On the regiochemistry of Mitsunobu alkylations of hydrazine derivatives

Article abstract of DOI:10.1016/j.tetlet.2012.10.084

That differentially protected hydrazines of the type TsNHNHCOR³ undergo regiospecific Mitsunobu reactions with a variety of alcohols, R ¹R²CHOH in a generalised manner, to give good to excellent yields of the mono-adducts Ts(R¹R²CH)NNHCOR³, has been proven by X-ray crystallographic analysis.

Full text of DOI:10.1016/j.tetlet.2012.10.084

Tetrahedron Letters 53 (2012) 7006–7009
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
On the regiochemistry of Mitsunobu alkylations of hydrazine derivatives
Damian G. Dunford a, Faryal Chaudhry a, Benson Kariuki a, David W. Knight a, , Robert C. Wheeler b
⇑
a School of Chemistry, Cardiff University, Main College, Park Place, Cardiff CF10 3AT, UK
b GSK, Medicines Research Centre, Stevenage ST1 2N, UK
a r t i c l e i n f o
a b s t r a c t
That differentially protected hydrazines of the type TsNHNHCOR3 undergo regiospecific Mitsunobu reac-
tions with a variety of alcohols, R1R2CHOH in a generalised manner, to give good to excellent yields of the
mono-adducts Ts(R1R2CH)NNHCOR3, has been proven by X-ray crystallographic analysis.
Ó 2012 Published by Elsevier Ltd.
Article history:
Received 6 September 2012
Revised 27 September 2012
Accepted 18 October 2012
Available online 26 October 2012
Keywords:
Mitsunobu
1,2-Hydrazine-1,2-dicarboxylates
Alkylation
Regiospecific
N-Sulfonylhydrazines
The Mitsunobu reaction (Scheme 1) is of crucial importance for
the introduction of a variety of heteroatomic species, including
oxygen-, nitrogen- and sulfur-based species and even some nucle-
ophilic carbon reagents by displacement of, overall, the hydroxyl
group in a precursor alcohol 1.1 Proceeding by an exquisitely con-
trolled SN2 inversion mechanism in the last step, perhaps its only
significant drawback is the sometimes Herculean efforts required
to separate the desired product 2 from the spent reagents, a phos-
phine oxide and a hydrazine dicarboxylate or relative, along with
the necessary slight excesses of the initial reagents. Recently, even
this feature has been addressed.2
Mitsunobu reactions.1 Subsequent addition of the precursor alco-
hol to this intermediate then leads to the penultimate species 4,
which can then reorganise, either intramolecularly or possibly
via the highly activated intermediate 7, to give the observed prod-
ucts—the alkylated hydrazine dicarboxylate 5 and the phosphine
oxide 6. It is interesting to note that this reaction does not interfere
OH
R2
XR3
R2
RO2CN NCO2R
PPh3, R3X-H
R1
R1
Just prior to its discovery, during studies carried out in Mukaiy-
ama’s laboratory which doubtless led to this major advance,3 what
can be regarded as a ‘halfway house’ or partial version of the Mits-
unobu reaction was logically developed wherein the final nucleo-
philic species, R3XH, was omitted. As now the only reactive
nucleophile is the ‘spent’ hydrazine, the overall reaction results in
mono-N-alkylation of the latter; ironically, at the time of its discov-
ery, there was more emphasis placed on the fact that this was a
method for the oxidation of trivalent phosphorus compounds
(Scheme 2).4 For a separate study, we required a diverse series of
such monoalkylated hydrazines and clearly this methodology could
provide a small number of these, starting from the limited range of
commercially available and symmetrical azodicarboxylates.
Mechanistically, this reaction presumably begins by nucleo-
philic addition of the phosphine to the azodicarboxylate to
generate the same initial adduct 3 which is involved in the normal
1
2
Scheme 1. A typical Mitsunobu reaction.
RO2CN NCO2R
PPh3
RO2CN NCO2R
RO2CN NCO2R
Ph3P
Ph3P H
H
O
O
R1
R1
3
4
PPh3
O
+
Ph3P
O
RO2CN NCO2R
R1
R1
H
7
6
5
⇑
Corresponding author. Tel.: +44 2920 874210; fax: +44 2920 874030.
E-mail address: knightdw@cf.ac.uk (D.W. Knight).
Scheme 2. Synthesis of a mono-N-alkyl-1,2-hydrazine dicarboxylate.
