P[(S,S,S)-CH3NCH(CH2Ph)CH2]3N: A new C3-symmetric enantiomerically pure proazaphosphatrane

Article abstract of DOI:10.1016/j.tet.2004.06.115

A new approach for the synthesis of C₃-symmetric proazaphosphatranes with chirality at the bridging methylene positions is reported. Graphical abstract

Full text of DOI:10.1016/j.tet.2004.06.115

Tetrahedron 60 (2004) 7877–7883
P[(S,S,S)-CH3NCH(CH2Ph)CH2]3N: a new C3-symmetric
enantiomerically pure proazaphosphatrane
†
Jingsong You, Andrzej E. Wroblewski and John G. Verkade
*
´
Department of Chemistry, 1275 Gilman Hall, Iowa State University, Ames, IA 50011, USA
Received 17 March 2004; revised 28 May 2004; accepted 10 June 2004
Available online 23 July 2004
Abstract—A new approach for the synthesis of C3-symmetric proazaphosphatranes with chirality at the bridging methylene positions is
reported.
q 2004 Elsevier Ltd. All rights reserved.
During the last ten years bicyclic proazaphosphatranes
2a–g, first synthesized in our laboratories, have proven to be
versatile very strong non-ionic bases and potent catalysts for
a variety of useful base-catalyzed reactions.1 These
transformations appear to be quite dependent upon the
occurrence of transannulation of the bridgehead nitrogen
lone pair to phosphorus that enhances the basicity of these
phosphines and the stability of reaction intermediates.
Moreover, the frameworks of these phosphines are fairly
rigid but strain-free in a bicyclic (approximately C3v)
structure, thus augmenting the lone pair electron density at
phosphorus. Additionally, the electronic and steric proper-
ties of these molecules can be easily tuned by introducing
suitable organic substituents at each N–P nitrogen. We have
used them successfully for the synthesis of b-nitroalkanols,2
b-hydroxy nitriles,3 silylated alcohols,4 glutaronitriles,5
oxazolines,6 a novel ylide,7 a, b-unsaturated nitriles8 and
homoallylic alcohols.9 These non-ionic bases have also
been successfully employed in the promotion of oxa-
Michael addition reactions,6 Knoevenagel condensations,10
Wittig olefination reactions11 and 1,2-addition reactions of
activated allylic synthons.12 Very recently, we reported that
bicyclic P(i-BuNCH2CH2)3N 2d serves as an effective
ligand for palladium-catalyzed Suzuki cross-coupling,13
aminations14,15 and a-arylations of nitriles16 with a wide
array of aryl chlorides, bromides and iodides. These
achievements stimulated our interest in exploring the
possibility of realizing asymmetric versions of some these
reactions.
Recently, we synthesized the C3-symmetric proazaphos-
phatrane 3 possessing non-racemic substituents at the PN3
nitrogen atoms.17 Enantiomerically pure 3, derived from
(S)-a-phenylethylamine, was found to be an efficient
derivatizing agent for the direct determination of enantio-
1
meric ratios of chiral azides by means of 31P and H NMR
spectroscopy.17 Derivatives of azaphosphatranes with non-
racemic substituents at the PN3 nitrogen atoms may not
exert good stereocontrol in catalytic reactions, because of
the flexibility of the C3 conformations of the cage moieties.
Keywords: Chiral; Proazaphosphatrane; Synthesis; Non-ionic base.
* Corresponding author. Tel.: C1-515-2945023; fax: C1-515-2940105; e-mail: jverkade@iastate.edu
† On leave of absence from the Medical University of Lodz, Poland.
0040–4020/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2004.06.115

7878
J. You et al. / Tetrahedron 60 (2004) 7877–7883
On the other hand, azaphosphatrane derivatives substituted
at either methylene position of the three CH2CH2 moieties
could be expected to be more rigid owing to intramolecular
steric hindrance on the three bridges of a C3-symmetric cage
and hence, may possibly be advantageous for efficient
chirality transfer in catalytic applications. Yamamoto’s
group successfully accomplished the synthesis of proaza-
phosphatrane 4 from (S)-proline.18 However, this compound
showed no asymmetric induction in the incomplete silyla-
tion of racemic 1-phenylethanol and when 4 was applied as
a catalyst in the addition of diethylzinc to benzaldehyde,
1-phenyl-1-propanol was obtained with a low ee value
(15%). Additionally, the synthesis of 4 is difficult and it
precludes the introduction of organic substituents on the
PN3 nitrogens aimed at fine tuning the electronic and steric
properties of 4. Moberg et al. have recently synthesized
C3-symmetric proazaphosphatranes19 from non-racemic
C3-symmetric TREN analogs, which were prepared from
enantiomerically pure aminoalcohols via aziridines.20
However, they were unable to purify the hydrochlorides
1h and 1i (ca. 90%), or the respective proazaphosphatranes
2h and 2i (ca. 70% purity) and, surprisingly, observed
incomplete deprotonation of the former even in the presence
of potassium tert-butoxide in acetonitrile.
in a multistep synthesis which included three LiAlH4
reductions, two condensations, and also protection and
deprotection steps.18 The most promising approach is the
third reported method, which depends on the reductive
amination of N-Boc-protected non-racemic a-amino-
aldehydes in the presence of ammonium acetate.21 Herein,
an extension of this methodology to the synthesis of
proazaphosphatrane 1j is presented.
