The coordination chemistry of cis-3,4-diaminopyrrolidine and related polyamines

FULL PAPER

The Coordination Chemistry of cis-3,4-Diaminopyrrolidine and Related

Polyamines^[‡]

Dirk Kuppert,^[a]Jürgen Sander,^[a]Christian Roth,^[a]Michael Wörle,^[b]

Thomas Weyhermüller,^[c]Guido J. Reiss,^[d]Uwe Schilde,^[e]Iris Müller,^[f]and

Kaspar Hegetschweiler*^[a]

Keywords: Chelates / N ligands / Ligand design / Stability constants / Coordination modes

cis-3,4-Diaminopyrrolidine (cis-dap), trans-3,4-diaminopyr- ligand. This was supported by molecular mechanics calcula-

rolidine (trans-dap), cis-1,2-cyclopentanediamine (cis-cptn),

and trans-1,2-cyclopentanediamine (trans-cptn) have been crystal X-ray diffraction studies of [Pt(Hcis-dap)Cl₄]Cl·H₂O,

tions (Cu^IIand Co^IIIcomplexes) and confirmed by single-

prepared in multigram quantities. The complexation of these

[Pd(Hcis-dap)₂](ClO₄)₄·2H₂O, and [Cu(Hcis-dap)₂(OH₂)₂]-

ligands and of 3-aminopyrrolidine (ampy) with Ni^II, Cu^II, Zn^II,

(SO₄)₂·3.5H₂O − 2x H⁺+ x Cu²⁺with 0.01 Յ x Յ 0.11. For

and Cd^IIhas been studied in solution by means of potentio- the diamine ligands, coordination through the two exocyclic

metric and spectrophotometric titrations. The complexes of

the triamines cis-dap and trans-dap show a pronounced amino group was established from the X-ray structure ana-

tendency to form protonated species such as [M^II(HL)]³⁺

lyses of [Ni(cis-cptn)₂](ClO₄)₂and [Cu(3R-ampy)(3S-ampy)]-

[M^II(HL)₂]⁴⁺, and [M^II(HL)L]³⁺, indicative of a bidentate coor- (ClO₄)₂, respectively. The crystal structure determination of

amino groups or through one exocyclic and one endocyclic

,

dination mode of the ligand L. The UV/Vis spectra of the

corresponding Cu^IIcomplexes further confirmed bidentate

coordination with trans-CuN₄geometry. The overall stabilit-

[Co(cis-dap)(tach)][ZnCl₄]Cl·C₂H₅OH (tach = cis-1,3,5-cyclo-

hexanetriamine) revealed tridentate, facial coordination of

cis-dap in this particular complex. However, the structural

ies of the bis complexes [ML₂]²⁺decrease in the order cis- parameters of the [Co(cis-dap)(tach)]³⁺moiety indicate signi-

cptn Ͼ cis-dap Ͼ trans-cptn Ͼ ampy Ͼ trans-dap. The con- ficant strain for this coordination mode. The coordinating

siderably lower stabilities of the ampy complexes as com- properties of the ligand cis-dap are compared with those of

pared to the corresponding cis-dap complexes indicate

other aliphatic and alicyclic triamines.

metal binding to the two primary amino groups of the latter

Introduction

ane-1,3,5-triamine backbone are well established in the lit-

erature.^[2,3]However, until recently, intermediate members

of this series, such as cis-3,5-diaminopiperidine (dapi) or

cis-3,4-diaminopyrrolidine (cis-dap), had not been investig-

ated to any great extent. Crystal structures of Pd^II(dapi)

complexes have been reported by Schwarzenbach,^[4]and we

recently published a comprehensive study of the com-

plexation of dapi with a variety of metal cations.^[5]The

metal-binding properties of the related triamine cis-3,4-di-

aminopyrrolidine (cis-dap) have not yet been described. We

have developed an efficient procedure for the preparation

of cis-dap and wish to report herein its coordination chem-

istry both in aqueous solution and in the solid state. Addi-

tionally, some related cyclic di- and triamine ligands with a

five-membered ring structure have also been studied

(Scheme 2).

StructureϪstability correlations are a subject of continu-

ing interest in coordination chemistry, and the dependence

of complex stability on the individual structural properties

of a ligand has been extensively discussed.^[1]In this respect,

cyclic polyamines that are restricted to a facial coordination

(Scheme 1) are of particular interest because their rigidity

considerably reduces the number of possible solution struc-

tures as compared to their open-chain analogues. Ligands

based on the 1,4,7-triazacyclononane or all-cis-cyclohex-

[‡]

Facially Coordinating Cyclic Triamines, 2. Ϫ Part 1: Ref.^[5]

[a]

Anorganische Chemie, Universität des Saarlandes,

Postfach 151150, 66041 Saarbrücken, Germany

Fax: (internat.) ϩ 49-(0)681/302-2663

E-mail: hegetschweiler@mx.uni-saarland.de

[b]

Laboratorium für Anorganische Chemie,

ETH-Zentrum, 8092 Zürich, Switzerland

[c]

Max-Planck-Institut für Strahlenchemie,

Stiftstraße 34Ϫ36, 45470 Mühlheim an der Ruhr, Germany

[d]

