An organocatalytic synthesis of cis-N-Alkyl- and N-arylaziridine carboxylates

Article abstract of DOI:10.1002/adsc.200900474

An extremely mild protocol that employs readily available starting materials, i.e., aldehyde, amine and alkyl diazoacetate, returns structurally diverse N-substituted-C-2/3-difunctionalised aziridines in excellent yields and stereoselectivities when pyrid

Full text of DOI:10.1002/adsc.200900474

COMMUNICATIONS
DOI: 10.1002/adsc.200900474
An Organocatalytic Synthesis of cis-N-Alkyl- and N-Arylaziridine
Carboxylates
Sean P. Bew,a,* Rachel Carrington,a David L. Hughes,a John Liddle,b
and Paolo Pescea
a
School of Chemistry, University of East Anglia, Norwich, NR4 7TJ, U.K.
Fax : (+44)-1603-592-003; phone: (+44)-1603-591-342; e-mail: s.bew@uea.ac.uk
Department of Medicinal Chemistry, GSK, Medicines Research Centre, Gunnels Wood Road, Stevenage, Herts, U.K.
b
Received: July 8, 2009; Published online: October 21, 2009
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/adsc.200900474.
Abstract: An extremely mild protocol that employs
readily available starting materials, i.e., aldehyde,
amine and alkyl diazoacetate, returns structurally
diverse N-substituted-C-2/3-difunctionalised aziri-
dines in excellent yields and stereoselectivities
when pyridinium triflate is incorporated as an orga-
nocatalyst. The reaction process is environmentally
benign affording water and nitrogen as the only by-
products. This racemic protocol paves the way for
the development of novel asymmetric organocata-
lysts capable of generating optically active aziri-
dines.
Keywords: aldehydes; aziridines; diazo compounds;
imines; organic catalysis
Figure 1. Natural product derived and synthetic aziridines.
velopment of interesting synthetic intermediates, effi-
Aziridines are important, synthetically versatile 3- cient protocols for aziridine synthesis continue to be
membered heterocycles with considerable Baeyer an important and very topical research area.[7]
strain.[1] As a consequence, aziridines are relatively re-
The last 10 years have seen enormous interest in
active and eminently capable of undergoing a ple- the development and use of organocatalysts capable
À
À
thora of chemical transformations that often afford an of mediating C C and C X (X=N, O, S) bond for-
assortment of different and useful chemical entities.[2] mation.[8] During this time a sizeable number of inno-
They have, for example, been used to generate natu- vative, often optically active and ever more structural-
ral and unnatural a- and b-amino acids,[3] azasugars[4] ly complex organocatalytic systems have been devel-
and chiral ligands.[5] Furthermore, aziridines are often oped and subsequently tested for their ability to me-
found as key bioactive structural elements within nat- diate reactions that, in many cases, afford increasingly
ural products, many of which have anticancer and or complex reaction products and or synthetic intermedi-
antibiotic properties, i.e., azinomycin A, maduropep- ates.[9]
tin[6] (Figure 1), mitomycin C, FK973 and ficellomy-
The continued construction of structurally unique
cin. In addition to the naturally occurring aziridines organocatalysts is critical if their utility within syn-
recounted, there have been numerous and in some thetic chemistry is to be maintained. It is evident,
cases potentially bioactive synthetic aziridines gener- therefore, that new organocatalysts are required to
ated, i.e., NSC676892, a tricyclic, aziridine-containing sustain and encourage the impressive developments
heterocycle. Thus, as key components of bioactive already achieved. At the same time it is equally im-
molecules and as useful precursor species for the de- portant that promising commercial or readily synthes-
Adv. Synth. Catal. 2009, 351, 2579 – 2588
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2579

Sean P. Bew et al.
COMMUNICATIONS
the application of strong, often metal-based Lewis
acids, i.e., Zn
ample, Jørgensen et al.[13] activated N-alkylimines
using Zn(OTf)2 (10 mol%) and similarly Templeton
ACHTNUGRTEN(NNUG OTf)2, AlCl3 or TiCl4. By means of ex-
AHCTUNGTRENNUNG
et al.[14] activated N-arylimines with BF3·OEt2
(10 mol%) in a process that allowed, after the addi-
tion of EDA, the formation of racemic cis- or cis/
trans-mixtures of N-alkyl- or N-aryl-C-2/3-disubstitut-
ed aziridines, respectively, in very poor to good yields,
i.e., 2–76%. The low yields can be partially accounted
for by the formation of isomeric enamines generated
via 1,2-aryl or 1,2-hydrogen shifts occurring on the
zwitterionic intermediate generated via the initial re-
action between the alkyl diazoacetate and the N-sub-
Scheme 1. Brçnsted acid-derived organocatalytic synthesis
of N-alkyl- or N-arylaziridines.
ised compounds that appear capable but yet uninves- stituted imine.
tigated or ꢁdormantꢂ as organocatalysts are examined
From a non-metal mediated viewpoint Armstrong
for their potential to mediate interesting and syntheti- et al.[15] reported the synthesis of trans-aziridinyl ke-
cally useful chemical transformations.
