Iridium-catalyzed condensation of amines and vicinal diols to substituted piperazines

Article abstract of DOI:10.1002/ejoc.201201099

A straightforward procedure is described for the synthesis of piperazines from amines and 1,2-diols. The heterocyclization is catalyzed by [Cp*IrCl₂]₂ and sodium hydrogen carbonate and can be achieved with either toluene or water as solvent. The transformation does not require any stoichiometric additives and only produces water as the byproduct. The reaction can be performed between a 1,2-diamine and a 1,2-diol or by a double condensation between a primary alkylamine and a 1,2-diol. At least one substituent is required on the piperazine ring to achieve the cyclization in good yield. The mechanism is believed to involve dehydrogenation of the 1,2-diol to the α-hydroxy aldehyde, which condenses with the amine to form the α-hydroxy imine. The latter rearranges to the corresponding α-amino carbonyl compound, which then reacts with another amine followed by reduction of the resulting imine. Piperazines are prepared by [Cp*IrCl ₂]₂-catalyzed heterocyclization of 1,2-diols with either 1,2-diamines or primary alkylamines. The reaction is performed in toluene or water and requires no stoichiometric additive. The key step in the mechanism is believed to be the isomerization of an α-hydroxy imine to the corresponding α-amino carbonyl compound. Copyright

Full text of DOI:10.1002/ejoc.201201099

FULL PAPER
DOI: 10.1002/ejoc.201201099
Iridium-Catalyzed Condensation of Amines and Vicinal Diols to Substituted
Piperazines
Linda L. R. Lorentz-Petersen,[a] Lars Ulrik Nordstrøm,[a] and Robert Madsen*[a]
Keywords: Alcohols / Amination / Cyclization / Homogeneous catalysis / Nitrogen heterocycles
A straightforward procedure is described for the synthesis of
piperazines from amines and 1,2-diols. The heterocyclization
is catalyzed by [Cp*IrCl2]2 and sodium hydrogen carbonate
and can be achieved with either toluene or water as solvent.
The transformation does not require any stoichiometric addi-
tives and only produces water as the byproduct. The reaction
can be performed between a 1,2-diamine and a 1,2-diol or
by a double condensation between a primary alkylamine and
a 1,2-diol. At least one substituent is required on the pipera-
zine ring to achieve the cyclization in good yield. The mecha-
nism is believed to involve dehydrogenation of the 1,2-diol
to the α-hydroxy aldehyde, which condenses with the amine
to form the α-hydroxy imine. The latter rearranges to the cor-
responding α-amino carbonyl compound, which then reacts
with another amine followed by reduction of the resulting
imine.
Introduction
Over the past decade, transition-metal-catalyzed alkyl-
ation of amines with alcohols has received significant atten-
tion as an environmentally benign method for carbon–
nitrogen bond formation.[1] The reaction does not require
any stoichiometric additives and produces only water as a
byproduct. The most effective metal catalysts are based on
iridium and ruthenium complexes such as [Cp*IrCl2]2,
[Ir(cod)Cl]2, [Ru(PPh3)3Cl2], [Ru(p-cymene)Cl2]2, and
[Ru3(CO)12].[1] Very recently, the mechanism for the
[Cp*IrCl2]2-catalyzed alkylation was investigated in detail
by a combination of experimental and theoretical meth-
ods.[2] It was shown that the reaction proceeds by dehydro-
genation of the alcohol to the aldehyde followed by hemia-
minal formation, dehydration to the imine, and reduction to
the product amine (Scheme 1). Most importantly, the entire
catalytic cycle takes place with all the intermediates bound
to the iridium catalyst.
The alkylation reaction has been applied in the synthesis
of pyrrolidines and piperidines by cyclization of α,ω-diols
with either primary amines[3] or ammonia.[4] Several phar-
macologically important heterocycles have also been pre-
pared from arylamines and diols including indoles,[5] quin-
olines,[6] and quinoxalines.[7] We were interested in using the
alkylation reaction for the synthesis of piperazines from
amines and 1,2-diols.[8] The piperazine moiety is an impor-
tant pharmacophore that is found in about 6% of all orally
Scheme 1. Cp*Ir-catalyzed alkylation of amines with alcohols.
administered drugs.[9] The synthesis of piperazines is usually
achieved by reduction of (di)oxopiperazines,[10] α-lithiation
of Boc-protected piperazines,[11] or by different cyclization
reactions such as dialkylation of amines with bis(2-chloro-
ethyl)amine,[12] reductive cyclization of diimines,[13] alkene
carboamination,[14] and the Mitsunobu reaction.[15] How-
ever, in virtually all these cases, various amounts of stoi-
chiometric reagents or additives are required. Accordingly,
a cyclocondensation approach from amines and 1,2-diols
constitutes a more atom-economical strategy for assembling
the piperazine skeleton. Previously, this reaction has been
achieved with simple 1,2-diamines and 1,2-diols in tetra-
hydrofuran (THF) solution with [Ru3(CO)12/PBu3] as the
catalyst, but these conditions require a temperature of
220 °C to furnish the piperazine moiety in good yield.[16]
Herein, we report a full account of our synthesis of sub-
stituted piperazines from amines and 1,2-diols. We outline
the scope and limitations of the reaction and provide sup-
port for a plausible mechanistic scenario.
[a] Department of Chemistry, Technical University of Denmark,
2800 Kgs. Lyngby, Denmark
Fax: +45-4593-3968
E-mail: rm@kemi.dtu.dk
Homepage: http://www.organic.kemi.dtu.dk
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201201099.
6752
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 6752–6759

Iridium-Catalyzed Synthesis of Substituted Piperazines
In addition to [Cp*IrCl2]2, several ruthenium catalysts
were also investigated in the heterocyclization. However,
and somewhat surprisingly, [Ru(PPh3)3Cl2] (2%) and com-
binations of RuCl3 (2%) and PPh3 (or xantphos) com-
pletely failed to give any significant conversion of the di-
amine and the diol even in mesitylene solution at 170 °C.
