A general method for the preparation of chiral TREN derivatives

Article abstract of DOI:10.1002/ejoc.200500094

A general procedure for the preparation of C₃-symmetric TREN derivatives with backbone chirality has been developed. Stereo- and regioselective ring opening by ammonia of (S)-N-tosyl-2-isopropylaziridine, obtained starting from either the corre

Full text of DOI:10.1002/ejoc.200500094

FULL PAPER
A General Method for the Preparation of Chiral TREN Derivatives
Yuxin Pei,[a] Katja Brade,[a] Emilie Brulé,[a] Lars Hagberg,[a][‡]
Fredrik Lake,[a][‡‡] and Christina Moberg*[a]
Keywords: Amines / Chirality / C3 symmetry / Tripodal ligands / TREN
A general procedure for the preparation of C3-symmetric
TREN derivatives with backbone chirality has been devel-
oped. Stereo- and regioselective ring opening by ammonia
of (S)-N-tosyl-2-isopropylaziridine, obtained starting from
either the corresponding amino alcohol or amino acid, fol-
lowed by deprotection of the amino groups afforded the par-
ent chiral TREN compound in high overall yield. In addition
to TREN compounds with primary amino groups, the syn-
thetic method employed provides easy access to a variety of
N,NЈ,NЈЈ-substituted derivatives.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2005)
Introduction
Tris(2-aminoethyl)amine, TREN (1, R = H), and its
N,NЈ,NЈЈ-substituted derivatives are widely employed as li-
gands for both transition metals[1] and main group ele-
ments.[2] Several of the TREN containing compounds exhi-
bit properties which make them useful for a variety of appli-
cations. Thus, proazaphosphatranes have been shown to be
extremely strong bases with wide synthetic applications[3] at
the same time as they serve as ligands for metals in catalytic
processes.[4] Molybdenum complexes of N,NЈ,NЈЈ-aryl-sub- Figure 1. Chiral TREN compounds.
stituted TREN[5] have been found to activate dinitrogen.[6]
Other recent applications include the fixation of CO2 by
zinc complexes[7] and the use of uncomplexed TREN deriv-
atives as receptors for anions.[8] In addition, TREN has
been used as a structural motif for the preparation of a
variety of more elaborate tripodal ligands (Figure 1).[9]
Due to our interest in C3-symmetric ligands[10] we
wanted to have easy access to chiral TREN compounds for
further functionalization. Therefore, we developed a pro-
cedure for the preparation of chiral N,NЈ,NЈЈ-substituted
TREN derivatives, with the chirality residing in the ligand
backbone, through aziridine ring opening by ammonia.[11]
Our synthesis was followed by the preparation by Yamam-
oto and co-workers of another chiral TREN compound
starting from proline using a stepwise procedure.[12] Sub-
sequently less rigid derivatives, with chiral substituents at-
tached to the equatorial nitrogen atoms, were described by
Verkade and co-workers.[13] Raymond and co-workers de-
veloped a method for the preparation of TREN compounds
with backbone chirality employing reductive amination of
α-amino aldehydes as the key step, which provided access
to derivatives with primary amino groups, isolated as their
trihydrochloride salts.[14] A recent report from Verkade’s
group[15] on the preparation of chiral TREN derivatives,
using the procedure developed by Raymond, prompts us to
report improvements on our previously disclosed procedure,
resulting in experimentally simple and convenient methods
for the preparation of parent chiral TREN compounds 1
(R = Me, iPr) as well as chiral N,NЈ,NЈЈ-alkyl and -aryl
derivatives.
Results and Discussion
Chiral enantiopure aziridines are valuable synthetic
building blocks due to their ability to undergo stereo- and
regioselective ring opening reactions.[16] The aziridines are
conveniently accessible by a variety of methods.[17] Starting
from chiral amino alcohols, commonly derived from natural
or synthetic amino acids, enantiopure aziridines are ob-
tained in high yields without the use of additional chiral
reagents. N-activated aziridines, such as N-sulfonylazirid-
ines, undergo regioselective ring opening using a wide range
[a] Royal Institute of Technology, Department of Chemistry, Or-
ganic Chemistry,
100 44 Stockholm, Sweden
Fax: +46-8-791-2333
E-mail: kimo@kth.se
[‡] Present address: Karo Bio AB, Novum,
141 57 Huddinge, Sweden
[‡‡]Present address: Boston University, Department of Chemistry,
590 Commonwealth Avenue, Boston, MA 02215, USA
Eur. J. Org. Chem. 2005, 2835–2840
DOI: 10.1002/ejoc.200500094
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2835

Y. Pei, K. Brade, E. Brulé, L. Hagberg, F. Lake, C. Moberg
FULL PAPER
of nucleophiles to provide ring-opened products in enantio- ation accompanied by base-mediated ring closure. The
pure form. For the preparation of chiral TREN compounds overall yield of 2a was 82%, which is lower than that ob-
we therefore found that a method based on aziridines would tained starting from the amino alcohol. Although tosyl-
be highly attractive. Our synthetic method thus uses the ation of 4b and reduction of the carboxylic acid function
ring opening of three equivalents of an N-(arylsulfonyl)azir- were reported to proceed smoothly (90% yield of 5 based
idine by ammonia as the key step.
