Pyridyllithium Compounds
Pyridyllithium Compounds
Pyridyllithium Compounds
PYRHDYLLITHIUM COMPOUTSDS
HENRY GILhIAN AND SYDNEY M. SPATilll
and/or IT.
Li Bu
BuLi + BuBr V.
3-PyLi
1
L
etc.
1 Present Address: Polak’s Frutal Works, Inc., Middletown, N. Y.
1485
1456 NENBY GILMAN AND SYDSEY M . SPATZ
In conjunction with the carbazole studies, these facts suggest a stepwise mecha-
nism for multiple interconversions in polyhalogenated compounds. Interestingly,
the application of the modified entrainment method to 2,6-dibromopyridine
yields the dimagnesium compounds (3).
TOdate, only the bromo- and iodo-pyridines and -quinolii:es have responded
t a the X-M interconversion. Compounds containing a- and ?-chlorine atoms,
recognized for their increased reactivity toward hydrolysis, ammonolysis, and
etherification by alkoxides, have not yet been found capable of the X-AI inter-
conversion reaction (lb, 13).
EXPERIMENTAL
S-Pyridyllithium from 8-bromopyridine. Of the several experiments summarized in Table
I, only experiment 1is described in detail. A solution of 10 g. (0.063mole) of 3-bromopyridine
in 50 ce. of ethyl ether was added dropwise under nitrogen to a rapidly stirred solution of
1485 HENRY G I L M d S AND SYDNEY M. SPdTL
0.07 mole of n-butyllithium in 150 cc. of ether a t -5" t o 0" over a period of 25 minutes. Each
addition of the 3-bromopyridine formed a yelloT,.; flocculent precipitate which gradually
changed t o a more compact red-brown body with the simultaneous coloration of the solution
to brick red. When all of the bromopyridine had been added, agitation of the mixture was
TABLE I
3-PYRIDYLLITHIUM
-
EXP
- YIELD
OB 3-
KO.
STABTIljCl X:ITEBIbL*
I
INIEECOXVEBTILW RZAGENT
Gi
2"
PyLi,
%
- -
1 3-BrPy n-CiHgLi 30 30.5
0.063 m. in 50 cc. 0.07 m. in 150 cc.
2 3-BrPy n-CAHoLi 15 16.8
0.063 m. in 100 cc. 0.07 m. in 150 cc.
3 3-BrPy n-CrH,Li Ethyl ethercbd -20" 16 16.9
0.045 m. in 50 cc. 0.052 m. in 100 c5.
4 3-BrPy WC6HllLi Ethyl etherold -20" 15 24.5
0.057 m . in 50 cc. 0.067 m . in 100 cc.
5 3-BrPy sec-CoHoLi Petroleum 15 35.7
0.057 rn. in 40 cc. 0.065 m. in 140 cc. ether (b.p.
20-28')
6
7
3-BrPy
0.Q47 m. in 40 cc.
3-BrPy
tert-CdHdi
. .
0.056 m. in 1419 cc.
n-CdH&i
1
Petroleum
etherd
Petroleum
'
' -20 to -15"
15
15
32.9
34.3
0.056 fii. i 3 40 cc. 0.062 m. in 140 ce. etherd
8 3-IPv n-C&Li Et5yl ether6 -20 t o -16" 41 41 . 3
0.017 m. in 30 cc. 0.027 m. in 80 cc.
9 3-IPy n-CiHsLi Ethyl ether" -20 t o -15" 25 42.2
0.02 m. ia 35 cc. 0.027 m. in 100 cc.
10 3-IPy n-C4HgLi Ethyl etherczd -20 t o -15" 16 46.3
0.02 m. in 3: cc. 0.027 m. in 100 cc.
11 3-IPy n-C4HgLi Ethyl ether6 -20 t o -15" ~ 33 40.2
0.013 m. in 25 cc. 0.018 m. in 75 cc,
brown residue was hydrolyzed t o a thick oil, but except for the isolation of a minute amount
of pyridine, t h e rest of the oily material remains t o be identified. It did not contain any
3-bromopyridine.
Bfter hydrolysis of the carbonation mixture with 1% potassium hydroxide solution, the
alkaline layer was boiled with Norit, filtered, acidified with hydrochloric acid, and boiled
tQremove odor traces of valeric acid. The solution, neutralized t o litmus, yielded via the
copper salt (14) 2.4 g. (30.8%) of nicotinic acid, melting a t 231-232" (15). The picrate melted
a t 214-215", which agrees with t h e reported value (16). The octadecylamine salt of this
acid, prepared by Dr. B. 9. Hunter (17), melted a t 78-79', and a mixture melting point with
an suthentic specimen showed no depression.