Because [Ru(PPh3)3Cl2] is known to catalyze the alkylation
of cyclohexylamine with primary alcohols,[19] the poor reac-
tivity in our case may be attributed to the coordination of
the diamine to the ruthenium atom and consequent inhibi-
tion of the alkylation reaction. Based on these results, the
[Cp*IrCl2]2 complex in the presence of sodium hydrogen
carbonate was selected for general use.
With this catalytic system in hand, the substrate scope of
the reaction could now be investigated with other amines
and 1,2-diols. Piperazine formation was studied in a closed
vial in both toluene and water to explore a possible differ-
ence between these highly different solvents. First, several
other 1,2-diamines were examined in the reaction with eth-
ylene glycol (Table 2). Unexpectedly, the reaction with eth-
ylenediamine failed to furnish more than trace amounts of
the unsubstituted piperazine, regardless of the reaction con-
ditions (Table 2, Entries 1 and 2). A temperature of 140 °C
was required for full conversion, but the outcome appeared
Results and Discussion
For the initial experiments (Ϯ)-trans-1,2-diaminocyclo-
hexane and ethylene glycol were employed as the test sub-
strates in equimolar amounts, and the reactions were per-
formed in a closed vial overnight (Table 1). [Cp*IrCl2]2 was
selected as the precatalyst, because this complex has gen-
erally been one of the most effective for mediating the
C–N bond formation from alcohols and amines.[1] Gratify-
ingly, the transformation proceeded well in toluene to afford
the desired piperazine in 78% isolated yield (Table 1, En-
try 1). Several weak bases have been shown to accelerate
the C–N bond formation in the presence of [Cp*IrCl2]2,
and sodium hydrogen carbonate has been the most effec-
tive.[17] This was also observed in the piperazine formation,
for which the yield improved to 94% with sodium hydrogen
carbonate as an additive (Table 1, Entry 2). The same yield
was obtained with triethylamine, whereas the use of sodium
carbonate or acetate both gave lower yields (Table 1, En-
tries 3–5). Other solvents were also investigated, but neither
dioxane nor heptane afforded comparable yields (Table 1,
Entries 6 and 7). Because water is released during the reac-
tion, we investigated whether the transformation could also
be achieved with water as the solvent. To our delight, the
reaction in water gave a yield similar to that in toluene
(Table 1, Entries 2 and 8).[18] Again, sodium hydrogen car-
bonate was necessary as an additive, and a significantly
slower reaction was observed without the additive or with
sodium carbonate (Table 1, Entries 9 and 10). Interestingly,
the reaction could also be accelerated by an acid, as demon-
strated when trifluoroacetic acid was employed as an addi-
tive (Table 1, Entry 11). The reaction temperature and time
were briefly examined, but attempts to decrease the tem-
perature or shorten the reaction time both resulted in in-
complete conversion (Table 1, Entries 12–14).
Table 2. Piperazine synthesis with ethylene glycol.
Table 1. Optimizing the piperazine formation.
Entry
Solvent
Temperature
Additive
Yield[a]
1
2
3
4
5
6
7
8
9
10
11
12
13[e]
14[e]
toluene
toluene
toluene
toluene
toluene
dioxane
heptane
H2O
H2O
H2O
H2O
toluene
toluene
H2O
110 °C
110 °C
110 °C
110 °C
110 °C
100 °C
98 °C
100 °C
100 °C
100 °C
110 °C
90 °C
none
NaHCO3
Et3N
Na2CO3
NaOAc
NaHCO3
NaHCO3
NaHCO3
none
78%
94%
94%
63%
53%
86%
13%
96%
41%[b]
24%[c]
98%
64%
32%
40%
Na2CO3
TFA[d]
NaHCO3
NaHCO3
NaHCO3
110 °C
100 °C
[a] Isolated yield. [b] At 160 °C. [c] Reaction time 64 h. [d] At
110 °C. [e] At 100 °C. [f] With 10% of TFA instead of NaHCO3.
[g] With 1.5 equiv. of ethylene glycol. [h] At 180 °C. [i] 30% of 2-
(hydroxymethyl)benzimidazole was also isolated.
[a] Isolated yield. [b] Diamine (59%) was recovered. [c] Diamine
(74%) was recovered. [d] Performed with TFA (10%). [e] Reaction
time 5 h.
Eur. J. Org. Chem. 2012, 6752–6759
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6753

L. L. R. Lorentz-Petersen, L. U. Nordstrøm, R. Madsen
FULL PAPER
to be oligomerization of the substrates rather than cycliza- same rate. This may not be completely unforeseen, because
tion. Interestingly, simple addition of a methyl group, as in the synthesis of 2-substituted benzimidazoles from 1,2-
(Ϯ)-1,2-diaminopropane, changed the direction of the reac- phenylenediamine and primary alcohols has previously
tion, and the target piperazine was now obtained in 69% been achieved with ruthenium catalysts in toluene solution
yield in either toluene or water (Table 2, Entries 3 and 4). (with [Ru(PPh3)3Cl2] at 200 °C[20] or with [Ru(PPh3)3(CO)-
Similar yields were obtained with N-benzyl-substituted eth- H2], xantphos, crotononitrile at 110 °C[21]).
ylenediamines, although the monobenzylated substrate re-
In addition to ethylene glycol, other 1,2-diols were also
quired a higher temperature of 160 °C for full conversion examined as substrates for the cyclocondensation with 1,2-
(Table 2, Entries 5–7). Optically pure (–)-1,2-diphenylethy- diamines (Table 3). 1,2-Propylene glycol displayed a reactiv-
lenediamine was transformed into the corresponding piper- ity very similar to that of ethylene glycol. (Ϯ)-trans-1,2-Di-
azine with no sign of racemization (Table 2, Entries 8–10). aminocyclohexane was transformed into the corresponding
The conversion was rather slow with sodium hydrogen car- piperazine in good yield at a temperature of 100–110 °C
bonate and a temperature of 100–110 °C, but a faster reac- (Table 3, Entries 1 and 2). A new stereocenter is formed
tion was observed when the additive was changed to tri- and, in this case, the solvent had a notable impact, because
fluoroacetic acid (Table 2, Entry 10). 1,2-Phenylenediamine, the most stable isomer with an equatorial methyl group was
on the other hand, was a poor substrate regardless of the formed to a much larger extent in water than in toluene.
reaction conditions. A temperature of 180 °C was necessary Ethylenediamine and 1,2-propylenediamine were also con-
to achieve full conversion of the diol, although some of the verted into the corresponding piperazines in good yields
diamine still remained. Eventually, the transformation was (Table 3, Entries 3–6). In the latter case, only the two most
carried out with 1.5 equiv. of ethylene glycol to afford stable isomers with equatorial methyl groups were formed.
