The bromodiazirinyl anion: A weakly bound complex of diazirinylidene and a bromide ion

Article abstract of DOI:10.1002/ejoc.201100529

Experimental and computational evidence suggests that the cleavage of n-butyl and tert-butyl 3-bromodiazirine-3-carboxylates by sodium n-butoxide in DMF, affording high yields of dibutyl carbonates, may proceed by nucleophilic acyl displacement of the bromodiazirinyl anion, a weakly bound complex of a cyclic carbene c-CN₂ (diazirinylidene) and a bromide ion. We explain the formation of substantial amounts of di-n-butoxymethane in the presence of n-butanol by a sequence of O-H insertion and denitrogenation reactions of the putative c-CN₂ carbene. This ground-state singlet species is the last undescribed member of the CN₂ family of reactive intermediates. It differs from classical N-heterocyclic carbenes by its high ionization potential and electron affinity.

Full text of DOI:10.1002/ejoc.201100529

FULL PAPER
DOI: 10.1002/ejoc.201100529
The Bromodiazirinyl Anion: A Weakly Bound Complex of Diazirinylidene and
a Bromide Ion
[a]
[a]
[a]
ˇ
˚
Tomás Martinu,* Stanislav Böhm, and Eva Hanzlová
Keywords: Carbenes / Carbenoids / Diazirinylidene / Nitrogen heterocycles / Diazirines
Experimental and computational evidence suggests that the
amounts of di-n-butoxymethane in the presence of n-butanol
cleavage of n-butyl and tert-butyl 3-bromodiazirine-3-carb- by a sequence of O–H insertion and denitrogenation reac-
oxylates by sodium n-butoxide in DMF, affording high yields
of dibutyl carbonates, may proceed by nucleophilic acyl dis-
placement of the bromodiazirinyl anion, a weakly bound
complex of a cyclic carbene c-CN2 (diazirinylidene) and a
bromide ion. We explain the formation of substantial
tions of the putative c-CN2 carbene. This ground-state sing-
let species is the last undescribed member of the CN2 family
of reactive intermediates. It differs from classical N-hetero-
cyclic carbenes by its high ionization potential and electron
affinity.
(5) under standard thermolytic or photolytic conditions,[3]
it can be considered as a chemically and geometrically or-
thogonal source of two different carbene species and, given
the reactivity of 4, as a single-carbon atom donor.
Introduction
3-Halo-3H-diazirines and their halogen-exchange deriva-
tives are well known nitrogenous precursors of carbenes.[1]
The halogen exchange usually proceeds through 1H-diazir-
ines, by two consecutive SN2Ј-like reactions on the diazirine
ring. In some cases, however, the initial nucleophilic attack
occurs in the vicinity of the diazirine ring, resulting in a
fragmentation reaction. For example, 3-chloro-3-(p-nitro-
phenoxy)diazirine was recently reported to fragment upon
nucleophilic attack on the phenyl ring (SNAr), competing
with the SN2Ј exchange on the diazirine ring.[2] Another
example is the smooth cleavage of butyl 3-bromodiazirine-
3-carboxylate (1) to dibutyl carbonate (2) and nitrogen gas
by the butoxide ion, with no halogen exchange observed.[3]
We report experimental and computational evidence that
the reaction of 1 may proceed by nucleophilic acyl displace-
ment of the bromodiazirinyl anion (3), a weakly bound
Results and Discussion
The known reaction of diazirine ester 1 with sodium but-
oxide (2 equiv.) in anhydrous N,N-dimethylformamide
(DMF) at –15 °C affords carbonate 2 (76% yield) and nitro-
gen gas (61% yield).[3] No other major products were de-
tected by NMR spectroscopy when we monitored the reac-
tion in [D7]DMF. As the ester group of 2 plausibly origi-
nates from that of 1, it is likely the diazirine ring is lost
during the reaction in the form of nitrogen gas and dark-
colored “polymers”. We hypothesized that the reactivity of
1 is dominated by the electrophilicity of its ester group, acti-
vated by three electronegative α-heteroatoms. In an attempt
to suppress the nucleophilic attack on the ester group in
favor of the halogen exchange on the diazirine ring, we have
prepared the sterically hindered tert-butyl ester 11 from cya-
nide 6[4] via trichloroamidine 10[3] (Scheme 2). However, the
exposure of 11 to sodium n-butoxide (4 equiv.) in DMF at
–15 °C resulted in an instantaneous cleavage to a 66:34 mix-
ture of essentially pure di-n-butyl (2) and n-butyl tert-butyl
(12) carbonates in 79% yield. Control experiments indi-
cated that the product composition is significantly affected
by fast transesterification with free butoxide ions, and,
therefore, the observed 2/12 ratio is unsuitable for mechan-
istic considerations.
complex of diazirinylidene (4) and
(Scheme 1). As 1 affords bromo(butoxycarbonyl)carbene
a bromide ion
Scheme 1. Orthogonal cleavage modes of 1.