Institut für Anorganische Chemie und Strukturchemie,

Results

Heinrich-Heine-Universität Düsseldorf,

Universitätsstraße 1, 40225 Düsseldorf, Germany

[e]

Synthesis of the Ligands

Institut für Anorganische Chemie und Didaktik der Chemie,

Universität Potsdam,

The only hitherto known synthesis of cis-3,4-diamino-

pyrrolidine involves a complicated multi-step procedure

and gives only unsatisfactory yields.^[6]We have developed

a new and efficient synthetic route that has allowed facile

Karl-Liebknecht-Straße 24Ϫ25, 14476 Golm, Germany

[f]

Institut für Analytische Chemie, Ruhr-Universität,

44780 Bochum, Germany

Supporting information for this article is available on the

WWW under http://ww w .eurjic.com or from the author.

Eur. J. Inorg. Chem. 2001, 2525Ϫ2542

WILEY-VCH Verlag GmbH, 69451 Weinheim, 2001

1434Ϫ1948/01/1010Ϫ2525 $ 17.50ϩ.50/0

2525

K. Hegetschweiler et al.

FULL PAPER

stituted derivatives have been described.^[7,8]We prepared

this ligand from the N-benzylimide derivative of tartaric

acid, which was again converted into the corresponding di-

azide. Subsequent hydrogenation resulted in reduction of

the azido groups and in the removal of the benzyl group.

The trans (3R,4R) configuration was again confirmed by a

single-crystal X-ray diffraction analysis (Figure 1, bottom).

˚

The puckering parameters Q ϭ 0.405(2) A, ϕ ϭ 83.0(4)°

˚

and Q ϭ 0.357(3) A, ϕ ϭ 238.3(4)° for the pyrrolidine rings

of H₃cis-dap^3ϩand H₃trans-dap^3ϩare indicative of a half-

chair conformation with a slight twist towards an envelope

form.^[9]As expected, the N···N distance between the two

primary amino groups in trans-dap is somewhat longer

˚

(3.28 A) than that in cis-dap (3.00 A).

Scheme 1

Figure 1. Molecular structures of the triply protonated triamines

H₃cis-dap^3ϩ(top) and H₃trans-dap^3ϩ(bottom) showing the atom

numbering scheme; thermal ellipsoids are drawn at a 50% probabil-

ity level; hydrogen atoms are shown as spheres of arbitrary size;

˚

range of bond lengths [A]: CϪC 1.520(3)Ϫ1.533(3), CϪN

1.478(3)Ϫ1.510(3); selected angles for H₃cis-dap^3ϩ[°]: C2ϪC3ϪC4

100.93(16), N3ϪC3ϪC4 114.11(16), N2ϪC4ϪC3 116.80(17)

Scheme 2. Structures of the ligands and their abbreviations

The structurally related diamines cis-1,2-cyclopentanedi-

preparation of this triamine in multigram quantities. Our amine (cis-cptn) and trans-1,2-cyclopentanediamine (trans-

procedure starts with 3-pyrroline, the amino group of cptn) were analogously prepared via the diazides using

which is protected by acetylation. The carbonϪcarbon the

corresponding

cyclopentanediols

as

starting

double bond is then stereoselectively cis-dihydroxylated materials.^[10Ϫ12]The chiral trans-cptn was obtained as a

with OsO₄to give 1-acetyl-cis-3,4-pyrrolidinediol. Sub- racemate. The two diastereomers can readily be distingu-

sequent conversion of the two hydroxy groups into amino ished by ¹H NMR spectroscopy. The cis isomer has C_ssym-

groups is achieved through the corresponding diazide. metry and shows four resonances due to the three methyl-

After removal of the protecting acetyl group, the desired ene groups, whereas the chiral trans isomer exhibits C₂sym-

cis-3,4-diaminopyrrolidine can be isolated in excellent yield metry showing only three signals for these six protons. The

as the trihydrochloride salt. The cis configuration of the diamine 3-aminopyrrolidine (ampy) is commercially avail-

product was established by a single-crystal X-ray diffrac- able and was again used as a racemate. A single-crystal X-

tion analysis (Figure 1, top). The corresponding chiral ray diffraction analysis of H₂ampy(ClO₄)₂has recently

trans isomer was hitherto unknown and only some 1-sub- been published.^[13]

2526

Eur. J. Inorg. Chem. 2001, 2525Ϫ2542

Coordination Chemistry of cis-3,4-Diaminopyrrolidine and Related Polyamines

FULL PAPER

Protonation Constants

tions indicated that the two microspecies with a protonated

primary amino group or a protonated secondary ring nitro-

gen atom are present in proportions of 57% and 43%, re-

spectively. This ratio corresponds closely to a statistical dis-

tribution of the proton over the three basic sites. A different

The acidity constants of the protonated amines are listed

in Table 1. For the two pairs of diastereomers H₃cis-dap^3ϩ

/

H₃trans-dap^3ϩand H₂cis-cptn^2ϩ/H₂trans-cptn^2ϩ, the cis

isomers of the fully protonated forms are slightly stronger

acids. The triamines have different basic sites (primary and

secondary amino groups) and this leads to different tauto-

mers for species with an intermediate degree of protonation.

