Regioselective hydroxylation of 2,4-lutidine: A practical synthesis of 4-hydroxymethyl-2-methylpyridine

Article abstract of DOI:10.1055/s-2002-20960

A practical synthesis of 4-hydroxymethyl-2-methylpyridine has been developed which makes use of Evans' regioselective lithiation of readily available 2,4-lutidine and trapping with dimethylformamide.

Full text of DOI:10.1055/s-2002-20960

PAPER
483
Regioselective Hydroxylation of 2,4-Lutidine: A Practical Synthesis of
4-Hydroxymethyl-2-methylpyridine
Synthesis
oof 4-Hydroxh
n
e
thylpyridine A. Ragan,* Brian P. Jones, Clifford N. Meltz, John J. Teixeira Jr.
Chemical Research and Development, Pfizer Global Research & Development, P.O. Box 8156-101, Groton, CT 06340, Canada
Fax +1(860)4416334; E-mail: john_a_ragan@groton.pfizer.com
Received 13 November 2001; revised 22 January 2002
electrophile trap for the anion that would install a hydrox-
yl moiety (i.e. an ‘
+’ equivalent). As summarized in
the Table, several electrophiles were examined
(Scheme 1).
Abstract: A practical synthesis of 4-hydroxymethyl-2-methylpyri-
dine has been developed which makes use of Evans’ regioselective
lithiation of readily available 2,4-lutidine and trapping with dimeth-
ylformamide.
Key words: lithiation, hydroxylation, regioselectivity, pyridines
Entries 1 and 23 were designed to directly provide either
alcohol 1 or the corresponding hydroperoxide. Others
were aimed at providing intermediates that could then be
converted to 1 [e.g. bromide (entries 3 and 4), organobor-
ane (entries 5 and 6), sulfide (entry 7), or pyridylacetic
acid (entries 8–10)]. Unfortunately, all of these electro-
philes failed to provide the desired product: starting mate-
rial was recovered unchanged, and in some cases
oxidative self-condensation product 2 was observed. This
type of oxidative dimerization of 4-methylpyridines has
several literature precedents, with oxidants such as ele-
mental sulfur or selenium,4 peroxides,5 KMnO4,6 iodine,7
bromine,8 and dibromoethane.8,9 In the case of dibromoet-
hane, the yield of dimeric products can be preparatively
useful; consistent with this, in entry 3, bis-pyridine 2 was
the sole product observed in the crude NMR.
We recently required a practical, scaleable synthesis of 4-
hydroxymethyl-2-methylpyridine (1, Figure). Previous
syntheses of this compound were deemed unsuitable for
our purpose, as they generate a mixture of regioisomers
and involve potentially high energy pyridine N-oxides.1
This paper describes the development of a practical, regi-
oselective synthesis of 1, which makes use of the selective
lithiation of 2,4-lutidine. Multigram quantities of 1 are
readily available using this method.
OH
N
Entries 11 and 12 show the successful trapping with sili-
con electrophiles: both Me3SiCl and Ph2MeSiCl provide
the target 4-pyridylmethylsilanes in good yields.10 Unfor-
tunately, all attempts to oxidatively cleave the carbon–sil-
icon bond were unsuccessful. Desilylation to form 3 was
the major pathway, with at most 15–20% levels of the de-
sired alcohol 1 being formed [several reagents were inves-
tigated: PhI(CF3CO2)2,11 Hg(OAc)2/CH3CO3H,12 and
mCPBA13].
1
Figure The structure of 4-hydroxymethyl-2-methylpyridine.