1,2,3,4-tetrahydroquinoxaline in 35% yield (Table 2, En- Similar results were obtained with phenylethylene glycol, al-
try 11). No improvement was observed when the reaction though a higher temperature was required for full conver-
was performed with other additives or under neat condi- sion (Table 3, Entries 7–9). 2,3-Butylene glycol could also
tions. The main side product was 2-(hydroxymethyl)benz- be employed as a substrate, as shown in the reaction with
imidazole, which was isolated in 30% yield. Thus, in this (Ϯ)-trans-1,2-diaminocyclohexane (Table 3, Entries 10 and
case, two competing cyclizations occur at approximately the 11). However, in this case, a temperature of 140 °C was re-
Table 3. Piperazine synthesis with other glycols.
[a] Isolated yield. [b] Determined by NMR spectroscopy.
6754
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 6752–6759

Iridium-Catalyzed Synthesis of Substituted Piperazines
quired for full conversion, which is higher than that for the trisubstituted piperazine was isolated in 63% yield, whereas
same reaction with ethylene and propylene glycol (Tables 1 an improved selectivity was observed with phenylethylene
and 3, Entries 1 and 2).
glycol, for which the product was obtained in 90% yield.
In all, the substrate screening summarized in Tables 1, 2,
and 3 illustrates that a number of 1,2-diamines and 1,2-
diols can be employed in the cyclization. The most reactive
diamine is 1,2-diaminocyclohexane, whereas ethylene and
propylene glycol are the most efficient diols. N-Substituted
diamines, i.e., secondary amines, are less reactive substrates,
and the same is observed with more substituted diols. Sub-
stituents, however, are required for the cyclization to pro-
ceed, because the parent piperazine could not be prepared
from ethylenediamine and ethylene glycol. When a substitu-
ent gives rise to a new stereocenter in the product, the ther-
modynamically most stable isomer is usually preferred.
Interestingly, no notable difference in reactivity was ob-
served when the reactions were conducted in toluene or
water.
Piperazine formation can also be envisioned from sub-
strates other than 1,2-diamines and 1,2-diols. First, it was
investigated whether a single primary amine could be used
instead of a 1,2-diamine. Accordingly, the condensation of
benzylamine with ethylene glycol was studied under dif-
ferent conditions in a closed vial (Scheme 2). In toluene at
140 °C the reaction only afforded a moderate yield of 1,4-
dibenzylpiperazine (45%). However, when the transforma-
tion was repeated at 160 °C in the absence of a solvent the
yield increased to 94% (Scheme 2). The high temperature
was necessary under the neat conditions, because the trans-
formation at 110 °C only afforded 30% yield after 18 h.
Aniline was also examined as a substrate under the same
conditions at 160 °C, but only trace amounts of 1,4-di-
phenylpiperazine were obtained in this case, presumably
due to the lower nucleophilicity compared to benzylamine.
Two other diols were also studied in the reaction with ben-
zylamine. However, with both phenylethylene glycol and cy-
clohexane-1,2-diol the reaction stopped at the amino
alcohol stage, and no dimerization to the corresponding
symmetric piperazines were observed. This illustrates once
more that the cyclization is hampered by additional substit-
uents.
Scheme 3. Synthesis of 2-substituted 1,4-dibenzylpiperazine.
Similar reactions may be possible with ammonia as the
amine component. N-Alkylation of ammonium salts with
primary and secondary alcohols has previously been carried
out in the absence of solvent with [Cp*IrCl2]2 as the precat-
alyst.[22] Under these conditions, however, we observed no
piperazine formation from ethylene glycol and ammonium
tetrafluoroborate. This may not be surprising, because the
lack of a substituent may hinder the synthesis of the unsub-
stituted piperazine. The transformation was therefore re-
peated with both 2,3-butylene glycol and cyclohexane-1,2-
diol. In both cases, a mixture of three products was formed,
which – according to GC–MS analysis – were believed to
be the corresponding piperazine, pyrazine, and morpholine.
The reaction with cyclohexane-1,2-diol was further opti-
mized, but it was not possible to obtain the piperazine as
the major product. Instead, the morpholine adduct could
be isolated in 63% yield when the condensation was carried
out with trifluoroacetic acid in mesitylene at 180 °C
(Scheme 4). Presumably, this transformation proceeds by a
double alkylation of ammonia with the diol to form bis(2-
hydroxycyclohexyl)amine, which then cyclizes to the
morpholine. The synthesis of morpholine derivatives from
di(ethanol)amines has previously been described in the
presence of RuCl3 and PPh3.[23]
Scheme 4. Synthesis of dodecahydro-1H-phenoxazine.
Ethanolamine is also a potential substrate that may form
a piperazine with the iridium catalyst. It was quickly discov-
ered that this reaction did not afford the parent piperazine,
but instead two substituted analogues were generated. After
some optimization, 1-(2-hydroxyethyl)piperazine and 1,4-
bis(2-hydroxyethyl)piperazine were isolated in 30 and 27%
yield, respectively (Scheme 5). The transformation required
Scheme 2. Synthesis of 1,4-dibenzylpiperazine.
It may, however, be possible to use two different diols in a rather long reaction time for full conversion, and it was
the reaction with benzylamine if one is ethylene glycol. To not possible to alter the ratio between the two products.
avoid 1,4-dibenzylpiperazine formation the substituted diol The exact pathway leading to the substituted piperazines
should first be allowed to react with 2 equiv. of benzyla- could not be identified, but 2-[(2-aminoethyl)amino]ethanol
mine, and – upon completion of this reaction – ethylene might be a potential intermediate. To probe this question
glycol should be added to the mixture (Scheme 3). It was the diamino alcohol was also subjected to the iridium cata-
found that this one-pot protocol could be achieved at lyst (Scheme 6). Interestingly, this afforded exclusively the
140 °C in aqueous solution. With 1,2-propylene glycol, the unsubstituted piperazine, which was isolated in 79% yield
Eur. J. Org. Chem. 2012, 6752–6759
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6755

L. L. R. Lorentz-Petersen, L. U. Nordstrøm, R. Madsen
FULL PAPER
as the hydrochloride salt. It was decided to isolate the prod-
uct as a salt in this case, because piperazine and water form
an azeotrope that is difficult to separate by other methods.