In the case of cis-dap, the equilibria between these different

result is obtained for the doubly protonated H₂cis-dap^2ϩ

,

for which the tautomer with a free primary amino group

clearly predominates (75%). This result can be explained in

terms of a simple electrostatic model (maximal separation

of the two positive charges).

1

microspecies were elucidated by H NMR titration (Fig-

ure 2). An NOESY experiment allowed the unambiguous

assignment of the two diastereotopic protons of the methyl-

ene groups. Only one proton (3-H) showed a significant

NOE with the adjacent 1-H proton. The observed pH de-

pendence of the chemical shifts could be reproduced using

pK_avalues of 2.15, 6.21, and 9.78 for the macrospecies

H_xcis-dap^xϩ(Figure 2; 3 Ն x Ն 1). These values (D₂O, 28

°C, no inert electrolyte) are in good agreement with the pK_a

values obtained from the potentiometric measurements. As

Metal Complex Formation

The coordination behavior of the polyamines in dilute

aqueous solution (25 °C) was studied by means of pH-met-

ric titration experiments. Formation constants of Ni^II, Cu^II,

and Cd^IIcomplexes were determined for ampy, cis-cptn,

and cis-dap. Additionally, the complexation of cis-dap with

Zn^IIwas studied. For the ligands trans-cptn and trans-dap,

our study was confined to complexation with Cu^II. All

formation constants are summarized in Tables 2 and 3. The

measurements with Ni^II, Cu^II, and Zn^IIwere performed in

0.1 mol dm^Ϫ3KCl. Due to the relatively high affinity of

Cd^IIfor Cl^Ϫ, these experiments were performed in 0.1 mol

dm^Ϫ3KNO₃.^[5]

Table 1. Protonation constants for the cyclic di- and triamine li-

gands (25 °C) at ionic strengths (I) of 0.1 mol dm^Ϫ3(KCl) and 1.0

mol dm^Ϫ3(KNO₃) as indicated

[a]

log K₁

0.1

log K₂

0.1

log K₃

0.1

I [mol dm^Ϫ3

]

1.0

Table 2. Formation constants of metal complexes with the triamine

ligands (25 °C, 0.1 mol dm^Ϫ3KCl or KNO₃)

ampy

10.39

9.74

9.92

9.66

9.58

10.66

Ϫ

9.90

9.80

6.82

6.13

7.22

6.25

6.34

7.20

Ϫ

6.63

6.72

Ϫ

2.42

3.67

Ϫ

2.96

4.19

cis-cptn

trans-cptn

cis-dap

Cu^2ϩ

Ni^2ϩ

Zn^2ϩ

Cd^2ϩ[a]

cis-dap

trans-dap cis-dap

cis-dap

trans-dap

[b]

log β_xyz

x,y,z

[a]

K_iϭ [H_iL] ϫ [H]^Ϫ1ϫ [H_iϪ1L]^Ϫ1; estimated standard deviations

1,1,0

1,1,1

1,2,2

1,2,1

Ϫ

16.8(1)

32.38(2) 28.40(2)

25.27(1) 21.59(2)

17.53(1) 14.05(2)

Ϫ

6.41(1)

14.14(2) 12.68(2)

27.06(2)

19.37(3)

11.24(2) 8.89(2)

4.93(2)

4.44(2)

12.33(1)

Ϫ

7.90(2)

Ϫ

Ͻ 0.01.

14.68(1)

Ϫ

described previously,^[5]the deshielding of the (CϪ)H pro- 1,2,0

tons by protonation of a nearby basic site in a cyclic polya-

1,3,0

1,2,Ϫ1

Ϫ

13.7(1)

Ϫ

Ϫ0.38(2)

Ϫ

mine can be modeled using a limited set of deshielding con-

stants. For the monoprotonated Hcis-dap^ϩ, such calcula-

[a]

Cd^2ϩ: 0.1 mol dm^Ϫ3KNO₃; all other metal ions: 0.1 mol dm^Ϫ3

[b]

KCl. Ϫ

β_xyzϭ [M_xL_yH_z] ϫ [M]^Ϫxϫ [L]^Ϫyϫ [H]^Ϫz; estimated

standard deviations (in parentheses) were calculated with HYPER-

QUAD^[36]and multiplied by a factor of three.

The results indicate that the two triamines cis-dap and

trans-dap show a pronounced tendency to form protonated

complexes of the compositions [M(HL)]^3ϩand [M(HL)₂]^4ϩ

in acidic solution (Figure 3). In the presence of excess li-

gand, the main species formed are the bis(complexes)

[ML₂H_x]^(2ϩx)ϩ. The formation of a tris(chelate) [ML₃]^2ϩ

was only verified in the case of Ni^II. [Ni(cis-dap)₃]^2ϩwas

seen as a minor species at the end of a titration using an

excess of the ligand. In the experiments with total Ni/total

cis-dap ϭ 1:2, formation of [Ni(cis-dap)(Hcis-dap)]^3ϩand

[Ni(cis-dap)]^2ϩwas observed in the same pH range, and a

fixed value of β_ML, as obtained from an independent 1:1

titration, was used in the evaluation to avoid mutual influ-

Figure 2. pD dependence of the ¹H NMR resonances of cis-dap;

squares correspond to the experimental values; the lines are calcu-

lated (minimization of [δ_obsϪ δ_calcd.]²) assuming a rapid equilib-

rium between the species H_xcis-dap^xϩ(1 Յ x Յ 3)

