α -Hydroxymethylation of pyridines at a frustrated lewis pair template

Article abstract of DOI:10.1002/chem.201405553

The "ν²-formylborane" moiety formed by CO reduction with HB(C₆F₅ )₂ at a P/B frustrated Lewis pair template undergoes a hydroxymethylation reaction at the aposition to nitrogen in pyridine or isoquinoline. The a

Full text of DOI:10.1002/chem.201405553

DOI: 10.1002/chem.201405553
Communication
&
Synthetic Methods |Hot Paper|
a-Hydroxymethylation of Pyridines at a Frustrated Lewis Pair
Template
Muhammad Sajid, Gerald Kehr, Constantin G. Daniliuc, and Gerhard Erker*[a]
mylborane” system 5 (Scheme 2).[6] This reaction is unusual
Abstract: The “h2-formylborane” moiety formed by CO re-
since [B]ÀH boranes were shown not to be able to reduce
duction with HB(C6F5)2 at a P/B frustrated Lewis pair tem-
carbon monoxide by themselves for thermodynamic reasons
plate undergoes a hydroxymethylation reaction at the a-
but just form the respective adducts, the borane carbonyls.[7]
position to nitrogen in pyridine or isoquinoline. The analo-
[HB(C6F5)2] is no exception. We could show that it forms the
gous reaction with pyrimidine revealed a mechanism re-
borane carbonyl [(C6F5)2BH·CO] which we could isolate at low
lated to the Tschitschibabin reaction.
temperature and characterize by X-ray diffraction.[8] The forma-
tion of 5 was thought to proceed by the FLP/borane carbonyl
addition product 4 as an intermediate which apparently lifts
Reactive arenes readily undergo single or even multiple hy-
droxymethylation reactions with suitable electrophilic reagents.
The situation is different with pyridines which only reluctantly
enter into the electrophilic aromatic substitution sequences.
When they do, they are usually converted to the 3-substituted
derivatives. Consequently, a-hydroxymethylated pyridines are
mostly prepared by principally different protocols. The parent
compound a-hydroxymethylpyridine (2) is for example, con-
veniently prepared by treatment of 2-picolin-N-oxide (1) with
acetic acid derivatives (Boekelheide reaction, Scheme 1).[1]
a-Substitutions of pyridines on the other hand can readily be
achieved by routes involving nucleophilic attack at the 2-posi-
tion with subsequent removal of hydride. The Tschitschibabin
(Chitchibabin) reaction[2] or the Ziegler alkylation and aryl-
ation[3] are prominent examples. We have now found a mecha-
nistically related a-hydroxymethylation reaction at pyridine
and related substrates using a carbon monoxide derived nucle-
ophilic reagent generated and controlled at a vicinal phos-
phane borane frustrated Lewis pair (P/B FLP).[4] First examples
and a mechanistic evaluation will be presented in this account.
We had recently shown that the vicinal P/B FLP 3[5] reacted
readily with a 1:1 mixture of Piers borane [HB(C6F5)2] and
carbon monoxide to give the FLP template stabilized “h2-for-
the thermodynamic restriction imposed for simple formylbor-
ane formation.[6] We had shown that the FLP/(h2-formyl)borane
adduct 5 is opened with pyridines to give 6 [R: H (a), Me (b)].[8]
Scheme 2. FLP route to the pyridine derivative 7.
We have now treated compound 5 with an excess of 4-
methylpyridine (ca. 5 molar equivalents) under slightly more
forcing conditions (Scheme 2). The reaction mixture (in di-
chloromethane) was kept for 2 days at 708C in a sealed NMR
tube. Workup including column chromatography (silica gel,
CH2Cl2) and crystallization then gave the hydroxymethylated
pyridine derivative 7 in 64% yield as a colorless crystalline
solid. Compound 7 was characterized by X-ray diffraction. The
X-ray crystal structure analysis showed that the (h2-formyl)bor-
ane of the precursor 5 had been cleaved from the FLP frame-
work and combined with the added pyridine reagent
(Figure 1). The strongly Lewis acidic boron center is found at-
tached to the Lewis basic pyridine nitrogen as expected and
the former formyl carbon atom has now formed a carbonÀ
carbon bond with the adjacent pyridine C2 center, but it was
at the same time further reduced. Overall, a selective boryloxy-
methylation reaction at the pyridine C2 carbon had been ach-
ieved. In solution compound 7 shows the 19F NMR signals of
a pair of symmetry-equivalent C6F5 substituents at boron (11B:
Scheme 1. The Boekelheide reaction.