Scheme 7. Condensations with acetoin.
Scheme 5. Piperazine synthesis from ethanolamine.
Scheme 6. Piperazine from 2-[(2-aminoethyl)amino]ethanol.
While this work was in progress, the [Cp*IrCl2]2-cata-
lyzed synthesis of various 1,4-dibenzylpiperazines was re-
ported from N-benzylethanolamines in toluene solution at
110 °C.[24] The homocoupling was achieved with 2.5%
[Cp*IrCl2]2 to afford the products in 45–63% isolated yield,
depending on the substituents on the benzyl group.[24] Only
N-benzyl-substituted ethanolamines were investigated in
this study, and no reactions were carried out with the un-
substituted ethanolamine.[24]
Scheme 8. Proposed mechanism for the piperazine formation.
Conclusions
We have presented an atom-economical procedure for the
synthesis of piperazines from amines and 1,2-diols in which
water is produced as the only byproduct. The reaction is
catalyzed by [Cp*IrCl2]2 and can be achieved in toluene as
well as in aqueous solution. Both primary alkylamines and
1,2-diamines can be employed in the condensation with 1,2-
diols to afford a variety of piperazines with different substi-
tution patterns. A mechanism is proposed in which the key
step is an isomerization of the initially formed hydroxy im-
ine to the corresponding amino aldehyde/ketone.
The mechanism for the piperazine formation from 1,2-
diols presumably involves initial formation of an α-hydroxy
aldehyde, which then reacts with the amine to furnish the
α-hydroxy imine. This can either be reduced to the corre-
sponding amino alcohol or undergo rearrangement to the
amino aldehyde/ketone. The isomerization of an α-hydroxy
imine to the α-amino carbonyl compound is a well-known
reaction that can be performed in benzene at reflux.[25] To
probe the ease of this rearrangement, we treated cyclohex-
ylamine with acetoin in toluene solution at various tem-
peratures (Scheme 7a). These experiments revealed that the
α-amino ketone could be obtained as the sole product at
60 °C after 1 h, and the same was observed when the reac-
tion was repeated in the presence of [Cp*IrCl2]2. The con-
densation was also performed with trans-1,2-diaminocyclo-
hexane, but in this case the only product formed after 4 h
was the corresponding diimine (Scheme 7b). Presumably,
the cyclic imine is formed and then oxidized by air to the
diimine.[26] Because the piperazines are prepared at 110–
140 °C for about 17 h, these experiments indicate that the
condensation proceeds by rearrangement of the initially
formed α-hydroxy imine and not by several dehydrogena-
tion/hydrogenation sequences (Scheme 8). Most likely, the
iridium atom remains coordinated during the isomerization,
because no byproducts were observed from the intermedi-
ates in the cyclization. This isomerization step may also ex-
plain why we have been unable to extend the cycloconden-
sation reaction to 1,3-diols.[27]
Experimental Section
General: Reactions were monitored by GC with a Shimadzu
GC2010 instrument equipped with an Equity 1 column
(15 mϫ0.1 mm, 0.1 μm film). Thin-layer chromatography was per-
formed on aluminum plates coated with silica gel 60. Visualization
was achieved by dipping in a solution of cerium(IV) sulfate (2.5 g)
and ammonium molybdate (6.26 g) in 10% sulfuric acid (250 mL),
or ninhydrin (10 g) in ethanol (300 mL). Flash chromatography was
performed on silica gel (220–440 mesh). GC–MS analysis was car-
ried out with a Shimadzu GC–MS QP2010S instrument equipped
with an Equity 5 column (30 mϫ0.25 mm, 0.25 μm film). NMR
spectra were recorded with a Varian Mercury 300 instrument, and
chemical shifts were measured relative to the residual solvent sig-
nal. Assignments were verified by COSY, HSQC, and HMBC spec-
tra.
General Procedure for Preparation of Piperazines: A 5 mL screw-
top vial was charged with [Cp*IrCl2]2 (8 mg, 0.01 mmol), NaHCO3
(10 mg, 0.1 mmol), diamine (2 mmol), diol (2 mmol), and solvent
(1 mL). The vial was flushed with argon, sealed and heated in an
aluminum block. After cooling to room temperature, the solvent
was removed in vacuo and the residue was purified by flash
6756
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 6752–6759

Iridium-Catalyzed Synthesis of Substituted Piperazines
chromatography (heptane/EtOAc or MeOH/CH2Cl2 with 1%
Et3N) to yield the desired product.
3/6eq-H), 2.65–2.56 (m, 2 H, 2/5-H), 2.32 (dd, J = 10.3, 11.7 Hz, 2
H, 3/6ax-H), 0.92 (d, J = 6.3 Hz, 6 H, Me) ppm. 13C NMR
(75 MHz, CDCl3): δ = 54.0, 50.8, 19.5 ppm. Minor cis-2,6 isomer:
1H NMR (300 MHz, CDCl3): δ = 2.84 (dd, J = 2.8, 11.9 Hz, 2 H,
3/5eq-H), 2.65–2.56 (m, 2 H, 2/6-H), 2.17 (dd, J = 10.3, 12.1 Hz, 2
H, 3/5ax-H), 0.93 (d, J = 6.3 Hz, 6 H, Me) ppm. 13C NMR
(75 MHz, CDCl3): δ = 52.8, 51.9, 19.7 ppm. MS: m/z = 114 [M+].
Identical to commercial compounds from Aldrich.
(؎)-trans-Decahydroquinoxaline: Table 1. M.p. 143–145 °C (ref.[28]
150–151 °C). 1H NMR (300 MHz, CDCl3): δ = 2.98–2.78 (m, 4 H),
2.25–2.10 (m, 2 H), 1.80–1.10 (m, 10 H) ppm. 13C NMR (75 MHz,
CDCl3): δ = 61.4, 47.1, 32.1, 25.0 ppm. MS: m/z = 140 [M+].