Eur. J. Inorg. Chem. 2001, 2525Ϫ2542

2527

K. Hegetschweiler et al.

Table 3. Formation constants for metal complexes with the diamine ligands (25 °C, 0.1 mol dm^Ϫ3KCl or KNO₃)

FULL PAPER

Cu^2ϩ

ampy

Ni^2ϩ

Cd^2ϩ[a]

ampy

cis-cptn

trans-cptn

ampy

cis-cptn

[b]

log β_xyz

x,y,z

1,1,0

1,2,0

1,3,0

8.24(1)

15.01(1)

10.59(1)

19.75(1)

8.58(1)

15.81(1)

4.1(1)

8.1(1)

7.02(1)

12.70(1)

15.98(4)

3.4(2)

5.20(1)

9.52(1)

1,2,Ϫ1

8.97(3)

Ϫ2.1(1)

[a]

[b]

Cd^2ϩ: 0.1 mol dm^Ϫ3KNO₃; all other metal ions: 0.1 mol dm^Ϫ3KCl. Ϫ

β_xyzϭ [M_xL_yH_z] ϫ [M]^Ϫxϫ [L]^Ϫyϫ [H]^Ϫz; estimated

standard deviations (in parentheses) were calculated with HYPERQUAD^[36]and multiplied by a factor of three.

Figure 3. Species distribution plots for an equilibrated aqueous solution with a total metal ion (Ni, Cu) concentration of 10^Ϫ3mol dm^Ϫ3

and a total ligand (L) concentration of 2 ϫ 10^Ϫ3mol dm^Ϫ3(Cu) or 3 ϫ 10^Ϫ3mol dm^Ϫ3(Ni); L ϭ cis-dap (top), L ϭ cis-cptn (bottom);

only metal-containing species are shown; the equilibrium constants listed in Tables 2 and 3 were used for the calculations

ence (correlation effects) of the two formation constants. In solution species.^[14,15]The d-d transitions of Ni^IIcomplexes

the case of Cu^II, the unprotonated [CuL]^2ϩcomplex was with saturated amines are of rather low intensity (ε Յ 10)

not observed.

and hence rather concentrated solutions (0.1 ) had to be

As expected, the diamines ampy, cis-cptn, and trans-cptn used. Under these conditions, both [Ni(cis-dap)₂]^2ϩand

do not form any protonated metal complexes and the titra- [Ni(Hcis-dap)₂]^4ϩform blue solutions, with A_2g-³T_1g(F)

3

tion curves could be evaluated simply in terms of [ML]^2ϩtransitions at 571 and 577 nm, respectively. These values are

and [ML₂]^2ϩformation. In the case of cis-cptn and Ni^II, consistent with an NiN₄O₂chromophore and the minor

the formation of a complex of composition [ML₃]^2ϩwas shift upon deprotonation clearly indicates that incorpora-

additionally established. The complexation of Ni^IIand Cd^IItion of an additional nitrogen atom into the chromophore

by ampy is rather weak and was only observed at pH Ͼ 8. does not occur to any significant extent. We can therefore

To avoid precipitation of solid metal hydroxides, a large ex- exclude both tridentate coordination and a bridging mode

cess of ampy had to be used.

with the secondary nitrogen atom coordinated to an adja-

UV/Vis data of Ni^II- and Cu^II(amine) complexes provide cent Ni^IIcenter. In contrast to cis-dap, [Ni(cis-cptn)₂]^2ϩ

a useful tool for assigning the structures of the different forms a bright-yellow solution (λ_maxϭ 444 nm), indicative

2528

Eur. J. Inorg. Chem. 2001, 2525Ϫ2542

Coordination Chemistry of cis-3,4-Diaminopyrrolidine and Related Polyamines

FULL PAPER

of a diamagnetic complex.^[10]It is noteworthy that the

slightly less basic cis-dap does not form such a low-spin

bis(complex). In the presence of excess cis-dap (Ni/L Ͼ 1:3),

purple solutions with λ_maxϭ 553 nm are observed. The

considerable shift to shorter wavelengths is indicative of a

tris(chelate) structure with an NiN₆chromophore. Such an

[Ni(cis-dap)₃]^2ϩcomplex had already been noted as a minor

species in the aforementioned titration experiments. Since

the significantly higher ligand concentration (0.3 mol

dm^Ϫ3) favors complexation, the tris complex becomes the

predominant species at pH Ͼ 10.