0040-4039/$ - see front matter Ó 2012 Published by Elsevier Ltd.
http://dx.doi.org/10.1016/j.tetlet.2012.10.084

D. G. Dunford et al. / Tetrahedron Letters 53 (2012) 7006–7009
7007
TsCl
DEAD, Ph3P
ROH
Boc2O
TsNH NHCO2But
13
TsN NHCO2But
ZNHNH2
ZNHNHTs
ZNHN(Boc)Ts
R
8
9
10
14
and not
TsNH NCO2But
PhCH2Br
R
15
Mg
ZN(Bn)NHBoc
ZN(Bn)N(Boc)Ts
MeOH
12
11
Scheme 4. The Ragnarsson regiospecific alkylation of an unsymmetrical hydrazine.
Scheme 3. A non-Mitsunobu approach to a mono-alkylated hydrazine dicarbox
ylate.
sulfonamides, ArSO2NHNH2, was 17.1,10 very much in agreement
with the measurements reported subsequently by the Ragnarsson
group.6,8,9 The sole formation of the monoalkylated products 14
was therefore ascribed, not unreasonably, to this large difference,
which predicts a product ratio of ca. 10,000:1, a difference which
we presumed could be extrapolated to doubly protected hydrazine
derivatives such as the sulfonamides 13.
Despite these compelling arguments, the structural assign-
ments were not supported by any other data beyond the apparent
homogeneity of the single products obtained in the two examples
reported. For at least two reasons, this seemed to us to be insuffi-
cient and indeed somewhat risky. Firstly, this type of hydrazine
derivative typically produces very poorly resolved NMR spectra,
even at elevated temperatures, due to extensive rotameric broad-
ening. The 1H NMR data reported for the products 14 (or 15) con-
sist of a series of broad resonances,9 which also proved to be a
serious problem in our own work and provided very little convinc-
ing evidence to distinguish the two possible products. The re-
ported9 presence of two separate conformers, as we also often
observed, merely increased this uncertainty. Secondly, there is
to any appreciable extent in the Mitsunobu reaction itself.1 It is
also notable that double alkylation does not occur and that hydra-
zine-1,2-dicarboxylates are very poor substrates for Mitsunobu
reactions in general, both of which observations support this intra-
molecular alkylation mechanism, rather than any involvement by
the activated species 7.
Later studies have established that a key rate determining step
in the Mitsunobu reaction is breaking of the X–H bond in the
incoming nucleophilic species. Hence, the pKa value of such a
nucleophile is a crucial determinant of its suitability, rather than
its inherent nucleophilicity. This is well illustrated by the work
of Martin and Dodge who showed that 4-nitrobenzoic acid is a
far superior component of a Mitsunobu reaction than the parent
benzoic acid.5 This feature was later quantified by the Ragnarsson
group, in seminal work which suggested an optimum value for
such a pKa of <13.5, for values determined in dimethyl sulfoxide.6
Subsequently, the same group returned to the original Mukaiy-
ama methodology (Scheme 2) for hydrazine alkylation and pro-
vided more examples which attested to the importance of this
relatively low pKa value in the incoming nucleophile. Prior to this,
a somewhat lengthier scheme was required to obtain a selectively
mono-alkylated, unsymmetrical hydrazine-1,2-dicarboxylate 12
(Scheme 3).7
Thus, starting with a hydrazine mono-carboxylate 8, sulfona-
tion8 of the free amino group gives the doubly protected species
9 which, in line with the following selective Mitsunobu method,
then undergoes selective alkoxycarbonylation at the more acidic
sulfonamide site to give the triply protected hydrazine derivative
10. This then can only be alkylated at the remaining free position;
subsequent detosylation of the resulting fully substituted hydra-
zine 11 finally yields the unsymmetrically protected mono-alkyl-
ated hydrazine-1,2-dicarboxylate 12.
When recently we required a series of such mono-alkylated-
1,2-hydrazine dicarboxylates, for example 12 and the related sulfo-
nyl derivatives 14, we were obviously attracted to the idea of using
the direct mono-alkylation approach, but using an unsymmetri-
cally protected hydrazine, rather than the method shown in
Scheme 2, which is limited to the formation of symmetrical 1,2-
dicarboxylates. We found that the Ragnarsson group had indeed
reported two such examples in a later publication,9 wherein it
was stated that the unsymmetrical hydrazine derivative 13 under-
went regiospecific alkylation at the sulfonamide function rather
than at the alternative carbamate site to give excellent yields of
the derivatives 14 [R = p-MeOC6H4CH2 and Me2CHCH2] (Scheme 4).
