A stereochemical approach to the Kabachnik-Fields reaction mechanism

Article abstract of DOI:10.1070/MC2003v013n03ABEH001763

A comparison of the diastereomeric composition of Kabachnik-Fields reaction products with those of two reactions simulating its stereo-controlling steps showed that only the 'imine' mechanism works in the (MeO) ₂P(O)H-PhCHO-[(S),(R,S)-H₂NCH(Ph)Me] system, like in the phosphite-imine system; in a similar system containing (R,S)-sec-butylamine, additional 'nucleophilic amination' of the initially formed α-hydroxybenzyl phosphonate occurs.

Full text of DOI:10.1070/MC2003v013n03ABEH001763

, 2003, 13(3), 150–151
A stereochemical approach to the Kabachnik–Fields reaction mechanism
Mudaris N. Dimukhametov,* Evgenia V. Bayandina, Elena Yu. Davydova, Aidar T. Gubaidullin, Igor A. Litvinov
and Vladimir A. Alfonsov
A. E. Arbuzov Institute of Organic and Physical Chemistry, Kazan Scientific Centre of the Russian Academy of Sciences,
420088 Kazan, Russian Federation. Fax: +7 8432 75 5322; e-mail: mudaris@iopc.knc.ru
10.1070/MC2003v013n03ABEH001763
A comparison of the diastereomeric composition of Kabachnik–Fields reaction products with those of two reactions simulating its
stereo-controlling steps showed that only the ‘imine’ mechanism works in the (MeO)2P(O)H–PhCHO-[(S),(R,S)-H2NCH(Ph)Me]
system, like in the phosphite–imine system; in a similar system containing (R,S)-sec-butylamine, additional ‘nucleophilic amina-
tion’ of the initially formed α-hydroxybenzyl phosphonate occurs.
1
The Kabachnik–Fields reaction (KFR) is important for the
synthesis of α-aminoalkyl phosphonates, which possess useful
properties.1 Although the synthetic potential of the KFR is
employed successfully, its mechanism is poorly known. We
performed the first comparative study of the stereochemical
result (de) of the KFR in the (MeO)2P(O)H–PhCHO–chiral
amines 1a,b system and of two-component reactions simulating
separate stereo-controlling steps in a three-component system,
viz., the ‘nucleophilic amination’ of α-hydroxybenzyl dimethyl
phosphonate 2 with amines 1a,b and the addition of (MeO)2P(O)H
to chiral imines 3a,b (the Pudovik reaction) (Scheme 1). The
expected products of all the three directions are α-aminobenzyl
phosphonates 4 and 5, each being a mixture of two diastereo-
mers A and B. To ensure the correct comparison of the results,
all the reactions were carried out under identical conditions.
The H NMR spectra allowed us to determine the diastereo-
mer ratio 5A/5B based on a comparison of the integral inten-
sities of two doublets of the HCP proton. This ratio is 1.19:1 for
the completed KFR with amine 1b.
Pudovik reactions in the (MeO)2P(O)H–chiral imines 3a,b
system carried out under the same conditions as the KFR give
compounds 4A and 4B in the ratios 1:4.12 (3a from enantio-
pure 1a) and 1:3.93 (3a from racemic 1a) (31P NMR) and also
5A and 5B in the ratio 1.52:1 (1H NMR). Furthermore, accord-
ing to 31P NMR data, prolonged refluxing (longer than 4 h) of a
mixture of compound 2 with amine 1a in benzene under KFR
conditions does not yield compound 4 in amounts that can be
detected spectroscopically. On the contrary, the model reaction
of compound 2 with amine 1b under KFR conditions took
place: the signal of compound 2 in the 31P NMR spectrum
almost disappeared after 8 h. The region around d 25 displays
O
Me
*
2: mp 101 °C (lit.,2 102 °C). 1H NMR [Bruker WM-250, 250 MHz,
(CD3)2CO, TMS] d: 3.66 (d, 3H, MeO, JHP 10.6 Hz), 3.71 (d, 3H,
Me'O, JHP 10.6 Hz), 5.13 (d, 1H, CHP, JHP 12.9 Hz), 7.32–7.58 (m,
5H, Ph).
4A (from a 4A:4B mixture, 3:1): 1H NMR, d: 1.36 (d, 3H, MeC, 3JHH
†
(MeO)2PH + PhCH=O + H2NCHR
3
3
2
1a,b
O
Me
*
O
Me
*
*
*
3
6.4 Hz), 2.54 (s, 1H, NH), 3.47 (d, 3H, MeOP, JHP 10.4 Hz), 3.77 (q,
(MeO)2PCHOH + H2NCHR
(MeO)2PCHNHCHR
Ph
