A new synthesis of cyclen (1,4,7,10-tetraazacyclododecane)

Full text of DOI:10.1021/jo9606665

5
186
J . Org. Chem. 1996, 61, 5186-5187
A New Syn th esis of Cyclen
Sch em e 1
(
1,4,7,10-Tetr a a za cyclod od eca n e)
Gary R. Weisman* and David P. Reed
Department of Chemistry, University of New Hampshire,
Durham, New Hampshire 03824-3598
Received April 10, 1996
Cyclen (1,4,7,10-tetraazacyclododecane, 1) is an im-
portant macrocyclic tetraamine that has been used
extensively in metal complexation1 and as a synthetic
2
precursor to related pendant-armed and bridged poly-
3
dentate ligands, some of which have biomedical applica-
tions.4
This route avoids, to a large extent, the above problems
1
2
and also avoids the use of transition metal templates.
The key step of the new synthesis is the double reductive
ring expansion of tricyclic bis-amidine 2 to cyclen (1) with
DIBALH. Yamamoto and Maruoka discovered the highly
regioselective reductive cleavage of amidines with DIBALH
and reported ring expansion of bicyclic amidines to
diazacycloalkanes (for example, the conversion of DBU
For many years, the best synthesis of 1 has been that
based upon the general Stetter-Richman-Atkins
5
-7
method.
The procedure worked out by Richman and
5
Atkins involves medium-dilution macrocyclization of the
disodium salt of tritosyldiethylenetriamine with N-tosyl-
diethanolamine ditosylate to give tetratosylcyclen, which
1
3
to 1,5-diazacycloundecane). Others have utilized this
reaction for the synthesis of diazacycloalkanes,1 but the
reaction has not to our knowledge been extended to the
reduction of bis-amidines of oxalic acid. Reduction of 2
4
is then detosylated with concentrated H
used a Kellogg-type Cs CO
2 3
(
2
SO
variation of this synthesis
for in situ deprotonation of tritosyldiethylenetriamine
in the cyclization) with good success and Sherry and co-
4
. We have
8
(100 mg to 5 g scale) with DIBALH in refluxing toluene
9
(15 h) gave good yields of crude (>90% purity) 1, which
was sublimed to >98% purity. No other purification step
was necessary. The largest scale reaction run to date
workers have employed K
this four-step route are (a) it is not “atom-economic”,
i.e. it requires tosylations and subsequent detosylations;
b) the cyclization requires large amounts of dry DMF;
2
CO
3
.
The disadvantages of
1
0
(see Experimental Section) gave an 83% yield of 1 after
(
sublimation, and we see no reason why the reaction
cannot be scaled up further.
and (c) it is labor intensive. The current high prices of
cyclen11 reflect the level of effort involved in its synthesis.
Cyclen precursor 2 was prepared in 69% yield by
We have developed an efficient two-step synthesis of
cyclen from triethylenetetraamine and dithiooxamide
1
5
S-alkylation of dithiooxamide with excess bromoethane
followed by reaction of the putative bis-thioimido ester
salt 3 with triethylenetetraamine. The procedure was
modeled upon that of Wang and Bauman, who prepared
2,2′-bi-2-imidazoline from ethylenediamine by an analo-
(
Scheme 1).
(1) (a) Bianchi, A.; Micheloni, M.; Paoletti, P. Coord. Chem. Rev.
1
991, 110, 17. (b) Dietrich, B.; Viout, P.; Lehn, J .-M. Macrocyclic
Chemistry; VCH: Weinheim, 1993. (c) Hancock, R. D.; Martell, A. E.
Chem. Rev. 1989, 89, 1875.
16
gous reaction. The only difficulty associated with the
(2) (a) Bernhardt, P. V.; Lawrance, G. A. Coord. Chem. Rev. 1990,
reaction is the production of byproduct ethanethiol
(stench), which must be flushed from the reaction mix-
1
04, 297. (b) Kaden, T. A. Adv. Supramol. Chem. 1993, 3, 65-96. (c)
Kaden, T. A. In Host Guest Complex Chemistry III. Topics in Current
Chemistry; V o¨ gtle, F., Weber, E., Eds.; Springer-Verlag: Berlin, 1984;
Vol. 121, Chapter 5, pp 157-79.
17
ture and oxidized.
In summary, the new synthesis affords cyclen (1) in
two synthetic steps with an overall yield of 57%. No
protecting groups are required and solvent requirements
are modest in comparison to the Richman-Atkins syn-
(3) (a) Micheloni, M. J . Coord. Chem. 1988, 18, 3. (b) Hancock, R.
D. In Crown Compounds; Cooper, S. R., Ed.; VCH: New York, 1992;
pp 167-90. (c) Hancock, R. D.; Pattrick, G.; Wade, P. W.; Hosken, G.