tones and esters in good, i.e., 62% to excellent, i.e.,
This communication reports our endeavours at uti- 97% yields via the aziridination of chalcones or acryl-
À
lising heterocyclic salts and, in particular, pyridinium ates using N N ylids generated in situ via amination
triflate as cheap, commercially available, structurally of NMO with O-(diphenylphosphinyl)hydroxylamine.
simple and efficient Brønsted acid organocatalysts. Of
Wulff et al.[16] reported an asymmetric catalyst ca-
the salts tested we demonstrate that pyridinium tri- pable of generating N-benzhydryl-C-2/3-disubstituted
flate is a remarkably efficient organocatalyst capable aziridines from the corresponding N-benzhydryl-
of mediating the synthesis of racemic N-alkyl- and N-
ACHTUNGTRENiNGNU mines using a catalyst derived from either vaulted
arylaziridines via an exceptionally mild reaction pro- biaryl ligand (S)-VANOL or (S)-VAPOL which are
tocol, with all of the pyridinium triflate-mediated azir- transformed prior to application into the correspond-
idination reactions tested to date proceeding to com- ing (S)-VANOL- or (S)-VAPOL-pyroborate com-
pletion quickly and efficiently at ambient temperature plexes.[17] Although the yields (51–77%) and ees (91–
and with only 1–10 mol% of the salt (Scheme 1).
Furthermore, our efficient conversion of N-substi- ceptible to enamine formation, i.e., 1–16%.
tuted imine starting materials into the corresponding In an important contribution Johnston et al. demon-
98%) of the aziridines are good, the reaction is sus-
aziridine generates only two by-products; water and strated that strong protic acids, i.e., TFA (pKa 0.5)
nitrogen. Thus utilizing these easily handled, non-hy- and triflic acid (pKa À14) catalyse, at a rather high
groscopic salts in conjunction with a wide variety of loading (7–25 mol%) the aza-Darzen reaction be-
structurally diverse primary N-alkyl- or N-arylamines tween N-benzhydrylimines and EDA affording mix-
with arylaldehydes, and alkyl diazoacetates, the corre- tures of cis-/trans-N-benzhydryl-C-2/3-disubstituted
sponding N-aryl or N-alkyl-cis-C-2/3-difunctionalized aziridines in 40–89% yields (Scheme 2).[18] The most
aziridines are afforded in good i.e., 71% to excellent, notable problems with this protocol are that aldimines
i.e., 88% yields.
generally afforded poor cis-/trans-aziridine regioselec-
Space considerations preclude going into detail on tivities [(60:40)–(90:10) cis-/trans, respectively] and
the synthesis of aziridines; a number of excellent re- that aldimines synthesised from cycloalkane, aliphatic
views[10] and monographs exist.[11] Instead, discussion or aromatic aldehydes afford the corresponding aziri-
will be limited to relevant protocols that employ dines in meagre 40–45% yields. Furthermore, al-
imines as the core starting materials[12] and the prob- though Schiff bases derived from glyoxalate esters, for
lems associated with these protocols. Traditional aziri- example, 1 afforded moderate to excellent yields (62–
dine synthesis via ꢁimine activationꢂ relies heavily on 89%, mass balance comprised enamide by-products)
Scheme 2. Triflic acid for the synthesis of aziridine cis-2.
2580
asc.wiley-vch.de
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2579 – 2588

An Organocatalytic Synthesis of cis-N-Alkyl- and N-Arylaziridine Carboxylates
COMMUNICATIONS
Scheme 3. Pyridinium triflate as an efficient organocatalyst for the synthesis of racemic aziridine 6.
and excellent regioisomer ratios (95:5) of racemic cis/ ported for cis- and trans-aziridines are 7Æ0.5 Hz[20]
trans-C-2/3-disubstituted aziridines these products, for and 3Æ0.5 Hz,[21] respectively. We were delighted to
example, 2 have limited synthetic applications. Fur- observe the formation of only one aziridine stereoiso-
ther additional drawbacks are evident. Thus, TfOH is mer with a J2,3 of 6.8 Hz indicating exclusive forma-
relatively expensive and both acids are extremely hy- tion of the cis-6 stereoisomer. Proof of the cis-rela-
groscopic making them difficult to handle and store, tionship between the C-2/3-substituents was sought.
both acids have to be dried and distilled before use, X-ray analysis conclusively proved our assignment,
their application on a large and small scale requires using J2,3 values, that cis-6 was correct (Figure 2).[22]
that strictly anhydrous solvents/reaction conditions be
Synthesising 7a–c to 10a–c (Figure 3), we investigat-
maintained at all times, if not degradation of the ed their ability to mediate the reaction outlined in
imine occurs, and use of either of these acids requires Scheme 3. Interestingly, quinuclidine (sp3-nitrogen)
À788C reaction conditions not, as would be prefera-
ble, ambient temperature. Furthermore, the use of
such strong acid precludes chemistry on substrates
that have acid-sensitive functional/protecting groups
and, in fact, the chemistry of diazocarbonyl com-
pounds with strong acids is dominated by reports of
their instability.[19]
Initiating our programme we sought to obviate the
problems recounted with transition metal-based
Lewis acids or strong protic acids and probe the abili-
ty of readily and/or commercially available heterocy-
clic salts to act as simple organocatalysts capable of
generating aziridines in high yield and in mild reac-
tion conditions. Employing heterocyclic salts confers
numerous benefits to our protocol. As stable solids
they are easily weighed and handled, are cheap, non-
hygroscopic, require ꢁno immediate to useꢂ drying or
crystallisation, are easily removed from the reaction
by filtration (plug of alumina) and their mildly acidic
nature should allow incorporation of acid-sensitive
functional and/or protecting groups.