[a] Department of Organic Chemistry, Institute of Chemical
Technology
The apparent propensity of diazirine esters 1 and 11 to
be attacked by an alkoxide ion at the ester group rather
than on the diazirine nitrogen atoms called for a computa-
tional study. Our DFT calculations[5] on the model reaction
Technická 5, 16628 Prague, Czech Republic
Fax: +420-220-444-288
E-mail: martinut@vscht.cz
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201100529.
6254
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 6254–6260

The Bromodiazirinyl Anion
of the corresponding TS2Ј with respect to the essentially
unaffected TS1Ј [relative free energies calculated at both
DFT levels are virtually equal (Ϯ0.6 kcal/mol)].
The second step of the halogen exchange pathway, SN2Ј
isomerization of 16 to 3H-diazirine 17, is strongly exer-
gonic, and we found that it proceeds without a barrier. The
fact that our experiments with 1 and 11 yielded only cleav-
age products, while we have observed no products attribut-
able to the halogen exchange, can be explained by facile
irreversible fragmentation of the possibly faster formed in-
termediate 14 to carbonate 15 (cf. carbonates 2 and 12 ob-
tained experimentally) and 3. The corresponding TS3 is cal-
culated to be higher in energy than TS2 by 7.4 (PBE) or
1.0 kcal/mol (B3LYP). Similar fragmentation of the halo-
gen-exchange product 17 to carbonate 15 and methoxydi-
azirinyl anion (19), proceeding via tetrahedral intermediate
18, has a much higher activation barrier (ca. 30–33 com-
pared to ca. 15–17 kcal/mol for 13).
Scheme 2. Preparation and butoxide cleavage of 11. (a) EtSH,
DIEA (N,N-diisopropylethanamine), 91%; (b) NH4Cl, MeOH,
57%; (c) aq. NaOCl, Et2O, 78%; (d) tBuOCl, 98%; (e) LiBr,
MeCN, 63%.
between methyl ester 13 and the methoxide ion in DMF
The results of our simplified model calculations thus
[CPCM (conductor-like polarizable continuum model) sol- indicate that the nucleophilic acyl substitution, explaining
vation model] showed that the addition of methoxide to the the observed alkoxide cleavage of diazirines 1 and 11,
ester carbonyl group of 13, leading to tetrahedral interme- should be able to effectively compete with the halogen-ex-
diate 14, is preferred to the SN2Ј bromide displacement, re- change reaction commonly observed with halodiazirines,
sulting in 1H-diazirine 16 (Scheme 3). Free energy differ- and in DMF the former process is probably somewhat fa-
ences between the corresponding transition structures TS2 vored. Calculations also show that if a halogen-exchange
and TS1, calculated at the PBE and B3LYP/6-311+G(d,p) product was formed, it would be unlikely to fragment to
levels, are 2.4 and 3.1 kcal/mol, respectively. As expected, the observed carbonate. Another conceivable mode of de-
the steric hindrance in tert-butyl ester 11 raises the energy composition of the 17-type exchange product is denitro-
Scheme 3. DFT study of methoxide attack on methyl ester 13. [a] Relative free energies [kcal/mol, 298 K] calculated at the PBE/6-
311+G(d,p) level by using the CPCM model for DMF as solvent. [b] Results at the B3LYP level. [c] Corresponding PBE and [d] B3LYP
results for tert-butyl ester 11 (TSЈ). [e] Anion 3 not found as energy minimum due to its dissociation. [f] Interatomic distances [Å]
calculated at the PBE level, hydrogen atoms omitted for clarity.
Eur. J. Org. Chem. 2011, 6254–6260
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6255

T. Martinu˚, S. Böhm, E. Hanzlová
FULL PAPER
genation to a stabilized (push–pull) carbene. As will be Table 1. Free energies of dissociation of 3 and 19–21.[a]
shown later, products derived from such a carbene should
have been detected if they were present.
Process
ΔGgas
ΔGDMF
–CN2Br (3) Ǟ CN2 (4) + Br–
–CN2OMe (19) Ǟ 4 + MeO–
–CF2Br (20) Ǟ CF2 + Br–
10.9
42.9
9.2
–3.2
25.0
–3.4
–3.7
In principle, the release of a diazirinyl anion (such as in
TS3 or TS5) may be feasible. Nonaromatic diazirinyl
anions[6] have been directly observed by mass spectrome- –CF2Cl (21) Ǟ CF2 + Cl–
12.2
try[7] and implicated in radical anion processes involving
[a] Values in kcal/mol at 298 K calculated at the PBE/6-311+G(d,p)
level.
halodiazirines.[8] The unusually long calculated C–Br dis-
tance, electron population, and bond orders in anion 3,
however, suggest that it can be interpreted as a complex of
the cyclic CN2 carbene (4) with a Br– ion (Figures 1 and 2).
The structure of 3 contrasts with that of 19, which possesses
only a slightly elongated C–O bond, and the majority of
negative charge is localized on the ring, as expected for a
regular diazirinyl anion. The features of anions 3 and 19
agree with free energies of their dissociation calculated at
the PBE/6-311+g(d,p) level (Table 1). Whereas the dissoci-
ation of 3 to 4 and a bromide ion in DMF is mildly exer-
gonic, the release of a methoxide ion from 19 is strongly
endergonic.