on 4b), mesylation and ring closure of the N-tosylated
For the synthesis of substituted TREN derivatives 1a and amino alcohol afforded merely 59% of 2b (i.e. 53% based
1b, (S)-N-tosyl-2-isopropylaziridine (2a) and (S)-N-tosyl-2- on 4b).[22] We found, however, that under Mitsunobu condi-
methylaziridine (2b) were required. By a recently published tions,[23] 5 was transferred to 2b in 74% yield, thus resulting
method these compounds were obtained in one step in 82 in an overall yield of 2b from 4b of 67%, which is lower
and 73% yield from commercially available (S)-valinol (3a) than that obtained in the one-pot procedure starting with
and (S)-alaninol (3b), respectively, using tosyl chloride, tri- (S)-alaninol. Scheme 1 summarizes the methods for the
ethylamine, and DMAP.[18] A two-step procedure, using two preparation of aziridines 2a and 2b from either the amino
equivalents of tosyl chloride, was reported to result in even alcohols or the amino acids which we found most suitable.
higher yields of the same compounds, 90 and 88%.[19] In
Our first published conditions for the stereo- and re-
our hands, this latter procedure afforded 2a and 2b in con- gioselective aziridine ring opening by ammonia resulted in
siderably lower yields. However, a slight modification, a rather sluggish reaction, requiring four days at 40 °C to
whereby the second equivalent of tosyl chloride was re- obtain a 70% yield of C3-symmetric 6b from 2b.[11] We later
placed by mesyl chloride, afforded higher yields of the de- found that efficient ring opening occurred when a 4:1 mix-
sired compounds. The isopropylaziridine 2a was thus ob- ture of 2a and ammonia in methanol was subjected to mi-
tained in a one-pot procedure by treatment of (S)-valinol crowave irradiation for 75 minutes, while the temperature
(3a) with 1.15 equivalents of tosyl chloride followed by 1.05 was maintained at 160 °C (Scheme 2).[24] We were particu-
equivalents of mesyl chloride in the presence of four equiva- larly pleased to find that cooling of the reaction mixture led
lents of triethylamine in 94% yield (Scheme 1). The same to the precipitation of white crystals, consisting of pure 6a,
procedure applied to (S)-alaninol (3b) gave 2b, albeit in according to NMR spectrosopy, in 85% yield. The same
lower yield, 74%.
procedure applied to 2b required chromatographic purifica-
tion of the product (6b), which did not precipitate after re-
action. Heating at 50 °C for four days in place of using mi-
crowave irradiation resulted in a similar yield (75%).
Scheme 1. Synthesis of chiral tosylaziridines.
Scheme 2. Synthesis of chiral TREN derivatives by aziridine ring
opening.
Although the amino alcohols are commercially available,
the corresponding amino acids are less expensive and may
Deprotection of compounds containing several N-sulfo-
therefore be found to be more suitable starting materials. nyl functions has occasionally been connected with prob-
The amino alcohols employed can be obtained from the lems.[25] In our previous procedure we employed N-nosyl-
corresponding amino acids. BH3 reduction of (S)-valine aziridines.[26] Deprotection was achieved using mercaptoace-
(4a) has been reported to afford (S)-valinol (3a) in 95% tic acid and LiOH,[27] thiophenol and sodium carbonate,[27]
yield,[20] which is a considerable improvement of previously or sodium in ammonia.[28] However, yields were not satis-
reported 44%.[21] However, to avoid difficulties encountered factory and not always reproducible, and for reasons which
during isolation of the water soluble products, we preferred are unknown to us, 1a and 1b were obtained at best only in
to employ a method published by Berry and Craig provid- very low yields by any of these methods. We have now
ing direct access to tosylaziridines from amino acids.[22] By found that the desired compounds are conveniently ob-
this method the desired aziridines were obtained from (S)- tained in higher and reproducible yields from the tosylated
valine (4a) and (S)-alanine (4b) by a three-step procedure amines using 48% HBr/phenol[29] for the deprotection.
consisting of N-tosylation, LAH reduction, and O-tosyl- Thus, deprotection of compounds 6a and 6b in refluxing
2836
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 2835–2840

A General Method for the Preparation of Chiral TREN Derivatives
FULL PAPER
HBr led to the parent chiral TREN derivatives 1a[30] and
1b in 90 and 38% yields, respectively (Scheme 2).
Our method thus provides easy access to chiral TREN
compounds 1a and 1b in three steps in 72 and 21% overall
With tosylated compounds 6a and 6b as well as depro- yields from (S)-valinol (3a) and (S)-alaninol (3b), respec-
tected TREN derivatives 1a and 1b in hand, we decided to tively. Alternatively, starting from the corresponding amino
prepare different types of N,NЈ,NЈЈ-substituted derivatives. acids, the same compounds are obtained in five steps in
For this purpose, three different routes were employed. The overall yields of 63 and 19%, respectively. The procedures
first consisted of N-alkylation of tosyl-protected TREN de- are experimentally simple. The preparation of 1a from 3a
rivatives 6a and 6b. Treatment of 6a and 6b with sodium requires column chromatography only after the first step,
hydride and methyl iodide gave 7a and 7b in yields of 97 the remaining purifications being accomplished by filtration
and 91%, respectively (Scheme 3). Deprotection, again or extraction. In contrast to some of the previous methods
using HBr and phenol, gave the desired methylated deriva- affording chiral TREN compounds with the stereocenters
tives 8a and 8b in 89 and 60% yields (69 and 30% from 3a located in the N-substituents[13] or in which the nitrogen
and 3b, respectively). Compound 8a was previously ob- atoms form part of a pyrrolidine ring,[12] our method gives
tained from the corresponding nosyl derivatives,[11] al- access to chiral parent TREN derivatives with primary
though in lower yield [53% from (S)-valine as compared to amino groups. In addition, a variety of N,NЈ,NЈЈ-substi-
60% employing the present method].
