, i t mas reported'? from these laboratories chloride4 while both diphenylmercury and phenyl- that ethereal methylrnagnesium iodide, cthylstannic chloride with iodine trichloride gave di- magnesium bromide. phenylmagnesium broiiiide phenyliodonium ~ h l o r i d e . ~ and phenyllithium a t 0' or above gave with diReaction B has been run with organometallic phenyliodonium bromide toluene (59%), ethylcompounds of m e r c ~ r y , silver,7 ~ . ~ tin6 and magne- benzene (3770) and biphenyl (33 and sium.8 Reactions with phenylstannic chloride6 At t h a t time it was thought t h a t these pr were convenient and successful while those with probably arose from nucleophilic attack of the Grignard reagentss gave iodoniurn salts in only organometallic compound on the I-carbon of the trace amounts. For example, iodosobenzene di- diphenyliodonium ion. The alternative route of chloride with ethylmagnesium bromide gave ethyl- formation via trisubstituted iodines was not explicitly considered. (1) This paper is taken from the M.S.thesis of Murray R'inicov and The first aim of the present work was to iiivestithe P h . D . dissertation of Joseph W. Ilehn, J r , and was presented in gate the synthesis of diaryliodoniutii salts froiii imrt a t the h-ew York City Meeting of the American Chemical Society i n September. 1{1.57 (Abstracts o f Papers, 11. :1GP). aryliodoso dichlorides a n d organolithiuni or nlagtlc( 2 ) Preceding paper: F. hl. Reringer and I. I,illien, 'l'111sI O U R N A L , siuin coinpounds a t low temperatures. The secontl aiiii was to investigate. i u r t h t r the products o f t h c ssiicinte I'rufesstrr. Yxle Unircriity, l ! l . X - l!jcj!l. reaction of diphenv1iodr)niLuii chloride ill cthcr i 4 ) A . N. Kesmeyanov, Bull. n c i i c i . s l i 1 - R..Y S . , ( : i m s c S c i iiiiii., 2,'jY (1Y431; C. A , , 40, 2122 (1946). with Grignard reagents.
+
( 5 ) R . K. Freidlina and A. N. A-esmeyantiv, C o i n p l . y e l i d . , o c l i d . s o . I' R . S . S . , 2 9 , 567 ( l Y 4 0 ) ; C. A , , 35, 3614 (1941). (6) C. Willgerodt, H K , 3 0 , 56 (1807); 3 1 , 915 (18!>8). ( 7 ) C Willgerodt, ihi!/, 28, 2107 f l 8 ! ) > ) ; .i 'l'hielc : ~ n dH.I l ; l a k h , .'I j l j l , 369, I :< I ' I < l [ I ! l J 1% I I l l ? p \ > ~ ~ l l l .I l , l ' l l r i i i S'ili , 119, I l ' i i , l ' l 2 l l .
t'
(II)G. Wittig and 51.Rieber, A X K ,662, 187 (1!)4!1). (10) G. Wittig and K , Clauss, ibhl., 578, 136 (1933). (11) K . Clauss, C h e w 1 3 ~ ; ,. 88, !?G8 f l < l G ) . ( 1 2 ) I?. R l 13rriiiurr. .4. H i i ~ . r l r v ,11 l~rt.xler, 1; 11. (>in(llw:in11 c I iiiiil,L.iii 'I.HI\ I O I , K \ > I , 7 5 , l 7 i i h f I
June 5 , 19tiO
"49
~ R G ~ ~ N O M E W L L IWITH C S IODOSOBENZENE D I C H L O R I D E S
Results and Discussion The Synthesis of Diaryliodonium Salts.-A single reaction of iodine trichloride with phenyllithium gave only a very small yield (lyo)of diphenyliodonium salt. More successful were the reactions of phenyliodoso dichloride with phenyllithium a t low temperatures (Table I).