(؎)-2-Methylpiperazine:[29] Tables 2 and 3, Entries 3 and 4. Isolated
1
by distillation and precipitation as the bis(hydrochloride) salt. H
(؎)-2-Phenylpiperazine: Table 3, Entry 7. 1H NMR (300 MHz,
NMR (300 MHz, D2O): δ = 2.71–2.62 (m, 3 H), 2.55–2.33 (m, 3
H), 2.06 (dd, J = 10.5, 12.5 Hz, 1 H), 0.78 (d, J = 6.4 Hz, 3 H)
ppm. 13C NMR (75 MHz, D2O): δ = 51.4, 50.2, 45.0, 44.0,
18.5 ppm. MS: m/z = 100 [M+].
CDCl3): δ = 7.40–7.20 (m, 5 H), 3.73 (dd, J2–3eq = 2.8, J2–3ax
=
10.2 Hz, 1 H, 2-H), 3.11–2.80 (m, 5 H, 3eq-H, 5ax-H, 5eq-H, 6ax-
H, 6eq-H), 2.69 (dd, J2–3ax = 10.2, J3gem = 11.9 Hz, 1 H, 3ax-H),
1.80 (br. s, 2 H, 1-H, 4-H) ppm. 13C NMR (75 MHz, CDCl3): δ =
142.8, 128.5, 127.5, 126.9, 62.1, 54.4, 47.9, 46.1 ppm. HRMS:
calcd. for C10H15N2 [M + H]+ 163.1235; found 163.0981.
1-Benzylpiperazine: Table 2, Entry 5. 1H NMR (300 MHz, CDCl3):
δ = 7.32–7.21 (m, 5 H), 3.49 (s, 2 H), 2.89 (t, J = 4.5 Hz, 4 H), 2.42
(br. s, 4 H), 2.33 (s, 1 H, NH) ppm. 13C NMR (75 MHz, CDCl3):
δ = 137.9, 129.2, 128.1, 126.9, 63.6, 54.3, 45.9 ppm. MS: m/z = 176
[M+]. NMR spectroscopic data are in accordance with literature
values.[30]
(؎)-trans-2-Methyl-5-phenyl- and (؎)-cis-2-Methyl-6-phenylpipera-
zine: Table 3, Entries 8 and 9. Isolated by distillation. Major trans-
2-methyl-5-phenyl isomer: 1H NMR (300 MHz, CDCl3): δ = 7.41–
7.22 (m, 5 H), 3.72 (dd, J5–6eq = 2.9, J5–6ax = 10.4 Hz, 1 H, 5-H),
3.09 (dd, J5–6eq = 2.9, J6gem = 12.0 Hz, 1 H, 6eq-H), 3.06 (dd,
J2–3eq = 2.9, J3gem = 12.1 Hz, 1 H, 3eq-H), 2.89 (m, 1 H, 2-H), 2.78
1,4-Dibenzylpiperazine: Table 2, Entries 6 and 7. 1H NMR
(300 MHz, CDCl3): δ = 7.35–7.21 (m, 10 H), 3.52 (s, 4 H), 2.49
(br. s, 8 H) ppm. 13C NMR (75 MHz, CDCl3): δ = 138.2, 129.4,
(dd, J5–6ax = 10.3, J6gem = 11.9 Hz, 1 H, 6ax-H), 2.55 (dd, J2–3ax
=
1
128.3, 127.1, 63.2, 53.2 ppm. MS: m/z = 266 [M+]. H NMR spec-
10.4, J3gem = 12.1 Hz, 1 H, 3ax-H), 1.01 (d, J = 6.2 Hz, 3 H, Me)
ppm. 13C NMR (75 MHz, CDCl3): δ = 142.4, 128.3, 127.3, 126.6,
61.2, 54.7, 54.4, 50.7, 19.8 ppm. Minor cis-2-methyl-6-phenyl iso-
mer: 1H NMR (300 MHz, CDCl3): δ = 7.41–7.22 (m, 5 H), 3.80
troscopic data are in accordance with literature values.[31]
(–)-(2S,3S)-2,3-Diphenylpiperazine: Table 2, Entries 8–10. [α]2D5
=
–102 (c = 1.0, CHCl3) {ref.[13a] [α]2D5 = –104.6 (c = 1.0, CHCl3)}.
M.p. 93–95 °C (ref.[13a] m.p. 94–96 °C). 1H NMR (300 MHz,
CDCl3): δ = 7.20–7.05 (m, 10 H), 3.71 (s, 2 H), 3.14 (s, 4 H), 2.01
(br. s, 2 H) ppm. 13C NMR (75 MHz, CDCl3): δ = 141.5, 128.1,
127.9, 127.3, 68.3, 47.2 ppm. MS: m/z = 238 [M+]. The starting
(–)-(1S,2S)-1,2-diphenyl-1,2-ethylenediamine was obtained by reso-
(dd, J6–5eq = 2.9, J6–5ax = 10.4 Hz, 1 H, 6-H), 2.98 (dd, J6–5eq
=
2.9, J5gem = 12.2 Hz, 1 H, 5eq-H), 2.95 (m, 1 H, 3eq-H), 2.89 (m,
1 H, 2-H), 2.62 (dd, J6–5ax = 10.3, J5gem = 12.0 Hz, 1 H, 5ax-H),
2.42 (dd, J2–3ax = 10.5, J3gem = 12.3 Hz, 1 H, 3ax-H), 1.01 (d, J =
6.2 Hz, 3 H, Me) ppm. 13C NMR (75 MHz, CDCl3): δ = 142.6,
128.3, 127.3, 126.9, 53.5, 53.0, 52.4, 52.6, 20.0 ppm. HRMS: calcd.
for C11H17N2 [M + H]+ 117.1386; found 117.1407.
lution[32] and showed [α]2D5 = –104 (c = 1.5, MeOH) {ref.[32] [α]2D3
–106 (c = 1.1, MeOH)}.