Focusing on pyridinium triflate (5), tosylate and tri-
Figure 2. X-ray crystal structure of racemic cis-6.
fluoroacetate, with 3 and diazo ester 4, we were de-
lighted to observe that at ambient temperature all
three salts catalysed (at 10 mol%) the reaction albeit
in variable reaction times, i.e., 2, 5 and 48 h, respec-
tively, with triflate 5 being the most efficient affording
an unoptimised 83% yield of racemic 6 (Scheme 3).
Conventional Lewis acid-mediated protocols often
afford stereoisomeric mixtures of cis-/trans-aziridines
as well as unwanted enamides (vida supra). Utilising
3, 4 and 5 the unpurified reaction mixture containing
1
rac-6 was scrutinised using H NMR (400 MHz) for
the formation of cis- and trans-6. Typical J2,3 values re- Figure 3. sp3 and sp2-nitrogen containing heterocyclic salts.
Adv. Synth. Catal. 2009, 351, 2579 – 2588
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2581

Sean P. Bew et al.
COMMUNICATIONS
salts 7a–c mediated the reaction in 4–12 h and re- Lewis acid-mediated aziridinations promotes enamine
turned cis-6 in 33%, 45% and 68% yields, respective- formation. Gratifyingly, the solvent study demonstrat-
ly. On the other hand catalysts 8a, b, 9a, b and 10a, b ed that utilising a diverse array of polar, apolar or
afforded, generally, higher yields of cis-6, i.e., 77– non-polar solvents there was little evidence for enam-
83%. Catalyst 9c afforded no product and 10c afford- ine formation.
ed cis-6 in a reduced 10% yield. The poor perfor-
In an effort to expand the structural diversity of the
mance of 9c and 10c was attributed to their low solu- aziridines created and explore the scope of the reac-
bility in dichloromethane. In significantly slower reac- tion a series of N-alkyl- and N-arylimines was syn-
tion times (12 h) the monotriflate salts of DBN and thesised. Utilising these Schiff bases, ethyl or tert-
TMG afforded cis-6 in 72% and 78% yields, respec- butyl diazoacetate and 5 (10 mol%) we observed the
tively.
formation of aziridines 11–31 (Table 2) in good (71%)
The reaction in Scheme 3 was repeated to confirm to excellent yields (88%). These preliminary results
the importance of 5 to act, presumably, as a catalyst. indicate that a catalytic quantity of 5 efficiently medi-
Initially in the absence of 5 a deuterated chloroform ates the high yielding synthesis of structurally diverse
solution of 3 and 4 was shaken in an NMR tube (3 h). N-alkyl- and N-arylaziridines.
1
Subsequent H NMR (400 MHz) analysis confirmed
Worthy of particular note, the unoptimised yields
no reaction had occurred. Subsequent addition of 5 of 11–31 are generally superior to those reported
1
(10 mol%) followed within one minute by H NMR using Lewis,[13,14] trifluoroacetic or trifluoromethane-
analysis confirmed two new doublets (J2,3 =6.8 Hz) at sulfonic acids.[18] Furthermore, the tolerance of our
2.7 ppm and 3.2 ppm had formed; clearly indicating aziridination reaction to N-substituent variation is im-
that rapid formation of cis-6 only occurs when catalyst pressive for, e.g., N-allylaziridine 11, N-octylaziridine
5 is added to the reaction.
12, N-tert-butylaziridine 13, N-benzhydrylaziridines
Using 3, 4 and 5 (10 mol%), the reaction in 14–16, N-triphenylmethylaziridine 17, N-benzylaziri-
Scheme 3 was repeated at À78 and À408C, there was dine 18, and N-arylaziridines 19–31. Substituting
no reaction. At À208 the reaction was significantly EDA for tert-butyl diazoacetate afforded cis-18 in an
slower (24 h) than the reaction at 08C which afforded 82% yield, a yield comparable to that afforded when
a 53% yield of racemic cis-6. This yield is, however, 4 was utilised cf. 6, 83%. Thus the reaction does not
significantly lower than the 83% yield afforded when appear to be adversely affected by the incorporation
the reaction was conducted at ambient temperature.
of a sterically encumbered, large tert-butyl group
Having established that heterocyclic salts and in within the alkyl diazoacetate component of the reac-
particular pyridinium triflate are viable catalysts for tion. The ability to utilise glyoxalate ester-derived
N-alkylaziridine synthesis, we elected, using 5 and the imines for the synthesis of 16 has been established.