1. Results and discussion
Our initial exploration of the synthesis of C(b)-substituted
enantiomerically pure TREN 10 is shown in Scheme 1.
In the first step, L-phenylalanine was readily converted to
N-Boc-L-phenylalanine methyl ester 7 in nearly quantitative
yield. The transformation of the ester 7 into aldehyde 8
was carried out in 90% yield in the presence of DIBAL-H at
K78 8C for 2 h. The 1H NMR spectrum of aldehyde 8 was
consistent with that previously reported for this com-
pound.22 Due to concern over the configurational stability of
aldehyde 8, the product was used immediately without
further purification in the subsequent reductive amination
reaction with NaHB(OAc)3,23 affording (S,S,S)-tris(N-tert-
butoxycarbonyl-2-amino-3-phenylpropyl)amine 9 as a
white solid in 50% yield. Vigorous stirring was critical for
achieving the highest possible yield. Unfortunately, the
subsequent reduction of 9 with LiAlH4 gave (S,S,S)-tris-
[2-(N-methylamino)-3-phenylpropyl]amine 10 in only 26%
yield.
It was very difficult to purify tetraamine 10 by chromato-
graphy on a silica gel column. However, when the crude
amine 10 was treated with 10% hydrochloric acid, the
corresponding hydrochloride (that separated immediately)
was easily purified by recrystallization from methanol/ether
to give pure (S,S,S)-tris[2-(N-methylamino)-3-phenylpro-
pyl]amine trihydrochloride 11 as white needles. Compound
11 was then neutralized with 20% aqueous KOH to produce
pure amine 10 quantitatively.
Known methods for the synthesis of tripodal tris(amino-
alkyl)amines possessing chiral centers at the carbon beta to
the tertiary nitrogen rely on the use of aminoacids as starting
materials. The first such method involved nucleophilic
opening of aziridines derived from b-aminoalcohols with
ammonia,20 while in the second route (S)-proline was used
Scheme 1.

J. You et al. / Tetrahedron 60 (2004) 7877–7883
7879
Scheme 2.
Although the use of the N-Boc group as a precursor to the
MeNH substituent afforded 10 in low yield, the use of
inexpensive ethyl chloroformate24 led to a significant
improvement of the overall yield (Scheme 2). Reductive
amination of 1325 gave (S,S,S)-tris[2-(N-ethoxycarbonyl-
amino)-3-phenylpropyl]amine 14 in 76% yield. Although
N-ethoxycarbonyl-(S)-phenylalaninol26,27 formed in vari-
able amounts as a by-product in the reductive amination
step, it was easily removed on a silica gel column. LiAlH4
reduction of 14 gave the desired compound 10 in 79% yield.
the case of the N-ethoxycarbonyl-protected TREN deriva-
tive 14, we found that all three carboxyethyl groups could be
removed quantitatively by refluxing in ethanol/aqueous
KOH for 20 h (Scheme 4). The successful preparation of
15 provides a route to the synthesis of a variety of
enantiomerically pure proazaphosphatranes by allowing
facile introduction of suitable organic substituents at each
PN3 nitrogen as described previously.28 Because the parent
proazaphophatrane P(HNCH2CH2)3N is very unstable with
respect to polymerization,1 we did not attempt to convert 15
to its corresponding proazaphosphatrane.
The intermediate N-protected a-aminoaldehydes 8 and 13
appeared to be configurationally stable under the conditions
of the reductive amination, since detailed analyses of the 1H
NMR spectra of the crude products 9 and especially 14
showed no presence of additional resonances that could be
assigned to diastereoisomeric (R,S,S)-9 or (R,S,S)-14.
When proazaphosphatrane 2j (0.1 equiv.) was used to
catalyze the addition of trimethylsilyl cyanide to benz-
aldehyde,29 mandelonitrile was obtained in 47% yield. The
product appeared to be a racemic mixture as shown by
derivatization with (4R,5R)-diisopropoxycarbonyl-2-
chloro-1,3,2-dioxaphospholane.30 Although quinine was
found useful as an enantiodifferentiating medium31 for
mandelonitrile, the estimation of the ee was not possible,
because in the presence of excess quinine, a retro-
cyanohydrin reaction occurred leading to the formation of
ca. 20% benzaldehyde within 2 h.
Azaphosphatrane 1j was synthesized according to our
established procedure for analogues of this compound as
shown in Scheme 3.24,28 The crude product was purified on
a silica gel column followed by crystallization from benzene
to give enantiomerically pure 1j in 71% yield. In the
presence of potassium tert-butoxide in THF, 1j was
converted in 2 h to proazaphosphatrane 2j in 82% yield.