In aqueous solution, the d-d transitions of Cu^II(amine)

complexes appear unresolved as a broad single band in the

range 500Ϫ750 nm.^[15]The intensity of this band is an order

of magnitude higher than that seen for the corresponding

Ni^IIcomplexes, and this allowed the recording of solution

spectra with a total Cu^IIconcentration of just 0.01Ϫ0.02

mol dm^Ϫ3. Measurements were made at an ionic strength

of 1 mol dm^Ϫ3KNO₃and were carried out in conjunction

with an additional series of potentiometric titrations (Fig-

ure 4). These data allowed the determination of the stability

constants by two independent methods. The agreement

found was generally good (Table 4). Additionally, the meas-

urements could be used for the calculation of spectra for

each individual species (Figure 4, b). It is well established

that for a tetragonally distorted geometry (C_4vor D_4h), suc-

cessive replacement of the four water ligands by nitrogen

donors in the equatorial plane results in a continuous shift

of λ_maxto shorter wavelengths, whereas substitution of the

loosely bound axial water ligands results in a red shift (the

‘‘pentaammine effect’’).^[16,17][Cu(Hcis-dap)₂]^4ϩ, [Cu(Hcis-

dap)(cis-dap)]^3ϩ, and [Cu(cis-dap)₂]^2ϩwere all found to ex-

hibit absorbance maxima at 552Ϫ567 nm, indicative of a

trans-CuN₄chromophore. Similarly, the λ_maxvalues of

560 nm (trans-dap), 550 nm (ampy), and 545 nm (cis-

cptn)^[18]indicate trans-CuN₄geometries with one or two

weakly bound water molecules at the apices in the corres-

ponding bis complexes [CuL₂]^2ϩ. The 1:1 complexes

[Cu(Hcis-dap)]^3ϩ(λ_maxϭ 688 nm) and [Cu(ampy)]^2ϩ

(λ_maxϭ 668 nm) exhibit absorbance maxima at consider-

ably longer wavelengths, as expected for cis-CuN₂coordina-

tion.^[14]

Figure 4. Spectral changes for the Cu^II(cis-dap) system during a

titration experiment with total Cu ϭ 10^Ϫ2mol dm^Ϫ3and total L ϭ

2 ϫ 10^Ϫ2mol dm^Ϫ3(a); the calculated spectra for the individual

species (b)

Table 4. Comparison of formation constants of Cu^IIcomplexes de-

rived from either potentiometric (pot) or spectrophotometric (spec)

titrations (25 °C, 1 mol dm^Ϫ3KNO₃) estimated standard deviations

are given in parentheses

pot^[a]

spec^[a]

[b]

log β_xyz

x,y,z

ampy

1,1,0

1,2,0

8.54(1)

15.68(1)

8.53(1)

15.65(2)

cis-dap

1,1,1

1,2,0

1,2,1

1,2,2

17.64(2)

18.41(2)

26.62(2)

34.23(2)

17.84(4)

18.9(3)

26.9(1)

34.45(5)

trans-dap

1,1,1

1,2,0

1,2,1

1,2,2

15.42(2)

14.78(1)

22.69(1)

29.99(1)

15.63(3)

14.9(2)

22.93(9)

30.27(5)

Structural Characterization of the Metal Complexes

^[a]Potentiometry: calculated with HYPERQUAD;^[36]spectrophoto-

In [Pt(Hcis-dap)Cl₄]Cl·H₂O, the ligand cis-dap coordin-

ates to the Pt^IVcenter in a bidentate fashion through the

two primary amino groups. The remaining secondary am-

ino group of the pyrrolidine ring is protonated (Figure 5).

The octahedral low-spin geometry of the Pt^IVcenter is com-

metry: calculated with SPECFIT;^[37]the obtained standard devi-

[b]

ations are multiplied by a factor of three. Ϫ

β_xyzϭ [M_xL_yH_z] ϫ

[M]^Ϫxϫ [L]^Ϫyϫ [H]^Ϫz

.

pleted by the coordination of four chloride ions. The posit- This packing is related to the GeS structure,^[19]although in

ive charge of the complex cation is balanced by an addi- GeS the six-membered Ge₃S₃rings have a chair conforma-

tional chloride counterion. Considering the positions of the tion, whereas in [Pt(Hcis-dap)Cl₄]Cl·H₂O the A₃B₃rings

cations (A), represented by their centers of gravity, and of have a distorted boat conformation. In both cases, the non-

the counterions (B), the resulting AB-type packing can be isometric packing can be explained simply in terms of the

described as a stack of layers consisting of fused six-mem- nonisotropic shape of the cations. In [Pt(Hcis-

bered rings with an alternating AϪBϪAϪBϪAϪB se- dap)Cl₄]Cl·H₂O, the anions and cations within a layer are

quence. The layers are oriented perpendicular to the b axis. connected through NϪH···Cl^Ϫhydrogen bonds and the dif-

Eur. J. Inorg. Chem. 2001, 2525Ϫ2542

2529

K. Hegetschweiler et al.

FULL PAPER

ferent layers are interconnected by NϪH···O···HϪN oid to the cisoid form in the course of deprotonation). The

bridges between the coordinated amino groups and water required breaking and reforming of Pd^IIϪN bonds was in-

molecules. The [Pt(Hcis-dap)Cl₄]^ϩcomplex cation is com- voked to account for the slow reaction rate. In our experi-

pletely asymmetric as a consequence of the puckered con- ments, we found a related steady shift of pH for a freshly

formation of the five-membered chelate ring. However, the prepared 0.1 mol dm^Ϫ3aqueous solution of [Pd(Hcis-dap)_2-

NMR-spectroscopic data (two signals in the ¹³C{¹H} NMR ](ClO₄)₄. However, this shift proved to be very slow (5 mV

1

and three signals in the H NMR spectrum) indicate the per 24 h) and the data set of a rapid titration could be

expected rapid ring inversion with average C_ssymmetry for evaluated in terms of a simple two-step deprotonation with

the complex in solution.

pK_avalues of 6.27(1) and 7.04(1).