protocol in Scheme 3, to screen solvents with dispa- Furthermore, the ability to append heteroaromatic
rate polarities for their effect, if any, on the cis-/trans- substituents on the C3-aziridine ring component, for
stereochemistry and yields of racemic 6. The aziridi- example, 11–14, 17–21 has particular appeal, as has
nation reaction was largely unaffected by the solvent the synthesis of N-sterically encumbered heteroaro-
employed, thus ethanol (m5.8, entry 4, Table 1) and matic species, for example, 17.
pentane (m0, entry 5) both afforded racemic cis-6 in,
On the other hand, addition of 5 (10 mol%) to a di-
essentially, identical yields, i.e., 69% and 68%, respec- chloromethane solution of the starting material imine
tively. It is known that the use of polar solvents in for N-arylaziridine 19 and tert-butyl diazoacetate re-
sulted in rapid evolution of nitrogen gas (~5 min, the
reaction rate with EDA was similar), consumption of
the starting materials and concomitant formation of
cis-19 (using H NMR no evidence for trans-19 could
Table 1. Solvent study for the synthesis of racemic cis-6.
1
Entry
Solvent
Yield of cis-6
be found) in an 87% yield. We decided to probe the
possibility of generating N-arylaziridines appended
with a variety of N-aryl substituents. Gratifyingly, our
organocatalytic protocol afforded N-arylaziridines ap-
pended with single, i.e., 19, 22–31 and multiple elec-
tron-rich substitents, i.e., 20 as well as aziridine 21,
derived from the aziridination of the imine generated
from condensing 2-pyridinecarbaldehyde and 2-
amino-4-nitrophenol.
Focussing on C-3 aryl substituents we demonstrated
that a series of diverse functional groups are readily
incorporated, e.g., halogens 23–28, strongly electron-
withdrawing substituents, e.g., 31, as well as weakly
1
2
3
4
5
6
7
8
Dichloromethane
Toluene
Chloroform
Ethanol
82%
76%
70%
69%
68%
68%
66%
65%
57%
56%
48%
0%
Pentane
Ethyl acetate
Propionitrile
Isopropyl alcohol
Ether
Tetrahydrofuran
tert-Butyl methyl ether
Pyridine
9
10
11
12
2582
asc.wiley-vch.de
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2579 – 2588

An Organocatalytic Synthesis of cis-N-Alkyl- and N-Arylaziridine Carboxylates
Table 2. N-Alkyl- and N-arylaziridines synthesised using 5.
COMMUNICATIONS
No. and Yield
Aziridine
No. and Yield
Aziridine
11, 92%
12, 90%
13, 86%
15, 74%
14, 80%
16, 86%
17, 78%
19, 97%
18, 82%
20, 94%
21, 79%
23, 76%
22, 71%
24, 75%
25, 79%
26, 92%
Adv. Synth. Catal. 2009, 351, 2579 – 2588
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2583

Sean P. Bew et al.
COMMUNICATIONS
Table 2. (Continued)
No. and Yield
Aziridine
No. and Yield
Aziridine
27, 99%
29, 77%
30, 91%
28, 79%
30, 95%
–
–
electron-donating groups, e.g., 29. An unprotected hy- starting materials that is, 4, 32 and 33 together in the
droxy group on 4-hydroxybenzaldehyde was not toler- presence of 5 (10 mol%). Gratifyingly, racemic cis-34
ated, simple O-protection of the phenol as the corre- was afforded in an excellent 80% yield (Scheme 4).
sponding O-Fmoc derivative allowed the correspond-
ing N-arylaziridine 30 to be efficiently generated in our organocatalytic aziridination protocol was sought.
an 85% yield (Table 2). The ability to incorporate an alcohol in place of an al-
Further exemplication of the scope and potential of
Synthesising aziridines directly from aldehydes and dehyde would be of significant utility for the synthesis
amines has significant experimental advantages over of aziridines. Transforming alcohol and amine core
imine isolating, multi-step procedures. Investigating starting materials into N-arylaziridines via a more ef-
the possibility of simplifying our aziridination proce- ficient two-stage three-component reaction process
dure, we reacted one equivalent each of 4-nitroben- was deemed worthy of investigation. Taylor et al. re-
zaldehyde (32) and 2-tert-butoxyaniline (33) in DCM ported a one-pot synthesis of imines using alcohols,
(with 4 ꢃ molecular sieves) the resulting imine was amines and manganese dioxide.[23] Undertaking the
subsequently reacted with 4 in the presence of 5 synthesis of cis-34 we synthesised N-arylimine 36
(10 mol%). Racemic cis-6 was afforded in an 82% from amine 33, alcohol 35 and manganese dioxide, we
yield. In our quest to reduce the number of reaction/ filtered off the excess oxidant and subsequently
isolation steps still further, we pondered the possibili- added 5 (10 mol%) and 4. Even though no special
ty of generating racemic cis-34 by mixing all three precautions were taken to exclude either water or air
Scheme 4. cis-34 via a one-pot three-component process.