dichloride have given the unsyiiiinetrical salts in 28 and 34Y0 yields after crystallization. 13,14 The related reaction in which 1-naphthyliodoso Li
fC12
TABLE I R DIPHEXYLIODONIUM BROMIDE FROM PHEXYLIODOSO DIdichloride13is treated with an aryllithium compound CHLORIDE AXD PHENYLLITHIUM AT Lorn TEMPERATURES~
is made difficult by the instability of this iodoso dichloride, which may decompose exothermically Ether -76 t o -70 1 .. 19 minutes after collection and washing on a Biichner -65 t o -55 1 27 22 funnel. The best yield by this route has been 8% -iO t o -55' 1 47 33 of 1,l'-dinaphthyliodonium salt. Dry Ice-bath 3 .. 29 As 2-naphthyliodoso dichloride'j is of about the -i6 t o -60 3 .. 35 same stability as the phenyl derivative, 2-naphthyl-67 t o -55" 3 49 31 iodonium saltsI5 may be prepared both from this THl?d - 6 j to -55 1 23 18 iodoso dichloride and from 2-naphthyllithium. Tolueiie -71 to -65 1 39 31 The two reagents together gave 2,2'-dinaphthyla Unless otherwise noted the ethereal solution of phenyliodonium salts (77,). lithium and lithium bromide was added to the phenylThus diaryliodonium salts may be prepared from iodoso dichloride. * Molar ratio of phenpllithium t o iodoso dichlorides by treatment not only with the phenyliodoso dichloride. Inverse addition. Tetrahyrelatively unreactive mercury and tin compounds drofuran. but also with the reactive lithium compounds. X reaction of phenyliodoso dichloride with Diphenyliodonium Halides with Grignard Rephenylmagnesium chloride in tetrahydrofuran at agents.-b-hen dry diphenyliodonium chloride or -72 t o -60" gave 37, of diphenyliodonium bromide was added t o ethereal Grignard reagent bromide. The reason for the superiority of the near the boiling point, there was local boiling and a lithium reagent is not presently known. fleeting yellow coloration. Addition below 0" gave One aspect of Table I requires further comment : a yellow-orange precipitate, which lost its color on the use in some reactions of phenyllithium in stirring a t room temperature. excess of that required by the stoichiometry of Products fiom the reactions of Grignard reagents reaction B. Under these conditions the first- with diphenyliodonium chloride or bromide are formed iodonium salt was converted t o triphenyl- given in Table 11. iodine (reaction C) from which the iodonium salt It is proposed that the variety of products could be recovered by cleavage with acid (reac- obtained arises from the decomposition of trition D). substituted iodines t o iodo conipounds and free (C~,H6)31f H X --j (CsHj)'I+X&He (D) radicals, which react with each other and with the In practice there was only a small increase in solvent. yield. I n a control experiment diphenyliodonium (CeH,)JR MgX? t- ( CsH6)iI +XRMgX chloride treated at low temperature first with phenyllithium and then with hydrogen chloride j. gave back 64YG of iodonium salt as the bromide; Cr,H:,. K . + CbHbC6H5 f CsHjK KR S.27, of diphenyliodonium bromide was recovered C ~ H J RI CeHs + R H + RH-1 from crystallization under comparable conditions. Another possible synthesis of diphenyliodonium C6Hj. (or R.)+ C&P, (or R H ) salts is the reaction of triphenyliodine with phenylC H ~ C H Z O C H ~ C H ~ CH3CHOCHZCHa iodoso dichloride (reaction E). I n a two-step The data of Table I1 indicate that there is a reaction phenyliodoso dichloride was first converted to yellow, crystalline triphenyliodine which tendency for R. t o be formed rather than C6H5' ; i . e . , the C6H61/RI ratio is always greater than two. Solvent
T e m p . , OC.
Molar ratio b
Yield, 5% Crude Crystd.