tuted derivatives, which are of interest for various potential
applications, are accessible in reasonable overall yields, as
demonstrated by the syntheses of N,NЈ,NЈЈ-substituted
compounds 8–13.
The location of chirality in the ligand backbone is ex-
pected to add rigidity to metal complexes and main group
element derivatives of the chiral TREN compounds,
thereby favoring one of the possible diastereomeric twisted
conformations. The stereocenters are, however, situated re-
mote from the metal center. Model studies indicate that in
1a and 1b as well as in N,NЈ,NЈЈ-alkyl derivatives, chiral
transfer to the region of a coordinated metal ion may be
inefficient. In contrast, aryl groups on nitrogen are thought
to serve as efficient transmittors of chirality. We therefore
consider the introduction of aryl groups on nitrogen by Pd-
catalyzed C–N coupling to be of special interest.
Our new method should be compared to that recently
published by Raymond[14] and subsequently extended by
Verkade.[15] Raymond prepared the parent compounds as
their hydrochloride salts in moderate overall yields from the
amino alcohols, whereas Verkade prepared an N,NЈ,NЈЈ-al-
kylsubstituted analogue of 1 (R = Bn) in five step from (S)-
phenylalanine in an overall yield of 48%. Our method thus
results in higher overall yields of the N,NЈ,NЈЈ-substituted
compounds starting from either (S)-valine (4a) or (S)-vali-
nol (3a). We have also demonstrated that derivatives with a
large variety of substituents on nitrogen are easily access-
Scheme 3. Synthesis of chiral N-alkylated TREN derivatives.
N,NЈ,NЈЈ-Tribenzyl derivative 9 was previously prepared
by an analogous procedure (45% from 3a).[11] However,
somewhat higher yield of the same compound was achieved
by reductive amination of three equivalents of benzaldehyde
with 1a and NaBH4 (72%, 52% from 3a, Scheme 4). Re-
ductive amination using cyclohexanecarboxaldehyde af-
forded 10 (75%, 54% from 3a). Finally, palladium-cata-
lyzed arylation of 1a using bromobenzene, 2-methylbro-
mobenzene, or 4-cyanobromobenzene and a catalyst pre-
pared from trisbenzylideneacetonedipalladium and rac-BI-
NAP provided N-aryl derivatives 11–13 in 69, 83 and 91%
yields (50, 60 and 66% overall from 3a), respectively.
Scheme 4. N-Functionalization of chiral TREN.
Eur. J. Org. Chem. 2005, 2835–2840
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2837

Y. Pei, K. Brade, E. Brulé, L. Hagberg, F. Lake, C. Moberg
FULL PAPER
ible. With the method based on reductive amination,[14]
there is a potential risk for epimerization of intermediates
having the stereocenter next to a carbonyl group, even if
that was not observed in the present case. In contrast, our
procedure gives access to stereochemically well-defined par-
ent chiral TREN compounds and their N,NЈ,NЈЈ-substi-
tuted derivatives in high yields.
In addition to C3-symmetric tripodal amines, our syn-
thetic strategy provides access also to C2-symmetric com-
pounds by employing primary amines for the initial azirid-
ine ring opening, or by using appropriate nucleophile/aziri-
dine ratios.[24,31] Alternatively, as was recently demon-
strated, the formation of products from single or double
ring-opening can be controlled by the choice of solvent.[32]
The presently described compounds are expected to serve
as useful ligands for metal ions and main group elements
in catalytic applications. In addition, the chiral TREN de-
rivatives constitute versatile building blocks for different
types of ligands, as demonstrated for example by the prepa-
ration of phosphanylamides with rotational symmetry.[33]
bath, which was slowly warmed to room temperature. The wine-
red reaction mixture was washed with 0.5 m HCl (2×225 mL). The
organic phase was washed with saturated aqueous Na2CO3
(2×225 mL) and dried with MgSO4. The solvent was removed in
vacuo. Purification by flash chromatography (silica gel, 10%
EtOAc in hexanes) afforded 2a (8.12 g, 94%) as white needles. The
1H NMR and 13C NMR spectroscopic data agreed with those pub-
lished for the same compound.[34]
Compound 2b (from 3b): The previously published method was
used.[11] By employing an exact 3:1 ratio of aziridine and ammonia
(concentration of ammonia in methanol determined by titration
using HCl, with methyl red as an indicator) the yield was increased
from 70 to 74%.