=
1,2,3,4-Tetrahydroquinoxaline: Table 2, Entry 11. 1H NMR
(300 MHz, CDCl3): δ = 6.61–6.56 (m, 2 H), 6.53–6.48 (m, 2 H),
3.42 (s, 4 H) ppm. 13C NMR (75 MHz, CDCl3): δ = 133.6, 118.7,
114.7, 41.3 ppm. MS: m/z = 134 [M+]. 13C NMR spectroscopic
data are in accordance with literature values.[33]
(؎)-(4aR,8aR)-2,3-Dimethyldecahydroquinoxaline: Table 3, En-
tries 10 and 11. Major (Ϯ)-(2S,3R) isomer: 1H NMR (300 MHz,
CDCl3): δ = 3.05 (dq, J = 3.6, 6.7 Hz, 1 H, 2-H or 3-H), 2.88 (dq,
J = 3.6, 6.7 Hz, 1 H, 2-H or 3-H), 2.44–2.15 (m, 2 H, 4a-H, 8a-H),
1.67–1.50 (m, 6 H, 1-H, 4-H, 5-H, 8-H), 1.30–1.15 (m, 4 H, 6-H,
7-H), 1.08 (d, J = 6.7 Hz, 3 H, Me), 0.91 (d, J = 6.7 Hz, 3 H, Me)
ppm. 13C NMR (75 MHz, CDCl3): δ = 62.6, 54.4, 53.4, 52.1, 31.2,
31.9, 25.0, 24.9, 19.2, 12.8 ppm. HRMS: calcd. for C10H21N2 [M +
H]+ 169.1705; found 169.1705. Minor (Ϯ)-(2S,3S) isomer: 13C
NMR (75 MHz, CDCl3): δ = 61.3, 57.9, 31.7, 19.0 ppm.
2-(Hydroxymethyl)benzimidazole: Table 2, Entry 11. 1H NMR
(300 MHz, CD3OD): δ = 7.56–7.50 (m, 2 H), 7.23–7.17 (m, 2 H),
4.89 (s, 1 H, NH), 4.83 (s, 2 H) ppm. 13C NMR (75 MHz, CD3OD):
δ = 156.2, 123.4, 59.0 ppm. MS: m/z = 148 [M+]. Identical to com-
mercial compound from Aldrich.
(؎)-1,4-Dibenzyl-2-methylpiperazine:
(300 MHz, CDCl3): δ = 7.35–7.20 (m, 10 H), 4.05 (d, J = 13.3 Hz,
1 H), 3.48 (s, 2 H), 3.19 (d, J = 13.2 Hz, 1 H), 2.75–2.60 (m, 3 H,
Scheme 3.
1H
NMR
(؎)-(4aR,8aR)-2-Methyldecahydroquinoxaline: Table 3, Entries 1–2.
Major (Ϯ)-(2S) isomer: 1H NMR (300 MHz, CDCl3): δ = 2.88 (dd,
J2–3eq = 2.9, J3gem = 11.6 Hz, 1 H, 3eq-H), 2.78 (ddq, J2–3eq = 2.9,
J2–Me = 6.3, J2–3ax = 10.0 Hz, 1 H, 2-H), 2.38 (dd, J2–3ax = 10.2,
J3gem = 11.6 Hz, 1 H, 3ax-H), 2.26–2.05 (m, 2 H, 4a-H, 8a-H),
1.75–1.55 (m, 6 H, 1-H, 4-H, 5-H, 8-H), 1.30–1.05 (m, 4 H, 6-H,
7-H), 0.96 (d, J2–Me = 6.3 Hz, 3 H, Me) ppm. 13C NMR (75 MHz,
CDCl3): δ = 61.7, 60.7, 54.2, 52.2, 32.2, 32.0, 25.2, 25.0, 20.0 ppm.
HRMS: calcd. for C9H19N2 [M + H]+ 155.1548; found 155.1556.
Minor (Ϯ)-(2R) isomer: 1H NMR (300 MHz, CDCl3): δ = 3.41 (tq,
J2–3 = 1.8, J2–Me = 7.0 Hz, 1 H, 2-H), 2.58–2.47 (m, 2 H, 3ax-H,
3eq-H), 2.16–2.07 (m, 2 H, 4a-H, 8a-H), 1.85–1.05 (m, 8 H, 5-H,
6-H, 7-H, 8-H), 0.96 (d, J2–Me = 6.6 Hz, 3 H, Me) ppm. 13C NMR
(75 MHz, CDCl3): δ = 62.9, 60.8, 56.5, 50.0, 32.3, 32.0, 25.1, 24.9,
18.3 ppm. MS: m/z = 154 [M+].
3eq-H, 5-H, 6-H), 2.50 (dqd, J2–3eq = 3.0, J2–Me = 6.2, J2–3ax
=
9.1 Hz, 1 H, 2-H), 2.26–2.12 (m, 2 H, 5-H, 6-H), 2.02 (dd, J2–3ax
= 9.7, J3gem = 10.5 Hz, 1 H, 3ax-H), 1.14 (d, J2–Me = 6.2 Hz, 3 H,
Me) ppm. 13C NMR (75 MHz, C6D6): δ = 140.1, 139.3, 129.2,
129.1, 128.5, 128.4, 127.2, 127.0, 63.3, 61.1, 58.5, 55.7, 53.9, 51.5,
16.7 ppm. HRMS: calcd. for C19H25N2 [M + H]+ 281.2018; found
281.2026.
(؎)-1,4-Dibenzyl-2-phenylpiperazine:
Scheme 3.
1H
NMR
(300 MHz, CDCl3): δ = 7.45–7.10 (m, 15 H), 3.71 (d, J = 13.4 Hz,
1 H), 3.43 (s, 2 H), 3.36 (dd, J2–3eq = 3.0, J2–3ax = 10.3 Hz, 1 H, 2-
H), 2.85–2.67 (m, 4 H, Bn, 3eq-H, 5-H, 6-H), 2.25–2.05 (m, 3 H,
3ax-H, 5-H, 6-H) ppm. 13C NMR (75 MHz, CDCl3): δ = 142.3,
(؎)-trans-2,5- and (؎)-cis-2,6-Dimethylpiperazine: Table 3, En- 139.2, 138.0, 129.3, 128.9, 128.6, 128.3, 128.2, 127.5, 127.1, 126.8,
tries 5 and 6. Isolated by distillation. Major trans-2,5 isomer: 1H
NMR (300 MHz, CDCl3): δ = 2.84 (dd, J = 2.8, 11.8 Hz, 2 H,
67.4, 63.1, 62.1, 59.1, 53.3, 51.9 ppm. HRMS: calcd. for C24H27N2
[M + H]+ 343.2174; found 343.2153.