With Cu^II, a solid sample of composition [Cu(Hcis-

dap)₂(OH₂)₂](SO₄)₂·3.5H₂O was obtained. This compound

contains two crystallographically independent [Cu(Hcis-

dap)₂(OH₂)₂]^4ϩcations (Figure 6). Cu1 is located on a two-

fold axis and its five-membered chelate rings thus have a λλ

or δδ conformation, whereas Cu2 resides on a center of

inversion with a λδ conformation of the rings. Again, the

Cu centers are exclusively bound to the primary amino

groups of the Hcis-dap^ϩligands (square-planar Cu^IIN₄

˚

geometry with an average CuϪN bond length of 2.02 A).

The coordination sphere is completed by two loosely bound

water ligands in the axial positions (average CuϪO bond

˚

length: 2.48 A), giving the well-known tetragonally dis-

torted CuN₄O₂octahedron. The noncoordinating ring ni-

trogen atoms are protonated and the pyrrolidine rings are

arranged in a transoid manner, akin to that described above

for the Pd complex. Deprotonation of [Cu(Hcis-

dap)₂(OH₂)₂]^4ϩcould occur either at the noncoordinating

ring nitrogen atoms or at the water ligands (with the forma-

tion of hydroxo complexes). In the first case, the liberation

of an additional amino group would generate a potentially

tridentate ligand. The additional amino group could either

coordinate in a tripodal intramolecular fashion or intermol-

ecularly as a bridging ligand to an adjacent metal center.

The crystal structure of [Cu(Hcis-dap)₂(OH₂)₂](SO₄)₂-

·3.5H₂O provided evidence that the latter mode is observed

with Cu^II. Close inspection of the difference Fourier map

revealed a residual peak at x ϭ 3/4, y ϭ 1/4, z ϭ 0.2736

Figure 5. Molecular structures of [Pt(Hcis-dap)Cl₄]^ϩ(top) and

[Pd(Hcis-dap)₂]^4ϩ(bottom); thermal ellipsoids are drawn at a 50%

probability level; hydrogen atoms are shown as spheres of arbitrary

˚

size; selected bond lengths [A] and angles [°]: (a) PtϪN2 2.075(10),

PtϪN3 2.007(9), PtϪCl1 2.333(3), PtϪCl2 2.311(3), PtϪCl3

2.322(3), PtϪCl4 2.313(3), N2ϪPtϪN3 83.8(5); (b) PdϪN2

2.035(2), PdϪN3 2.039(2), N2ϪPdϪN3 83.86(9), N2ϪPdϪN3A

96.14(9)

3

˚

with an electron density of 1.6 e/A . This peak is located on

a twofold axis in a cavity formed by the noncoordinating

A similar bidentate coordination of the protonated Hcis- ammonium groups of two complex cations, two sulfate

dap^ϩligand is observed for [Pd(Hcis-dap)₂](ClO₄)₄·H₂O counterions, and a water molecule. The distances to the ne-

˚

(Figure 5). The crystal structure of this compound can be arest atoms are 2 ϫ 1.96 A (O22 & O22A), 2.00 A (O60), 2

˚

described as a packing of neutral {[Pd(Hcis-dap)₂](ClO₄)₄} ϫ 2.19 A (N1 & N1A). Further studies of additional speci-

aggregates

that

are

hydrogen-bonded

through mens showed considerable variation in the intensity of this

NHϪH···OHϪH···OClO₃···HϪNH and NHϪH···OClO₃··· peak. Although the cell parameters and atom positions

HϪNH bridges. The Pd^IIatom resides on a center of inver- were almost the same for all the samples under investi-

sion and exhibits the well-known square-planar PdN₄geo- gation, the intensity of this peak corresponded to an elec-

metry (λδ conformation for the two five-membered chelate tron density of up to 12.6 eA^Ϫ3. Based on the chemical

˚

rings). The pyrrolidine rings are located at opposite sides of composition of the mother liquor, the charge balance, and

the PdN₄plane and thus adopt a transoid orientation. A the distances to the five nearest neighboring atoms, we in-

related structure with protonated endocyclic nitrogen atoms terpret this peak in terms of some additional incorporation

in a transoid geometry has been reported by Schwarzenbach of Cu^II, which is bound to the two pyrrolidine nitrogen

et al. for the Pd complex of cis-3,5-diaminopiperidine atoms with the abstraction of two protons (Scheme 3). In

(dapi).^[4]These authors performed a potentiometric study fact, it became clear that this compound has a nonstoichi-

to determine the acidity constants of the noncoordinating ometric (berthollide) composition, for which the correct

ammonium groups of [Pd(Hdapi)₂]^4ϩ; an unusually slow formula would be [Cu(Hcis-dap)₂(OH₂)₂](SO₄)₂·3.5 H₂O Ϫ

equilibration was noted, which was discussed in terms of a 2x H^ϩϩ x Cu^2ϩ. In this study, the data for x ϭ 0.005,

slow rearrangement of the ligands (a change from the trans- 0.08, and 0.11 are reported. Although the cavity appears to