Compound 2b (from 5): DEAD (0.39 mL, 2.48 mmol) was added
dropwise to a solution of N-tosylalaninol[22] (500 mg, 2.18 mmol)
and PPh3 (629 mg, 2.40 mmol) in THF (20 mL) at 0 °C under ni-
trogen. The reaction mixture was stirred at room temperature for
3 h, and then the solvent was evaporated. The crude product was
purified by chromatography (silica gel, hexane/EtOAc, 3:2) to give
2b (343 mg, 74%). The 1H NMR and 13C NMR spectroscopic data
agreed with those published for the same compound.[34]
Compound 1a: Deprotection of the tosyl functions was achieved
according to a modified literature procedure.[29] HBr (48% aq.,
64 mL) was added to a flask containing 6a (3.69 g, 5.02 mmol) and
phenol (4.56 g, 48.44 mmol) and the mixture was refluxed for 48 h.
Water (88 mL) and 2 m NaOH (38 mL) were added at room tem-
perature to adjust the pH to about 1. The aqueous phase was
washed with EtOAc (3×50 mL), then its pH adjusted with 2 m
NaOH (about 350 mL) to Ͼ 13. The aqueous phase was extracted
with CH2Cl2 (5×100 mL). The combined organic phases were
dried with Na2SO4. The solvent was removed in vacuo to give 1a
(1.24 g, 90%) as a light yellow oil. [α]2D0 = +179 (c = 0.72, MeOH).
1H NMR (400 MHz, CDCl3): δ = 2.72–2.67 (m, 3 H), 2.29–2.26
(m, 6 H), 1.89 (br. s, 6 H), 1.50 (octet, J = 6.6 Hz, 3 H), 0.91 (d, J =
6.7 Hz, 9 H), 0.90 (d, J = 6.7 Hz, 9 H) ppm. 13C NMR (100.6 MHz,
CDCl3): δ = 59.6, 53.5, 32.3, 19.3, 18.4 ppm. C15H36N4 (272.47):
calcd. C 66.12, H 13.32, N 20.56; found C 65.93, H 13.28, N 20.42.
Conclusions
An efficient and experimentally simple route for the
preparation of enantiopure C3-symmetric TREN analogues
has been developed starting from the appropriate amino
alcohol or amino acid. The synthesis proceeds via N-sulfo-
nylaziridines, which are ring-opened by ammonia. In ad-
dition to tripodal compounds with primary amino func-
tions, a variety of N,NЈ,NЈЈ-alkyl- and aryl-substituted
derivatives are conveniently accessible using our method.
Experimental Section
Compound 1b: Tetraamine 1b was prepared analogously to tet-
raamine 1a from tris(sulfonamide) 6b (3.50 g, 5.38 mmol). Distil-
lation (105–110 °C/0.13 Torr) gave the product in 38% yield
General Remarks: All reactions were run under dry nitrogen unless
otherwise indicated. Glassware was oven- or flame-dried. Sodium
tert-butoxide was sublimated before use. Dichloromethane, triethyl-
amine, and toluene were distilled from CaH2. Methanol and all
other reagents were used as received.
(0.39 g) as a clear oil which partly solidified upon standing. [α]2D0
=
1
+183 (c = 0.73, MeOH). H NMR (500 MHz, CDCl3): δ = 3.09–
3.04 (m, 3 H), 2.24–2.14 (m, 6 H), 1.57 (br. s, 6 H), 0.98 (d, J =
6.3 Hz, 9 H) ppm. 13C NMR (125.8 MHz, CDCl3): δ = 64.1, 43.9,
24.4 ppm.
1H NMR and 13C NMR spectra were recorded with a Bruker Av-
ance 400 or a Bruker DMX 500 instrument at room temperature
in CDCl3, using the residual signals from CHCl3 (1H: δ = 7.27 ppm;
13C: δ = 77.2 ppm) as internal standard. Microwave heating was
performed in a SmithCreatorTM single-mode cavity from Biotage
AB, Sweden. Optical rotations were recorded with a Perkin–Elmer
343 polarimeter at the sodium D line at ambient temperature. Flash
chromatography was carried out using SDS silica gel 60 (40–
63 μm). Melting points were measured in open capillary tubes using
an Electrothermal instrument and are uncorrected. Elemental
analyses were performed by Analytische Laboratorien, Lindlar,
Germany. The preparation of 6a[24] and 6b[11] from 2a and 2b,
respectively, has been described previously.