The rearrangement of meso-epoxides to allylic alcohols is
an important transformation in organic chemistry since
chiral products are formed. Enantioselective versions of this
reaction using a variety of non-racemic ionic bases have
been extensively investigated in recent years32–35 and the
enantioselective deprotonation of cyclohexene oxide
appeared to be a good model reaction to try.32,36–39
Additionally, the synthesis of the N-unsubstituted tetra-
amine 15 was also developed as outlined in Scheme 4.
Treatment of 9 with trifluoroacetic acid for several hours
removed the Boc protecting groups. Subsequent neutrali-
zation with base afforded tetraamine 15 in excellent yield. In
Scheme 3.
Scheme 4.

7880
J. You et al. / Tetrahedron 60 (2004) 7877–7883
However, when cyclohexene oxide was treated with 2j
(0.2 equiv.) at room temperature for 27 h, only a negligible
amount of 2-cyclohexenol was detected.
ester 7 as a colorless oil (20.0 g, 98%). To a cold (K78 8C)
solution of 7 (5.10 g, 18.3 mmol) in CH2Cl2 (60 mL) was
added dropwise a 1.0 M solution of DIBAL-H in hexane
(40 mL, 40 mmol) over 1 h. After 2 h, the reaction mixture
was quenched with AcOH (34 mL, 5 M in benzene) at
K78 8C and then warmed to room temperature. The mixture
was poured into 10% aqueous tartaric acid (100 mL) and
extracted with hexane/AcOEt (1:1) (3!60 mL). The
combined organic layers were washed with brine and
dried over Na2SO4. The solvent was removed to leave crude
N-Boc-L-phenylalaninal 8 as a white solid (4.5 g), which
usually was immediately used in the next step without
further purification. This material was once purified by
chromatography on silica gel using AcOEt/hexane (1:3, v/v)
Proazaphosphatrane 2j was also unsuccessful in catalyzing
the addition of allyl cyanide to benzaldehyde, leading
instead to a complex mixture of products. However, it was
pointed out that 2b also gave low conversions in this
transformation.12
In summary, a general route to C3-symmetric enantiomeric-
ally pure TRENs from L-phenylalanine is reported, thus
facilitating access to a wide variety of enantiomerically pure
proazaphosphatranes. Proazaphosphatrane 2j was found
ineffective in inducing asymmetry in the formation of
mandelonitrile, not sufficiently basic to execute the
rearrangement of cyclohexene oxide to 2-cyclohexenol,
and (not unexpectedly) inefficient in catalyzing the addition
of allyl cyanide to benzaldehyde. The lack of observable ee
in the reaction products reported here may be attributable to
an excessive distance of the chiral centers on 2j to a
substrate in an intermediate that might be formed. It is also
possible that the strong basicity of a proazaphosphatrane
racemizes the chiral center in the product via a deprotona-
tion equilibrium. Results of further experiments with other
PN3 nitrogen-substituted analogues of 2j will be reported in
due course.
1
as eluent to give a white solid (3.64 g, 80%). H NMR
(300 MHz, CDCl3): dZ1.43 (s, 9H, C(CH3)3), 3.11 (d, JZ
6.5 Hz, 2H, PhCH2), 4.40 (m, 1H, NHCHCH2Ph), 5.04 (br s,
1H, NH), 7.16–7.31 (m, 5H, Ph), 9.63 (s, 1H, CHO).
2.1.2. (S,S,S)-Tris(N-tert-butoxycarbonyl-2-amino-3-
phenylpropyl)amine 9. N-Boc-L-phenylalaninal 8 (4.88 g,
19.6 mmol, 4.0 equiv.), ammonium acetate (0.38 g,
4.9 mmol, 1.0 equiv.) and NaHB(OAc)3 (6.23 g,
29.4 mmol, 6.0 equiv.) were placed in a flask under an
argon atmosphere and THF (200 mL) was added via canula.
The reaction mixture was stirred overnight and then was
quenched with 10% AcOH/MeOH (50 mL). The solvent
was removed in vacuo, and the residue was dissolved in
methylene chloride (120 mL) and washed with KOH (4%,
2!50 mL). The organic phase was washed with brine, dried
over Na2SO4 and evaporated to dryness leaving a yellow
solid. The residue was first purified by chromatography on
silica gel with ether/hexanes (1:1, v/v) as eluent and then
recrystallized from ether/hexanes to give the tetraamine 9 as
2. Experimental
2.1. General methods
1H and 13C NMR spectra were recorded at 400 or 300 MHz,
and at 100.6 or 75.5 MHz, respectively. Elemental analyses
were performed by Desert Analytics (Tucson, Arizona).
Mass spectra were recorded on a Kratos MS 50 instrument.
Melting points were determined in unsealed capillary tubes
and are uncorrected. All reactions were performed under an
argon atmosphere. Solvents were dried over P4O10 (dichloro-
methane), CaH2 (acetonitrile) or sodium/benzophenone
(diethyl ether, THF, dioxane and toluene) and were freshly
distilled prior to use. Chemicals employed were purchased
from Aldrich Chemical Co. and were used without further
purification.