Evans has reported the regioselective lithiation of 2,4-lu-
tidine with lithium diethylamide, and trapping of the re-
sulting anion with benzaldehyde.2 The same selectivity
can be realized by metalation with BuLi, followed by ad-
dition of 3 equivalents of diethylamine. The role of the
amine is to facilitate equilibration between the 2- and 4-
lithio species, leading to the thermodynamically more sta-
ble 4-lithio species. We reasoned that we could take ad-
vantage of this regioselective metalation by identifying an
Entries 10–12 show the result of trapping with carbon
electrophiles: while CO2 and ClCO2Et returned un-
changed starting material, DMF provided an excellent
yield of dimethyl enamine 4, along with smaller quantities
Li
n-BuLi
Et2NH
E
N
E+
+
N
N
THF
N
N
2
3
Scheme 1
Synthesis 2002, No. 4, 15 03 2002. Article Identifier:
1437-210X,E;2002,0,04,0483,0486,ftx,en;M05501SS.pdf.
© Georg Thieme Verlag Stuttgart · New York
ISSN 0039-7881

484
J. A. Ragan et al.
PAPER
Table Reactions of Electrophiles with Lithiated 3
Entry Electrophile Target Result
N
n-BuLi
OH
CHO
NaBH4
NaIO4
Et2NH
1
O2
E = OH
(or OOH)
SM 3 + dimer 2, ca. 1:1
N
aq NaOH
THF
aq THF
N
THF;
N
N
23
E = OH
(or OOH)
Recovered SM 3
DMF trap
5
4
1
3
aq SO2
N
THF
S
O
O
O
O
O
3
Br(CH2)2Br
E = Br
Dimer (2) was the only
product
S
OH
HO
4
5
6
7
8
9
NBS
E = Br
Recovered SM 3
Recovered SM 3
Recovered SM 3
Recovered SM 3
Recovered SM 3
Recovered SM 3
N
B(OPr-i)3
B(OMe)3
PhSSPh
(MeO)2CO
CO2
E = B(OPr-i)2
E = B(OMe)2
E = SPh
6
Scheme 2
E = CO2Me
E = CO2H
E = CO2Et
of the corresponding diethyl enamine (from exchange of
diethylamine with dimethylamine). This was viewed as a
promising intermediate, as oxidative cleavage of enam-
ines to aldehydes with NaIO4 is well precedented in relat-
ed pyridines.14
10
ClCO2Et
Recovered SM 3 +
Et2NCO2Et
11
12
13
Me3SiCl
Ph2MeSiCl
DMF
E = SiMe3
Desired product. Subse-
quent oxidation unsuc-
cessful (see text)
As summarized in Scheme 2, this strategy was successful-
ly executed to provide a three step, one-pot synthesis of 4-
hydroxymethyl-2-methylpyridine (1) from 2,4-lutidine
(3).
E = SiPh2Me
Desired product. Subse-
quent oxidation unsuc-
cessful (see text)
Enamine 4 was formed in reasonably pure form following
an aqueous workup. Although it was stable to chromatog-
raphy, this did not provide an efficient purification, and it
was deemed preferable to carry the crude enamine directly
into the oxidative cleavage to generate aldehyde 5. On
laboratory scale, it was found most convenient to directly
reduce aldehyde 5 and isolate the target alcohol 1 by chro-
matography. The overall yield for this three-step sequence
was 71% (Scheme 2).
Desired product 4. Suc-
cessfully converted to
1, see Scheme
N
N
which can be reduced in situ to form 1. Although the effi-
ciency of isolating bisulfite adduct 6 is less than going di-
rectly to 1 (45% vs. 71%), it avoids the silica gel
chromatrography used in the latter procedure, and is thus
viewed as the preferred method for preparing kilogram
quantities of 1. Indeed, on 10 kg scale, the overall yield for
1 from 2,4-lutidine was 64%.
Since 1 is not a readily crystallizable material, we were in-
terested in developing non-chromatographic methods of
purification on a larger scale prior to the isolation of 1.