B3LYP/6-311+g(d) level, we were unable to locate any of
the three anions as an energy minimum in DMF. This result
is consistent with the known lability of 20 and 21 towards
dissociation; deprotonation of the corresponding haloforms
HCF2Br (20-H) and HCF2Cl (21-H) probably leads directly
to the CF2 carbene by a concerted α-elimination.[9a] If the
same behavior applies to our bromodiazirine system, car-
bene 4 could be formed from a 14-type adduct without the
intermediacy of 3. The similarity of 3 and fluorodihalome-
thides is further documented by their similar way of genera-
tion from α-trihetero-substituted esters. For example,
dichlorofluoroacetates are cleaved by alkoxides to dialkyl
carbonates (cf. the carbonate formation from esters 1 and
–
11) and the CFCl2 ion, which subsequently dissociates to
the CFCl carbene and a Cl– ion.[10]
If carbene 4 is indeed an intermediate in the alkoxide
cleavage of diazirines 1 and 11, it should be detectable by a
suitable reaction. Thus far, our attempts to trap 4 by cyclo-
addition to an alkene (2,3-dimethyl-2-butene, cyclohexene)
have only resulted in intractable mixtures of products. How-
ever, we believe we have found an alternative way of tracing
the fate of the diazirine ring carbon atom: The reaction of
1 with butoxide, when performed under strictly aprotic con-
ditions, affords carbonate 2 as the only organic product
(vide supra), whereas in the presence of n-butanol, we have
isolated a virtually unchanged amount of 2 (75%) along
with dibutoxymethane (22) in 35% yield (Scheme 4). Opti-
mized conditions involve slow addition of 1 to a 0.5 m solu-
tion of sodium butoxide (4 equiv.) and n-butanol (4 equiv.)
in DMF at 0 °C. Under these conditions, diazirine 11 also
afforded 2 in 82% yield and 22 in 31% yield. In this case
only traces of tert-butyl ester 12 (likely converted by trans-
esterification to 2) and virtually no n-butyl/tert-butyl scram-
bling in the acetal was observed.
Figure 1. Calculated bond lengths [Å] and angles [°] for 3, 4, 19,
and 20. [a] MP2 level (gas phase). [b] B3LYP level (gas phase).
[c] PBE level (DMF). The 6-311+G(d,p) basis set was used in all
calculations.
Figure 2. Natural bond orbital analysis of 3 and 19 at the PBE/6-
311+G(d,p) level in DMF. [a] Natural charges. [b] Bond orders in
terms of Wiberg bond indices.
Scheme 4. Cleavage of 1 and 11 by n-butoxide and n-butanol.
The diazirine–butoxide/butanol reaction is instantaneous
There is a striking resemblance between 3 and difluo- (nitrogen evolution) in DMF even at –40 °C, but also pro-
–
–
rohalomethide ions such as CF2Br (20) and CF2Cl (21), ceeds slowly in pure n-butanol. We detected no deuterium
known to be weakly bound complexes of the CF2 carbene incorporation in acetal 22 formed from diazirine 1 in [D7]-
with a bromide and chloride ion, respectively (Figure 1).[9] DMF. Based on these observations, the central carbon
Free energies of dissociation of 3, 20, and 21 are very sim- atoms of 2 and 22 do not originate from the solvent, and
ilar both in the gas phase and in DMF (Table 1). At the mass balance indicates they must come from different parts
6256
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 6254–6260

The Bromodiazirinyl Anion
of the starting material (with both 1 and 11, the sum of
The slow reaction of diazirine 1 with butoxide in pure n-
yields of 2 and 22 is over 100%). We propose that the cen- butanol did not reach full conversion after 24 h at 0 °C, and
tral carbon atom of 22 originates from the diazirine ring a mixture of 1, 2, 22, and dibutoxyacetate 29 in a ca.
(Scheme 5). The formation of 22 from 1 can be rationalized 2:2:1:1.5 ratio was isolated. In this case, nucleophilic acyl
by the exchange of bromide in anion 3 to butoxide via substitution (producing 2 and 22) and halogen exchange
strongly electrophilic (vide infra) carbene 4. The resulting appear to proceed concurrently, as the formation of 29 can
3-butoxydiazirinyl anion (23) is then protonated at the ring be explained by the insertion of stabilized (push–pull) car-
carbon atom to form the putative 3-butoxydiazirine (23-H), bene 28, resulting from denitrogenation of the substitution
which is expected to undergo nitrogen extrusion. The gener- product 27, into the solvent O–H bond (Scheme 6). As di-
ated butoxymethylene (24) inserts into the O–H bond of n- azirine 1 is thermally stable at 0 °C, it is unlikely that acet-
butanol, which completes the formation of 22. Alterna- ate 29 is formed via carbene 5 (Scheme 1). Importantly, we
tively, protonation of anion 3 could lead to 3-bromodiazir- have verified that an authentic sample of 29[11] does not
ine (3-H), further converted into acetal 22 through halogen cleave to 2 and 22 upon treatment with butoxide/butanol in
exchange to diazirine 23-H, or through denitrogenation to DMF or in pure alcohol. This observation corroborates the
bromomethylene (25), insertion of butanol to bromomethyl idea that carbonates 2 or 12 and acetal 22 do not result
ether 26, and bromide displacement.
from halogen exchange of diazirines 1 or 11, and had halo-
gen exchange proceeded in DMF to an appreciable extent,
its products should be observable (in fact, we have detected
traces of 29 by GC–MS in the crude products of reactions
of 1 or 11 in DMF).