+
+
+
(C~,H5)31 CsHjIClz
+2(C6H6)21iC1-
(E)
+
+
+
(13) 1-h-aphthylphenyliodoniuin salts have been made from 1-
then was cleaved with an equivalent of phenyliodoso naphthyliodoso dichloride and diphenylmercury : C. Willgerodt and P. dichloride t o the iodonium salt, isolated as the Schlosser, Ber., 33, 6Q2 (1900). (14) Similar salts substituted in the phenyl group have been precrude bromide (4iyc). pared by t h e condensation of 1-iodosonaphthalene with substituted Two symmetrical salts were prepared by the iodoxybenzenes in the presence of silver oxide: C .Willgerodt and W. addition of a substituted phenyliodoso dichloride Ernst, i b i d . , 34, 3406 (1901); C.Willgerodt and I;.I h m m a n , ibid.,34, to a similarly substituted phenyllithium. Both 3678 (1901); C Willgerodt and W. Bergdult, A n i ? . ,327, 286 (1903); C. and P. Lewinu, J . profit. Chem., [ 2 ] 6 9 , 321 (1904). -I,l'-dimethyl- and 4.-I'-dichlorodiphenyliodoniuni Willgerodt (15) 2-Naphthglphenyliodonium chloride was prepared from 2 bromide were obtained in 30YG yield after crystal- naphthyliodoso dichloride and diphenylmercur) : C.Willgerodt, Bet'., lization. 31, 915 (1898). T h e 2-naphthyliodoso dichloride is more stable t h a n Some preliminary attempts t o prepare naphthyl- t h e 1-isomer but less stable t h a n t h e phenyl derivative. (16) A related reaction scheme helps t o explain some of t h e products iodonium salts by reaction B have been success- from diphenylirrdrmium ions w-ith carhanions from di- and triketones: ful. Reactions of 1-naphthyllithium with phenyl- 117. hl. Reringer, 1' S !*orxiune a n d fir. I). Yurlis. ?'plua/rrdron, 8 , 49 iotloso dichloritlc C I I I ~ with / ~ - c h l o r o ~ ~ h r ~ i y l i o ~ Ol o! i sf XoJ ~ a~irl ] estrrr 11'. S Furgiut.,Ph I ) . di~\rrlilliiin.lWX11. lrtliii
2950
17.
MARSHALL BERINGER,JOSEPH W. DEIIN,JK.,
PRODUCTS FROM THE
AND
MURRAY M'INICOV
VOl. 82
T A B L E 11 REACTIONS O F DIPHENYLIODONIUM HALIDES WITH GRIGXARD REACEXTS IS b;THEK
Starting materials and products in m i l l i m o l e s ~b ~
c
R M g X from RX
--
OC2Ha
PhiPh U-
Teow., C.
Y-
PhI
PhH
PhPh
PhK
KI
I
PhCHCHs
CHSI 200 Cl 100 90 85 .. 10 .. CH3Br 200 c1 160 35 135 50 10 40 .. 20 C2HjCl 100 c1 100 5 TO 43 5 15 .. CzH5Br 200 C1 100 - 5 80 95 5 15 20 CfHsBr 100 c1 100 35 80 60 5 35 Trace CLHSB~ 200 Br 140 35 130 130 .. 20 10 20OC - 40 130 1'0 15 21) 25 CzHjBr 200 Cl 55 10 IO I racc CzHJ 100 CI 100 2 I U 35 153 9.> 1I1 25 .. 10 i-CsH;Br 200 C1 160 t-c~HgCl 100 Cl 100 35 95 70 J 0-10 5 .> r n-CaH,;Br 100 c1 130d 35 90 3.1 10 .. .. 5 a \'slues rounded off t o the nearest five millimoles. The leaders ( , , signify t h a t none of the substance was found. * Four %dobenzene fractions were analyzed to determine t h e amount of PhCH( CH3)OCH2CI13;these are shown in the last eohtnm. Iodoniuni salt was recovered (20 mmoles\: also. a considerable amount of what is believed t o be butane was collected. Twenty millimoles of octene-1 was recovered.
--
Next, i t is seen that more phenyl radicals abstract a hydrogen from the ether than couple with another radical. The number of millimoles of iodobenzene recovered is close to that of the starting reagent present in smaller amount. Furthermore, the material balance for the other phenyl group, the sum of the millimoles of benzene and alkylbenzene and twice the millimoles of biphenyl, is often 80100% of the theoretical. I n accord with these observations, reactions in which alkyl iodides were formed also produced larger amounts of products other than iodobenzen e. I t has been found t h a t some of the missing phenyl groups are in the ethyl ether of methylphenylcarbinol which arises from interaction with solvent. The simplest route for its formation would be radical coupling. Alternatively, this product might CH,
C6116.