Salts of both 5 and 1 were studied, but none were found
which provided a convenient crystallization. However,
treatment of aldehyde 5 with aqueous sulfur dioxide
(H2SO3) generated the bisulfite adduct 6, which is a well-
behaved, crystalline, free-flowing solid, and provided a
convenient purification point in the sequence. We also
studied the sodium salt of 6, formed by treatment of alde-
hyde 5 with NaHSO3 in aqueous organic media. While the
desired sodium bisulfite adduct was generated as a white
solid from EtOAc–EtOH–H2O (5:3:1), combustion analy-
sis indicated that it was not a well-defined 1:1 adduct. The
In summary, a practical, regioselective synthesis of 4-hy-
droxymethyl-2-methylpyridine (1) has been developed
via Evans’ selective lithiation of 2,4-lutidine and trapping
with DMF. Crystalline bisulfite adduct 6 can be isolated
to avoid chromatographic purification of 1.
Reagents were purchased from commercial suppliers and used as
sodium bisulfite adduct was also less well-behaved in sub- received unless otherwise noted. DMF and THF were purchased in
anhyd form from Aldrich in ‘Sure-Seal’ glass bottles; all other sol-
sequent chemistry, providing lower yields than the ‘free
vents were reagent grade. Reactions were run under a positive pres-
acid’ bisulfite adduct. There are several literature prece-
sure of N2 glassware, which was flame-dried under N2. Reaction
dents for related bisulfite adducts of pyridinealdehydes.15
progress was monitored by TLC or GC/MS. TLC was performed on
precoated sheets of 60 F254 (Merck Art. 5719), and visualized by
Treatment of 6 with aqueous base regenerates aldehyde 5,
Synthesis 2002, No. 4, 483–486 ISSN 0039-7881 © Thieme Stuttgart · New York

PAPER
Synthesis of 4-Hydroxymethyl-2-methylpyridine
485
UV, and/or staining with I2, phosphomolybdic acid, ceric ammoni-
um molybdate, or p-anisaldehyde solutions and heating. GC analy-
ses were performed on a Hewlett-Packard 5890 GC/MS, HP-1
column (12 m 0.2 mm 0.33 m), 1 mL/min flow rate, injector
temp. 280 °C, oven temp. 133 °C for 0.1 min, then ramp 19 °C/min
to 310 °C, hold for 1.65 min). Retention times are as follows: enam-
ine 4 (2.40 min), aldehyde 5 (0.63 min), alcohol 1 (0.90 min), bis-
pyridine 2 (3.49 min). Mass spectral data was collected on either a
Hewlett-Packard 5890 GC/MS (electron impact ionization), or a
Micromass (Fisons) Platform II mass spectrometer (atmospheric
pressure chemical ionization). Combustion analyses were per-
formed by Schwarzkopf Microanalytical Laboratory, Woodside,
New York.
IR (film): 3212, 3035, 2948, 1644, 1630, 1233, 1206, 1169, 1097,
1029 cm–1.
1H NMR (D2O): = 8.37 (d, J = 6 Hz, 1 H), 7.80 (s, 1 H), 7.77 (d,
J = 6 Hz, 1 H), 5.58 (s, 1 H), 2.58 (s, 3 H).
13C NMR (D2O): = 155.9, 153.8, 140.1, 126.4, 122.8, 83.7, 19.2.
Anal. Calcd for C7H9NSO4: C, 41.37; H, 4.46; N, 6.89. Found: C,
41.72; H, 4.47; N, 6.76.
1,2-Bis(2-methyl-4-pyridyl)ethane (2)
A flame-dried, 50 mL, single neck round bottom flask was charged
with 2,4-lutidine (0.50 mL, 4.3 mmol) and THF (30 mL), and then
cooled to –50 °C. BuLi (1.9 mL of a 2.5 M solution in hexanes, 4.8
mmol) was added via syringe, and the resulting solution was stirred
at –50 °C for 25 min. Diethylamine (0.67 mL, 6.5 mmol) was then
added via syringe, and the stirring was continued for 20 min. 1,2-
Dibromoethane (0.75 mL, 8.7 mmol) was then added via a syringe.