Tris(sulfonamide) 7a: A solution of 6a (4.00 g, 5.44 mmol) in DMF
(33 mL) was added dropwise during 30 min via a cannula to NaH
(1.43 g of a 60% dispersion in oil, 35.8 mmol, which had been
washed with 3×10 mL of pentane) in DMF (20 mL) at 0 °C. The
suspension was stirred for 10 min at this temperature and then MeI
(2.03 mL, 32.7 mmol) was added during 10 min. The suspension
was stirred overnight while the temperature was allowed to reach
room temperature. The reaction was quenched by careful addition
of water (40 mL). CH2Cl2 (50 mL) was added, the phases separated
and the aqueous phase was extracted with CH2Cl2 (4×50 mL). The
combined organic phases were washed with brine, dried (MgSO4),
filtered, and concentrated to give a sticky solid. The solid was
Aziridine 2a: TsCl (7.90 g, 41.4 mmol) was added to a solution of
Et3N (20.2 mL, 144 mmol) and (S)-valinol (3a, 3.72 g, 36 mmol) in washed with diethyl ether to yield 7a (3.95 g, 97%), which was used
CH2Cl2 (300 mL) at –25 °C. The reaction mixture was first stirred
for 45 min in the cooling bath, then for another 2.5 h at room tem-
perature. MsCl (2.93 mL, 37.8 mmol) was then added dropwise
over 10 min at –25 °C. The reaction was run for 24 h in the cooling
directly in the next step. M.p. 172–174 °C. [α]2D0 = –32.1 (c = 0.70,
CH2Cl2). Rf = 0.35 (40% EtOAc in hexanes). 1H NMR (500 MHz,
CDCl3): δ = 7.70 (d, J = 8.2 Hz, 6 H), 7.27 (d, J = 8.2 Hz, 6 H),
3.75 (app q, J = 6.6 Hz, 3 H), 2.95 (dd, J = 13.6 and 6.7 Hz Hz, 3
2838
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 2835–2840

A General Method for the Preparation of Chiral TREN Derivatives
FULL PAPER
H), 2.70 (s, 9 H), 2.41 (s, 9 H), 2.07 (dd, J = 13.6 and 6.7 Hz Hz,
3 H), 1.79 (octet, J = 6.6 Hz, 3 H), 0.97 (d, J = 6.7 Hz, 9 H), 0.85
Compound 10: Cyclohexanecarboxaldehyde (0.16 mL, 1.32 mmol)
was added to tetraamine 1a (88.5 mg, 0.33 mmol) in methanol
(d, J = 6.9 Hz, 9 H) ppm. 13C NMR (125.8 MHz, CDCl3): δ = (2 mL) and the reaction mixture was stirred at room temperature
143.2, 137.3, 129.6, 127.5, 60.0, 55.2, 30.5, 30.2, 21.7, 20.6,
19.8 ppm. C39H60N4O6S3 (777.12): calcd. C 60.28, H 7.78, N 7.21;
found C 60.36, H 7.81, N 7.17.
for 3 h. Sodium borohydride (91 mg, 2.41 mmol) was added and
the reaction mixture stirred at room temperature for an additional
3 d. Water (3 mL) was carefully added and the methanol was evap-
orated. The aqueous phase was extracted with CH2Cl2 (5×3 mL)
and the organic phase was washed with brine (5 mL) and dried
with MgSO4. The solvent was removed and the resulting crude pro-
duct was purified by column chromatography (silica gel, hexane/
EtOAc, 7:3 + 1% triethylamine) to yield tetraamine 10 (136.4 mg,
75%) as a colourless solid. Rf = 0.4 (hexane/EtOAc, 7:3 + 1% tri-
Tris(sulfonamide) 7b: Compound 7b was prepared analogously to
compound 7a. Trissulfonamide 6b (1.51 g, 2.32 mmol) gave tri-
methylated 7b as a white solid in 99% crude yield (1.59 g). The
solid was pure enough to be used directly in the subsequent depro-
tection step. An analytical sample was obtained by flash
chromatography (silica gel, 40% EtOAc in hexanes) in 91% yield
(1.45 g). M.p. 57–59 °C. [α]2D0 = –55.3 (c = 0.90, CHCl3). Rf = 0.35
ethylamine). [α]2D0
=
+142.5 (c = 0.57, CH2Cl2). 1H NMR
(400 MHz, CDCl3): δ = 2.39–2.28 (m, 4 H), 2.38 (dd, J = 12.3,
1.9 Hz, 1 H), 1.91–1.81 (m, 2 H), 1.75–1.67 (m, 4 H), 1.36–1.15 (m,
5 H), 0.92–0.82 (m, 8 H) ppm. 13C NMR (101 MHz, CDCl3): δ =
60.6, 56.6, 55.7, 39.3, 32.1, 28.8, 27.1, 26.4, 26.3, 19.2, 16.8 ppm.
1
(30% EtOAc in hexanes). H NMR (500 MHz, CDCl3): δ = 7.67
(d, J = 8.2 Hz, 6 H) 7.30 (d, J = 8.1 Hz, 6 H), 4.00 (app sextet, J
= 6.7 Hz, 3 H), 2.72 (s, 9 H), 2.51 (dd, J = 13.0 and 5.8 Hz Hz, 3
H), 2.42 (s, 9 H), 2.37 (dd, J = 12.9 and 8.4 Hz Hz, 3 H), 0.84 (d,
J = 6.6 Hz, 9 H) ppm. 13C NMR (125.8 MHz, CDCl3): δ = 143.3,
137.1, 129.8, 127.2, 59.2, 51.0, 28.4, 21.7, 15.1 ppm. C33H48N4O6S3
(692.96): calcd. C 57.20, H 6.98, N 8.09; found C 57.39, H 7.10, N
8.02.
Compound 11: N-Arylation was achieved by a modified literature
procedure.[35] The Pd catalyst was pre-formed by vigorously stirring
Pd2(dba)3 (27.6 mg, 0.03 mmol) and rac-BINAP (56 mg,
0.09 mmol) in toluene (11 mL) at room temperature for 3.5 h. A
separate flask was charged with 1a (544 mg, 2 mmol), bromoben-
zene (0.63 mL, 6 mmol), sodium tert-butoxide (669 mg,
6.96 mmol), and toluene (11 mL). The wine-red catalyst solution
was added and the reaction mixture was heated at 80 °C for 20 h.