+ CH3CHOCH2CH3
-3 C6H6CHOCH2CH3
have been formed from trisubstituted iodine by attack of solvent radical or its derived carbonium ion.17-1 Y CH~ CH3CHOCH2CH3+ CsH5CHOCH2CH3
-t( CJ-Id21R
+ CsHbIK (ref. 17) +
CH3CIIOCH2CII3 + CH3CIIOCH~CHa
+
-t C ~ I I J C ~ H ~
+ C&SIC&,
(ref. 17, 18)
CHI
I
CH3bHOCH2CH3--+ CsH5CHOCI-12CH8(ref. 19)
+ (C6Hj)zIR
+GH~R
Finally, it should be noted that in the present work the yields of alkylbenzenes (Table 11) are substantially lower than those previously reported. ( 1 7 ) For discussions of t h e formation and reactions of diphenyliodine and related radicals see (a) H. E. Bachofner, F. M. Beringer and L. RIeites, THISJOURNAL, 80, 4269 (1958); (b) F. hl. Beringer, E. L f . Girtdler, M . Kapoport and R. J . Taylor. ibid., 81, 351 (1959). (IS) T h e rtxidatihn of sulvent radical by iodonium ion was proposed in ref. 171) arid is annlogous t o - ,washcd with 300 ml. of cold ether and used imtnedintely; see prepar3 tion of di-1-naphthyliodonium bromide. Diphenyliodonium chloride*j and bron1ide24 were prcparctl directly from benzene. Ethereal solutions of cirganoniagnesium halides and organolithium compi~undsjeycept 1chlorophenyllithium, l - n a p h t h ~ - l l i t h i uand n ~ 2-nap11tliyllith(20) Capillary melting points were taken in a Hershberg apparatus. For a discussion of techniques and their effects on observed ranges of melting and decomposition see F. M . Beringer, R . A . Falk, Rf. Karniol, I. Lillien, G. hlasullo, M. Mausner and E. Sommer, THIS J O U R N A L , 81,342 (1959). (21) J. Schmidlin and M .Huber, Ber., 43, 2829 (1910). (22) L. C. Raiford and H. P. Lankelma, Tms JOURNAL, 47, 1118 (1925). (23) H. J. Lucas and E, R. Kennedy, "Organic Syntheses," Coll. Vol. 111, Johu Wiley and Sons,Inc., New York, N. Y . , 1ll5.5, p. 482. (24) F.M. Beringer, h l . Drexler, E. XI. Gindler and C. C. Lumpkin, THIS JOURNAL, 75, 2705 (1953). (25) F. M. Beringer, E. J . Geering, I. K u n t z and M . Mausner, J . Phys. Chem.. 60, 141 (19.56).
June 5 , 1960
ORGANOMETALLICS WITH IODOSOBENZENE DICHLORIDES
i u m ) were made from the alkyl or aryl halide with magnesium or lithium. 4-Chloroplirnyllithium , z 6 l-nnphthyllithiunizi* and 2-naphtliyllithiunll' wcre fiirrrirtl by thc it1tc.raction of n-butyllithiumz8 with 4-bromochlorobenzene, 1bromonaphthalene and 2-bromonaphthalene. The alkyl halides were of the highest purity commercially available (Distillation Products, Inc., or Matheson Co., I n c . ) and were redistilled before use. Merck magnesium turnings were used for t h e first runs; later, Dow magnesium of very high purity became available. With t h e latter, slightly higher yields of Grignard reagents were obtained. The ether and other solvents used were the best grades obtainable and were used without further purification. The preparation of aryllithium compounds and their reactions with aryliodoso dichlorides were all conducted under an atmosphere of dry nitrogen. Iodosobenzene Dichloride.-Iodobenzene (102 g., 0.5 mole) was treated with chlorine in 150 nil. of cold, dry chlortrform by the method of Lucas and KennedyZ3 t o give 130.2 g. (97y0) of iodosobenzene dichloride, m.p. 104-107°; reported m .p . 115-127 ,z9 110-136 O ,3" 120-12 1'. 31 Tripheny1iodine.