After an additional 35 min at –50 °C, the reaction was quenched
with aq NH4Cl (5 mL), and allowed to warm to r.t. The mixture was
transferred to a separatory funnel and extracted with CH2Cl2 (3 20
mL). The combined organic extracts were dried (MgSO4), filtered,
Dimethyl-[2-(2-methylpyridin-4-yl)vinyl]amine (3)
A flame-dried, 50 mL, single neck round bottom flask was charged
with 2,4-lutidine (2.00 mL, 17.3 mmol) and THF (20 mL), and then
cooled to –50 °C. BuLi (7.6 mL of a 2.5 M solution in hexanes, 19
mmol) was added via syringe, and the resulting solution was stirred
at –50 °C for 35 min. Diethylamine (2.68 mL, 26.0 mmol) was then
added via syringe, and the stirring was continued for 25 min. DMF
(2.68 mL, 34.6 mmol) was then added via syringe. After an addi-
tional 70 min at –50 °C, the reaction was quenched with aq NH4Cl
(5 mL), and allowed to warm to r.t. The mixture was transferred to
a separatory funnel and extracted with CH2Cl2 (3 20 mL). The
combined organic extracts were dried (MgSO4), filtered, and con-
centrated to provide the product as an orange oil (3.51 g, 125% of
theory due to presence of the diethylenamine and residual DMF).
1
and concentrated to provide 0.88 g of a pale yellow oil. H NMR
analysis showed dimer 2 plus unreacted dibromoethane ( = 3.64).
1H NMR (CDCl3): = 8.37 (d, J = 5 Hz, 2 H), 6.94 (s, 2 H), 6.87 (d,
J = 5 Hz, 2 H), 2.86 (s, 4 H), 2.51 (s, 6 H).
MS (EI): m/z (%) = 212 (100), 106 (85).
Large Scale Preparation of Hydroxy-(2-methylpyridin-4-yl)me-
thanesulfonic Acid (6)
Enamine 4
1H NMR (CDCl3): = 8.11 (d, J = 5 Hz, 1 H), 7.03 (d, J = 14 Hz,
1H), 6.83 (s, 1 H), 6.80 (d, J = 5 Hz, 1 H), 4.95 (d, J = 14 Hz, 1 H),
2.87 (s, 6 H), 2.43 (s, 3 H).
A 75 L Hastalloy stainless steel reaction vessel was charged with
THF (22.5 L, Karl–Fischer analysis 0.02% H2O) and 2,4-lutidine
(3.0 L, 26 mol). After cooling to –75 °C, BuLi (2.5 M in hexanes,
13.5 L, 34 mol) was added over 35 min (internal temp was kept be-
low –50 °C). Diethylamine (4.0 L, 38 mol) was then charged over
15 min, followed by DMF (4.0 L, 52 mol), also over 15 min. After
60 min, GC/MS analysis of an aliquot indicated complete consump-
tion of starting material. The reaction was quenched by addition of
sat. aq NH4Cl (9 L) over 5 min. The internal temperature rose to
–30 °C during the addition. The cooling was stopped, and the mix-
ture was allowed to warm to 15 °C. CH2Cl2 (12 L) was added, the
mixture was stirred for 10 min, then allowed to settle. The organic
phase was removed, and another portion of CH2Cl2 (12 L) was add-
ed. After stirring and allowing to settle, the organic phase was re-
moved and combined with the first extract. This solution was
concentrated under partial vacuum to a volume of ca. 8 L, then di-
luted to 60 L with fresh CH2Cl2 and held overnight at –12 °C prior
to use in the next reaction.
MS (EI): m/z = 162 (100), 147 (25).
4-Hydroxymethyl-2-methylpyridine (1)
NaIO4 (11.1 g, 51.9 mmol) was slurried in MeOH (10 mL). Enam-
ine 3 (17.3 mmol, the entirety of previous reaction) was dissolved
in MeOH (10 mL), and added to the periodate slurry dropwise. The
slurry was filtered through sintered glass, then added to a solution
of NaBH4 (0.72 g, 19 mmol) in MeOH (3 mL) at 0 °C. After 30 min,
GC/MS analysis indicated incomplete reduction, so an additional
portion of NaBH4 was added (0.72 g, 19 mmol). After an additional
30 min, the mixture was filtered, treated with silica gel (5 g) , and
concentrated to provide an orange solid, which was placed on top of
a short silica gel column (2.5 g) and eluted with 0 to 2 to 5%
MeOH–CH2Cl2 (avoiding an aqueous workup is advantageous due
to the high water solubility of 1). The product-containing fractions
are combined and concentrated to provide alcohol 1 as a pale orange
oil; yield: 1.52 g (71% yield from 2,4-lutidine).