1H, HCPh, 3JHH 6.4 Hz), 3.81 (d, 1H, HCP, JHP 20.0 Hz), 3.83 (d, 3H,
2
1a,b
Ph
Me'OP, 3JHP 10.4 Hz), 7.25–7.42 (m, 5H, Ph).
4B: mp 96.5–97.5 °C, {[a]D20 –15.5° (c 3.4, C6H6)}. 1H NMR (Bruker
WM-250, 250 MHz, CD3CN, TMS) d: 1.35 (d, 3H, MeC, 3JHH 6.4 Hz),
2.52 (s, 1H, NH), 3.55 (d, 3H, MeOP, 3JHP 10.4 Hz), 3.79 (d, 3H, Me'OP,
3JHP 10.4 Hz), 3.86 (q, 1H, HCPh, 3JHH 6.4 Hz), 4.21 (d, 1H, HCP, 2JHP
20.2 Hz), 7.27–7.39 (m, 5H, Ph).
2
4 R = Ph
5 R = Et
O
Me
*
(MeO)2PH + PhCH=NCHR
3a,b
Mixture 5A and 5B (1.5:1): nD20 1.7138. H NMR (Bruker WM-250,
1
3
a R = Ph (S), (R,S)
250 MHz, CCl4, TMS) d: 0.85 (t, 3H, MeCCN, JHH 7.2 Hz), 0.88 (t,
3H, Me'CCN, 3JHH 7.2 Hz), 0.98 (d, 3H, MeCN, 3JHH 5.8 Hz), 1.36 (m,
R = Et ( , )
R S
b
Scheme 1
2H, CH2CN), 2.07 (m, 1H, NH), 2.38 (m, 1H, CCHN), 2.51 (m, 1H,
3
3
CCH'N), 3.41 (d, 3H, MeO, JHP 10.7 Hz), 3.75 (d, 3H, Me'OP, JHP
10.7 Hz), 4.06 (d, 1H, HCP, 2JHP 22.5 Hz), 4.12 (d, 1H, H'CP, 2JHP 22.3 Hz),
7.25–7.41 (m, 5H, Ph).
According to 31P {1H} NMR spectra of the reaction mix-
tures, the KFR in both systems (reagent ratio of 1:1:1; boiling
benzene; 2 h; water removal by azeotropic distillation) results
in three phosphorus-containing products [amine 1a: dP 23.5
(2), 25.4 (4A), 25.8 (4B) in the ratio 2.70:1:3.75 (S-amine),
2.34:1:3.21 (R,S-amine), respectively; amine 1b: dP 23.50 (2),
25.15 (5A), 25.19 (5B) in the ratio 1 (2) : 2.17 (5A + 5B)]. In
the case of amine 1a, the subsequent refluxing of the reaction
mixture does not change the number of phosphorus-containing
products or the intensity ratio of their signals in the 31P NMR
spectra. However, in the case of amine 1b, the intensity of the
signal with dP 23.5 in the 31P {1H} NMR spectrum under
similar conditions decreased gradually and almost disappeared
after 4 h.
‡
X-Ray diffraction study of compound 4B. Crystal of 4B, C17H22NO3P,
monoclinic, space group P21. At 20 °C a = 10.362(2) Å, b = 5.9637(4) Å,
c = 13.860(5) Å, b = 95.08(2)°, V = 853.1(4) Å3, Z = 2, M = 319.34,
dcalc = 1.24 g cm–3, m(Cu) = 15.16 cm–1, F(000) = 340. The intensities of
1993 reflections were measured on an Enraf-Nonius CAD-4 diffracto-
meter at 20 °C [l(CuKα irradiation, w/2q-scanning, 2qmax = 148°]; of
these, 1650 reflections with I ³ 3s were observed. The structure was
solved by the direct method using the SIR program3 from the MolEN
software package.4 The structure was refined by a full-matrix least-
squares method in an anisotropic approximation; all hydrogen atoms
were located by different synthesis and refined isotropically in final
least-squares iterations. The absolute crystal structure and absolute mole-
cule configuration were determined by the Hamilton test5 with 95%
probability. The final divergence factors are R = 0.049, Rw = 0.075 based
on 1608 reflections with F2 ³ 3s.
Atomic coordinates, bond lengths, bond angles and thermal param-
eters have been deposited at the Cambridge Crystallographic Data Centre
(CCDC). These data can be obtained free of charge via www.ccdc.cam.uk/
conts/retrieving.html (or from the CCDC, 12 Union Road, Cambridge
CB2 1EZ, UK; fax: +44 1223 336 033; or deposit@ccdc.cam.ac.uk).
Any request to the CCDC for data should quote the full literature citation
and CCDC reference number 213979. For details, see ‘Notice to Authors’,
Mendeleev Commun., Issue 1, 2003.
Column chromatography in the C6H6–MeOH (5:1) (amine
1a) and C6H6–diethyl ether (1:1) (amine 1b) systems gave pure
products of both KFRs. The structures of compounds 2, 4A,
1
4B, 5A and 5B were established by H NMR.† The absolute
configuration (R) of the carbon atom at the α-position with
respect to the phosphorus atom in 4B (from optically pure 1a)
was determined by X-ray diffraction analysis‡ in coordinates of
the second chiral centre at the C(5) atom, which is known to
have (S) configuration.
– 150 –