Scheme 6. Proposed mechanism of formation of 29 from diazirine
1.
From an experimental point of view, 4 has been an unde-
scribed reactive intermediate, whereas its silicon analog, 1-
sila-2,3-diazacyclopropenylidene (30), was matrix-isolated a
decade ago (Figure 3).[12] Cyclopropenylidene (31), a carba
analogue of 4, is abundant in the interstellar medium and
was matrix-isolated in the 1980s.[13] Carbene 4 has appeared
in several theoretical studies,[14,15] but has been mostly con-
sidered in the broader context of the other CN2 species,
cyanonitrene (32) and diazomethylene (33), which were
both matrix-isolated in the 1960s.[16] Although 4 is calcu-
lated to be a ground-state singlet (ΔES–T ≈ –30 to –35 kcal/
mol), 32 and 33 have been found to be ground-state triplets
experimentally (ΔES–T = 23.3 and 19.5 kcal/mol, respec-
tively).[17] The order of calculated relative gas-phase ener-
Scheme 5. Possible mechanisms of cleavage of 1 by butoxide and
butanol.
Unlike the bromide–carbene complex 3, we expect the
butoxy anion 23 to be a regular diazirinyl carbanion[6–8] (as
shown above for the simplified methoxy species 19). Our
calculations indicate that the proton affinities of anions 3,
20, and 21 are similar and much lower than that of 23 or
19 (Table 2). As the calculated dissociation and protonation
energetics of 3, 20, and 21 are very similar, and the latter
two species are known to dissociate faster than they are
protonated,[9a] it is expected that 3 will exhibit the same
behavior. We therefore believe that the dissociation of anion
3 to carbene 4 followed by the addition of butoxide, which
results in anion 23 (from data in Table 1, ΔG = –28 kcal/
mol for 3 Ǟ 19 in DMF), is more likely than the initial
protonation of 3 (Scheme 5), and the intermediacy of car-
bene 4 is thus plausible. The possibility of direct formation
of 4 by a concerted α-elimination from the tetrahedral inter-
mediate upon the addition of alkoxide to the diazirine ester
group should also be considered (Scheme 3).
3
1
3
gies (in kcal/mol) is 32 (0), 4 (25.3), and 33 (28.9).[14e]
Figure 3. Species related to carbene 4.
Table 2. Relative proton affinities of anions 3 and 19–21.[a]
Process
ΔG
–
–
HCN2OMe (19-H) + CN2Br (3) Ǟ CN2OMe (19) + HCN2Br (3-H)
15.7
–1.7
1.7
–
HCF2Br (20-H) + 3 Ǟ CF2Br (20) + 3-H
HCF2Cl (21-H) + 3 Ǟ CF2Cl (21) + 3-H
–
[a] Values in kcal/mol at 298 K calculated at the PBE/6-311+G(d,p) level in DMF.
Eur. J. Org. Chem. 2011, 6254–6260
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6257

T. Martinu˚, S. Böhm, E. Hanzlová
FULL PAPER
Carbene 4 differs from classical N-heterocyclic carbenes
(NHCs) by its very high ionization potential (IP) – calcd.
IP = 12.3 eV, typical NHC calcd. IP ≈ 8 eV – and high
electron affinity (EA) – calcd. EA = 0.89 eV, typical NHC
calcd. EA ≈ –0.5 eV (Figure 4).[18] The high IP and EA of
4 are more features it shares with fluorocarbenes (for CF2,
calcd. IP = 12.3 eV, EA = 0.0 eV; for CFCl, calcd. IP =
11.0 eV, EA = 0.84 eV).
Experimental Section
Caution! Neat diazirines are potentially shock-sensitive and may
violently decompose without warning. All operations with neat diazir-
ines should be carried out behind a safety shield. However, we have
not experienced any violent decomposition of diazirines 1 and 11.
General Methods: Lithium bromide was dried at 250 °C under a
2ϫ10–2 Torr vacuum for 48 h and was used immediately. n-Buta-
nol, DMF, and acetonitrile were distilled from calcium hydride. All
other chemicals were used as purchased.
tert-Butyl 2-Ethylthio-2-iminoacetate (7): A mixture of tert-butyl
cyanoformate (6)[4] (9.0 g, 70.8 mmol) and ethanethiol (10.5 mL,
142 mmol) in a 100 mL round-bottomed flask was cooled to 0 °C.