The explanation, and indeed structural assignments as products
14 and not 15, for these potentially very useful results were based
entirely on considerations of pKa values. Thus, as argued above,
X–H bonds with lower values are more reactive in Mitsunobu
and related reactions as breaking of this bond is a key rate
determining step. Prior work by Bordwell had established a
relative pKa value of 22.2 for the N–H bond in carbamate deriva-
tives, EtO2CNHNH2, while the corresponding value for similar
considerable mechanistic ambiguity associated with many details
1b,c
of the Mitsunobu reaction in general (See Ref.
for an extensive
discussion of this aspect). For example, it is not inconceivable that,
while initial reaction occurs preferentially at the sulfonamide N–H
group, during the later stages this could give rise to an unantici-
pated intermediate, which could react by subsequent transfer of
the alkylating group to the other N–H group, by reason of its great-
er nucleophilicity (cf. Scheme 2). These two factors understandably
also made us nervous about using alternative, multistep synthetic
routes to provide structural proof.
The particular series of hydrazine derivatives that was required
for a separate study,11 was based on general formula 17a, the opti-
mum precursors being the corresponding alcohols 16 as these were
readily available from condensations between acetylides and alde-
hydes (Scheme 5). In fact, the relative sensitivity of these secondary
propargylic alcohols 16 made us wonder if the reaction would work
at all, prior to any considerations regarding regioselectivity.
In the event, we found that the reaction proceeded slowly but
very cleanly12 and delivered an excellent yield of 91% of the first
example [17a; R1 = iBu, R2 = Bu, R3 = tBu, Ar = p-MeC6H4] of this
reaction, quite clearly as a single product.13 A considerable bonus
was that the product was a crystalline solid, suitable for X-ray
crystallographic analysis. The refined ORTEP diagram subsequently
obtained is shown in Figure 1.14
?
ArSO2
R1
NHCO2R3
R2
OH
N
ArSO2NHNHCO2R3
R1
Ph3P, iPrO2CN=NCO2 Pr
i
R2
16
17a
Scheme 5. The desired Mitsunobu alkylation.

7008
D. G. Dunford et al. / Tetrahedron Letters 53 (2012) 7006–7009
Figure 1. ORTEP diagram of the Mitsunobu adduct 17a [R1 = iBu, R2 = Bu, R3 = tBu,
Ar = p-MeC6H4]. CCDC 783430.
Figure 2. ORTEP diagram of the Mitsunobu adduct in Table 1, entry 6. CCDC
783434.
Table 1
Mitsunobu reactions using N-tosyl-N0-carbonylhydrazines
worthwhile to check that this, albeit small, structural alteration
had not affected the regiochemical outcome of the alkylation
which, once again, appeared to give a single compound. The crys-
talline product from Table 1, entry 6 gave crystals amenable to
X-ray analysis and the refined ORTEP diagram is shown in Figure
2,16 which confirms that there was indeed no alteration in
regiochemistry.
Ts
R1
NHCOR3
R2
OH
R2
N
p-TolSO2NHNHCOR3
R1
i
Ph3P, iPrO2CN=NCO2 Pr
18
19
Entry
R1
R2
R3
Yield (%)
The presence of an alkyne function was not essential for suc-
cessful alkylations, as expected. Thus, cinnamyl alcohol, prenyl
alcohol and geraniol all underwent smooth alkylation of the
unsymmetrical hydrazine TsNHNHCO2Me (entries 8–10), as did a
purely saturated alcohol (entry 11).
We have therefore shown that the original Ragnarsson conclu-
sion, based entirely on considerations of pKa values and two
examples, was indeed correct despite an absence of a rigorous
spectroscopic and analytical proof of structure. This should there-
fore encourage greater use of this efficient methodology for the syn-
thesis of differentially substituted and differentially protected
hydrazines, in this controlled, general and predictable manner.
iBu
iBu
OtBu
OtBu
91 (= 17a)
78 (= 17b)
Bu
Ph
1
2
3
4
Me
Me
OtBu
OtBu
75
66
S
N
N
N
5
Me
OtBu
64
iBu
iBu
Ph
Me
77
67
89
Bu
Ph
6
7
8
Me
H
OMe
Acknowledgments
9
H
OMe
90
We thank Peter Rutkowski for skilled experimental assistance
and advice and GSK and the EPSRC for financial support (DGD)
and the Higher Education Commission, Pakistan for the award of
a six month fellowship (to FC).
10
11
H
H
OMe
OMe
91
75
Bu
References and notes
In just the same way, the related Mitsunobu product 17b ob-
tained from the corresponding phenyl substituted alcohol, derived
from phenylacetylene and isovaleraldehyde, also gave excellent
crystals for X-ray analysis. Once again, the only product was that
arising from reaction at the NHTs function (ORTEP diagram not
shown).15 This and other results obtained for the reaction shown
in Scheme 5, along with similar products obtained from other alco-
hols, are collected in Table 1.