The precipitate (NaBr) was removed by centrifugation. The organic
phase was concentrated and the crude product was purified by flash
chromatography (silica gel, hexane/EtOAc, 9:1 + 0.5% Et3N) to
give 11 (687 mg, 69%) as a light yellow solid. Rf = 0.30 (hexane/
EtOAc, 9:1 + 1.0% Et3N and 1.0% MeOH). [α]2D0 = –181.4 (c =
0.30, CHCl3). M.p. 87–89 °C. 1H NMR (400 MHz, CDCl3): δ =
7.11 (t, J = 7.9 Hz, 6 H), 6.64 (t, J = 7.2 Hz, 3 H), 6.26 (d, J =
7.8 Hz, 6 H), 3.53 (d, J = 5.5 Hz, 3 H), 3.43–3.28 (m, 3 H), 2.54
[dd (resolution enhancement), J = 12.7 and 11.0 Hz Hz, 3 H], 2.31
(dd, J = 12.7 and 3.9 Hz Hz, 3 H), 1.99–1.82 (m, J = 7.1 and
3.8 Hz Hz, 3 H), 0.89 (d, J = 7.1 Hz, 9 H), 0.79 (d, J = 7.1 Hz, 9
H) ppm. 13C NMR (100.6 MHz, CDCl3): δ = 149.0, 129.7, 117.2,
113.4, 54.9, 54.4, 30.4, 18.8, 18.2 ppm. C33H48N4 (500.76): calcd.
C 79.15, H 9.66, N 11.19; found C 78.91, H 9.81, N 11.09.
Tetraamine 8a:
A mixture of tris(sulfonamide) 7a (149 mg,
0.19 mmol), phenol (156 mg, 1.66 mmol), and HBr (2.3 mL, 48%
in water) was refluxed for 24 h. Water and NaOH (s) were carefully
added to the cooled dark red mixture until pH Ϸ 1. The aqueous
phase was washed with EtOAc (10×8 mL), the pH was increased
to Ͼ 13 with NaOH (s), and the aqueous phase was extracted with
CH2Cl2 (5×8 mL). The combined organic phases were dried with
Na2SO4, filtered, and concentrated in vacuo leaving a pink oil. Ad-
dition of diethyl ether caused formation of a pink precipitate, which
was filtered off leaving 8a as a colourless oil in 89% yield (54 mg).
Spectral data were in accordance with those previously pub-
lished.[11]
Tetraamine 8b: Tetraamine 8b was prepared analogously to com-
pound 8a from tris(sulfonamide) 7b (4.34 g, 6.26 mmol). Distil-
lation (110 °C/0.05 Torr) gave the gave tetraamine 8b in 60% yield
(0.87 g) as a colourless oil which partially solidified upon standing.
1
[α]2D0 = +155 (c = 0.26, CHCl3). H NMR (500 MHz, CDCl3): δ =
2.62-2.58 (m, 3 H), 2.41 (s, 9 H), 2.33 (dd, J = 12.9 and 10.0 Hz Hz,
3 H), 2.19 (dd, J = 12.9 and 3.2 Hz Hz, 3 H), 1.75 (br. s, 3 H), 0.95
(d, J = 6.2 Hz, 9 H) ppm. 13C NMR (125.8 MHz, CDCl3): δ =
63.2, 53.4, 35.4, 19.5 ppm. C12H30N4 (230.39): calcd: C 62.56, H
13.12, N 24.32; found C 62.43, H 12.95, N 24.17.
Compound 12: This compound was synthesized analogously to 11
using Pd2(dba)3 (201 mg, 0.22 mmol), rac-BINAP (410 mg,
0.66 mmol), 1a (996 mg, 3.66 mmol), 2-bromotoluene (13.13 g,
76.75 mmol), sodium tert-butoxide (2.63 g, 27.41 mmol), and tolu-
ene (146 mL). The reaction mixture was heated at 100 °C for 96 h.
The crude product was purified by flash chromatography with a
gradient eluent (silica gel, hexane/EtOAc, 9:1 + 0.5% Et3N and 0–
1.0% MeOH) to give 12 (1.65 g, 83%) as a light yellow solid. Rf =
0.34 (hexane/EtOAc, 85:15 + 1.0% Et3N and 1.0% MeOH). [α]2D0
= –74.9 (c = 0.90, CHCl3). M.p. 82–84 °C. 1H NMR (400 MHz,
CDCl3): δ = 7.07 (t, J = 7.2 Hz, 3 H), 6.97 (d, J = 7.3 Hz, 3 H),
6.64–6.53 (m, 6 H), 3.58–3.45 (m, 6 H), 2.64 (dd, J = 13.2 and
8.2 Hz Hz, 3 H), 2.57 (dd, J = 13.4 and 5.3 Hz Hz, 3 H), 2.24–2.13
(m, J = 7.0 and 4.1 Hz Hz, 3 H), 1.81 (s, 9 H), 0.97 (d, J = 7.0 Hz,
9 H), 0.84 (d, J = 7.0 Hz, 9 H) ppm. 13C NMR (125.8 MHz,
CDCl3): δ = 146.2, 131.3, 130.7, 128.0, 127.4, 122.9, 116.9, 116.6,
110.4, 56.4, 55.3, 31.7, 30.2, 29.7, 29.1, 18.2, 18.0, 17.5 ppm.