Examination of the ether layers of experiments 3 and 4, subsequent t o carbonation and
hydrolysis, yielded 0.07 g. and 0.04 g . of pyridine, respectively, but no 3-bromopgridine.
The preparation of the n-amyllithium (12b) for experiment 4 was effected in 84Q/, yield
from 8.5 g. (0.08 mole) of n-amyl chloride in 60 cc. of petroleum ether (b.p. 20-28") and
1.25 g. (0.18 g.-atom) of lithium in 60 cc. of the same solvent.
The sec-hutyEithium (1%) for experiment 5 was prepared in 65% yield from 0.1 mole of
sec-butyl chloride and 0.22 g.-atom of lithium. The lert-butyllithium for experiment 6 was
prepared by Dr. W. F. Moore from ierl-butyl chloride (12b).
I n experiments 5 a r d 7, one-half t o one-cc. samples of the reaction mixture were with-
drawn jnst prior t o carbonation for Color Tests I (18) and IIa (19). I n both experiments, the
former test was positive (deep, dark green) and the latter negative, indicating t h a t tha
orgenolithiurn G Q E I ~ Q Uwas~ not an alkyilithium compound. sec-Butyllithium itself in
petrolemi ether gave a red-violet Color Test I and a bright salmon red Color Test I I a .
FIG.I
The 3-bromopyridlne for experiments 2 and 4 was provided by the kindness of Dr. 8 . $.
Harris of Merck tk Co. Eastman's 3-bromopyridine was employed in the other experiments.
Wniike the preparation of the other pyridyllithium compounds, t h a t of the 3-
pyridyllithium is accompanied by precipitate formation, presumably originating from
either simple or multiple RLi additions t o t h e --N=C< group (reactions IV and V I I I ) ,
The absence of precipitate formation does not necessarily infer the absence of ani1 addition.
For example, t h e reaction between o-anisyllithium and quinoline may or may not precipi-
tate the intermediary l-lithio-2-o-anisyl-I, 2-dihydroquinoline a i t h o u t affecting materially
the uitir-ate isolation of 2-o-anisylquinoline (13). Since the alkyllithium compound is con-
sumed comp:etely n-ithin 15 minutes, the occurrence of the rapid reactions I, I Y , and V I I I
is strongly suggested. The product of reaction I, as already indicated, is dissolved in the
solvent. The insoluble body, since it yielded neither pyridine nor 3-bromopyridine subse-
quent t o hydrolysis, is neither 3-pyridyllithiuni nor an addition complex (Fig. 1) of the
type postulated for pyridine and phenylmagnesium bromide (20) ; the insoluble body must,
therefore, hzve resulted from reactions I V and/or VIII.
d-P~ridylZithiunzfrom 3-iodepyridine. I n experiment 8, Table I, the ethereal solution of
iodopyridine4 was added dropwise over B 36-minute period t o a vigorously stirred solution
of n-butyllithiurn. The pale yellow, voluminous precipitate gradually changed t o a compact
body. The soIution darkened from yellow through orange t o brick red. When all of the
iodopyridine had been added, stirring v a s arrested t o permit settling of the dark red precipi-
4 Provided through the courtesy of the late Dr. F. C. Schmelkes, Wallace and Tiernan
t a t e ; this was accomplished within five minutes. Also, from the clear supernatant liquid a
white, crystalline material settled out in increasing amount toward the end of the reaction
period. The supernatant solution mas easily decanted upon D r y Ice and processed for t h e
nicotinic acid.
I n experiment 11, the white, crystalline material was rinsed off the dark red gum with
dry ether. After one recrystallization from water, i t rvas identified as lithium iodide, t h e
inorganic product of coupling. I t s chronological appearance, largely toward the end of t h e
reaction period, suggests t h a t the primary coupling (reaction 111) is of minor significance.
Of the three secondary coupling reactions, VI and/or VI1 appear to be the dominant ones
since the color test I I a (19) for alkyllithium compounds is negative long before the major
portion of the lithium iodide has precipitated.
It is interesting t o note t h a t the pyridyllithium solution from 3-iodopyridine and n-
butyllithium in ethyl ether is more easily decanted than the pyridyllithium solution from
3-bromopyridine in t h e same solvent.
b-P2/ridgllithium and pyridyl-l-phenylcarbinol. -4solution of 12.6 g. (0.08 mole) of 2-
bromopyridine6 in 50 cc. of ether was added slowly over a three-minute period t o 0.09 mole
of n-butyllithium in 200 cc. of ether a t -Bo, and the red solution stirred for seven minutes.