D. Pure Appl. Chem. 1993, 65, 473.
(
4) (a) Alexander, V. Chem. Rev. 1995, 95, 273. (b) Parker, D. In
Crown Compounds; Cooper, S. R., Ed.; VCH: New York, 1992; pp 51-
7. (c) Kumar, K.; Tweedle, M. F. Pure Appl. Chem. 1993, 65, 515. (d)
J urisson, S.; Berning, D.; Wei, J .; Dangshe, M. Chem. Rev. 1993,
3,1137.
5) Richman, J . E.; Atkins, T. J . J . Am. Chem. Soc. 1974, 96, 2268.
Atkins, T. J .; Richman, J . E.; Oettle, W. F. Org. Synth. 1978, 58, 86.
6
(12) (a) Barefield, E. K.; Wagner, F.; Herlinger, A. W.; Dahl, A. R.
Inorg. Synth. 1976, 16, 220. (b) Hoss, R.; V o¨ gtle, F. Angew. Chem.,
Int. Ed. Engl. 1994, 33, 375.
9
(
(13) Yamamoto, H.; Maruoka, K. J . Am. Chem. Soc. 1981, 103, 4186.
(14) Alder, R. W.; Heilbronner, E.; Honegger, E.; McEwen, A. B.;
Moss, R. E.; Olefirowicz, E.; Petillo, P. A.; Sessions, R. B.; Weisman,
G. R.; White, J . M.; Yang, Z.-Z. J . Am. Chem. Soc. 1993, 115, 6580.
(15) Kantlehner, W. In Iminium Salts in Organic Chemistry; B o¨ hme,
H., Viehe, H. G., Eds.; Advances in Organic Chemistry: Methods and
Results; Taylor, E. C., Series Ed.; Wiley: New York, 1979; Vol. 9; Part
2, pp 279-320.
(16) Wang, J . C.; Bauman, J . E., J r. Inorg. Chem. 1965, 4, 1613.
(17) We have used commercial laundry bleach solution for this
purpose, but even the resulting disulfide is very odoriferous. Professor
Peter A. Petillo (University of Illinois), who has graciously checked
(
6) Stetter, H.; Roos, E.-E. Chem. Ber. 1954, 566.
(7) Macrocyclic polyamine synthesis reviews: (a) Bradshaw, J . S.;
Krakowiak, K. E.; Izatt, R. M. Aza-Crown Macrocycles, The Chemistry
of Heterocyclic Compounds; Taylor, E. C., Series Ed.; Wiley: New York,
1
Krakowiak, D. J . Chem. Rev. 1989, 89, 929. (c) Gokel, G. W.; Dishong,
D. M.; Schultz, R. A.; Gatto, V. J . Synthesis 1982, 997.
993; Vol. 51. (b) Krakowiak, K. E.; Bradshaw, J . S.; Zamecka-
(8) Vriesema, B. K.; Buter, J .; Kellogg, R. M. J . Org. Chem. 1984,
4
9, 110.
(
(
(
9) Chavez, F.; Sherry, A. D. J . Org. Chem. 1989, 54, 2990.
10) Trost, B. M. Science 1991, 254, 1471.
11) Major specialty chemical suppliers’ cyclen (1) prices: $213-
our cyclen synthesis, used 30% aqueous H
2
O
2
successfully, and we have
subsequently found 15% aqueous H
reaction.
2 2
O to work well in a related
2
90 (U.S.) per gram (March, 1996).
S0022-3263(96)00666-4 CCC: $12.00 © 1996 American Chemical Society

Notes
J . Org. Chem., Vol. 61, No. 15, 1996 5187
MHz) δ 2.69 (s, 16H), 1.92 (s, 4H); 13C NMR (CDCl
, 90.56 MHz)
3
thesis.5 The sequence can be regarded as involving a
permanent covalently-bound template”12b consisting of
δ 46.1. A mixed melting point (mp 103-109 °C) of resublimed
product and authentic pure sublimed 1 prepared by the Rich-
man-Atkins procedure (mp 105-109 °C) further confirmed the
“
the two carbon unit that is introduced in the first step
and ultimately converted to the new ethylene bridge of
19
identity of product.
1
. Additional methods for synthesis of 2 and related
2
,3,5,6,8,9-Hexa h yd r od iim id a zo[1,2-a :2′,1′-c]p yr a zin e (2).
compounds are being actively investigated, as is the
applicability of the approach presented here to the
synthesis of other macrocyclic tetraamines.
Bromoethane (19.0 mL, 257 mmol) was added in one portion to
a suspension of dithiooxamide (5.15 g, 42.9 mmol) in absolute
EtOH (19 mL). The slurry was heated to 60 °C for 4 h and then
cooled to rt. The reaction flask was equipped with a short-path
distillation head, and excess bromoethane and EtOH were
removed by vacuum distillation (water aspirator) until 2-3 mL
of liquid remained. EtOH (18 mL) was added, and the mixture
was again concentrated by vacuum distillation. The nitrogen
manifold exit line was then routed through a fritted gas washing
Exp er im en ta l Section
Gen er a l. Reactions were run under nitrogen with stirring.