Eur. J. Org. Chem. 2012, 6752–6759
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6757

L. L. R. Lorentz-Petersen, L. U. Nordstrøm, R. Madsen
FULL PAPER
Dodecahydro-1H-phenoxazine: Scheme 4. 1H NMR (300 MHz,
CDCl3): δ = 3.19–3.11 (m, 2 H), 2.63 (s, 1 H, NH), 2.54–2.47 (m,
2 H), 1.90–1.83 (m, 2 H), 1.74–1.66 (m, 6 H), 1.35–1.23 (m, 8 H)
ppm. 13C NMR (CDCl3, 75 MHz): δ = 80.9, 60.7, 31.0, 30.8, 24.5,
24.5 ppm. HRMS: calcd. for C12H22NO [M + H]+ 196.1696; found
196.1698.
[6]
a) R. N. Monrad, R. Madsen, Org. Biomol. Chem. 2011, 9,
610–615; b) H. Aramoto, Y. Obora, Y. Ishii, J. Org. Chem.
2009, 74, 628–633.
C. S. Cho, S. G. Oh, Tetrahedron Lett. 2006, 47, 5633–5636.
For a preliminary communication, see: L. U. Nordstrøm, R.
Madsen, Chem. Commun. 2007, 5034–5036.
M. Vieth, M. G. Siegel, R. E. Higgs, I. A. Watson, D. H. Rob-
ertson, K. A. Savin, G. L. Durst, P. A. Hipskind, J. Med.
Chem. 2004, 47, 224–232.
a) J. I. Jang, S. Y. Kang, K. H. Kang, Y. S. Park, Tetrahedron
2011, 67, 6221–6226; b) M. Jida, M. Soueidan, N. Willand,
F. Agbossou-Niedercorn, L. Pelinski, G. Laconde, R. Deprez-
Poulain, B. Deprez, Tetrahedron Lett. 2011, 52, 1705–1708; c)
G. Reginato, B. Di Credico, D. Andreotti, A. Mingardi, A.
Paio, D. Donati, B. Pezzati, A. Mordini, Tetrahedron: Asym-
metry 2010, 21, 191–194; d) C. J. Dinsmore, D. C. Beshore, Tet-
rahedron 2002, 58, 3297–3312; e) C. J. Dinsmore, D. C. Be-
shore, Org. Prep. Proced. Int. 2002, 34, 367–404.
a) G. Barker, P. O’Brien, K. R. Campos, Org. Lett. 2010, 12,
4176–4179; b) M. Berkheij, L. van der Sluis, C. Sewing, D. J.
den Boer, J. W. Terpstra, H. Hiemstra, W. I. I. Bakker, A.
van den Hoogenband, J. H. van Maarseveen, Tetrahedron Lett.
2005, 46, 2369–2371.
a) R. Gao, D. J. Canney, J. Org. Chem. 2010, 75, 7451–7453;
b) K. G. Liu, A. J. Robichaud, Tetrahedron Lett. 2005, 46,
7921–7922.
a) P. Vairaprakash, M. Periasamy, J. Org. Chem. 2006, 71,
3636–3638; b) G. J. Mercer, M. S. Sigman, Org. Lett. 2003, 5,
1591–1594.
[7]
[8]
[9]
1-(2-Hydroxyethyl)piperazine: Scheme 5. 1H NMR (300 MHz,
CD3OD): δ = 3.68 (t, J = 5.9 Hz, 2 H), 2.93–2.91 (m, 4 H), 2.56–
2.51 (m, 6 H) ppm. 13C NMR (75 MHz, CD3OD): δ = 61.7, 59.7,
54.4, 45.8 ppm. MS: m/z = 130 [M+]. Identical to commercial com-
pound from Aldrich.
[10]
1,4-Bis(2-hydroxyethyl)piperazine: Scheme 5. 1H NMR (300 MHz,
CD3OD): δ = 3.70 (t, J = 5.9 Hz, 4 H), 2.73 (br. s, 8 H), 2.65 (t, J
= 5.9 Hz, 4 H) ppm. 13C NMR (75 MHz, CD3OD): δ = 60.9, 59.4,
53.6 ppm. HRMS: calcd. for C8H19N2O2 [M + H]+ 175.1447;
found 175.1448.
[11]
Piperazine: Scheme 6. Isolated by distillation and precipitation as
1
bis(hydrochloride) salt. H NMR (300 MHz, CDCl3): δ = 2.78 (s,
8 H) ppm. 13C NMR (75 MHz, CDCl3): δ = 47.2 ppm. MS: m/z =
86 [M+].
[12]
[13]
[14]
1
3-(Cyclohexylamino)butan-2-one: Scheme 7a. H NMR (300 MHz,
C6D5CD3): δ = 3.12 (q, J = 6.9 Hz, 1 H), 2.25–2.17 (m, 1 H), 1.80
(s, 3 H), 1.69–1.53 (m, 6 H), 1.12–1.00 (m, 4 H), 0.93 (d, J = 6.9 Hz,
3 H) ppm. 13C NMR (75 MHz, C6D5CD3): δ = 210.2, 60.6, 55.4,
34.3, 33.9, 26.5, 25.7, 25.2, 25.1, 18.7 ppm. HRMS: calcd. for
C10H20NO [M + H]+ 170.1539; found 170.1539.
a) E. Borsini, G. Broggini, A. Fasana, S. Galli, M. Khansaa,
U. Piarulli, M. Rigamonti, Adv. Synth. Catal. 2011, 353, 985–
994; b) J. S. Nakhla, D. M. Schultz, J. P. Wolfe, Tetrahedron
2009, 65, 6549–6570; c) B. M. Cochran, F. E. Michael, Org.
Lett. 2008, 10, 329–332.
a) K. Samanta, G. Panda, Chem. Asian J. 2011, 6, 189–197; b)
F. Crestey, M. Witt, J. W. Jaroszewski, H. Franzyk, J. Org.
Chem. 2009, 74, 5652–5655.