2584
asc.wiley-vch.de
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2579 – 2588

An Organocatalytic Synthesis of cis-N-Alkyl- and N-Arylaziridine Carboxylates
COMMUNICATIONS
Scheme 5. Synthesis of cis-34 via an oxidation/imine/aziridination reaction process.
Scheme 6. One-pot synthesis of cis-34 using polymer 37.
(no dry solvents were used and the reaction was per- group affording the desired NH-aziridine 41 in an
formed in a screw-top vial) we were delighted to iso- 89% yield. Reaction of 41 with PTSA in aqueous ace-
late cis-34, a precursor to chloramphenicol, in an tonitrile at 408C mediated the hydrolytic ring-opening
overall 80% yield. The fact that no precautions to of the aziridine affording racemic anti-b-hydroxy-a-
protect the reaction against air or water during the amino acid 42 in a 64% yield. Confirmation of the
highly efficient synthesis of cis-34 indicates the robust anti-stereoselectivity in the ring opening of the aziri-
nature of our aziridination protocol (Scheme 5).
dine was sought. Reacting b-hydroxy-a-amino ester
The use of a polymer-bound Brønsted acid would 42 with CDI in THF afforded oxazolidin-2-one 43.
1
confer several benefits to our protocol; these include Analysis of the H NMR spectrum of 43 confirmed
an easier work-up of the reaction, i.e., simple filtra- the trans relationship of the C-4/5 protons (J4,5
=
tion removes 37 and the potential for recycling 37. 5.1 Hz) which is in close agreement to J values report-
We were delighted that a straightforward synthesis of ed for similar oxazolidin-2-ones.[25] Further evidence
37 [addition of TfOH (1 equiv.) to polystyrene-immo- for the trans-relationship of the C-2/3-substituents on
bilised DMAP in DCM at 08C] allowed the genera- 43 was gathered via an nOe study. A correlation was
tion of cis-34 in a pleasing 80% yield (Scheme 6).
observed beween the C-2 hydrogen and the ortho-hy-
The racemic synthesis of fluorinated tyrosine 42, an drogens on the aryl ring and between the C-3 hydro-
important component of the potent antibiotic fluoro- gen and the hydrogen on the NH group. No correla-
balhimycin, was undertaken.[24] Commercially avail- tion was observed between the C-2/3 hydrogens
able 3-fluoro-4-hydroxybenzaldehyde 38 was trans- (Scheme 7).
formed into the corresponding O-pivaloyl derivative
39 via reaction with pivaloyl chloride and pyridine. synthesis of structurally diverse N-alkyl- and N-aryl-
The subsequent transformation of 39 into racemic cis- aziridines has been developed with the protocol
In summary, an efficient organocatalytic racemic
AHCTUNGTRENNUNG
aziridine 40 was performed in a one-pot reaction suited to a wide range of N-substituents. Worthy of
(75% over the 2 steps from 38, Scheme 7) by reacting particular note, our aziridination reaction is environ-
39 with EDA in the presence of 5 (10 mol%) and 2- mentally benign generating only water and N2 by-
tert-butoxy-4-methoxyaniline. Gratifyingly, racemic products. Furthermore, using an in situ tandem oxida-
cis-aziridine 40 (J2,3 =6.6 Hz) was afforded in an uno- tion protocol we demonstrate the feasibility and effi-
pitimised 78% yield. Subsequent oxidation using ciency of taking alcohols through to aziridines via a
CAN efficiently cleaved the electron-rich N-aryl one-pot reaction that does not require aldehyde isola-
Adv. Synth. Catal. 2009, 351, 2579 – 2588
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2585

Sean P. Bew et al.
COMMUNICATIONS
Scheme 7. Synthesis of rac-42 from aldehyde 38.
semi-solid. Purification of the reaction mixture via flash
column chromatography (SiO2, elution with hexane/diethyl
ether, 4:1) afforded a yellow solid; yield: 90 mg (84%). Sub-
sequent analysis indicated this to be 34; mp 117.4–119.38C.
1H NMR (CDCl3, 400 MHz): d=8.17 (d, J=8.75 Hz, 2H),
7.69 (d, J=8.88 Hz, 2H), 7.02–6.98 (m, 1H), 6.95–6.90 (m,
3H), 3.50 (d, J=6.76 Hz, 1H), 3.11 (d, J=6.77 Hz, 1H),
1.32 (s, 9H), 1.19 (s, 9H); 13C NMR (CDCl3, 300 MHz): d=
166.5, 148.2, 147.6, 145.6, 143.2, 129.2, 123.7, 123.3, 123.2,
123.1, 120.9, 82.1, 80.6, 47.9, 46.9, 28.9, 28.0; IR (film): n=
2978, 2933, 1743, 1716, 1603, 1520, 1490, 1451, 1392, 1367,
1344, 1262, 1224, 1160, 1111, 1048, 888, 854, 753 cmÀ1; MS
(EI)+: m/z=412.1 (100%) [M]+; HR-MS (EI)+: m/z=
412.1998, exact mass calculated for [C23H28N2O5]+: 412.1993.
tion and can be conducted without the exclusion of
water or air. Exemplifying our protocol, the first race-
mic synthesis of 3-fluoro-b-hydroxytyrosine is report-
ed in four steps.