-To 106.5 mmoles of phenyllithium in 200 ml. of anhydrous ether at -60" under nitrogen 14.65 g . (0.0533 mole) of iodosobenzene dichloride was added gradually. While first additions dissolved, in time bright yellow crystals of triphenyliodine formed from deep yellow solution. The reaction was stirred 30 min. longer a t -70 t o -65". Diphenyliodonium Bromide from Triphenyliodine with Iodosobenzene Dichloride.-An additional 14.7 g . (0.0533 mole) of iodosobenzene dichloride was added gradually a t -70" t o the above mixture. The yellow crystals gradually changed over 45 min. t o a white precipitate. The reaction was allowed t o stir another 50 hours (over a weekend) and come slowly t o room temperature. The thick white precipitate was collected and washed with 75 ml. of ether t o give 30 g. of solid which was suspended in 100 ml. of water and ground with a mortar and pestle. The aqueous suspension was filtered and washed with 50 ml. of water t o give 18.2 g. (47Yc) of crude diphenyliodonium bromide, which was recrystallized from 1000 ml. of water t o yield 12.57 g. (32.7y0), m . p . 229-230'. One gram of this salt was recrystallized from 60 ml. of methyl alcohol t o give 0.65 g. of diphenyliodonium bromide, m.p. 233-234'; reported m . p . 208°,24208209',2O 210°,32 230°.33 Anal. Calcd. for C12HloIBr: C , 39.92; H, 2.79. Found: C , 40.42; H, 2.92. Diphenyliodonium Bromide from Triphenyliodine with Hydrogen Chloride.-To 2.75 g. (10 mmoles) of iodosobenzene dichloride suspended in 15 ml. of anhydrous ether at -60" under nitrogen, 31 mmoles of phenyllithium in ether was added. Practically all the solid dissolved in about 3 min., and after about 8 min. yellow crystals precipitated. After 2 hours the yellow solid was allowed t o settle, t h e supernatant liquid was decanted, and the solid was washed by decantation with 25 ml. of cold anhydrous ether. (Lithium hydroxide in t h e supernatant a n d wash liquids was titrated with 11.3 mmoles of hydrochloric acid, which is the excess phenyllithium employed.) Another 25 ml. of cold anhydrous ether was added t o the residual solid and with stirring 32 mmoles of hydrogen chloride in ether was added gradually. T h e yellow needles immediately changed t o a white precipitate. T h e b a t h was filled with D r y Ice, and the reaction mixture was stirred overnight and allowed t o come gradually to room temperature. The thick white precipitate was filtered, suspended in 30 ml. of ether, filtered again and washed with ether. T h e 3.18 g. obtained was suspended in 15 ml. of water, filtered, washed, and then recrystallized from 80 ml. of water t o give 1.04 g. (29y0) of diphenyliodonium bromide as white needles, m . p . 223-226'.
-
(26) H. Gilman, W. Langham and F. W. Moore, THISJ O U R N A L62, , 2327 (1940). (27) (a) H . Gilman and F. W. Moore, ibid., 62, 1843 (1940); (b) H . Gilman and C. G. Brannen, i b i d . , 72, 4280 (1950). (28) H. Gilman, "Organic Reactions," Vol. V I I I , John Wiley and Sons, Inc., New York, N. Y., 1954, p. 285; H. Gilman, et al., THIS J O U R N A L71, , 1499 (1949). (29) C. Willgerodt, Bev., 41, 1097 (1908). (30) W. Caldwell and E. A. Werner, J . Chem. Soc.. 91, 240 (1907). (31) J. C. Nichol and R . B. Sandin, THISJOURNAL, 67, 1307 (1945). (32) hl. 0. Forster and J . H. Schaeppi, J . Chem. Soc., 101, 382 (1912). (33) C. Hartman and V. Aleyer, B e y . , 27, 502 (1804).