After addition of DIEA (12.2 mL, 71.2 mmol), the reaction mix-
ture was stirred at 0 °C for 30 min and then at room temperature
for 17 h. Unreacted thiol was evaporated at 0 °C (water aspirator
vacuum) through a trap at –78 °C containing diethyl ether (50 mL),
and the residual amine was removed at room temperature
(3ϫ10–1 Torr). This procedure afforded 12.2 g (91%) of 7 as a yel-
low liquid. 1H NMR (300 MHz, CDCl3): δ = 10.38 (br. s, 1 H,
NH), 2.96 (q, J = 7.0 Hz, 2 H, CH2S), 1.52 (s, 9 H, tBuCH3), 1.30
(t, J = 7.5 Hz, 3 H, CH3) ppm. 13C NMR (75 MHz, CDCl3): δ
= 168.9 (C=O), 157.6 (C=N), 84.4 (tBuCO), 27.6 (tBuCH3), 24.1
(CH2S), 13.2 (CH3) ppm. HRMS (TOF ES+): calcd. for
C8H15NO2SNa [M + Na]+ 212.0721; found 212.0738.
Figure 4. IP and EA values of carbenes calculated at the
(U)B3LYP/6-311+G(d) level.[18]
(tert-Butoxycarbonyl)formamidine Hydrochloride (8): A mixture of
7 (11.0 g, 58.1 mmol), finely powdered ammonium chloride (3.11 g,
58.1 mmol), and methanol (30 mL) was heated to reflux in a
100 mL round-bottomed flask for 1 h, and the escaping ethanethiol
was condensed in a –78 °C trap containing diethyl ether (50 mL).
The reaction mixture was cooled to room temperature, and meth-
anol was removed under reduced pressure. The residue was thor-
oughly triturated with diethyl ether and the resulting solid dried in
vacuo. This procedure afforded 7.20 g of crude 8 with 83.3% (w/
w) purity by 1H NMR spectroscopy, the remaining 16.7% being
the unreacted ammonium chloride (the total yield of 8 was 57 %).
The crude product was used directly in the next step (preparation
of N-chloroamidine 9). A purified sample of 8 (m.p. Ͼ250 °C) was
obtained by several cycles of fractional dissolution in tert-butyl
By nature, carbene 4 is expected to be an elusive species.
We are currently searching for its direct trapping products,
which are spirocyclic alkene adducts (34) and transition
metal complexes (35). The major issue in finding such un-
precedented species is the inherent lability of their diazirine
ring, complicating their isolation and characterization. The
viability of searching for 35 is supported by a recent compu-
tational study indicating that the binding energy of 4 to
some transition metals (AuCl test particle) is comparable to
that of phosphanes (but smaller than that of NHCs).[15]
1
alcohol followed by concentration of the mother liquor. H NMR
(300 MHz, [D6]DMSO): δ = 9.77 (br. s, 2 H, NH2), 9.66 (br. s, 2
H, NH2), 1.50 (s, 9 H, tBuCH3) ppm. 13C NMR (75 MHz, [D6]-
DMSO): δ = 155.53 (C=O), 155.45 (C=N), 86.8 (tBuCO), 27.3
(tBuCH3) ppm. HRMS (TOF ES+): calcd. for C6H13N2O2 [M –
Cl]+ 145.0977; found 145.0985.
Conclusions
We have found experimental and computational evidence
for the competition between bromide exchange and nucleo-
philic acyl displacement of the bromodiazirinyl anion (3) in
the reactions of n- or tert-butyl 3-bromodiazirine-3-carb-
oxylates (1 or 11) with sodium n-butoxide. Although in n-
butanol both reactions proceed concurrently, the latter pre-
dominates in DMF. We interpret anion 3 as a weakly bound
(tert-Butoxycarbonyl)-N-chloroformamidine (9): To a vigorously
stirred suspension of crude 8 [7.00 g, corresponding to 5.83 g
(32.3 mmol) of pure 8 and 1.17 g (21.9 mmol) of ammonium chlo-
ride] in water (20 mL) and diethyl ether (250 mL) in a 500 mL
round-bottomed flask cooled below 5 °C was added a 5.04% (iodo-
metric) solution of commercial bleach (74.4 g, 50.4 mmol of so-
complex of the cyclic carbene c-CN2 (4) with a bromide dium hypochlorite) dropwise maintaining a temperature below
10 °C. When the addition was completed, the mixture was stirred
for another 5 min. The two layers were separated, and the aqueous
phase was extracted with diethyl ether (2ϫ300 mL). The combined
organic phases were dried with MgSO4 and concentrated under
reduced pressure below room temperature to afford 4.52 g (78%
yield) of 9 as a white solid. M.p. 70.4–71.0 °C (pentane/diethyl
ether). 1H NMR (300 MHz, CDCl3): δ = 6.01 (br. s, 1 H, NH),
5.75 (br. s, 1 H, NH), 1.56 (s, 9 H, tBuCH3) ppm. 13C NMR
(75 MHz, CDCl3): δ = 158.0 (C=O), 155.7 (C=N), 85.3 (tBuCO),
27.8 (tBuCH3) ppm. HRMS (TOF ES+): calcd. for
C6H11ClN2O2Na [M + Na]+ 201.0407; found 201.0410.
ion. The intermediacy of 4 is supported by the isolation of
substantial amounts of dibutoxymethane (22) when the n-
butoxide substitution reactions are carried out in the pres-
ence of n-butanol. We propose a mechanism in which 22 is
formed from carbene 4 by a sequence of O–H insertions
and denitrogenation. The ground-state singlet 4 (diazirinyli-
dene) is the last experimentally undescribed member of the
CN2 family of reactive intermediates, which differs from
classical NHCs by its high ionization potential and electron
affinity.