1
a white foamy solid (1.76 g, 50%). Mp 108–110 8C. H
NMR (300 MHz, CDCl3): dZ1.39 (s, 27H, 3 C(CH3)3),
2.20 (dd, JZ13.0, 4.1 Hz, 3H, 3 NCHaHb), 2.52 (dd, JZ
13.0, 9.5 Hz, 3H, 3 NCHaHb), 2.68 (dd, JZ13.7, 6.6 Hz,
3H, 3 PhCHaHb), 2.79 (dd, JZ13.7, 6.7 Hz, 3H, 3
PhCHaHb), 3.87 (m, 3H, 3 CHNH), 4.83 (brs, 3H, 3 NH),
7.14–7.28 (m, 15H, 3 Ph). 13C NMR (100.6 MHz, CDCl3):
dZ28.8, 40.4, 49.8, 57.8, 79.5, 126.5, 128.6, 129.4, 138.44,
156.1.
2.1.3. N-Ethoxycarbonyl-L-phenylalaninal 13. L-Phenyl-
alanine methyl ester hydrochloride 6 (20.2 g, 93.6 mmol)
was dissolved in water (300 mL) and NaHCO3 (38.3 g,
456 mmol) was added portion-wise. Ethyl chloroformate
(25.86 mL, 270.6 mmol) was then added drop wise with
vigorous stirring at 0 8C. After 5 h the mixture was allowed
to warm to room temperature and was allowed to stir
overnight. The mixture was extracted with AcOEt (3!
100 mL). The organic layer was dried over MgSO4 and the
solvent was removed under reduced pressure to leave
N-ethoxycarbonyl-L-phenylalanine methyl ester 12 as a
colorless oil in a quantitative yield. To a cold (K78 8C)
solution of 12 (16.22 g, 64.61 mmol) in CH2Cl2 (150 mL)
was added drop wise a 1.0 M solution of DIBAL-H in
hexane (100 mL, 100 mmol) over 1 h. After 2 h, the
reaction mixture was quenched with AcOH (64 mL, 5 M
in benzene) at K78 8C and then warmed to room
temperature. The mixture was poured into 10% aqueous
tartaric acid (200 mL) and extracted with hexane/AcOEt
2.1.1. N-Boc-L-phenylalaninal 8. To a suspension of
L-phenylalanine 5 (10.0 g, 60.5 mmol) in ice-cooled dry
methanol (100 mL) was added dropwise thionyl chloride
(10.0 g, 6.15 mL, 84.2 mmol) and the solution was stirred at
room temperature overnight. Removal of the volatiles under
reduced pressure gave L-phenylalanine methyl ester hydro-
chloride 6 as a colorless crystalline solid quantitatively. To
the mixture of the methyl ester hydrochloride 6 (15.77 g,
73.10 mmol) in 150 mL of aqueous NaHCO3 (13.51 g,
160.8 mmol) was added dropwise a solution of (Boc)2O
(17.55 g, 80.41 mmol) in dioxane (150 mL). The mixture
was stirred at room temperature for 12 h, then concentrated
to half of its original volume and extracted with ether (3!
100 mL). The combined organic layers were dried over
Na2SO4 and concentrated to leave a colorless oil, which was
purified by chromatography on silica gel using ether/hexane
(1:1, v/v) as eluent to give N-Boc-L-phenylalanine methyl

J. You et al. / Tetrahedron 60 (2004) 7877–7883
7881
(1:1, v/v) (3!60 mL). The combined organic layers were
washed with brine and dried over Na2SO4. The solvent was
removed to give crude N-ethoxycarbonyl-L-phenylalaninal
13 as a white solid (14.12 g, 99%), which was immediately
was adjusted to pH 10 by slow addition of 20% aqueous
KOH. The organic phase was dried over MgSO4 and the
solvent was removed under reduced pressure to give 10 as a
1
pale yellow oil (0.5 g, 26%). H NMR (300 MHz, CDCl3):
1
used in the next step without further purification. H NMR
dZ1.8–2.1 (br s, 3H, 3 NH), 2.13–2.24 (m, 6H, 3 NCH2),
2.42 (s, 9H, 3 CH3), 2.44 (dd, JZ13.2, 7.2 Hz, 3H, 3
PhCHaHb), 2.68 (m, 3H, 3 CHNH), 2.81 (dd, JZ13.2,
8.4 Hz, 3H, 3 PhCHaCHb), 7.1–7.3 (m, 15H, 3 Ph). 13C
NMR (75.5 MHz, CDCl3): dZ34.2, 39.0, 59.5, 59.5, 126.3,
128.6, 129.5, 139.4. HRMS (EI) calcd for C30H42N4:
458.34095. Found: 458.34180.
(400 MHz, CDCl3): dZ1.21 (t, JZ7.0 Hz, 3H, OCH2CH3),
3.12 (d, JZ6.6 Hz, 2H, PhCH2), 4.09 (q, JZ7.0 Hz, 2H,
OCH2CH3), 4.48 (q, JZ6.6 Hz, 1H, NHCHCH2Ph), 5.18
(br s, 1H, NH), 7.18–7.32 (m, 5H, Ph), 9.64 (s, 1H, CHO).