2951
Ai-al. Cdlcd. for ClzHloIBr: C, 92; H , 2 . 7 9 . Found: C , 40.00, 40.10; H , 2.79, 2.68. Reaction of Diphenyliodonium Chloride with Phenyllithium and its Recovery.-To 11.07 g. (35 mmoles) of diphenyliodonium chloride in 80 ml. of anhJ-drous ether 70 mmoles of phenyllithium in ether was added dropwise a t - 70". The solution a n d the precipitate gradually changed t o a bright )-ellow. After 2 hr., 83 nlmoles of hydrogen chloride in ether solution was added dropwise a t below -60'; The resulting white precipitate was stirred 30 min. a t -70 and then allowed t o stand overnight and gradually come t o room temperature. The product was filtered, washed with ether, and recrystallized from 600 ml. of water to yield 8.10 g . (22.4 mmoles) of diphenyliodonium bromide as white needles, m . p . 219-221'. There was thus of recovery of the diphenyliodonium cation. Iodine Trichloride with Phenyl1ithium.-Treatment of 23.4 g. (100 nimolcs) of iodine trichloride in 400 ml. of ether under nitrogen a t -60" with 200 mmoles of ethereal p t r n ? 1lithium was followed by prolonged stirring at -70 and warming t o room temperature. The usual work-up gave 2.94 g. of red-brown solid. Crystallization from water gave 0.73 g. of impure, light yellow diphenyliodonium iodide, Recrystallization from methanol gave 0.40 m . p . 147-153'. g. ( 1 mmole, lye) of colorless crystals, m . p . 156-157"; n o depression of m . p . on admixture with an authentic sample. Iodosobenzene Dichloride with Phenylmagnesium Chloride.-To 27.5 g. (100 mmoles) of iodosobenzene dichloride in 200 ml. of tetrahydrofuran under nitrogen a t -60" there was gradually added 100 mmoles of a tetrahydrofuran solution of phenylmagnesium chloride34 a t -70 t o -60'. After 45 minutes the color of the initially deep bright blue solutiiin began t o change to green and t o fade, while precipitation of a white solid commenced. After 44 hours a t this temperature and warming t o room temperature the precipitate was collected, ground with 20 ml. of water, collected again, washed with water and dried t o give 2.54 g . (8 mmoles, 8 % ) of diphenyliodonium chloride, m.p. 204-209". Crystallization from water and once from methanol gave 0.45 g. ( 1.4 mmoles, 1.4y0) of colorless crystals, m . p . 223-224'. 4,4'-Dimethyldiphenyliodonium Bromide.-To 100 mmoles of p-tolyllithium in 200 ml. of ether a t -60' 28.9 g. (100 mmoles) of p-tolyliodoso dichloride, m.p. 94-97.35 was added. The precipitate was bright yellow initially b u t then became colorless. The precipitate was collected and washed with water t o give 16.1 g . (41.4 mmoles, 41%) of crude 4,4'-dimethyldiphenyliodoniumbromide, m.p. 176.5-178.5". Crystallizations from water and methanol (three times) raised the n1.p. t o 181-182038; reported m.p. 171-173°,20 178°.37 Anal. Calcd. for Cl4HI4IBr:C , 43.21; H , 3.62. Found: C, 42.82; II, 3.56. 4,4'-Dichlorodiphenyliodonium Bromide.-To 4-chlorophenyllithium prepared from 100 mmoles each of 4-bromochlorobenzene and n-butyllithium in 100 ml. of ether there was added 30.94 g. (100 mmoles) of 4-chlorophcnyliodoso dichloride, m.p. 110-112,88 at -60" over 10 minutes. The yellow precipitate had made the mixture too thick t o stir, so an additional 100 ml. of ether was added. A4ftera n additional 10 minutes at -60" the precipitate became colorless. T h e temperature rose t o -22' over 48 hr. and then t o ambient. The collected solid (35.5 g . ) was ground with 100 ml. of water, collected again, washed with water and dried t o give 21.6 g. (50.2 mmoles, 50Yc) of crude 4,4'-dichlorodiphenyliodonium bromide. Crystallization from three liters of water gave 13.1 g. (30.5 mmoles, 31y0) of salt, m.u. 187-189': two further crvstallizations from methannl reported m.p. 187-188°,20 raised the m . p . t o 189-190'; 190", 3 8 204-205O. 2' Anal. Calcd. for C12HsC12BrI: C , 33.52; H , 1.88. Found: C,33.33,33.38; H,2.06, 1.91. (34) H. E. Ramsden. A. E. Balint, W. R . Whitford, J . J . Walburn and R . Cserr. J . Ovg. Chem., 22, 1202 (1957). (35) Reported m.p. 88-90°: R . A l . Keeler and L. J. Andrews, THIS JOURNAL 80, , 277 (1938). (36) See discussion of this melting point in Experimental section of ref. 20. (37) J. McCrae, Bey., 28, 97 (18%). ( 3 8 ) Reported" m.p. 115'. (39) L. Wilkinson, Bev., 28, 09 (1895).