6258
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 6254–6260

The Bromodiazirinyl Anion
(tert-Butoxycarbonyl)-N,N,NЈ-trichloroformamidine (10): A solu-
tion of 9 (4.00 g, 22.4 mmol) in tBuOCl (36 g, 332 mmol) was
stirred at room temperature in the dark for 24 h. Upon concentra-
tion under reduced pressure at room temperature, 5.42 g (98%
yield) of 10 was obtained as a yellow liquid with a strong chlorine-
like odor, which solidified below room temperature. 1H NMR
oxide/butanol results in the formation of dimethylamine, causing
lower reaction yields). Nitrogen evolution and frothing occurred
upon each addition of starting material solution. After the addition
was complete, the reaction mixture was stirred at room temperature
for 20 min, and was worked up as described above. The resulting
yellow-brown oil was identified by comparison to authentic sam-
(300 MHz, CDCl3): δ = 1.59 (s, 9 H, tBuCH3) ppm. 13C NMR ples to be a mixture of 2 and dibutoxymethane (22),[19c] 90–94%
(75 MHz, CDCl3): δ = 166.5 (C=O), 155.8 (C=N), 87.7 (tBuCO),
27.8 (tBuCH3) ppm. HRMS (EI+): calcd. for C6H9Cl3N2O2 [M]+
245.9730; found 245.9719.
pure by 1H NMR spectroscopy. Yields were determined by 1H
NMR spectroscopy with naphthalene as an internal standard.
Crude product from reaction with 1: 248 mg, of which 163 mg
(75% yield) was 2 and 70 mg (35% yield) was 22; crude product
from reaction with 11: 268 mg, of which 179 mg (82% yield) was 2
tert-Butyl 3-Bromodiazirine-3-carboxylate (11): To a vigorously
stirred solution of dry lithium bromide (26.0 g, 299 mmol) in dry
acetonitrile (450 mL) in a 1 L round-bottomed flask in the dark
was added 10 (5.00 g, 20.2 mmol) in acetonitrile (40 mL) dropwise
over 15 min. When the addition was completed, the mixture was
stirred for another 10 min and then extracted into pentane
(6ϫ500 mL). The combined pentane layers were concentrated un-
der reduced pressure below 10 °C affording 2.81 g (63%) of a pale-
yellow liquid, essentially pure 11 by 1H NMR spectroscopy. For
UV/Vis spectroscopic purposes the product was further purified by
fractional vacuum transfer (4ϫ10–2 Torr) at room temperature into
a U-trap cooled to –196 °C. 1H NMR (300 MHz, CDCl3): δ = 1.46
(s, 9 H, tBuCH3) ppm. 13C NMR (75 MHz, CDCl3): δ = 162.2
(C=O), 85.9 (tBuCO), 31.3 (CN2), 27.6 (tBuCH3) ppm. IR (film,
1
and 62 mg was 22 (31% yield). 22: H NMR (300 MHz, CDCl3):
δ = 4.65 (s, 2 H, OCH2O), 3.52 (t, J = 6.6 Hz, 4 H, CH2O), 1.60–
1.51 (m, 4 H, CH2), 1.43–1.31 (m, 4 H, CH2), 0.92 (t, J = 7.3 Hz,
3 H, CH3) ppm. 13C NMR (75 MHz, CDCl3): δ = 95.2 (OCH2O),
67.5 (CH2O), 31.8 (CH2), 19.4 (CH2), 13.8 (CH3) ppm.
Reaction of 1 with Sodium Butoxide in n-Butanol: Sodium metal
(57 mg, 2.50 mmol) was dissolved in n-butanol (4.0 mL) in a 10 mL
round-bottomed flask (magnetically stirred, under argon). After
the solution had been cooled to 0 °C, 1[3] (276 mg, 1.25 mmol) in
butanol (0.5 mL) was added dropwise with vigorous stirring over
5 min. After stirring at 0 °C for 24 h, the reaction mixture was di-
luted with water (70 mL), and extracted with pentane (5ϫ100 mL).