2.1.4. (S,S,S)-Tris(N-ethoxycarbonyl-2-amino-3-phenyl-
propyl)amine 14. N-Ethoxycarbonyl-L-phenylalaninal
(5.40 g, 24.4 mmol, 4.0 equiv.), ammonium acetate
(0.47 g, 6.1 mmol, 1.0 equiv.) and NaHB(OAc)3 (6.46 g,
30.5 mmol, 5.0 equiv.) were placed in a flask under an argon
atmosphere and THF (150 mL) was added via canula. The
reaction mixture was stirred overnight at room temperature
and was quenched with 10% AcOH/MeOH (50 mL). The
solvent was removed in vacuo, the residue was dissolved in
methylene chloride (120 mL) and washed with KOH (4%,
2!50 mL). The organic phase was washed with brine, dried
over Na2SO4 and evaporated to dryness leaving a yellow
solid. The residue was first purified by chromatography on
silica gel using ether/hexanes (1:1, v/v) as eluent, and then
recrystallized from ether/hexanes to give the amine 14 as a
(S,S,S)-Tris[2-(N-methylamino)-3-phenylpropyl]amine tri-
1
hydrochloride 11. H NMR (300 MHz, CDCl3): dZ1.95
(d, JZ13.8 Hz, 3H, 3 PhCHaHb), 2.73 (dd, JZ12.9,
11.4 Hz, 3H, 3 NCHaHb), 2.98 (s, 9H, 3 CH3), 3.05 (t, JZ
12.9 Hz, 3H, 3 NCHaHb), 3.33 (d, JZ13.8 Hz, 3H, 3
PhCHaHb), 4.27 (m, 3H, CHNH), 7.2–7.4 (m, 15H, 3 Ph),
9.21 (s, 3H, NHaHCb ), 9.95 (s, 3H, NHaHCb ).
Method B. To a suspension of LiAlH4 (4.50 g, 118 mmol)
in dry THF (150 mL) was added drop wise a solution of
N-ethoxycarbonyl protected amine 14 (9.80 g, 15.5 mmol)
in THF (50 mL) at 0 8C and the reaction mixture was
refluxed overnight. Aqueous KOH [from KOH (7.5 g) and
water (15 mL)] was carefully added at 0 8C. The mixture
was refluxed for 2 h and the solution was decanted from the
inorganic gel. The gel was refluxed with THF (100 mL) for
1 h. Evaporation of solvents from the combined extracts
gave a pale yellow oil, which was treated with a 10%
aqueous HCl (100 mL). The mixture was stirred at room
temperature for 4 h, the solid was filtered and recrystallized
from hot methanol/ether to give the trihydrochloride 11 as
white needles. The solid was dissolved in CH2Cl2 (150 mL)
and H2O (50 mL) was added. The biphasic mixture was
adjusted to pH 10 by slow addition of 20% aqueous KOH.
The organic phase was dried over MgSO4, and the solvent
was removed under reduced pressure to afford 10 as a pale
yellow oil (5.6 g, 79%) identical in all respects with the
material described in Method A.
1
white amorphous solid (2.93 g, 76%). Mp 108–110 8C. H
NMR (300 MHz, CDCl3): d 1.16 (s, JZ7.2 Hz, 9H, 3
OCH2CH3), 2.06 (br d, JZ11.7 Hz, 3H, 3 NCHaHb), 2.49
(br t, JZ12.0 Hz, 3H, 3 NCHaHb), 2.59 (dd, JZ13.8,
7.2 Hz, 3H, 3 PhCHaHb), 2.83 (dd, JZ13.8, 6.3 Hz, 3H, 3
PhCHaCHb), 3.90–4.08 (m, 6H, 3 OCH2CH3), 4.08–4.25
(m, 3H, 3 CHNH), 5.16 (br d, JZ7.5 Hz, 3H, 3 NH), 7.10–
7.25 (m, 15H, 3 Ph). 13C NMR (75.5 MHz, CDCl3): dZ
14.8, 40.5, 49.4, 57.4, 60.9, 126.6, 128.6, 129.3, 138.3,
157.5. Anal. calcd for C36H48N4O6: C, 68.33; H, 7.65; N,
8.85. Found: C, 68.58; H, 7.61; N, 8.73.
From more polar fractions, variable amounts of N-ethoxy-
carbonyl-L-phenylalaninol26,27 were obtained. Mp 66–
67 8C. 1H NMR (300 MHz, CDCl3): dZ1.21 (t, JZ
7.2 Hz, 3H, CH3CH2), 2.1–2.4 (br s, 1H, OH), 2.85 (d,
JZ7.2 Hz, 2H, PhCH2), 3.56 (dAB, JZ11.1, 8.1 Hz, 1H,
CHaHbOH), 3.67 (dAB, JZ11.1, 3.9 Hz, 1H, CHaHbOH),
3.85–3.98 (m, 1H, HCN), 4.08 (q, JZ7.2 Hz, 2H, CH2CH3),
4.85–4.95 (br s, 1H, HN), 7.18–7.34 (m, 5H, Ph).