Although the use of asymmetric heterocyclic salts
has not, to date, returned optically active aziridines
we are currently investigating this aspect of our reac-
tion and our preliminary racemic results pave the way
for an asymmetric variant of our reaction to be devel-
oped. The results of these studies will be published in
due course.
Experimental Section
Typical Procedure
Acknowledgements
To a stirred solution of 4-nitrobenzyl alcohol (35, 766 mg,
5.0 mmol), 2-tert-butoxyaniline (33, 785 mg, 4.75 mmol) and
4 ꢃ molecular sieves (ca. 1 g) in DCM (125 mL) was added
activated manganese dioxide (625 mg, 7.5 mmol, 1.5 equiv.,
Aldrich 21,764-6) and the mixture heated at reflux. A
second equivalent of manganese dioxide (435 mg, 5 mmol)
The authors thank the EPSRC, University of East Anglia and
GSK (Stevenage, UK) for financial support.
was added to the reaction after one hour and the resulting References
mixture stirred and refluxed overnight. Spent and excess
manganese dioxide was removed by filtration through Celite
and then washed with DCM (3ꢄ25 mL). The combined
DCM fractions were concentrated under vacuum affording
36 as a pale yellow oil; yield: 95%.
(E)-2-tert-Butoxy-N-(4-nitrophenylmethylene)phenylam-
ACHTUNGTRENNUNGine (36, 78 mg, 0.26 mmol) and pyridinium triflate (5, 6 mg,
0.026 mmol, 10 mol%) were dissolved in dichloromethane
(1 mL). The reaction mixture was stirred at room tempera-
ture for 2 min. tert-Butyl diazoacetate (0.038 mL, 0.28 mmol,
1.1 equiv.) was added (flurry of bubbles observed, N2
evolved) and the resulting solution stirred at ambient tem-
perature for 5 h. Subsequent TLC analysis (eluent, hexane:
diethyl ether, 4:1) indicated complete consumption of 36.
Removal of the solvent under vacuum afforded a brown
[1] J. B. Sweeney, Chem. Soc. Rev. 2002, 31, 247.
[2] X. Eric Hu, Tetrahedron 2004, 60, 2701.
[3] B. Zwanenburg, P. T. Holte, Top. Curr. Chem. 2001,
216, 93.
[4] a) D. Tanner, Angew. Chem. 1994, 106, 625; Angew.
Chem. Int. Ed. Engl. 1994, 33, 599; b) T. K. Chakrabor-
ty, S. Jaiprakash, Tetrahedron Lett. 1997, 38, 8899;
c) G. J. Joly, K. Peeters, H. Mao, T. Brossette, G. J.
Hoornaert, F. Compernolle, Tetrahedron Lett. 2000, 41,
2223.
[5] P. Somfai, Latvijas Kimijas Zurnals 1999, 59.
[6] K. Komano, S. Shimamura, M. Inoue, M. Hirama, J.
Am. Chem. Soc. 2007, 129, 14184.
[7] W. McCoull, F. A. Davis, Synthesis 2000, 1347.
2586
asc.wiley-vch.de
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2579 – 2588

An Organocatalytic Synthesis of cis-N-Alkyl- and N-Arylaziridine Carboxylates
COMMUNICATIONS
[8] a) A. Dondoni, A. Massi, Angew. Chem. 2008, 120,
4716; Angew. Chem. Int. Ed. 2008, 47, 4638; b) H. Pel-
lissier, Tetrahedron 2007, 63, 9267; c) P. I. Dalko, L.
Moisan, Angew. Chem. 2004, 116, 5248; Angew. Chem.
Int. Ed. 2004, 43, 5138; d) P. I. Dalko, L. Moisan,
Angew. Chem. 2001, 113, 3840; Angew. Chem. Int. Ed.
2001, 40, 3726.
[9] a) D. Enders, O. Niemeier, A. Henseler, Chem. Rev.
2007, 107, 5606; b) H. Kotsuki, H. Ikishima, A. Okuya-
ma, Heterocycles 2008, 75, 757; c) W. R. Roush, J. L.
Methot, Adv. Synth. Catal. 2004, 346, 1035; d) Q. Tao,
G. Tang, K. Lin, Y.-F. Zhao, Chirality 2008, 20, 838;
e) M. Shi, X.-G. Liu, Org. Lett. 2008, 10, 1043; f) S. H.
McCooey, S. J. Connon, Angew. Chem. 2005, 117, 6525;
Angew. Chem. Int. Ed. 2005, 44, 6367; g) N. Dahlin, A.
Bøgevig, H. Adolfsson, Adv. Synth. Catal. 2004, 346,
1101.
Wulff, J. Am. Chem. Soc. 1999, 121, 5099; aa) W. Xie, J.
Fang, J. Li, P. G. Wang, Tetrahedron 1999, 55, 12929;
ab) K. Juhl, R. G. Hazell, K. A. J. Jørgensen, J. Chem.
Soc. Perkin Trans. 1 1999, 2293; ac) J. H. Espenson,
Chem. Commun. 1999, 479; ad) M. F. Mayer, M. M.