2530

Eur. J. Inorg. Chem. 2001, 2525Ϫ2542

Coordination Chemistry of cis-3,4-Diaminopyrrolidine and Related Polyamines

FULL PAPER

have an almost ideal size for accommodating Cu^II, some the same metal center, could be obtained in the form of the

minor structural rearrangements are evidently required for heteroleptic [Co^III(cis-dap)(tach)]^3ϩ. Attempts to prepare a

the uptake of an additional Cu^IIion. This is indicated by a homoleptic complex [Co(cis-dap)₂]^3ϩby aerial oxidation of

disorder that is found for those samples containing a large a solution of Co^IIand cis-dap failed and the use of Co^III

amount of additional Cu. This disorder is manifested in precursors such as K₃[Co(CO₃)₃] or trans-[CoCl₂(py)₄]Cl

strong anisotropy of the displacement parameters for the led to unresolved reaction mixtures. [Co(cis-dap)(tach)]^3ϩ

atoms of the outer part of the pyrrolidine rings and the was prepared from [Co(tach)Cl₃], in which the relatively

observation of two partially occupied positions for one of labile Cl^Ϫligands define a set of leaving groups having the

the oxygen atoms of the sulfate residue (Figure 6). The ad- required preformed facial geometry. The resulting mononu-

ditional Cu^IIcenter has trigonal-bipyramidal coordination clear Co^IIIcomplex was crystallized as a chloride tetrachlo-

with the two pyrrolidine nitrogen atoms in the axial posi- rozincate salt. Its hexaamine geometry exhibits some rather

tions and two sulfate ions and a water molecule in the unusual features. Bond lengths and angles for the Co(tach)

equatorial positions. This additional Cu^IIcenter intercon- moiety all fall within the expected ranges (Figure 7),

nects two mononuclear [Cu(Hcis-dap)₂(OH₂)₂]^4ϩcations whereas the Co^IIIϪN bonds in the Co(cis-dap) moiety are

giving

a

trinuclear [(Hcis-dap)Cu(OH₂)₂(µ-cis-dap)- remarkably long. This is particularly true for the secondary

Cu(OH₂)(SO₄)₂(µ-cis-dap)Cu(Hcis-dap)(OH₂)₂]^4ϩ

entity. nitrogen atom, which has a CoϪN distance of 2.068(5) A.

˚

The observation of such trinuclear species nicely illustrates The puckering parameters [Q ϭ 0.601(4) A and ϕ ϭ 0.8(4)°]

the ability of the neutral cis-dap molecule to participate in for the pyrrolidine ring correspond to an almost ideal envel-

bridging interactions, and we regard this ligand as an inter- ope conformation. This conformation implies an eclipsed

esting building block for the construction of coordination orientation of the two CH protons with a torsional angle

polymers.

of only 8.7°, a rather short (NϪ)H···H(ϪN) separation of

˚

A complex featuring a tridentate coordination mode of 2.24 A for the primary amino groups, and a significantly

cis-dap, with all the nitrogen atoms of the ligand bound to elongated C7ϪC8 bond (1.60 A). Owing to the rather long

˚

Figure 6. Structural representations of [Cu(Hcis-dap)₂(OH₂)₂](SO₄)₂·3.5H₂O Ϫ 2x H^ϩϩ x Cu^2ϩ: the two [Cu(Hcis-dap)₂(OH₂)₂]^4ϩcations

˚

for x Ͻ 0.01 (top); hydrogen atoms are shown as spheres of arbitrary size; selected bond lengths [A] and angles [°]: Cu1ϪN3 2.0174(13),

Cu1ϪN4 2.0271(13), Cu1ϪO10 2.4660(12), Cu2ϪN13 2.0144(13), Cu2ϪN14 2.0186(13), Cu2ϪO20 2.4985(13), N3ϪCu1ϪN4 84.40(5),

N13ϪCu2ϪN14 84.73(5); view of the trimer with the partially occupied Cu3 position (x ϭ 0.11) (bottom); hydrogen atoms are omitted

2Ϫ

for clarity; in the absence of Cu3 the ring nitrogen atom N12 is protonated, forming a hydrogen bond to the SO₄counterion (O32);

only one of the two positions of the disordered O32 is shown; thermal ellipsoids are drawn at a 50% probability level

Eur. J. Inorg. Chem. 2001, 2525Ϫ2542

2531

K. Hegetschweiler et al.

FULL PAPER

Scheme 3

Co^IIIϪN bonds in the Co(cis-dap) moiety, a decrease in li- one exocyclic and one endocyclic nitrogen donor, respect-

gand field and an increased tendency to form the corres- ively. Toftlund et al. have previously reported the prepara-

ponding Co^IIcomplex could be expected.^[20]However, [Co- tion of the yellow, diamagnetic complex [Ni(cis-

(cis-dap)(tach)]^3ϩexhibits the usual yellow color with ab- cptn)₂]^{2ϩ [10]}

We have crystallized this species as [Ni(cis-

.

sorbance maxima at 351 and 475 nm, as expected for a nor- cptn)₂](ClO₄)₂and report its crystal structure (Figure 8). As

mal Co^III(hexaamine) chromophore.^[21]Cyclic voltammetry expected for a diamagnetic complex, the Ni center has a

revealed quasi-reversible redox behavior with a redox po- square-planar NiN₄geometry with an average NiϪN bond

˚

tential of Ϫ0.21 V (versus NHE). Compared to the range length of 1.92 A. Again, the two cyclopentane rings have a

usually observed for Co^III(hexaamine) complexes (0.3Ϫ0.5 transoid arrangement with respect to the NiN₄plane.