2.1.6. (3S,7S,10S)-3,7,10-Tribenzyl-2,8,9-trimethyl-
2,8,9-triaza-5-azonia-1l5-phosphatricyclo[3.3.3.01,5]un-
decane chloride 1j. To a solution of P(NMe2)3 (1.51 g,
9.25 mmol) in CH3CN (15 mL), PCl3 (405 mL, 4.67 mmol)
was syringed at 0 8C. The resulting solution was stirred at
0 8C for 30 min and then a solution of (S,S,S)-tris[2-
(N-methylamino)-3-phenylpropyl)amine 10 (5.73 g,
12.5 mmol) in CH3CN (10 mL) was added via canula at
5 8C. The mixture was stirred at room temperature overnight
and the volatiles were removed under vacuum to leave a
white foam, which was purified by chromatography on silica
gel using CHCl3/CH3OH (20:1, v/v) as eluent. The
appropriate fractions were collected and recrystallized
from benzene to give 1j (4.63 g, 71%) as colorless plates.
Mp 84–86 8C. 1H NMR (400 MHz, CDCl3): dZ2.17 (d, JZ
18.0 Hz, 9H, 3 CH3), 2.53 (t, JZ12.4 Hz, 3H, 3 NCHaHb),
2.55 (br d, JZ14.4 Hz, 3H, 3 PhCHaHb), 2.70–2.77 (m, 3H,
3 NCHaHb), 2.85 (dd, JZ14.4, 5.6 Hz, 3H, 3 PhCHaCHb),
3.85 (dt, JZ12.4, 5.6 Hz, 3H, 3 NCH), 4.92 (d, JZ
515.2 Hz, 1H, PH), 6.99–7.02 (m, 6H, Ph), 7.20–7.30 (m,
9H, Ph). 13C NMR (100 MHz, CDCl3): dZ32.1 (d, JZ
14.2 Hz), 35.6 (d, JZ6.7 Hz), 49.8 (d, JZ7.4 Hz), 51.2
2.1.5. (S,S,S)-Tris[2-(N-methylamino)-3-phenylpropyl]-
amine 10. Method A. To a suspension of LiAlH4 (4.14 g,
109 mmol) in dry THF (130 mL) was added drop wise a
solution of the N-Boc protected amine 9 (3.00 g,
4.19 mmol) in THF (30 mL) at 0 8C and the reaction
mixture was refluxed overnight. Aqueous KOH [KOH
(5.0 g) dissolved in H2O (10 mL)] was carefully added at
0 8C. After addition, the mixture was refluxed for 2 h, and
the solution was decanted from the inorganic gel. The gel
was refluxed with THF (100 mL) for 1 h, and the solution
was decanted. Evaporation of the combined extracts gave a
pale yellow oil. The crude 10 was placed in a 10% aqueous
HCl (50 mL) solution, the mixture was stirred at room
temperature for 4 h, and the solid was filtered off. This pale
yellow solid was recrystallized from hot methanol/ether to
give 11 as white needles. The solid was dissolved in CH2Cl2
(80 mL) and H2O (30 mL) was added. The biphasic mixture

7882
J. You et al. / Tetrahedron 60 (2004) 7877–7883
(d, JZ2.1 Hz), 127.2, 128.6, 130.4, 134.9. 31P NMR
(162 MHz, CDCl3): dZK6.76.
2.2. ee Estimation of mandelonitrile
Method A. To a solution of (4R,5R)-diisopropoxycarbonyl-
2-chloro-1,3,2-dioxaphospholane (33.0 mg, 110 mmol) in
chloroform-d (0.7 mL) containing triethylamine (17 mL,
120 mmol), mandelonitrile (12.8 mg, 96.1 mmol) was added
and the reaction mixture was analyzed by 31P NMR
(162 MHz). Two resonances at dZ144.38 and 142.28 ppm
in a 1:1 ratio were observed.
2.1.7. (3S,7S,10S)-3,7,10-Tribenzyl-2,8,9-trimethyl-
2,5,8,9-tetraaza-1-phosphabicyclo[3.3.3]undecane 2j. A
suspension of the hydrochloride 1j (5.20 g, 9.97 mmol) and
potassium tert-butoxide (2.24 g, 19.9 mmol) in THF
(40 mL) was stirred at room temperature for 2 h. The
volatiles were removed in vacuo, and toluene (150 mL) was
added to the residue. The mixture was left without stirring
overnight, inorganic salts were filtered off, and the solvent
was removed in vacuo (0.1 mm Hg) to give proazaphos-
phatrane 2j as a white solid (4.0 g, 82%). 1H NMR
(400 MHz, C6D6): dZ2.25 (dd, JZ14.0, 11.6 Hz, 3H, 3
NCHaHb), 2.30 (dd, JZ14.4, 6.8 Hz, 3H, 3 PhCHaHb), 2.54
(d, JZ10.0 Hz, 9H, 3 NCH3), 2.57 (dd, JZ14.0, 2.4 Hz, 3H,
3 NCHaHb), 2.69 (dd, JZ14.4, 7.8 Hz, 3H, 3 PhCHaCHb),
3.25 (dddd, JZ11.6, 7.8, 6.8, 2.4 Hz, 3H, 3 NCH), 7.04–
7.20 (m, 15H, 3 Ph). 13C NMR (75.5 MHz, C6D6): dZ28.7
(d, JZ34.0 Hz), 36.1 (d, JZ2.3 Hz), 54.4 (d, JZ8.4 Hz),
54.9 (d, JZ3.0 Hz), 126.1, 128.5, 128.7, 141.0. 31P NMR
(162 MHz, C6D6): dZ124.66. MS (EI): m/z (MC, 486).