Triethylenetetraamine (>97%) and dithiooxamide were pur-
chased from Fluka. All other reagents and solvents were
obtained from Aldrich.
17
bottle charged with aqueous (household) bleach solution.
EtOH (10 mL) was added to the reaction flask, followed by
triethylenetetraamine (6.27 g, 42.9 mmol), followed by EtOH (10
mL). The solution was stirred at rt for 30 min and then at 80
1
,4,7,10-Tetr a a za cyclod od eca n e, Cyclen (1). DIBALH
160 mL of 1.5 M solution in toluene, 240 mmol) was added over
h to a slurry of 2 (4.93 g, 30.0 mmol) in dry toluene (160 mL)
and then refluxed for 15 h. The mixture was then cooled to 0
C, NaF (40.2 g, 960 mmol) and water (13.0 mL, 722 mmol) were
(
1
°
C for 20 min. Ethanethiol (stench) was then purged from the
solution by entrainment with N , which was bubbled through
2
°
the reaction mixture (fritted gas dispersion tube) into the bleach-
filled gas washing bottle. After 15 h, the reaction mixture was
concentrated to 2-3 mL by short-path vacuum distillation (40
added in alternating small portions with stirring, and the slurry
was allowed to warm to rt. Solids were removed by vacuum
filtration and washed with CHCl
3
(5 × 50 mL). Filtrate and
°
C, water aspirator). (A dry-ice cooled trap was used to assure
washings were combined and dried (Na
2
SO ), and solvents were
4
that any residual ethanethiol was trapped; the trap was
1
removed to yield 2.40 g of 1 (>98% purity by H NMR). Soxhlet
extraction of the reaction solids with toluene for 24 h, drying,
and toluene removal yielded an additional 2.55 g of crude 1
17
subsequently cleaned with bleach solution. ) Toluene (50 mL)
was then added, undissolved solids were removed by filtration,
the filtrate was concentrated, and resultant crude 2 was
recrystallized from hot toluene (hot filtration necessary) to give
(
0
>90% purity). Combined crude product was sublimed (120 °C,
1
13
.05 Torr) to yield 4.32 g (83%) of 1 (>98% purity by H and
C
5
.83 g of yellow solid (mp 146-149 °C). A second recrystalli-
zation from toluene gave 4.86 g (69%) of off-white 2 (mp 149-
50 °C). This material was shown by NMR to be pure enough
>98%) for subsequent conversion to cyclen. If desired, it can
be further purified by sublimation (100 °C, 0.01 Torr): white
NMR; mp 103-107 °C). Slower resublimation (65 °C, 0.02 Torr;
9
4% recovery) gave 1 of even greater purity according to NMR
1
(
1
8
19 1
3
(mp 103-107 °C, Lit. mp 119-120 °C). H NMR (CDCl , 360
1
(
18) Buøen, S.; Dale, J .; Krane, J . Acta Chem. Scand. 1984, B38,
73.
19) There has been a great deal of confusion about the mp of 1 in
3
solid; mp 150-151 °C; H NMR (CDCl , 360 MHz) δ 3.26 (s, 4H),
3.35 (apparent t (XX′ of AA′XX′), 4H, J appar) 9.6Hz), 3.86
7
(
(
(
(
apparent t (AA′ of AA′XX′), 4H, J appar ) 9.6 Hz); C NMR
13
2
0
the literature. Stetter and Mayer originally reported mp 35 °C. Buøen
et al. reported mp 119-120 °C. Aldrich and Fluka list melting point
ranges of 110-113 °C (97%) and 105-110 °C (g97%) repectively in
-1
3
CDCl , 90.56 MHz) δ 45.3, 52.1, 53.9, 155.4; IR (KBr) 1629 cm
1
8
+
8 12 4
CdN); MS (EI) 164.15 (M) . Anal. Calcd for C H N : C, 58.52;
2
1
H, 7.37; N, 34.12. Found: C, 58.38; H, 7.55; N, 34.22. (Note:
is unstable toward hydrolysis and should be stored in a
their catalogs. Confusing matters further, Zhang and Busch subse-
quently reported mp 36-38 °C. Our mp range for 1 (calibrated
thermometer) is lower than that reported in ref 18, but is consistent
with the mp range of sublimed material (no detectable impurities by
high S/N NMR) we have prepared by the Richman-Atkins method
2
desiccator.)
1
Ack n ow led gm en t. G.R.W. gratefully acknowledges
helpful discussions with Roger W. Alder.
(
mp 105-109 °C). H NMR relative integrations of the material
reported in this note are consistent with anhydrous 1.
(20) Stetter, H.; Mayer, K.-H. Chem. Ber. 1961, 94, 1410.
21) Zhang, R.; Busch, D. H. Inorg. Chem. 1993, 32, 4920.
(
J O9606665