8688 J . Org. Chem., Vol. 66, No. 25, 2001
Notes
stream of dry nitrogen. From a pressure equalizing dropping
funnel p-nitrobenzesulfonyl chloride (27.86 g, 125.7 mmol) in
dichloromethane (75 mL) was added dropwise. The reaction
mixture was left stirring for 1 h. The product precipitated and
was collected by filtration. The solid was recrystallized from
dichloromethane to yield light yellow crystals10 of 5c (21.3 g,
42%); mp 76-79 °C (CH2Cl2), (found: C, 35.46; H, 2.85; N, 7.06;
(C22H20O14N4S3)5(CH2Cl2)810 requires C, 35.58; H, 2.93; N, 7.03);
ν
max/cm-1 1609 (CdC st.), 1520 (NO2 st.); 1H (250 MHz, [2H6]-
nor the tosylamide could be deprotected using the usual
metal-mediated reductive methods.8
DMSO), 3.49 (4 H, t, J 5), 4.17 (4 H, t, J 5), 8.02 (2 H, AA′ of
AA′BB′ J 8), 8.08 (4 H, AA′ of AA′BB′, J 8), 8.27 (2 H, BB′ of
AA′BB′, J 8), 8.42 (4 H, BB′ of AA′BB′, J 8); 13C (100 MHz, [2H6]-
DMSO) 48.51, 69.91, 125.41, 125.66, 129.64, 130.38, 141.78,
145.21, 151.29, 152.03; m/z (FAB, 70 meV) 661.0219 (MH+
requires 661.0216).
As we were unable to deprotect either the triflamide
or the tosylamide, we turned to the 4-nitrophenylsulfonyl
(nosyl) group, sulfonamides of which are readily cleaved
by reaction with thiophenol in a nucleophilic aromatic
substitution process.9 The trinosyl derivative 5c of di-
ethanolamine proved simple to prepare and also reacted
readily with the carbamate anion derived from diethyl-
amine and carbon dioxide (eq 4).
Pleasingly, deprotection of the nosylamide 6c was high-
yielding (Scheme 1), and the product was simple to
separate from the sulfide byproduct 8 using column
chromatography. To prove the method was applicable to
dendrimer synthesis, we needed to show that the depro-
tected amine 7 could be further reacted to produce a
higher generation dendron 9 and also that it could be
coupled to a “core” molecule 10 to produce a small
dendrimer 11. Both these reactions, as shown in Scheme
1, were trouble free.
N,N-Bis-(2-d iet h ylca r b a m oyloxyet h yl)-N-p -n it r ob en -
zen esu lfon a m id e 6c. A solution of diethylamine (3.32 g, 45.4
mmol) and DBU (6.91 g, 45.4 mmol) in acetonitrile (100 mL) in
a pressure vessel fitted with a pressure equalizing funnel was
put under 40 psi CO2 pressure. Addition of CO2 resulted in a
rise of temperature to ca. 40 °C. The solution was stirred for 1
h. A solution of trinosyl compound 5c (10.00 g, 7.50 mmol)10 in
acetonitrile (20 mL) was added dropwise. The reaction mixture
was warmed overnight at 80 °C. The pressure was released and
the acetonitrile was evaporated in a rotary evaporator. The crude
material was dissolved in ethyl acetate and washed with 0.5 M
HCl solution, water, and brine. The organic layer was dried over
MgSO4 and the solvent was evaporated in a rotary evaporator
to dryness to give 6c as a dark brown oil (3.55 g, 96%). νmax
/
cm-1 1693 (CdO st.), 1607 (CdC st.), 1531 (NO2 st. as.); 1H (250
MHz, CDCl3) 1.09 (12 H, s br.), 3.22 (8 H, m br.), 3.54 (4 H, t, J
8), 4.22 (4 H, t, J 8) 8.02 (2 H, AA′ of AA′BB′, J 7), 8.35 (2 H,
BB′ of AA′BB′, J 7); 13C (100 MHz, CDCl3) 13.37, 14.00, 41.23,
41.89, 47.79, 62.22, 124.47, 128.26, 145.70, 150.01; m/z (CI, 70
meV) 489.2033 (MH+ requires 489.2019).
Con clu sion
In conclusion, an approach to dendritic polyurethanes
has been developed that has three main advantages.
First, cheap, nontoxic carbon dioxide is used as an
alternative to toxic phosgene or reagents derived from
it. Second, commercially available diethanolamine is used
as the branch unit. Finally, the chemistry of the nosyl
group permits simultaneous activation of alcohols and
protection of amines, thus simplifying the syntheses.