T o the solution of 2-pyridyllithium was added, over a five-minute period, 10.6 g. (0.1 mole)
of freshly purified benzaldehyde in 50 cc. of ether. When most of the aldehyde had been
added, t h e red color suddenly disappeared and a gray precipitate formed which gradually
turned t o yellow by the end of 30 minutes. The temperature was kept a t -18" throughout.
At t h e end of this time, a sample of the reaction contents gave a negative color test (18).
After hydrolysis a i t h an ice-cold solution of ammonium chloride, t h e ether layer wa8 filtered
from a small amount of insoluble residue and extracted with 10% hydrochloric acid. Neu-
tralization of t h e acid extract gave a dark oil, which was dissolved in ether and the solution
dried over potassium carbonate. Fractional distillation gave a pale yellow oil (b.p. 143'/2
mm.) which gradually solidified t o a white, crystalline mass, melting a t 68-74", The yield
of crude carbinol was 10.1 g. or 68.7% on the basis of the 2-bromopyridine used.
A sample of t h e crude product, dissolved in 95% ethanol, gave a picrate, melting a t
169O, which is in agreement with the value reported by Ashworth, et al. (21).
Recrystallization of the carbinol from an ether-ligroin mixture gave heavy, white crys-
tals, melting a t 76-78". Ashworth reported the melting point as 78". The yield of pure product
was 9.3 g. (62.7%).
b-Bromo-9-pyridylZithiz~m.A solution of 10 g. (0.042 mole) of 3,5-dibromopyridine (East-
man) in 100 cc. of ether, cooled t o -30", was added with stirring over a two-minute period
t o a filtered, similarly cooled solution of n-butyllithium (0.11 mole) in 160 cc. of ether. The
reactants were stirred for ten more minutes a t -30' and for five minutes with the cooling
bath removed. After carbonation, 4.35 g. of crude acid was isolated from the alkaline layer.
The acidic material was redissolved in dilute base and then acidified with glacial acetic
acid, which precipitated 0.1 g. of a yellow colored product, melting a t 213-217'. Further
acidification of the filtrate with dilute hydrochloric acid yielded 3.5 g . (41%) of 5-bromo-
nicotinic acid, melting, as reported (22,23), a t 182-183".
When 9.9 g. (0.4 mole) of 3,5-dibromopyridine3 in 150 cc. of ether was added gradually
t o 0.12 mole of n-butyllithium in 200 cc. of ether and stirred at -16" for 50 minutes, t h e
transparent orange colored solution gave, subsequent to carbonation, a 64.1% yield of crude
5-bromonicotinic acid. The yield of pure product was 46.3%.
-4nother 9.9 g. of the 3,5-dibromopyridine3 and 0.11 mole of n-butyllithium was kept a t
-35" for ten minutes, and then refluxed for 30 minutes in an attempted di-interconversion.
After carbonation, only a minute quantity of highly impure acidic material mas isolated.
Substitution of n-propyl- for n-butyl-lithium also failed t o effect di-interconversion. A
solution of 0.04 mole of the pyridine compound in 150 oc. of ether was treated with 0.12 mole
of n-propyllithium in 20!3 cc. of the same solvent for 25 minutes at -35 t o -10' t o yield,
~
subsequently, 4.0 g. (47.2%) of crude 5-bromonicotinic acid. The yield of pure product was
2.7 g. (32.1%).
Methyl 5-bromonicotinate. The methyl ester was prepared quantitatively by the action of
excess diazomethane on an ethereal solution of 0.1 g. of the acid. The ester melted a t 48-99",
as reported (23).
B-Brorno-2-pyridyllithium. In this case, (10 g., 0.042 mole) of 2,6-dibromopyridine was
treated with 0.11 mole of n-butyllithium in the manner described for the preparation of
the 5-bromo-3-pyridyllithium. The purified acid, isolated after carbonation, melted a t 192-
194'. Obtained in a yield of 3.85 g. (45.270), t h e compound gave satisfactory bromine and
nitrogen analyses for 6-bromopicolinic acid, as yet unreported in the literature. Likewise,
t h e methyl ester derivative was found t o have the correct nitrogen content. However, the
acid itself gave unsatisfactory neutral equivalent data when titrated with dilute alkali
against phenolphthalein as indicator.
Anal. Calc'd for CJ%BrKOz: Br, 39.57; N, 6.95; Neut. equiv., 202.
Found: Br, 39.13, 39.34; N, 6.88, 6.52; Neut. equiv., 181, 182.2.