[a] Dr. M. Sajid, Dr. G. Kehr, Dr. C. G. Daniliuc,+ Prof. G. Erker
Organisch-Chemisches Institut, Westfꢀlische Wilhelms-Universitꢀt
Corrensstr. 40, 48149 Mꢁnster (Germany)
E-mail: erker@uni-muenster.de
[+] X-ray crystal structure analyses.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201405553.
1
d=6.4 ppm). It shows the H NMR feature of the newly formed
Chem. Eur. J. 2014, 20, 1 – 5
1
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
Figure 1. A view of the molecular structure of the hydroxymethylation prod-
uct 7 (thermal ellipsoids are shown with 30% probability). Selected bond
lengths [ꢁ] and bond angles [8]: N1-B1 1.619(4), B1-O1 1.453(4), O1-C7
1.404(3), C2-C3 1.381(4), C2-C7 1.492(4), C2-N1 1.342(3), N1-C6 1.349(4), C6-
C5 1.372(4), N1-B1-O1 100.0(2), B1-O1-C7 113.0(2), O1-C7-C2 106.7(2), C7-C2-
N1 109.8(2), C2-N1-B1 109.1(2).
Figure 2. Molecular structure of compound 8 (thermal ellipsoids are shown
with 30% probability). Selected bond lengths [ꢁ] and bond angles [8]: N2-B1
1.611(3), B1-O1 1.459(3), O1-C9 1.400(2), C3-N2 1.378(3), N2-C1 1.318(2), C1-
C9 1.488(3), C4-C4A 1.416(3), N2-B1-O1 99.5(2), B1-O1-C9 111.4(2), O1-C9-C1
106.4(2), C9-C1-N2 109.5(2), C3-N2-B1 129.2(2).
CH2 group at d=5.28 ppm (13C: d=69.6 ppm). Compound 7
was deborylated by treatment with aqueous NaOH to give the
respective free hydroxymethylpyridine derivative (for details
see the Supporting Information).
We achieved the CÀC coupling reaction between the h2-
formyl borane unit with the pyrimidine reagent in the follow-
ing way: treatment of 6b (in situ generated from 5 and 4-
methylpyridine) with a large excess of pyrimidine (ca. 13 molar
equivalents) followed by thermolysis of the mixture for 3 days
at 708C resulted in the formation of the N-borylated dihydro-
pyrimidine product 10. It was isolated in 16% yield after
column chromatography and washing with CH2Cl2.
We then treated 5 with isoquinoline (5 molar equivalents).
Under the analogous conditions we obtained the hydroxy-
methylation derivative 8 selectively, isolated as a colorless crys-
talline solid in 80% yield (Scheme 3). The X-ray crystal structure
The X-ray crystal structure analysis of 10 revealed the forma-
tion of the annulated five-membered heterocycle that was
formed by addition of the components of the (h2-formyl)bor-
ane unit to the N=C bond of pyrimidine (Figure 4). The 2-hy-
drogen is still attached and it is trans-oriented to the CÀH hy-
drogen of the former formyl group. The remaining
[P]ÀCH2CH2À[B] FLP framework is still present, probably caused
by the strong internal B2ÀN3 coordination.
Scheme 3. Formation of the isoquinoline derivative 8.
analysis confirmed the attachment of the CH2ÀO[B] group
(originally from the (h2-formyl)B(C6F5)2 unit in 5) at the C1 posi-
tion of the isoquinoline fragment (Figure 2). Again, the boron
atom is found attached to the nitrogen. Compound 8 shows
the expected NMR features [CH2ÀO[B]: d=5.86 ppm (1H), d=
70.0 ppm (13C), d=7.6 ppm (11B)].