N,N-Bis-(2-d ieth ylca r ba m oyloxyeth yl)a m in e 7. To 6c
(5.00 g, 10.2 mmol) in DMF (100 mL) were added K2CO3 (2.83
g, 20.5 mmol) and PhSH (2.1 mL, 20.5 mmol). The reaction was
stirred for 2 h at room temperature under a stream of dry
nitrogen. The reaction mixture was filtered, and the DMF was
removed by vacuum distillation. Flash chromatography on silica
gel using 10% methanol:ethyl acetate (1:1) as an eluent gave 7
as a transparent oil (2.51 g, 81%). νmax/cm-1 1691 (CdO st.); 1H
(250 MHz, CDCl3) 1.07 (12 H, t, J 7), 2.94 (4 H, t, J 5), 3.22 (8
H, m, J 7), 4.18 (4 H, t, J 5); 13C (100 MHz, CDCl3) 13.44, 14.05,
41.27, 41.85, 48.55, 64.14, 155.86; m/z (CI, 70 meV) 304.2230
(MH+ requires 304.2236).
Exp er im en ta l Section
N,N-Bis-[N′,N′-b is-2-(2-d iet h ylca r b a m oyloxyet h yl)ca r -
ba m oyloxyeth yl-N-p-n itr oben zen esu lfon a m id e 9. A mix-
ture of amine 7 (4.00 g, 13.2 mmol) and DBU (1.99 g, 13.2 mmol)
in acetonitrile (50 mL) in a pressure vessel fitted with a pressure
equalizing funnel was put under 40 psi CO2 pressure. Addition
of CO2 resulted in a rise of temperature to ca. 40 °C. The solution
was stirred for 1 h. A solution of trinosyl compound 5c (2.90 g,
2.20 mmol)10 in acetonitrile (20 mL) was added dropwise. The
reaction mixture was warmed overnight at 80 °C. After cooling
to room temperature, the pressure was released and the aceto-
nitrile was evaporated in a rotary evaporator. The reaction
mixture was dissolved in ethyl acetate and washed with 0.5 M
HCl solution, water, and brine. Flash chromatography on silica
gel using methanol:ethyl acetate (1:9) as an eluent gave 9 as a
dark brown oil (1.83 g, 88%); νmax/cm-1 1689 (CdO); 1H (300
MHz, CDCl3) 1.10 (24 H, m. br.), 3.20 (16 H, m. br.), 3.50 (12 H,
m), 4.20 (12 H, m), 8.01 (2 H, AA′ of AA′BB′, J 8.9), 8.35 (2 H,
BB′ of AA′BB′, J 8.9); 13C (300 MHz, CDCl3) 13.46, 14.05, 41.26,
41.86, 43.90, 47.14, 47.82, 62.58, 62.69, 63.03, 124.59, 128.38,
145.11, 150.16, 155.50, 155.56; m/z (CI, 70 meV) 949.4568 (MH+
requires 949.4552).
Carbon dioxide was supplied from BOC GASES Ltd. and used
without any further purification. The carbon dioxide reactions
were performed in a glass and PTFE pressure vessel manufac-
tured by Ken Kimble (Reactor Vessels) Ltd. rated to 10 bar at
100 °C or 6 bar at 150 °C with a capacity of 250 mL. For safety
reasons, this pressure vessel was equipped with a bursting disk.
The bursting pressure of the reactor was a multiple of the
maximum allowable working conditions. The glass pressure tube
was mounted in a protective cage. If, during the reaction, stirring
was required, a PTFE coated magnetic stirring bead was placed
in the vessel to allow agitation via a magnetic stirrer. The
miniclave was sealed by the PTFE coverplate, with a Viton
O-ring being clamped to the vessel using a stainless steel
threaded clamp. The complete miniclave was heated or cooled
by placing it in a temperature-controlled bath. Heating and
stirring were performed simultaneously by using a bath on a
magnetic hotplate. Pressure and temperature were monitored
using the pressure gauge and a thermometer mounted in the
thermopocket on the coverplate. Carbon dioxide was dosed or
vented via the needle valve also mounted on the coverplate.
N ,N -Bis-(2-p -n it r ob e n ze n e su lfon yloxye t h yl)-N -p -n i-
tr oben zen esu lfon a m id e 5c. To diethanolamine (4.00 g, 38.1
mmol) in dichloromethane (150 mL) were added triethylamine
(17.5 mL, 125.7 mmol) and a few drops of (dimethylamino)-
pyridine. The solution was stirred at 0 °C for 30 min under a
(10) Combustion analysis of a number of independently prepared
samples of 5c consistently showed the presence of the same percentage
of dichloromethane. That the solvent was included in the crystals was
demonstrated by prolonged exposure to high vacuum, after which the
correct amount of dichloromethane was still observed in the 1H NMR
spectrum. Since the syntheses of 6c and 9 were completed before this
fact had been confirmed, the amount of 5c added was half that planned.
This resulted in the other reagents inadvertently being used in greater
molar excesses than intended.
(8) Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic
Synthesis, 2nd ed.; Wiley: New York, 1991.
(9) Fukuyama, T.; J ow, C. K.; Cheung, M. Tetrahedron Lett. 1995,
36, 6373.