An attempted di-interconversion between 12.4 g. (0.05 mole) of 2,6-dibromopyridine in
150 cc. of ether and 0.16 mole of n-butyllithium, maintained a t -16" for 55 minutes, gave,
instead, a 39.8% yield of 6-bromopicolinic acid. I n a run carried out at the temperature of
t h e refluxing ether, no acidic material was isolated after carbonation.
Methyl 6-bromopicolinute. Methylation of 6-bromopicolinic acid with diazomethane gave
a crystalline ester, which softened a t 92-93" and melted at 93-94'.
Anal. Calc'd for C7HeBrN02:N , 6.41. Found: N, 6.49.
DISCUSSION
.. ..
FIG. 2
by less ani1 additisii and accordingly more X-M interconversion in the molecule
represented by Fig, 2 than in 2-bromopyridine.
The superior X-M interconversion yield from 3-iodopyridine, compared ‘io
that from 3-bromopyridine, is in accord with the greater polarizability of the
iodine atom (- I > - Br > - C1 > - F).6
SCMMARY
REFERENCES
(1) (a) GILMANASD SPATZ,J. Am. Chem. SOC.,62, 446 (1940); (b) 63, 1553 (1941); ( e )
SPATZASD GILNAN,Proc. Iowa Acad. Sci., 47, 262 (1940); (d) SFATZ, Iowa State
Coll. J . Sci., 17, 129 (1942).
(2) GILMANIND ?~IELSTROM, J . Am. Chem. SOC.,68, 103 (1946).
(3) OVERHQFF ASD PRQOST, Rec. trav. chim., 57, 179 (1938); PROQST AND WIBATT, Rec.
trav.chim., 59, 971 (1940).
(4) ALLENAND F ~ a u n J, . Am. Chem. SOC., 62, 1301 (1940).
(5) SACHSARD SACHS, Ber., 37, 3088 (1904); ROWITII AX’D KQPHE.A m . , 396, 38 (1913);
WIBAUTA N D OVERHOFF,Rec. trav. chim., 47,761 (1928); S. A. HAZEIS,Iowa State
Coll. J . Sci., 6, 425 (1932); HERTOG AS> WIBAVT,Rec. trav.chim., 66, 122 (1936).
( 6 ) GiL&1.4N, LANGHAM, . ~ X DJACOBY, J . A m . Chem. Xoc., 61, 106 (1939); WITTIG, POCICELS,
AND DROGE, Ber., 71, 1903 (1938).
( 7 ) I. BANNER, M.S. Thesis, Iowa State College, Axes, Iowa, 1939.
(8) Unpublished work of Drs. S. &I. Spatz and Leo Tolman.
(9) Z I E G L E R AND Z E I S E B , Ber., 83, 1847 (1930); Ann., 485, 174 (1931); BERQMSWN, et d.,
Ann., 483, 80 (1930); J . prakt. Chem., [2] 135, 267 (1932).
(10) Unpublished work of Dr. J. A. V. Turck, Jr.
(11) Unpublished work of Dr. J. T. Edward.
(12) (a) GILMANAXD R I O D R ~J, . Am. Cheni. SOC., 629 1843 (1940) ; (a) GILUAN,MDQIZE, AND
BAINE,J . Am. Chem. Soc., 83, 2479 (1941).
(13) GILMANAND SD.4TZ. J . Am. Chem. SOC., 66, 621 (1944).
(14) VON PECHNANN ASD WCLSH,Ber., 17, 2392 (1884); S C H E I BASD ~ E MNOTIIC, B e y . , 46,
2258 (1912).
(16) Beilstein, 22, 38 (1935).
(16) SUZUKI, SHIMAYURA, A N D QDAKE, Biocham. z., 43, 99 (1912).
(17) HCNTER, Iowa State Coll. J. Sci., 15, 223 (1941).
(18) GILXAN-4ND SCHULZE, J . Am. Chem. SOC., 47, 2302 (1925).
(19) GILXANASD SWISS,J . Am. Chem. SOC.,62, 1847 (1943).
(20) BERGSTIZOM A N D MCALLISTER,J . Am. Chem. SOC., 52, 2845 (1940).
(21) ~ ~ S B W O R T HDIFFERN,
, A R D HAMMICK, J . Chem. SOC., 809 (1939).
(22) CLAUSA N D PYCHLAU, J . prakt. Chem. [2] 47, 414 (1893).
ieaction, particularly in respect t o the nature of the organolithium compound, the kind
and number of halogen atoms present, and the Iiuclear positions occupied by the halogen
atoms, see Sunthankar and Gilman, J. Org. Chem., 16, 8 (1951).
1494 HENRY GILMAS AND SYDNEY M . SPAT2