, 2003, 13(3), 150–151
O
O
C(12)
C(13)
Me
C(17)
C(20)
C(11)
(MeO)2PH
C(6)
(MeO)2PH
C(18)
C(19)
*
PhCH=NCHEt
C(16)
C(5)
C(21)
– H2O
C(10)
C(9)
3b
C(15)
C(3)
PhCH=O
Me
5
N(4)
O
1b
*
P(2)
O(4)
C(8)
(MeO)2PCHOH
O(1)
*
– H2O
H2NCHEt
Ph
1b
O(3)
2
C(7)
Scheme 2
Figure 1 Molecular geometry of 4B in a crystal. Selected bond lengths
(Å): P(2)–O(1) 1.470(4), P(2)–O(3) 1.574(3), P(2)–O(4) 1.558(3), P(2)–
C(3) 1.805(5), O(3)–C(7) 1.444(9), O(4)–C(8) 1.438(9), N(4)–C(3)
1.486(6), N(4)–C(5) 1.466(7), C(3)–C(9) 1.519(6), C(5)–C(6) 1.513(9),
C(5)–C(16) 1.518(7); selected bond angles (°): O(1)–P(2)–O(3) 114.3(2),
O(1)–P(2)–O(4) 113.2(2), O(1)–P(2)–C(3) 112.5(2), O(3)–P(2)–O(4)
103.3(2), O(3)–P(2)–C(3) 106.4(2), O(4)–P(2)–C(3) 106.3(2), P(2)–O(3)–
C(7) 119.9(4), P(2)–O(4)–C(8) 123.3(4), C(3)–N(4)–C(5) 115.3(4), P(2)–
C(3)–N(4) 105.2(3), P(2)–C(3)–C(9) 111.3(3), N(4)–C(3)–C(9) 116.9(4),
N(4)–C(5)–C(6) 111.2(4), N(4)–C(5)–C(16) 108.8(4), C(6)–C(5)–C(16)
110.7(4).
by the initial formation of compound 2 from PhCHO and
(MeO)2P(O)H. Its subsequent amination with amine 1b results
in the KFR product 5.
Thus, depending on the nature of reagents, the KFR can
involve either one of the known mechanisms exclusively or two
alternative mechanisms simultaneously.
This study was supported by the Russian Foundation for
Basic Research (grant no. 03-03-33082), the Research and
Development Foundation of the Tatarstan Academy of Sciences
(grant no. 07-7.2-129/2002) and a Scientific Programme of the
Russian Academy of Sciences.
signals corresponding to compounds 5A and 5B in the ratio
1:1.59. Note that, first, this reaction results in some prevalence
of diastereomer 5B, and second, the stereoselectivity of the
Pudovik reaction is noticeably higher than that of the KFR.
A comparison of the stereochemical results of the three
reactions in each of the two reaction series allowed us to make
the conclusions given below. The KFR with amine 1a in boiling
benzene occurs via the addition of (MeO)2P(O)H to imine 3a,
which is initially formed from PhCHO and amine 1a (‘imine’
mechanism). The basicity of the amine is sufficient to catalyse
the addition of (MeO)2P(O)H to PhCHO to give compound 2;
however, the thermodynamic stability of 2 under the reaction
conditions and its low reactivity in the ‘nucleophilic amination’
step make compound 2 a by-product. The KFR with amine 1b
mainly occurs through the initial formation of imine 3b from
PhCHO and amine 1b, which then enters the Pudovik reaction
with (MeO)2P(O)H (‘imine’ mechanism) (Scheme 2). On the
other hand, the overall result of the KFR is noticeably affected
References
1 R. A. Cherkasov and V. I. Galkin, Usp. Khim., 1998, 67, 940 (Russ.
Chem. Rev., 1998, 67, 857).
2 F. Texier-Boullet and A. Foucaud, Synthesis, 1982, 165.
3 A. Altomare, G. Cascarano, C. Giacovazzo and D. Viterbo, Acta
Crystallogr., Sect. A, 1991, 46, 744.
4 L. H. Straver and A. I. Schierbeek,
MolEN. Structure Determination
System, Nonius B.V., Delft, Netherlands, 1994, vols. 1 and 2.
5 W. C. Hamilton, Acta Crystallogr., 1965, 18, 502.
Received: 16th April 2003; Com. 03/2089
– 151 –