Hossain, J. Org. Chem. 1998, 63, 6839; ae) S. Nagaya-
ma, S. Kobayashi, Chem. Lett. 1998, 685; af) K. G. Ras-
mussen, K. Juhl, R. G. Hazell, K. A. J. Jørgensen,
Chem. Soc. Perkin Trans. 2 1998, 1347; ag) H.-J. Ha J.-
M. Suh, K.-H. Kang, Y.-G. Ahn, O. Han, Tetrahedron
1998, 54, 851; ah) J. M. Mohan, B. S. Uphade, V. R.
Choudhary, T. Ravindranathan, A. Sudalai, Chem.
Commun. 1997, 1429; ai) T. B. Gunnoe, P. S. White,
J. L. Templeton, L. Casarrubios, J. Am. Chem. Soc.
1997, 119, 3171; aj) V. K. Aggarwal, A. Thompson,
R. V. H. Jones, M. C. H. Standen, J. Org. Chem. 1996,
61, 8368; ak) Z. Zhu, J. H. Espenson, J. Am. Chem.
Soc. 1996, 118, 9901; al) K. G. Rasmussen, K. A. J. Jør-
gensen, J. Chem. Soc. Chem. Commun. 1995, 1401;
am) Z. Zhu, J. H. Espenson, J. Org. Chem. 1995, 60,
7090; an) K. B. Hansen, N. S. Finney, E. N. Jacobsen,
Angew. Chem. 1995, 107, 750; Angew. Chem. Int. Ed.
Engl. 1995, 34, 676; ao) M. Moran, G. Bernardinelli, P.
Mꢈller, Helv. Chim. Acta 1995, 78, 2048; ap) P. Baret,
H. Buffet, J.-L. Pierre, Bull. Soc. Chim. Fr. 1972, 2493;
aq) T. Hashimoto, N. Uchiyama, K. Maruoka, J. Am.
Chem. Soc. 2008, 130, 14380; ar) T. Akiyama, T.
Suzuki, K. Mori, Org. Lett. 2009, 11, 2445; as) X. Zeng,
Z. Zeng, Z. Zu, M. Lu, G. Zhong, Org. Lett. 2009, 11,
3026.
[10] a) H. M. Osborn, J. B. Sweeney, Tetrahedron: Asymme-
try 1997, 8, 1693; b) G. S. Singh, M. Dꢂhooghe, N. De
Kimpe, Chem. Rev. 2007, 107, 2080; c) V. H. Dahanu-
kar, I. A. Zavialov, Curr. Opin. Drug. Discovery and
Development, 2002, 5, 918.
[11] V. K. Aggarwal, D. M. Badine, V. A. Moorthie, in:
Aziridines and Epoxides in Organic Synthesis, (Ed.: A.
K. Yudin), Wiley-VCH, Weinheim, 2006, pp 1–35.
[12] a) A. Takahiko, S. Tohru, M. Keiji, Org. Lett. 2009, 11,
2445; b) Z. Lu, Y. Zhang, W. D. Wulff, J. Am. Chem.
Soc. 2007, 129, 7185; c) A. Mazumdar, Z. Xue, M. F.
Mayer, Synlett 2007, 2025; d) Y. Deng, Y. R. Lee, C. A.
Newman, W. D. Wulff, Eur. J. Org. Chem. 2007, 2068;
e) S. Zhu, Y. Liao, S. Zhu, Synlett 2005, 1429; f) S.-H.
Lee, I.-W. Song, Bull. Korean Chem. Soc. 2005, 26, 223;
g) A. F. Khlebnikov, M. S. Novikov, R. R. Kostikov,
Russ. Chem. Rev. 2005, 74, 171; h) B. Vanderhoydonck,
C. V. Stevens, Synthesis 2004, 722; i) Y. Li, P. W. H.
Chan, N.-Y. Zhu, C.-M. Che, H.-L. Kwong, Organome-
tallics 2004, 23, 54; j) M. Redlich, H. H. Hossain, Tetra-
hedron Lett. 2004, 45, 8987; k) J. S. Yadav, B. V. S.
Reddy, M. S. Rao, P. N. Reddy, Tetrahedron Lett. 2003,
44, 5275; l) W. Sun, C.-G. Xia, H.-W. Wang, Tetrahe-
dron Lett. 2003, 44, 2409; m) T. Akiyama, S. Ogi, K.
Fuchibe, Tetrahedron Lett. 2003, 44, 4011; n) J. R.
Krumper, M. Gerisch, J. M. Suh, R. J. Bergman, T. D.
Tilley, J. Org. Chem. 2003, 68, 9705; o) D. Morales, J.
Pꢅrez, L. Riera, V. Riera, R. Corzo-Suꢆrez, S. Garcꢇa-
Granda, D. Miguel, Organometallics 2002, 21, 1540;
p) M. V. Spanedda, B. Crousse, S. Narizuka, D. Bonnet-
Delpon, J.-P. Bꢅguꢅ, Collect. Czech. Chem. Commun.