C36H54N4 (542.84): calcd. C 79.65, H 10.03, N 10.32; found C
79.47, H 9.92, N 10.22.
Compound 9: Benzaldehyde (0.14 mL, 1.38 mmol) was added to tet-
raamine 1a (89.7 mg, 0.33 mmol) in methanol (1.5 mL) and the re-
action mixture was stirred at room temperature for 2 h. Sodium
borohydride (92.4 mg, 2.44 mmol) was added and stirring was con-
tinued for an additional 20 h. Water (2 mL) was then carefully
added and the methanol was evaporated on the rotary evaporator.
The aqueous phase was extracted with CH2Cl2 (5×2 mL). The or-
ganic phase was washed with brine (5 mL) and dried with MgSO4.
The solvent was removed, the resulting crude product was dissolved
in pentane and the formed colourless solid was filtered off. The
filtrate was concentrated and the residue purified by column
chromatography (silica gel, hexane/EtOAc, 7:3 + 1% triethylamine)
to yield 9a (127.8 mg, 72%) as a colourless solid. [α]2D0 = +152.4 (c
= 0.88, CH2Cl2) (this value was determined from an analytical sam-
ple and is higher than that previously published[11]). 1H NMR and
13C NMR spectroscopic data were in agreement with those pre-
viously published.[11] C36H54N4 (542.84): calcd. C 79.65, H 10.03,
N 10.32; found C 79.53, H 10.07, 10.29.
Compound 13: This compound was synthesized analogously to 11
using Pd2(dba)3 (24.7 mg, 0.027 mmol), rac-BINAP (50.4 mg,
0.081 mmol), 1a (493 mg, 1.81 mmol), 4-bromobenzonitrile
(988 mg, 5.43 mmol), sodium tert-butoxide (606 mg, 6.3 mmol) and
Eur. J. Org. Chem. 2005, 2835–2840
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2839

Y. Pei, K. Brade, E. Brulé, L. Hagberg, F. Lake, C. Moberg
FULL PAPER
[9] a) M. K. Thompson, M. Botta, G. Nicolle, L. Helm, S. Aime,
A. E. Merbach, K. N. Raymond, J. Am. Chem. Soc. 2003, 125,
14274–14275; b) G. Serratrice, F. Biaso, F. Thomas, C. Béguin,
Eur. J. Inorg. Chem. 2004, 1552–1565; c) C. Bazzicalupi, A.
Bencini, E. Berni, A. Bianchi, S. Ciattini, C. Giorgi, S. Maoggi,
P. Paoletti, B. Valtancoli, J. Org. Chem. 2002, 67, 9107–9110;
d) M. Heitzmann, J.-C. Moutet, J. Pécaut, O. Reynes, G. Royal,
E. Saint-Aman, G. Serratrice, Eur. J. Inorg. Chem. 2003, 3767–
3773.
[10] C. Moberg, Angew. Chem. Int. Ed. 1998, 37, 248–268.
[11] M. Cernerud, H. Adolfsson, C. Moberg, Tetrahedron: Asym-
metry 1997, 8, 2655–2662.
[12] K. Ishihara, Y. Karumi, S. Kondo, H. Yamamoto, J. Org.
Chem. 1998, 63, 5692–5695.
toluene (18.8 mL). The reaction mixture was heated at 80 °C for
24 h to yield a dark brown slurry. Toluene was removed and the
residue was dissolved in CHCl3. The precipitated fine powder
(NaBr) was removed by centrifugation. The organic phase was con-
centrated and the crude product was purified by flash chromatog-
raphy (hexane/EtOAc, 9:1 + 0.5% Et3N, followed by CHCl3) to
give 13 (949 mg, 91%) as a light yellow solid. Rf = 0.1 (hexane/
EtOAc, 3:1 + 2.0% Et3N and 2.0% MeOH). [α]2D0 = –722 (c = 0.24,
1
CHCl3). M.p. 112–114 °C. H NMR (400 MHz, CDCl3): δ = 7.36
(d, J = 8.6 Hz, 6 H), 6.14 (d, J = 8.6 Hz, 6 H), 3.90 (d, J = 7.6 Hz,
3 H), 3.41 (ddd, J = 10.9, 4.1 and 4.0 Hz, 3 H), 2.49 (dd (resolution
enhancement), J = 13.2 and 10.9 Hz Hz, 3 H), 2.41 (dd, J = 13.2
and 4.0 Hz Hz, 3 H), 1.97–1.83 (m, J = 6.9 and 4.1 Hz Hz, 3 H),
0.91 (d, J = 6.9 Hz, 9 H), 0.84 (d, J = 6.9 Hz, 9 H) ppm. 13C NMR
(100.6 MHz, CDCl3): δ = 151.7, 134.4, 120.5, 112.7, 99.0, 54.6,
30.6, 18.6, 18.3 ppm. C36H45N7 (575.79): calcd. C 75.09, H 7.88, N
17.03; found C 74.85, H 7.77, N 16.76.
[13] X. Liu, P. Ilankumaran, I. A. Guzei, J. G. Verkade, J. Org.
Chem. 2000, 65, 701–706.
[14] S. P. Hajela, A. R. Johnson, J. Xu, C. J. Sunderland, S. M. Co-
hen, D. L. Caulder, K. N. Raymond, Inorg. Chem. 2001, 40,
3208–3216.