The outcome of the reaction of 5 with pyrimidine gave us
an indication about the pathway followed in this hydroxy-
methylation reaction. The treatment of 5 with pyrimidine at
room temperature followed by crystallization at À358C gave
the ring opened addition product 9 which still contained the
intact (h2-formyl)borane subunit attached at phosphorus
(Scheme 4). Compound 9 was characterized by X-ray diffrac-
tion (Figure 3) and by spectroscopy (for details see the Sup-
porting Information).
Scheme 4. Reaction of 5 or 6 with pyrimidine.
&
&
Chem. Eur. J. 2014, 20, 1 – 5
www.chemeurj.org
2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communication
by coordination of the respective pyridine reagent to the
boron atom of the FLP framework of 5 to give 6 (Scheme 2)[8]
followed by opening of the (h2-formyl)borane subunit by
a second equivalent of the pyridine nucleophile to give a stabi-
lized carbanion (i.e., the phosphorus ylide intermediate 11).
This would activate this moiety for nucleophilic attack at the
pyridine a-position to give 12. Hydride migration with extru-
sion of the good FLP phosphane leaving group would provide
an attractive short route to the a-hydroxymethylated pyridine
product with concurrent formation of the FLP pyridine adduct
3·pyridine.
Figure 3. The molecular structure of the pyrimidine adduct 9 (thermal ellip-
soids are shown with 30% probability). Selected bond lengths [ꢁ] and bond
angles [8]: N1-B1 1.634(4), C9-B1 1.634(5), C8-C9 1.536(5), P1-C8 1.819(3), P1-
C7 1.783(4), C7-O1 1.435(4), C7-B2 1.596(5), B2-O1 1.471(5), B2-C7-O1 57.7(2),
C7-O1-B2 66.6(2), O1-B2-C7 55.6(2).
Scheme 5. Mechanistic scheme of the formation of compound 7.
Our study shows that frustrated Lewis pair chemistry has
provided a pathway for hydroxymethylation of pyridines by
means of a typical nucleophilic aromatic substitution pathway,
thereby offering a potential alternative method to the Boeckel-
heide reaction. Our reaction is similar in its essential course to
the Tschitschibabin amination of pyridines or the Ziegler pyri-
dine alkylation and arylation reactions. In those classical reac-
tions hydride is the leaving group, in our closely related case
hydride is not set free but utilized for reduction of the CO-
derived formyl group by 1,2-hydride shift. We think that this is
another example that demonstrates the potential frustrated
Lewis pair chemistry has in developing new chemical
reactions.
Experimental Section
A solution of pyridine/isoquinoline (ca. 10 equiv) in dichloro-
methane (ca. 0.9 mL) was added to powdered compound 5 (up to
0.3 mmol) to give a clear reaction solution. The resulting colorless
solution was sealed in an NMR tube and heated at 708C (oil-bath
temperature) for 2 days. The dark brown reaction mixture was di-
luted with CH2Cl2 (2 mL) and loaded onto a column (SiO2, CH2Cl2).
The obtained product (7 or 8) was further purified by crystalliza-
tion (toluene/n-pentane; 1:4) at À358C.
For the detailed reaction procedures and analytical data please see
the Supporting Information. CCDC-1028144 (7), -1028145 (8),
-1028146 (9) and -1028147 (10) contain the supplementary crystal-
lographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Figure 4. A view of the molecular structure of compound 10 (thermal ellip-
soids are shown with 30% probability). Selected bond lengths [ꢁ] and bond
angles [8]: N3-B2 1.601(3), N1-B1 1.567(3), N3-C2 1.467(3), N1-C2 1.458(2), B1-
O1 1.473(2), P1-C7 1.858(2), N3-C4 1.324(3), N1-C6 1.311(3), B2-N3-C2
124.5(2), N1-B1-O1 100.7(2), C2-C7-O1 107.1(2), C6-N1-B1 133.7(2), C4-N3-C2
111.5(2), N1-C2-N3 112.4(2), C2-N1-B1 110.6(2), C7-O1-B1 113.4(2).