Di-1-naphthyliodonium Bromide.--To roughly 20 inmoles ethereal I-naphthyllithium (from 20 mmoles each o f 1bromonaplithalene and n-butyllithiunl) there was d d e d roughly 20 mmoles of 1-naphthyliodoso dichloride (from 20 mmoles of I-iodonaphthalene). The yellow color persisted a f t e r 3 hours a t -60” and 42 hours between -60 and -40’. The reaction mixture was treated a t - 70”with 40 mmoles of cthercal hydrogen chloride ( t o convert any tri-l-naphthyliodine t o the iodonium salt). .After 2 hours at this tempernture, the white mixture was warmed to ambient. The collected precipitate was ground with 12 ml. of water, filtered, waslied n i t h water and dried t o give 0.71 g. (1.54 mmoleq, X?;) of crude di-1-naphthyliodonium bromide, n . p . 116147’. Cry,stalliz:iti~~~~ from rnethanol raised the r r i . p . t o I 49-t5lo. ;Inii/. Calcd. for C2:>HI4BrI: C, 52.09; H, 3.06. Found: C,*?l.91; 13, 3.12. Phenyl-1-naphthyliodonium Bromide.-To roughly 100 ininoles of ethereal 1-naphthyllithiurn (from 100 mmoles each of 1-bromonaphthalene and n-butyllithiurnj there was added 2’7.5 g. (100 mmoles) of phenyliodoso dichloride a t -75“. After 0.5 hour the yellow color had disappeared, hut stirring a t -75 t o -45” was continued for 48 hours. The precipitate was collected, ground with 150 ml. of water, colIccted, washed with water and dried t o give 22.6 g . (55 rnmoles, 5.57,) of crude phenyl-1-naphthyliodonium bromide, m . p . 148-156’. One crystallization from water and two from methanol raised the n1.p. to 168-189”. Anal. Calcd. for CloH12BrI:C , 46.74; H, 2.91. Found: C, 46.74; H , 2.94. 4-Chlorophenyl-1-naphthyliodoniumBromide.-The reaction was run and worked u p as for the preceding euperinient (except t h a t 4chlorophenyliodoso dichloride was used ) and gave 2 i . 1 g. (60.8 mmoles, 610/,) of crude 4-chlorophenyl-1-naphthyliodoniumbromide as a light yellow solid. While three crystallizations from water gave colorless cr?-stals, m.p. 150-154’, a further crystallization from methanol raised the m.p. to 161-162’. 4 ? i a l . Calcd. for CleHIIBrCII: C, 43.13; H , 2.49. Found: C, 43.32; H, 2.63. (if
Reactions with Iodonium Salts General Procedure.--.%bout 250 ml. of ethereal solution of Grignard reagent was prepared in the usual way from 0.1 mole of alkyl or aryl halide and 0.1 g. atom of magnesium. To the cooled solution, iodonium salt was added in about 3 g. portions. Each addition caused superficial ether boiling and a fleeting yellow coloration. -1ddition of excess iodoniurn salt fovcr the molar equivalent) did not cause further hoiling. T h e iodonium salt dissolved in the ethereal solution. X heavy sticky white solid of smaller total bulk appeared from tlie solution as addition was continued. Depending on the state of subdivision, this new solid sometimes appeared t o be a second, heavier liquid phase. Stirring mas usually continued for 0.5 hour after the addition was complete. Several of tlie reactions were carried out with the Grignard solution kept under 0” by an ice-salt-bath. The iodonium salt was added a t intervals sufficiently spaced apart t o prevent an>- disturbance a t the surface of the ether. I n these reactions an intensely yellow-orange colored precipitate formed during the addition, which reverted t o a white sticky solid upon stirring a t room temperature. S o apparent evolution of heat accompanied this transition. To the reaction mixtures 5-10 ml. of 1Oyosulfuric acid was added dropwise, causing vigorous boiling. T h e ether layer was separated, washed twice with bicarbonate solution and dried over anhydrous sodium sulfate or potassium carbonate Iwfore distillation. .I small amount of free iodine was usually present in the ether layer at this point. T h e dried ether solutions containing the reaction products werc firqt distilled to remove ether using a 21-hall Snyder