[15] J. You, A. E. Wróblewski, J. G. Verkade, Tetrahedron 2004, 60,
7877–7883.
[16] X. E. Hu, Tetrahedron 2004, 60, 2701–2743.
[17] H. M. I. Osborn, J. Sweeney, Tetrahedron: Asymmetry 1997, 8,
1693–1715.
Acknowledgments
[18] B. M. Kim, S. M. So, H. J. Choi, Org. Lett. 2002, 4, 949–952.
[19] G. Argouarch, C. L. Gibson, G. Stones, D. C. Sherrington, Tet-
rahedron Lett. 2002, 43, 3795–3798.
[20] M. J. McKennon, A. I. Meyers, K. Drauz, M. Schwarm, J. Org.
Chem. 1993, 58, 3568–3571.
The project was financed by EU grant HPRN-CT-2001–00187 and
by the Göran Gustafsson Foundation for Research in Sciences and
Medicine.
[21] G. A. Smith, R. E. Gawley, Org. Synth. 1985, 63, 136–139.
[22] M. B. Berry, D. Craig, Synlett 1992, 41–44.
[23] O. Mitsunobu, Synthesis 1981, 1–28.
[24] F. Lake, C. Moberg, Eur. J. Org. Chem. 2002, 3179–3188.
[25] R. Alul, M. B. Cleaver, J.-S. Taylor, Inorg. Chem. 1992, 31,
3636–3646.
[1]
[2]
R. R. Schrock, Acc. Chem. Res. 1997, 30, 9–16.
a) J. G. Verkade, Acc. Chem. Res. 1993, 26, 483–489; b) J.
Pinkas, J. Tang, Y. Wan, J. G. Verkade, Phosphorus, Sulfur, Sili-
con 1994, 87, 193–207.
a) J. G. Verkade, P. B. Kisanga, Tetrahedron 2003, 59, 7819–
7858; b) J. G. Verkade, Topics Curr. Chem. 2003, 223, 1–44.
a) S. Urgaonkar, M. Nagarajan, J. G. Verkade, Tetrahedron
Lett. 2002, 43, 8921–8924; b) J. You, J. G. Verkade, Angew.
Chem. Int. Ed. 2003, 42, 5051–5053; c) J. You, J. Xu, J. G. Verk-
[3]
[4]
[26] P. E. Maligres, M. M. See, D. Askin, P. J. Reider, Tetrahedron
Lett. 1997, 38, 5253–5256.
[27] T. Fukuyama, C.-K. Jow, M. Cheung, Tetrahedron Lett. 1995,
36, 6373–6374.
ade, Angew. Chem. Int. Ed. 2003, 42, 5054–5056; d) J. You, J. G. [28] D. A. Alonso, P. G. Andersson, J. Org. Chem. 1998, 63, 9455–
Verkade, J. Org. Chem. 2003, 68, 8003–8007; e) S. Urgaonkar,
J.-H. Xu, J. G. Verkade, J. Org. Chem. 2003, 68, 8416–8423; f)
S. Urgaonkar, M. Nagarajan, J. G. Verkade, Org. Lett. 2003, 5,
815–818; g) W. Su, S. Urgaonkar, J. G. Verkade, Org. Lett.
2004, 6, 1421–1424.
9461.
[29] R. S. Compagnone, H. Rapoport, J. Org. Chem. 1986, 51,
1713–1719.
[30] This compound has previously been prepared as its trishyd-
rochloride salt, but no experimental details or characterization
data were given (ref. 14).
[31] M. Cernerud, A. Skrinning, I. Bérgère, C. Moberg, Tetrahe-
dron: Asymmetry 1997, 8, 3437–3441.
[32] J. E. W. Scheuermann, G. Ilyashenko, D. Vaughan Griffiths, M.
Watkinson, Tetrahedron: Asymmetry 2002, 13, 269–272.
[33] E. Brulé, Y. Pei, F. Lake, F. Rahm, C. Moberg, Mendeleev
Commun. 2004, 14, 276–277.
[5] G. E. Greco, A. I. Popa, R. R. Schrock, Organometallics 1998,
17, 5591–5593.
[6] a) D. V. Yandulov, R. R. Schrock, J. Am. Chem. Soc. 2002, 124,
6252–6253; b) D. V. Yandulov, R. R. Schrock, Science 2003,
301, 76–78; c) V. Ritleng, D. V. Yandulov, W. W. Weare, R. R.
Schrock, A. S. Hock, W. M. Davis, J. Am. Chem. Soc. 2004,
126, 6150–6163.
[7] M. M. Ibrahim, K. Ichikawa, M. Shiro, Inorg. Chem. Commun.
2003, 6, 1030–1034.
[8] M. A. Hossain, J. A. Liljegren, D. Powell, K. Bowman-James,
Inorg. Chem. 2004, 43, 3751–3755.
[34] G. Argouarch, G. Stones, C. L. Gibson, A. R. Kennedy, D. C.
Sherrington, Org. Biomol. Chem. 2003, 1, 4408–4417.
[35] G. E. Greco, R. R. Schrock, Inorg. Chem. 2001, 40, 3850–3860.
Received: February 4, 2005
2840
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 2835–2840