This leads us to propose a general reaction pathway as sche-
matically depicted in Scheme 5. The reaction could be initiated
Chem. Eur. J. 2014, 20, 1 – 5
www.chemeurj.org
3
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
[3] a) G. Fraenkel, J. C. Cooper, Tetrahedron Lett. 1968, 9, 1825–1830. See for
comparison: b) J. A. Bull, J. J. Mousseau, G. Pelletier, A. B. Charette, Chem.
Rev. 2012, 112, 2642–2713.
[4] a) D. W. Stephan, G. Erker, Angew. Chem. Int. Ed. 2010, 49, 46–76; Angew.
Chem. 2010, 122, 50–81; b) G. Erker, D. W. Stephan, Frustrated Lewis
Pairs I, Top. Curr. Chem., Springer, Heidelberg, 2013, p.332; c) G. Erker,
D. W. Stephan, Frustrated Lewis Pairs II, Top. Curr. Chem., Springer, Heidel-
berg, 2013, p.334.
Acknowledgements
Financial support from the European Research Council is grate-
fully acknowledged.
Keywords: boron
·
frustrated
Lewis
pairs
·
[5] P. Spies, G. Erker, G. Kehr, K. Bergander, R. Frçhlich, S. Grimme, D. W. Ste-
phan, Chem. Commun. 2007, 5072–5074.
hydroxymethylation · phosphorus · pyridine
[6] M. Sajid, L. M. Elmer, C. Rosorius, C. G. Daniliuc, S. Grimme, G. Kehr, G.
Erker, Angew. Chem. Int. Ed. 2013, 52, 2243–2246; Angew. Chem. 2013,
125, 2299–2302.
[7] A. B. Burg, H. I. Schlesinger, J. Am. Chem. Soc. 1937, 59, 780–787.
[8] M. Sajid, G. Kehr, C. G. Daniliuc, G. Erker, Angew. Chem. Int. Ed. 2014, 53,
1118–1121; Angew. Chem. 2014, 126, 1136–1139.
[1] a) V. Boekelheide, W. J. Linn, J. Am. Chem. Soc. 1954, 76, 1286–1291; See
also b) B. Klein, J. Berkowitz, N. E. Hetman, J. Org. Chem. 1961, 26, 126–
131; c) D. Andreotti, E. Miserazzi, A. Nalin, A. Pozzan, R. Profeta, S. Spada,
Tetrahedron Lett. 2010, 51, 6526–6530; d) A. Massaro, A. Mordini, A. Min-
gardi, J. Klein, D. Andreotti, Eur. J. Org. Chem. 2011, 271–279; e) S. K.
Das, J. Frey, Tetrahedron Lett. 2012, 53, 3869–3872.
[2] a) A. E. Tschitschibabin, O. A. Zeide, J. Russ. Chem. Soc. 1914, 46, 1216;
b) D. R. Bragg, D. G. Wibberley, J. Chem. Soc. 1963, 3277–3281; c) R. A.
Abramovitch, F. Helmer, J. G. Saha, Can. J. Chem. 1965, 43, 725–731; d) T.
Kametani, H. Nemoto, S. Takano, Chem. Pharm. Bull. 1968, 16, 367–370;
e) E. I. Kostik, A. Abiko, A. Oku, J. Org. Chem. 2001, 66, 2618–2613;
f) G. S. Singh, E. E. Mmatli, Eur. J. Med. Chem. 2011, 46, 5237–5257.
Received: October 7, 2014
Published online on && &&, 0000
&
&
Chem. Eur. J. 2014, 20, 1 – 5
www.chemeurj.org
4
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communication
COMMUNICATION
&
Synthetic Methods
M. Sajid, G. Kehr, C. G. Daniliuc, G. Erker*
&& – &&
a-Hydroxymethylation of Pyridines at
a Frustrated Lewis Pair Template
The “h2-formylborane” moiety, formed
by CO reduction with HB(C6F5)2 and a P/
B frustrated Lewis pair template, hy-
droxymethylates pyridine or isoquino-
line by a reaction pathway related to
the Tschitschibabin reaction (see
scheme).
Chem. Eur. J. 2014, 20, 1 – 5
www.chemeurj.org
5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