(؎)-trans-2,3-Dimethyl-5,6,7,8,9,10-hexahydroquinoxaline:
Scheme 7b. 13C NMR (75 MHz, C6D5CD3): δ = 158.0, 59.2, 33.8,
25.7, 22.3 ppm. MS: m/z = 164 [M+]. NMR spectroscopic data are
in accordance with literature values.[34]
[15]
Supporting Information (see footnote on the first page of this arti-
[16]
[17]
G. Jenner, G. Bitsi, J. Mol. Catal. 1988, 45, 165–168.
K. Fujita, Y. Enoki, R. Yamaguchi, Tetrahedron 2008, 64,
1943–1954.
1
cle): Copies of H and 13C NMR spectra for all products.
[18]
The use of water as a solvent in the N-alkylation of amines
with alcohols was first reported by us[8] in 2007. Since then,
several modified Cp*Ir catalysts have been developed for the
alkylation in aqueous media; see Kawahara et al.[3a] and: a) H.
Ohta, Y. Yuyama, Y. Uozumi, Y. M. A. Yamada, Org. Lett.
2011, 13, 3892–3895; b) O. Saidi, A. J. Blacker, M. M. Farah,
S. P. Marsden, J. M. J. Williams, Chem. Commun. 2010, 46,
1541–1543.
Acknowledgments
We thank the Danish National Research Foundation for financial
support.
[1] a) S. Bähn, S. Imm, L. Neubert, M. Zhang, H. Neumann, M.
Beller, ChemCatChem 2011, 3, 1853–1864; b) Y. Obora, Y.
Ishii, Synlett 2011, 30–51; c) R. Yamaguchi, K. Fujita, M. Zhu,
Heterocycles 2010, 81, 1093–1140; d) G. Guillena, D. J. Ramón,
M. Yus, Chem. Rev. 2010, 110, 1611–1641; e) G. E. Dobereiner,
R. H. Crabtree, Chem. Rev. 2010, 110, 681–703; f) M. H. S. A.
Hamid, P. A. Slatford, J. M. J. Williams, Adv. Synth. Catal.
2007, 349, 1555–1575; g) K. Fujita, R. Yamaguchi, Synlett
2005, 560–571.
[19]
a) Y. Tsuji, K.-T. Huh, Y. Ohsugi, Y. Watanabe, J. Org. Chem.
1985, 50, 1365–1370; b) A. Arcelli, Bui-the-Khai, G. Porzi, J.
Organomet. Chem. 1982, 235, 93–96.
T. Kondo, S. Yang, K.-T. Huh, M. Kobayashi, S. Kotachi, Y.
Watanabe, Chem. Lett. 1991, 1275–1278.
A. J. Blacker, M. M. Farah, M. I. Hall, S. P. Marsden, O. Saidi,
J. M. J. Williams, Org. Lett. 2009, 11, 2039–2042.
R. Yamaguchi, Z. Mingwen, S. Kawagoe, C. Asai, K. Fujita,
Synthesis 2009, 1220–1223.
W.-Y. Su, C. L. LeBas, A. C. Kopecky, J. F. Knifton, Tetrahe-
dron Lett. 1992, 33, 871–874.
[20]
[21]
[22]
[23]
[24]
[2] P. Fristrup, M. Tursky, R. Madsen, Org. Biomol. Chem. 2012,
10, 2569–2577.
[3] a) R. Kawahara, K. Fujita, R. Yamaguchi, Adv. Synth. Catal.
2011, 353, 1161–1168; b) M. H. S. A. Hamid, C. L. Allen,
G. W. Lamb, A. C. Maxwell, H. C. Maytum, A. J. A. Watson,
J. M. J. Williams, J. Am. Chem. Soc. 2009, 131, 1766–1774; c)
K. Fujita, T. Fujii, A. Komatsubara, Y. Enoki, R. Yamaguchi,
Heterocycles 2007, 74, 673–682.
[4] R. Yamaguchi, S. Kawagoe, C. Asai, K. Fujita, Org. Lett. 2008,
10, 181–184.
[5] a) M. Tursky, L. L. R. Lorentz-Petersen, L. B. Olsen, R.
Madsen, Org. Biomol. Chem. 2010, 8, 5576–5582; b) Y. Tsuji,
K.-T. Huh, Y. Watanabe, J. Org. Chem. 1987, 52, 1673–1680.
K. Fujita, Y. Kida, R. Yamaguchi, Heterocycles 2009, 77,
1371–1377.
[25]
[26]
E. F. Pratt, M. J. Kamlet, J. Org. Chem. 1963, 28, 1366–1368.
F. Pan, T.-M. Chen, J.-J. Cao, J.-P. Zou, W. Zhang, Tetrahedron
Lett. 2012, 53, 2508–2510.
Treatment of trans-1,2-diaminocyclohexane with 1,3-propylene
glycol or 1,1-bis(hydroxymethyl)cyclohexane gave, in both
cases, a complex mixture in which the desired seven-membered
1,4-diazepane could not be identified.
[27]
[28]
E. Brill, H. P. Schultz, J. Org. Chem. 1964, 29, 579–581.
6758
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 6752–6759

Iridium-Catalyzed Synthesis of Substituted Piperazines
[29] K. V. S. Rao, M. Subrahmanyam, Photochem. Photobiol. Sci.
2002, 1, 597–599.
[30] T. Myochin, K. Kiyose, K. Hanaoka, H. Kojima, T. Terai, T.
Nagano, J. Am. Chem. Soc. 2011, 133, 3401–3409.
[31] J. Alcázar, J. Comb. Chem. 2005, 7, 353–355.
[32] S. Pikul, E. J. Corey, Org. Synth. 1993, 71, 22–29.
[33] S. F. Nelsen, L. A. Grezzo, W. C. Hollinsed, J. Org. Chem.
1981, 46, 283–289.
[34] T. Yamaguchi, N. Kashige, N. Mishiro, F. Miake, K. Watanabe,
Biol. Pharm. Bull. 1996, 19, 1261–1265.
Received: August 13, 2012
Published Online: October 23, 2012
Eur. J. Org. Chem. 2012, 6752–6759
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6759