The combined extracts were washed with water (3ϫ250 mL), dried
with magnesium sulfate, and concentrated under reduced pressure
at 0 °C. The resulting yellow oil (96 mg) was identified by compari-
son to authentic samples to be a mixture of unreacted 1, 2, 22, and
butyl dibutoxyacetate (29)[11] in a ca. 2:2:1:1.5 ratio. 1: 1H NMR
(300 MHz, CDCl3): δ = 4.22 (t, J = 6.6 Hz, 2 H, CH2O), 1.65 (m,
2 H, CH2), 1.38 (m, 2 H, CH2), 0.94 (t, J = 7.4 Hz, 3 H, CH3)
ppm. 13C NMR (75 MHz, CDCl3): δ = 163.6 (C=O), 68.2 (CH2O),
30.3 (CH2), 30.0 (CN2), 18.8 (CH2), 13.5 (CH3) ppm. 29: 1H NMR
(300 MHz, CDCl3): δ = 4.87 (s, 1 H, OCHO), 4.18 (t, J = 6.7 Hz,
2 H, CH2O2C), 3.65–3.51 (m, 4 H, CH2O), 1.70–1.55 (m, 6 H, CH2,
CH2), 1.45–1.32 (m, 6 H, CH2, CH2), 0.93 (t, J = 7.3 Hz, 3 H,
CH3), 0.91 (t, J = 7.3 Hz, 6 H, CH3) ppm. 13C NMR (75 MHz,
CDCl3): δ = 167.7 (C=O), 97.8 (OCHO), 66.6 (CH2O), 65.1
(CH2O2C), 31.6 (CH2), 30.5 (CH2), 19.2 (CH2), 19.0 (CH2), 13.8
(CH3), 13.6 (CH3) ppm.
NaCl):
ν
= 1754, 1735, 1605 cm–1. UV (pentane): λmax (ε,
˜
Lmol–1 cm–1) = 308 (ca. 16), 323 (ca. 29), 335 (ca. 32) nm.
Reaction of 11 with Sodium Butoxide in DMF: Sodium metal
(115 mg, 5.00 mmol) was dissolved in dry n-butanol (10 mL) in a
50 mL round-bottomed flask (magnetically stirred, under argon).
Excess butanol was evaporated below room temperature, and the
resulting solid sodium butoxide was dried at 45 °C for 12 h (both
operations were performed under a 2ϫ10–2 Torr vacuum). The so-
dium butoxide was quickly dissolved in dry DMF (9.0 mL), and the
solution was cooled to –15 °C. Diazirine 11 (276 mg, 1.25 mmol)
in DMF (0.5 mL) was added dropwise with vigorous stirring over
20 min. Nitrogen evolution and frothing occurred upon each ad-
dition of starting material solution. Immediately after the addition
was complete, the reaction mixture was diluted with water (50 mL)
and extracted with pentane (4ϫ50 mL). The combined extracts
were washed with water (2ϫ100 mL), dried with magnesium sul-
fate, and concentrated under reduced pressure at 0 °C. The re-
sulting yellow-brown oil (182 mg) was identified by comparison to
authentic samples to be a 95% pure mixture of di-n-butyl carbon-
ate (2)[19a] (114 mg, 52% yield) and n-butyl tert-butyl carbonate
(12)[19b] (58 mg, 27% yield). The yields were determined by 1H
NMR spectroscopy with naphthalene as an internal standard. 2:
1H NMR (300 MHz, CDCl3): δ = 4.10 (t, J = 6.7 Hz, 4 H, CH2O),
1.63 (quint, J = 6.7 Hz, 4 H, CH2), 1.38 (sext, J = 7.3 Hz, 4 H,
CH2), 0.91 (t, J = 7.3 Hz, 3 H, CH3) ppm. 13C NMR (75 MHz,
CDCl3): δ = 155.4 (C=O), 67.6 (CH2O), 30.7 (CH2), 18.9 (CH2),
Supporting Information (see footnote on the first page of this arti-
cle): NMR spectra for 1, 2, 7–12, 22, and 29; IR and UV spectra
for 11; details of DFT calculations.
Acknowledgments
This work was funded by the Research Centre of the Ministry of
Education, Youth, and Sports of the Czech Republic (LC06070).
We would like to thank Ms. Kveˇta Bártová for technical assistance.
1
13.6 (CH3) ppm. 12: H NMR (300 MHz, CDCl3): δ = 4.03 (t, J
[1] R. A. Moss, Acc. Chem. Res. 2006, 39, 267–272.
= 6.7 Hz, 2 H, CH2O), 1.67–1.57 (m, 2 H, CH2), 1.46 (s, 3 H,
tBuCH3), 1.45–1.32 (m, 2 H, CH2), 0.92 (t, J = 7.3 Hz, 3 H, CH3)
ppm. 13C NMR (75 MHz, CDCl3): δ = 153.7 (C=O), 81.7 (tBuCO),
66.9 (CH2O), 30.7 (CH2), 27.7 (tBuCH3), 18.9 (CH2), 13.6 (CH3)
ppm.
[2] a) R. A. Moss, G. Chu, R. R. Sauers, J. Am. Chem. Soc. 2005,
127, 2408–2409; b) C. J. Shaffer, B. J. Esselman, R. J. McMa-
hon, J. F. Stanton, R. C. Woods, J. Org. Chem. 2010, 75, 1815–
1821; c) R. A. Moss, R. R. Sauers, Tetrahedron Lett. 2010, 51,
3266–3268.
[3] T. Martinu, W. P. Dailey, J. Org. Chem. 2004, 69, 7359–7362.
[4] L. A. Carpino, J. Am. Chem. Soc. 1960, 82, 2725–2727.
[5] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B.
Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li,
H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Son-
nenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hase-
Reaction of 1 or 11 with Sodium Butoxide and n-Butanol in DMF:
Sodium butoxide (5.00 mmol), prepared as described above, was
quickly dissolved in dry DMF (9.0 mL) and n-butanol (0.458 mL,
5.00 mmol), and the solution was cooled to 0 °C. Diazirine 1[3] or
11 (276 mg, 1.25 mmol) in DMF (0.5 mL) was added dropwise with
vigorous stirring over 20 min (prolonged exposure of DMF to but-
Eur. J. Org. Chem. 2011, 6254–6260
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6259

T. Martinu˚, S. Böhm, E. Hanzlová
FULL PAPER
gawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai,
T. Vreven, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro, M.
Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Starov-
erov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell,
J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M.
Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Ad-
amo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev,
A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Mar-
tin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B.
Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian 09W,
version 7.0, Gaussian, Inc., Wallingford, CT, 2009.
[13] a) R. I. Kaiser, Chem. Rev. 2002, 102, 1309–1358; b) H. P. Re-
isenauer, G. Maier, A. Riemann, R. W. Hoffmann, Angew.
Chem. Int. Ed. Engl. 1984, 23, 641.
[14] a) W. A. Lathan, L. Radom, P. C. Hariharan, W. J. Hehre, J. A.
Pople, Top. Curr. Chem. 1973, 40, 1–45; b) J. M. L. Martin,
P. R. Taylor, J. P. François, R. Gijbels, Chem. Phys. Lett. 1994,
226, 475–483; c) B. S. Jursic, THEOCHEM 1999, 491, 33–40;
d) J. Armstrong, L. Degoricija, A. Hildebrand, J. Koehne, P. E.
Fleming, Chem. Phys. Lett. 2000, 332, 591–596; e) L. V. Mos-
kaleva, M. C. Lin, J. Phys. Chem. A 2001, 105, 4156–4163; f)
M. Z. Kassaee, M. Ghambarian, S. M. Musavi, F. A. Shakib,
M. R. Momeni, J. Phys. Org. Chem. 2009, 22, 919–924; g) S.
Gronert, J. R. Keeffe, R. A. More O’Ferrall, J. Am. Chem. Soc.
2011, 133, 3381–3389.
[6] M. S. Gordon, S. R. Kass, J. Phys. Chem. 1995, 99, 6548–6550.
[7] a) R. L. Kroeker, S. R. Kass, J. Am. Chem. Soc. 1990, 112,
9024–9025; b) R. A. Seburg, B. T. Hill, R. A. Jesinger, R. R.
Squires, J. Am. Chem. Soc. 1999, 121, 6310–6311.
[8] a) X. Creary, A. F. Sky, G. Phillips, D. E. Alonso, J. Am. Chem.
Soc. 1993, 115, 7584–7592; b) R. A. Moss, S. Xue, W. G. Liu,
J. Am. Chem. Soc. 1994, 116, 10821–10822.
[15] P. Pyykkö, N. Runeberg, Chem. Asian J. 2006, 1, 623–628.
[16] D. E. Milligan, M. E. Jacox, J. Chem. Phys. 1966, 44, 2850–
2856.
[17] a) T. R. Taylor, R. T. Bise, K. R. Asmis, D. M. Neumark,
Chem. Phys. Lett. 1999, 301, 413–416; b) E. P. Clifford, P. G.
Wenthold, W. C. Lineberger, G. A. Petterson, K. M. Broadus,
S. R. Kass, S. Kato, C. H. DePuy, V. M. Bierbaum, G. B. El-
lison, J. Phys. Chem. A 1998, 102, 7100–7112.
[9] a) J. Hine, P. B. Langford, J. Am. Chem. Soc. 1957, 79, 5497–
5500; b) J. A. Paulino, R. R. Squires, J. Am. Chem. Soc. 1991,
ˇ
113, 1845–1847; c) J. Kvícˇala, J. Stambaský, S. Böhm, O. Pal-
eta, J. Fluorine Chem. 2002, 113, 147–154; d) R. A. Moss, M.
Zhang, K. Krogh-Jespersen, Org. Lett. 2009, 11, 5702–5704.
[10] a) T. Ando, H. Yamanaka, S. Terabe, A. Horike, W. Funasaka,
Tetrahedron Lett. 1967, 12, 1123–1126; b) D. L. S. Brahms,
W. P. Dailey, Chem. Rev. 1996, 96, 1585–1632.
[11] D. Papillon-Jegou, B. Bariou, N. Soyer, M. Kerfanto, Bull. Soc.
Chim. Fr. 1977, 977–982.
[18] W. Sander, C. Kötting, R. Hübert, J. Phys. Org. Chem. 2000,
13, 561–568.
[19] a) A. I. Vogel, W. T. Cresswell, G. H. Jeffrey, J. Leicester, J.
Chem. Soc. 1952, 514–549; b) E. H. White, K. W. Field, J. Am.
Chem. Soc. 1975, 97, 2148–2153; c) C. L. Fisher, S. D. Kahn,
W. J. Hehre, M. C. Caserio, J. Am. Chem. Soc. 1989, 111, 7379–
7387.
[12] G. Maier, H. P. Reisenauer, J. Glatthaar, Organometallics 2000,
19, 4775–4783.
Received: April 15, 2011
Published Online: September 2, 2011
6260
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 6254–6260