colunin. iV1ie1i the volume was reduccd to about 35 nil., tlie solution was transferred to a 5O-ml. flask and the distillatifin w a s coiitinuctl utilizing a smaller, #-ball column. P‘roiii Soinc runs biphenyl was isolated from thc small residue. Identification of the Products.-Preliminary identification of liquids ivas made on the basis of hoiling point, delisit!and refractive index. The infrared spectrum of most of t h c cuts and residues was taken on a Beckman single beam aut[)niatic recording infrared spectrophotometer between 15.1) :ind 2.3 p . IZefereiice spectra of pure benzene, alkylbenzenes and iodobenzene wcre taken under the same operating ciinditions. Senii-quantitative estimates were made o n t h c I)a.;ii uf tlirse stantlards. 13iphenyl was identified b y recry.;tallizati(in of reiitlues froni rnethanol and dctcrrnining t lic capillary melting point a n t i tnixcd melting point with am :IIItheiitic sample. Diphenyliodonium Chloride with Methylmagnesium Bromide.-Ysing the procedure described previouslJ-, 50 g . i160 rnmoles) of diphenyliodonium chloride wv3s added ti) 201) mmnles of inethylmagiiesiuin bromide in 300 nil. of ether with mciderate ice-bath cooling t o prevent excessivc ether hoiling. The reaction was “decomposed” with 10 g. of sulfuric acid in 100 nil. of aqueous solution, washed twice with j C,fc ‘ sodium . bicarbonate, twice with water, dried over anhj,drous sodium sulfate and distilled in a 21-hall Snyder column t o remove the ether. h sample of the recovered ether did not gi\-e a positive test for methyl iodide using alcoholic silver nitrate acidified with nitric acid. The 43.0 g . of product remaining after the ether was removed was further fractimated in a n 8-ball Snyder column t o give three main cuts (henzene, toluene, iodobenzene and a biphenyl-containing resid u e ) . Infrared analysis of the iodobenzene (cut 3 ) fraction in this experiment and in three others noted in Tahle I1 indicated t h a t the same impurity (identified as ethJ-l a-phenylethyl ether I was present in all four cases. Superimposed on the normal iodohenzene ipectrum in each case were strong peaks a t 1104, 701,762 and 2’990 c m . -1 (arranged in order ( i f decreasing strength). Isolation of Ethyl a-Phenvlethyl Ether from the Iodobenzene Cut.-To 1 . 5 g. of magnesiim turnings (60 nig. atoms) in a 50-1111. two-necked flask were added 20 nil. of anhydrous ethyl ether. d 5.0-nil. portion of the iodobenzene (cut 3 ) fraction was added to the flask, and the Grignard reaction mas initiated with a drop of methyl iodide. I n about 1 minute the reaction started and refluxed without added heat for about 30 minutes. The mixture was allowed t o stand one week a t room temperature hefore “decomposing” with 25 nil. of 1 -Y wlfuric acid. .lfter washing and drying in the usual xnaiincr, the ether solution was distilled in a microapparatus yielding 1.2 nil. of an oil of distillation range 17020Oo, predominantly a t 185’; d25251.00;infrared spectrum / v . solution of this material in carbon disulfide el\- identical witli an authentic sample of 7.5yo phcnJ-lcthyl ether in the same solvent with respect t o exactness of position and relative heights of the characteristic peaks. A slight amount of extraneous absorption in the recovered sample was classified as “very weak” and was ascribable to a trace of unreacted iodubenzene. The amount of this ether found hy isolation and infrared anal agreed with the amount calculated using the 1104 cni: ether peak in the untreated iodobenzene (cut 3 ) fractiou. The other three values shiiwn in Table I1 were similarly ohtained using the absorption a t 1104 cni.-’ as the analytical peak. hIislow40 has reproduced the spectrum of ethyl aphenyletliyl ether in carbon tetrachloride; the use of this solvent which absorbs very strongly in the 760 cm.-’ region resulted i n a spectrum which is qualitatively similar t o t h a t obtained in this study, but in which the peak a t i 6 2 c m - I was completely masked. Tliii ctlicr 1i:is a reported b . p . 78” 120 m m . ) , d25r0.9131. BROOKLYS, S.Y . (40) K. RIislow, THISJ O U R N A I . , 73, :(Si54 (1951)