The first two entries are as described above, the products of
which have been subjected to X-ray analysis. Other aryl-substi-
tuted alkyne derivatives were obtained in slightly lower yields (en-
tries 3–5), most likely due, in the latter two cases, to the additional
polarity induced by the pyridine and pyrimidine groups, which
rendered chromatographic separation more difficult.
1. For reviews, see: (a) Mitsunobu, O. Synthesis 1981, 1–28; (b) Hughes, D. L. Org.
React. 1992, 42, 335–656; (c) Swamy, K. C. K.; Kumar, N. N. B.; Balaraman, E.;
Pavan Kumar, K. V. P. Chem. Rev. 2009, 109, 2551–2651.
2. (a) Proctor, A. J.; Beautement, K.; Clough, J. M.; Knight, D. W.; Li, Y. Tetrahedron
Lett. 2006, 47, 5151–5154; For a review, see: (b) Dembinski, R. Eur. J. Org. Chem.
2004, 2763–2772; See also: (c) Barrett, A. G. M.; Roberts, R. S.; Schröder, J. Org.
Lett. 2000, 2, 2999–3001; (d) Valeur, E.; Roche, D. Tetrahedron Lett. 2008, 49,
4182–4185.
3. For a fascinating description of the contributions to synthetic chemistry made
by the Mukaiyama group, and especially the thinking behind the discovery of
the Mitsunobu reaction, see: Mukaiyama, T. Challenges in synthetic organic
chemistry. In International Series of Monographs on Chemistry; Baldwin, J. E., Ed.;
Oxford University Press: Oxford, 1990.
4. For later applications, see for example: (a) Mitsunobu, O.; Yamada, M.;
Mukaiyama, T. Bull. Chem. Soc. Jpn. 1967, 40, 935; (b) Stanek, J.; Frei, J.; Mett, H.;
Schneider, P.; Regenass, U. J. Med. Chem. 1992, 35, 1339–1344; (c) Okitsu, T.;
Sato, K.; Wada, A. Org. Lett. 2010, 12, 3506–3509.
Replacing the N-Boc group with N-acetyl (entries 6 and 7) had
little effect, and only slightly lower yields were obtained under
the same, non-optimised alkylation conditions. Again, we felt it
5. (a) Martin, S. F.; Dodge, J. A. Tetrahedron Lett. 1991, 32, 3017–3020; (b) Dodge, J.
A.; Trujillo, J. I.; Presnell, M. J. Org. Chem. 1994, 59, 234–236; (c) Dodge, J. A.;
Nissen, J. S.; Presnell, M. Org. Synth. 1996, 73, 110–115.

D. G. Dunford et al. / Tetrahedron Letters 53 (2012) 7006–7009
7009
6. (a) Koppel, I.; Koppel, J.; Degerbeck, F.; Grehn, L.; Ragnarsson, U. J. Org. Chem.
1991, 56, 7172–7174; For a more recent and very extensive assessment of such
pKa values, see: (b) Koppel, I.; Koppel, J.; Leito, L.; Pihl, V.; Wallin, A.; Grehn, L.;
Ragnarsson, U. J. Chem. Soc. Perkin Trans. 2 1993, 655–658; (c) Milletti, F.;
Storchi, L.; Goracci, L.; Bendels, S.; Wagner, B.; Kansy, M.; Cruciani, G. Eur. J.
Med. Chem. 2010, 45, 4270–4279.
7. Grehn, L.; Nyasse, B.; Ragnarsson, U. Synthesis 1997, 1429–1432.
8. Nyasse, B.; Grehn, L.; Maia, H. L. S.; Monteiro, L. S.; Ragnarsson, U. J. Org. Chem.
1999, 64, 7135–7139.
9. (a) Ragnarsson, U.; Grehn, L.; Koppel, J.; Loog, O.; Tšubrik, O.; Bredikhin, A.;
Mäeorg, U.; Koppel, I. J. Org. Chem. 2005, 70, 5916–5921; (b) Ragnarsson, U.;
Grehn, L.; Maia, H. L. S.; Monteiro, L. S. J. Chem. Res. (S) 2000; (c) Ragnarsson, U.
Chem. Soc. Rev. 2001, 30, 205–213.
10. (a) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456; (b) Zhao, Y.; Bordwell, F. G.;
Cheng, J.-P.; Wang, D. J. Am. Chem. Soc. 1997, 119, 9125–9129.