IR (film): 3217 (br), 1608, 1562, 1446, 1405, 1063, 824 cm–1.
1H NMR (CDCl3): = 8.33 (d, J = 5 Hz, 1 H), 7.14 (s, 1 H), 7.06 (d,
J = 5 Hz, 1 H), 4.67 (s, 2 H), 4.05 (br s, 1 H), 2.49 (s, 3 H).
Aldehyde 5
A 200 L glass-lined reactor was charged with H2O (94 L) and NaIO4
(11.1 kg, 52 mol). To this stirred slurry was added the CH2Cl2 solu-
tion of crude enamine prepared above, from a cold (–12 °C) tank.
The addition was moderated such that the internal temperature of
the periodate slurry did not rise above 38 °C (addition took 30 min).
After 60 min, GC/MS analysis of an aliquot showed complete con-
sumption of the enamine. The resulting slurry was treated with suf-
ficient 2 N NaOH to bring the aqueous pH from 6.4 to 7.9 (ca. 250
mL were required). The mixture was then filtered through a Buch-
ner funnel (enclosed filter with a polyethylene filter paper), rinsing
with an additional amount of H2O (10 L). The lower organic phase
was separated, and the aqueous layer was extracted with an addi-
tional amount of CH2Cl2 (3 L). The organic extracts were combined,
and concentrated under partial vacuum to a volume of ca. 8 L (as an
amber oil). The above procedure was repeated twice more, such that
a total of 8.34 Kg of 2,4-lutidine were processed to provide a theory
13C NMR (CDCl3): = 157.6, 153.1, 147.8, 121.3, 118.9, 62.5, 23.8.
MS (EI): m/z = 123 (100), 94 (95).
Hydroxy-(2-methylpyridin-4-yl)methanesulfonic Acid (6)
From an earlier run of the previous reaction, a portion of the solu-
tion after filtration of the NaIO4 and prior to addition of the NaBH4
was concentrated to provide 1.00 g (8.25 mmol) of an orange oil
(GC/MS and 1H NMR indicated this to be reasonably pure aldehyde
5). This material was dissolved in THF (9 mL), cooled to 0 °C, and
treated with aq SO2 (sulfurous acid, H2SO3, 9 mL). After stirring for
60 min, the solids were collected, rinsed with 50% aq THF, and
dried in a vacuum oven to provide bisulfite adduct 6 as a pale yellow
solid (0.752 g, 45% yield).
Synthesis 2002, No. 4, 483–486 ISSN 0039-7881 © Thieme Stuttgart · New York

486
J. A. Ragan et al.
PAPER
(7) Wright, M. E.; Yamada, L. Y. Synth. Commun. 1988, 18,
of 78 mol of aldehyde 5 in a total volume of 24 L as a CH2Cl2 so-
lution.
1449.
(8) Lehn, J.-M.; Ziessel, R. Helv. Chim. Acta 1988, 71, 1511.
(9) (a) Sun, L.; Berglund, H.; Davydov, R.; Norrby, T.;
Hammarstroem, L.; Korall, P.; Borje, A.; Philouze, C.; Berg,
K.; Tran, A.; Andersson, M.; Stenhagen, G.; Mortensson, J.;
Almgren, M.; Styring, S.; Akermark, B. J. Am. Chem. Soc.
1997, 119, 6996. (b) Ito, Y.; Kobayashi, K.; Saegusa, T. J.
Organomet. Chem. 1986, 303, 301. (c) Elliott, C. M.;
Freitag, R. A.; Blaney, D. D. J. Am. Chem. Soc. 1985, 107,
4647.