2002, 67, 1359; q) M. F. Mayer, Q. Wang, M. M. Hos-
sain, J. Organomet. Chem. 2001, 630, 78; r) C. Loncaric,
W. D. Wulff, Org. Lett. 2001, 3, 3675; s) V. K. Aggarwal,
M. Ferrara, C. J. OꢂBrien, A. Thompson, R. V. H.
Jones, R. J. Fieldhouse, J. Chem. Soc. Perkin Trans. 1
2001, 1635; t) M. P. Doyle, W. Hu, D. J. Timmons, Org.
Lett. 2001, 3, 933; u) S.-H. Lee, T.-D. Han, K. Yu, K. H.
Ahn, Bull. Korean Chem. Soc. 2001, 22, 449; v) B.
Crousse, S. Narizuka, D. Bonnet-Delpon, J.-P. Bꢅguꢅ,
Synlett 2001, 679; w) J. C. Antilla, W. D. Wulff, Angew.
Chem. 2000, 112, 4692; Angew. Chem. Int. Ed. 2000, 39,
4518; x) S. Sengupta, S. Mondal, Tetrahedron Lett.
2000, 41, 6245; y) T. Kubo, S. Sakaguchi, Y. Ishii,
Chem. Commun. 2000, 625; z) J. C. Antilla, W. D.
[13] K. G. Rasmussen, K. A. Jørgensen, J. Chem. Soc.
Perkin Trans. 1 1997, 1287.
[14] L. Casarrubios, J. A. Pꢅrez, M. Brookhart, J. L. Temple-
ton, J. Org. Chem. 1996, 61, 8358.
[15] A. Armstrong, C. A. Baxter, S. G. Lamont, A. R. Pape,
R. Wincewicz, Org. Lett. 2007, 9, 351.
[16] J. C. Antilla, W. D. Wulff, Angew. Chem. 2000, 112,
4692; Angew. Chem. Int. Ed. 2000, 39, 4518.
[17] Y. Zhang, A. Desai, Z. Lu, Z. Ding, W. D. Williams,
Chem. Eur. J. 2008, 14, 3785.
[18] A. L. Williams, J. N. Johnston, J. Am. Chem. Soc. 2004,
126, 1612.
[19] Diazo compounds: properties and synthesis (Eds.: M.
Regitz, G. Maas), Orlando, London, Academic Press,
1986.
[20] K. G. Rasmussen, K. A. Jørgensen, J. Chem. Soc.
Chem. Commun. 1995, 1401.
[21] W. Xie, J. Fang, J. Li, P. G. Wang, Tetrahedron 1999, 55,
12929.
[22] Crystal data for compound cis-6: C17H18N2O2, M=
282.3. Orthorhombic, space group P212121 (no. 19), a=
7.5305(5),
1489.66(14) ꢃ3. Z=4, 1cald. =1.259 gcmÀ3, F
T=140(1) K, lCAHTUNGRTENNGUN
(Mo-Ka)=0.83 cmÀ1,
b=10.8160(6),
c=18.2893(9) ꢃ,
(000)=600,
(Mo-Ka)=
V=
AHCTUNGTRENNUNG
mG
0.71069 ꢃ. Colourless plate crystal. Intensity data mea-
sured on an Oxford Diffraction Xcalibur-3 CCD dif-
fractometer (Mo-Ka radiation, graphite monochroma-
tor). 4334 unique reflections (Rint =0.044) to qmax =308;
4218 ꢁobservedꢂ with I>2sI. Data processed using the
CrysAlis-CCD and -RED[26] programmes. Structure de-
termined by direct methods in SHELXS[27] and refined
in SHELXL.[27] Non-hydrogen atoms were refined ani-
Adv. Synth. Catal. 2009, 351, 2579 – 2588
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2587

Sean P. Bew et al.
COMMUNICATIONS
sotropically. Hydrogen atoms in idealised positions
with Uiso values riding. At conclusion, wR2 =0.099 and
R1 =0.045 for all 4334 reflections; for ꢁobservedꢂ data
only, R1 =0.043. Flack x=À0.9(10); absolute configura-
tion not certain.CCDC 609715 contains the supplemen-
tary crystallographic data for product cis-6 in 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.
[24] S. Weist, B. Bister, O. Puk D. Bischoff, S. Pelzer, G. J.
Nicholson, W. Wohlleben, G. Jung, R. D. Sꢈssmuth,
Angew. Chem. 2002, 114, 3531; Angew. Chem. Int. Ed.
2002, 41, 3383.
[25] T. Manaka, S-I. Nagayama, W. Desadee, N. Yajima, T.
Kumamoto, T. Watanabe, T. Ishikawa, M. Kawahata,
K. Yamaguchi, Helv. Chim. Acta 2007, 90, 128–142.
[26] Programmes CrysAlis-CCD and -RED, Oxford Diffrac-
tion Ltd., Abingdon, UK, 2005.
[23] L. Blackburn, R. J. K. Taylor, Org. Lett. 2001, 3, 1637.
[27] G. M. Sheldrick, Acta Crystallogr. 2008, A64, 112.
2588
asc.wiley-vch.de
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2579 – 2588