Anal. calcd for C30H39N4P: C, 74.04; H, 8.08; N, 11.51.
Found: C, 74.65; H, 7.82; N, 10.99.
1
Method B. The H NMR spectrum of a solution of the
racemic mandelonitrile (8.1 mg, 61 mmol) in chloroform-d
(0.7 mL) was recorded and quinine (39.5 mg, 122 mmol)
was added. A clear solution was obtained which was
immediately analyzed by 1H NMR spectroscopy
(400 MHz). Two resonances for the methine proton in
mandelonitrile at dZ5,512 and 5.501 ppm were observed in
a 1:1 ratio.
Acknowledgements
The authors are grateful to the National Science Foundation
for grant support of this research.
2.1.8. (S,S,S)-Tris(2-amino-3-phenylpropyl)amine 15.
Method A. The N-Boc protected amine 9 (2.0 g,
2.8 mmol) was dissolved in trifluoroacetic acid (15 mL)
and stirred at room temperature for 2 h. All volatiles were
evaporated, the residue was dissolved in methylene chloride
(30 mL) and evaporated several times with fresh portions of
this solvent in order to remove excess acid leaving finally a
pale yellow foamy solid. The salt was dissolved in
methylene chloride (50 mL) and water (50 mL) was
added. While stirring, the biphasic solution was adjusted
to pH 10 by slow addition of 20% KOH. The layers were
separated and the aqueous phase was extracted with
additional methylene chloride (50 mL). The combined
organic layers were dried over Na2SO4 and evaporated to
give the tetraamine 15 as a pale yellow solid (1.15 g)
quantitatively. Mp 47–49 8C. 1H NMR (400 MHz, CDCl3):
dZ1.30–1.70 (br s, 6H, 3 NH2), 2.29 (dd, JZ12.5, 2.8 Hz,
3H, 3 NCHaHb), 2.39 (dd, JZ12.5, 10.3 Hz, 6H, 3
NCHaHb), 2.44 (dd, JZ13.3, 8.9 Hz, 3H 3 PhCHaHb),
2.66 (dd, JZ13.3, 4.5 Hz, 3H, 3 PhCHaHb), 3.15–3.30 (m,
3H, 3 CHNH), 7.19–7.29 (m, 15H, 3 Ph). 13C NMR
(75.5 MHz, CDCl3): dZ42.6, 49.9, 61.9, 126.5, 128.7,
129.4, 139.3. HRMS (EI) calcd for C27H36N4: 416.29400,
found 416.29495. Anal. calcd for C27H36N4: C, 77.84; H,
8.71; N, 13.45. Found: C, 77.29; H, 8.70; N, 13.19.
References and notes
1. For recent reviews, see: (a) Verkade, J. G. Top. Curr. Chem.
2003, 233, 1–44. (b) Verkade, J. G.; Kisanga, P. Tetrahedron
2003, 59, 7819–7858. (c) Verkade, J. G. Kisanga. Aldrichi-
mica Acta 2004, 37, 3–14.
2. Kisanga, P.; Verkade, J. G. J. Org. Chem. 1999, 64, 4298–
4303.
3. Kisanga, P.; McLeod, D.; D’Sa, B.; Verkade, J. G. J. Org.
Chem. 1999, 64, 3090–3094.
4. D’Sa, B.; Verkade, J. G. J. Org. Chem. 1996, 61, 2963–2966.
5. Kisanga, P.; D’Sa, B.; Verkade, J. G. J. Org. Chem. 1998, 63,
10057–10059.
6. Kisanga, P.; Ilankumaran, P.; Verkade, J. G. Tetrahedron Lett.
2001, 42, 6263–6266.
7. Wang, Z.; Verkade, J. G. Tetrahedron Lett. 1998, 39, 9331–
9334.
8. D’Sa, B.; Kisanga, P.; Verkade, J. G. J. Org. Chem. 1998, 63,
3691–3967.
9. Wang, Z.; Kisanga, P.; Verkade, J. G. J. Org. Chem. 1999, 64,
6459–6461.