(Na2SO4), filtered and evaporated. The product was separated using silica gel
column chromatography (10% EtOAc-hexanes) to give the alkylated hydrazine
17 [R1 = iBu, R2 = Bu, R3 = tOBu, Ar = p-MeC6H4] as a yellow solid (11.82 g, 91%),
mp 59–60 °C (EtOAc-petrol), m
max/cmꢀ1 (CHCl3) 3314, 2958, 2871, 1759, 1710,
1598, 1469, 1366, 1233, 1165, 1092; dH (400 MHz, CDCl3) 7.76 (2H, d,
J = 8.0 Hz, 2 x ArH), 7.41 (2H, d, J = 8.0 Hz, 2 ꢁ ArH), 4.65 (1H, app. t, J = 7.4 Hz,
4-H), 2.42 (3H, s, Me), 2.05–1.97 (2H, m, 7-CH2), 1.95–1.81 (1H, m, 2-H), 1.56
(2H, quin, J = 7.4 Hz, 8-CH2), 1.49–1.28 (4H, m, 3- and 9-CH2), 1.37 (9H, s, tBu),
0.89 (3H, t, J = 7.1 Hz, 10-Me), 0.88 (6H, d, J = 7.0 Hz, 2 ꢁ Me); dC (100 MHz,
CDCl3) 144.3 (C), 129.5 (C), 129.2 (2 ꢁ ArCH), 128.5 (2 ꢁ ArCH), 89.1 (C), 81.3
(C), 50.4 (4-CH), 43.1 (CH2), 30.4 (CH2), 27.9 (3 ꢁ Me), 24.0 (2-CH), 21.9 (CH2),
21.6 (CH2), 21.4 (ArMe), 18.1 (Me), 13.5 (Me); m/z (ES) 454 (M + NH4+, 100%),
381 (25). Found: [M+NH4+] 454.2732. Calcd for C23H40N3O4S: M, 454.2740. Line
broadening prevented definite observation of the C@O and quaternary tBu
carbon resonances.
11. Dunford, D. G.; Knight, D. W.; Song, C. unpublished results.
12. Brosse, N.; Pinto, M. F.; Jamart-Gregoire, B. Tetrahedron Lett. 2002, 43, 2009–
2011.
14. The compound (Table 1, entry 1) crystallised in the triclinic space group P-1,
with cell dimensions a = 8.58800(10) Å, b = 11.8690(2) Å and c = 13.1390(2) Å.
The data were refined to an R value of 0.0467 and have been deposited in full at
the Cambridge Crystallographic Data Centre, reference CCDC 783430,
accessible at http://www.ccdc.cam.ac.uk.
15. The product 17b (Table 1, entry 2) mp 134–135 °C, crystallised from ether–
petrol in the monoclinic P21/a space group with cell dimensions of
a = 8.9670(2) Å, b = 22.3460(6) Å and c = 12.9480(4) Å. The structure was
refined to a final R value of 0.0537 and the full data set has been deposited
at the Cambridge Crystallographic Data Centre, reference CCDC 783431.
16. The product from Table 1, entry 6, mp 62–63 °C, crystallised from EtOAc-petrol
in the monoclinic P2/1n space group with cell dimensions of a = 15.7349(4) Å,
b = 13.4711(2) Å and c = 20.1130(6) Å. The structure was refined to a final R
value of 0.0586 and the full data set has been deposited at the Cambridge
Crystallographic Data Centre, reference CCDC 783434.
13. A typical procedure is as follows: N-t-Butyloxycarbonyl-N’-(2-methyldec-5-yn-
4-yl)-N’-p-toluenesulfonylhydrazine [17; R1 = iBu, R2 = Bu, R3 = tOBu, Ar = p-
MeC6H4] (Table 1, entry 1)—PPh3 (11.71 g, 44.64 mmol) was added to stirred,
anhydrous THF (450 ml) maintained under nitrogen and cooled in an ice-water
bath. Once the solid had dissolved, diisopropyl azodicarboxylate (8.8 ml,
44.6 mmol) was added dropwise and the resulting solution stirred with
continued cooling for 0.25 h. A white precipitate formed. 2-Methyldec-5-yn-4-
ol [16; R1 = iBu, R2 = Bu] (5.00 g, 29.8 mmol) was then added followed, after an
additional 0.25 h, by N-Boc-N’-tosylhydrazine 13 (11.52 g, 44.64 mmol)
resulting in formation of a clear orange solution. This was stirred overnight
without the addition of further coolant and then the THF was evaporated. The
residue was partitioned between EtOAc and H2O. The separated aqueous layer
was extracted with EtOAc (2 x) and the combined organic solutions were dried