(10) Trapping of a lithiated 4-methylpyridine with TMSCl is
precedented in a related system: Fraser, C. L.; Anastasi, N.
R.; Lamba, J. J. S. J. Org. Chem. 1997, 62, 9314.
(11) Andrews, I. P.; Lewis, N. J.; McKillop, A.; Wells, A. S.
Heterocycles 1996, 43, 1151.
Bisulfite Adduct 6
The enamine solution from above (combination of 3 runs, 78 mol)
was added to THF (94 L) in a 200 L glass-lined reaction vessel. To
this solution was slowly added 6% aq H2SO3 (SO2-saturated H2O,
84 L, 79 mol). The addition, which took 60 min, was moderated
such that the internal temperature remained below 35 °C. The re-
sulting solids were granulated for 2 h, then collected on a Büchner
funnel fitted with a polyethylene filter paper. The solids were
washed with isopropyl ether (1 L), transferred to a vacuum oven,
and dried until a constant weight was achieved, at which point Karl–
Fisher analysis indicates 0.4% of H2O. 10.10 kg (49.7 mol) of a
white solid was obtained (spectral and analytical data same as de-
scribed above for 6), with an overall yield of 64% from 2,4-lutidine.
(12) Fleming, I.; Sanderson, P. E. J. Tetrahedron Lett. 1987, 32,
4229.
(13) Ishihara, T.; Miwatashi, S.; Kuroboshi, M.; Utimoto, K.
Tetrahedron Lett. 1991, 36, 1069.
References
(1) (a) Furukawa, S. Pharm. Bull. (Japan) 1955, 3, 413; Chem.
Abstr. 1955, 50, 13926. (b) Katz, R. B.; Mistry, J.; Mitchell,
M. B. Synth. Commun. 1989, 19, 317.
(2) Evans, D. A.; Cee, V. J.; Smith, T. E.; Santiago, K. J. Org.
Lett. 1999, 1, 87.
(3) Davis, F. A.; Sheppard, A. C. Tetrahedron 1989, 54, 5703.
(4) (a) Thayer, H. I.; Corson, B. B. J. Am. Chem. Soc. 1948, 70,
2330. (b) Gleich, E.; Warnke, Z.; Szafranek, J.; Malinski, E.
Phosphorus Sulfur 1986, 28, 315.
(5) (a) Jahangir Wang, J.; Smith, R. W.; MacLean, D. B.;
Holland, H. L. Can. J. Chem. 1990, 68, 587. (b) Johnston,
K. M.; Williams, G. H. J. Chem. Soc. 1960, 1168.
(6) Kaiser, E. M. Tetrahedron 1983, 39, 2055.
(14) (a) Vetelino, J. G.; Coe, J. W. Tetrahedron Lett. 1994, 35,
219. (b) Dupau, P.; Renouard, T.; Le Bozec, H. Tetrahedron
Lett. 1996, 37, 7503. (c) Feliu, L.; Ajana, W.; Alvarez, M.;
Joule, J. A. Tetrahedron 1997, 53, 4511. (d) Niu, C.; Li, J.;
Doyle, T. W.; Chen, S.-H. Tetrahedron 1998, 54, 6311.
(e) Sasaki, I.; Daran, J. C.; Balavoine, G. G. A. Synthesis
1999, 815.
(15) (a) Henze, H. R.; Knowles, M. B. J. Org. Chem. 1954, 19,
1127. (b) Mathes, W.; Sauermilch, W. Chem. Ber. 1960, 93,
286. (c) Mathes, W.; Sauermilch, W. Chem. Ber. 1951, 84,
648. (d) Krasavin, I. A.; Radin, Y. P.; Ryabokokylko, Y. S.;
Parusnikov, B. V.; Dziomko, V. M. Chem. Heterocycl.
Compd. (Engl. Transl.) 1978, 14, 235.
Synthesis 2002, No. 4, 483–486 ISSN 0039-7881 © Thieme Stuttgart · New York