10. McLaughlin, P.; Kisanga, P.; Verkade, J. G. In preparation.
11. Wang, Z.; Verkade, J. G. Heteroatom Chem. 1998, 9, 687–689.
12. Kisanga, P.; Verkade, J. G. J. Org. Chem. 2002, 67, 426–430.
13. Urgaonkar, S.; Nagarajan, M.; Verkade, J. G. Tetrahedron
Lett. 2002, 43, 8921–8924.
Method B. The mixture of the N-ethoxycarbonyl protected
amine 14 (6.0 g, 9.5 mmol), KOH (17.0 g), ethanol (48 mL)
and H2O (12 mL) was refluxed under an argon
atmosphere for 20 h. The volatiles were removed and
chloroform (100 mL) was added. The organic phase was
washed with water (2!50 mL), brine (50 mL) and was
dried over Na2SO4. After evaporation of solvents the
tetraamine 15 was obtained (4.0 g, 100%) as a pale yellow
solid identical in all respects with the material obtained by
Method A.
14. Urgaonkar, S.; Nagarajan, M.; Verkade, J. G. J. Org. Chem.
2003, 68, 452–459.
15. Urgaonkar, S.; Nagarajan, M.; Verkade, J. G. Org. Lett. 2003,
5, 815–818.
16. You, J.; Verkade, J. G. J. Org. Chem. 2003, 68, 8003–8007.
17. Liu, X.; Ilankumaran, P.; Guzei, I. A.; Verkade, J. G. J. Org.
Chem. 2000, 65, 701–706.
18. Ishihara, K.; Karumi, Y.; Kondo, S.; Yamamoto, H. J. Org.
Chem. 1998, 63, 5692–5695.

J. You et al. / Tetrahedron 60 (2004) 7877–7883
7883
19. Lake, F.; Hagberg, L.; Svensson, M.; Moberg, C. Collect.
Czech. Chem. Commun. 2000, 65, 570–576.
29. Wang, Z.; Fetterly, B.; Verkade, J. G. J. Organomet. Chem.
2002, 646, 161–166.
20. Cernerud, M.; Adolfsson, H.; Moberg, C. Tetrahedron:
Asymmetry 1997, 8, 2655–2661.
30. Brunel, J.-M.; Pardigon, O.; Maffei, M.; Buono, G.
Tetrahedron: Asymmetry 1992, 3, 1243–1246.
31. Uccello-Barretta, G.; Pini, D.; Mastantuono, A.; Salvadori, P.
Tetrahedron: Asymmetry 1995, 6, 1965–1975 and references
cited therein.
21. Hajela, S. P.; Johnson, A. R.; Xu, J.; Sunderland, C. J.; Cohen,
S. M.; Caulder, D. L.; Raymond, K. N. Inorg. Chem. 2001, 40,
3208–3216.
22. Salvino, J. M.; Mervic, M.; Mason, H. J.; Kiesow, T.; Teager,
D.; Airey, J.; Labaudiniere, R. J. Org. Chem. 1999, 64, 1823–
1830.
32. Malhotra, S. V. Tetrahedron: Asymmetry 2003, 14, 645–647.
33. Hodgson, D. M.; Cameron, I. D.; Christlieb, M.; Green, R.;
Lee, G. P.; Robinson, L. A. J. Chem Soc., Perkin Trans. 1
2001, 2161–2174.
23. Abdel-Magid, A. F.; Carson, K. G.; Harris, B. D.; Maryanoff,
C. A.; Shah, R. D. J. Org. Chem. 1996, 61, 3849–3862.
24. Schmidt, H.; Lensink, C.; Xi, S. K.; Verkade, J. G. Z. Anorg.
Allg. Chem. 1989, 578, 75–80.
34. Hodgson, D. M.; Robinson, L. A. Chem. Commun. 1999, 309–
310.
35. Soedergren, M. J.; Anderson, P. G. J. Am. Chem. Soc. 1998,
120, 10760–10761.
25. Acton, J. J., III; Adams, A. D.; Hermes, J. D.; Jones, A. B.;
Parsons, W. H.; Sinclair, P. J. U.S. Patent US 5,719,296, 1998;
Chem Abstr. 1998, 128, 1929–1942.
36. Amedjkouh, M.; Pettersen, D.; Lill, S. O. N.; Davidsson, O.;
Ahlberg, P. Chem. Eur. J. 2001, 7, 4368–4377.
37. Khan, A. Z.-Q.; De Groot, R. W.; Arvidsson, P. I.; Davidsson,
O. Tetrahedron: Asymmetry 1998, 9, 1223–1229.
38. Bhuniya, D.; Singh, V. K. Synth. Commun. 1994, 24, 1475–
1481.
26. Wu, Y.; Shen, X. Tetrahedron: Asymmetry 2000, 11, 4359–
4363.
27. Lewis, N.; McKillop, A.; Taylor, R. J. K.; Watson, R. J. Synth.
Commun. 1995, 25, 561–568.
28. Kisanga, P.; Verkade, J. G. Tetrahedron 2001, 57, 467–475.
39. Bhuniya, D.; Singh, V. K. Synth. Commun. 1994, 24, 375–385.