V versus NHE), the potential of [Co(cis-dap)(tach)]^2/3ϩin- [Cu(ampy)₂]^2ϩ, which has also been crystallized as its per-

dicates only a minor degree of stabilization of the Co^IIspe- chlorate salt (Figure 8), represents an example of a struc-

cies.^[5,20,21]

ture where both endo- and exocyclic nitrogen donors are

The diamine ligands cis-cyclopentane-1,2-diamine (cis- used for metal binding.^[22Ϫ24]The ligand was used as a race-

cptn) and 3-aminopyrrolidine (ampy) have been included in mate and the crystal structure of [Cu(3R-ampy)(3S-ampy)]-

this investigation as structural models for a coordination of (ClO₄)₂showed the incorporation of both enantiomers in

a metal cation to either two exocyclic nitrogen donors or to one complex entity. The Cu^IIcenter has square-planar

2532

Eur. J. Inorg. Chem. 2001, 2525Ϫ2542

Coordination Chemistry of cis-3,4-Diaminopyrrolidine and Related Polyamines

FULL PAPER

Cu(ampy) fragment can be regarded as a metalla derivative

of norbornane. By analogy with the structure of this hydro-

carbon,^[25]the geometry at the bridging methylene groups

deviates significantly from a tetrahedral arrangement

(NϪCϪC angles: 98.9° and 97.6°) and, compared to the

corresponding cis-cptn system, the M(ampy) structure ap-

pears to be considerably strained.

Discussion

Equilibria in Aqueous Solution

A comparison of the protonation constants of cis-dap,

trans-dap, cis-cptn, trans-cptn, ampy, and related amines

(Table 5) indicates some characteristic trends. Compared to

related monoamines such as cyclopentylamine (pK_aϭ 10.6)

or pyrrolidine (pK_aϭ 11.2),^[26]the triamines cis-dap and

trans-dap are weaker bases. On statistical grounds, the pres-

ence of three basic sites should increase the overall basicity;

however, it seems that the electron-withdrawing influence of

additional amino groups outweighs this effect. In contrast

Figure 7. Molecular structure of [Co(cis-dap)(tach)]^3ϩ; thermal el-

lipsoids are drawn at a 50% probability level; hydrogen atoms are

˚

shown as spheres of arbitrary size; selected bond lengths [A] and

angles [°]: CoϪN2 1.958(6), CoϪN3 1.964(5), CoϪN1 1.966(4),

CoϪN6 1.975(4), CoϪN5 2.001(7), CoϪN4 2.068(5), C7ϪC8

1.598(5); the other CϪC bonds range from 1.518(6) to 1.532(5);

N6ϪCoϪN5 76.27(15), N5ϪCo1ϪN4 85.2(3), N6ϪCo1ϪN4

84.4(2); intraligand NϪCoϪN angles in the Co(tach) moiety range

from 89.4(3) to 92.40(14)

1

to the monoamine analogues, the H NMR titration did

not show a significant difference for the intrinsic basicities

of the primary and secondary amino groups in cis-dap^[5]

and it appears that the different behavior of the individual

diastereomers is based on steric rather than electronic ef-

fects. The cyclopentane or pyrrolidine frameworks give rise

to particularly rigid molecules and the different basicities

of the diastereomers can simply be explained in terms of

electrostatic repulsion. In H₂trans-cptn^2ϩ, the two amonio

groups adopt a staggered (antiperiplanar) conformation

and the relevant pK_avalues of H₂trans-cptn^2ϩare in close

agreement with those of the open-chained H₂en^2ϩ(en ϭ

ethane-1,2-diamine): the difference pK₁Ϫ pK₂ϭ ∆pK is

2.70 for H₂trans-cptn^2ϩand 2.80 for H₂en^2ϩ. In H₂cis-

cptn^2ϩ, the interatomic distance between the two positive

charges is shorter and consequently ∆pK_afor this isomer

is increased to 3.60. Analogously, the acidity constants of

monoprotonated Hcis-dap^ϩand Htrans-dap^ϩare in close

agreement, whereas those of the triply protonated forms

differ considerably with the cis isomer being the stronger

acid.

Analysis of the formation constants together with the

spectroscopic properties of cis-dap complexes in aqueous

solution points to a bidentate rather than a tridentate coor-

dination mode for this ligand:

Figure 8. Molecular structures of [Ni(cis-cptn)₂]^2ϩ(top) and

[Cu(3R-ampy)(3S-ampy)]^2ϩ(bottom); thermal ellipsoids are drawn

at a 30% probability level; hydrogen atoms are shown as spheres

(i) Ni^IIforms a tris(chelate) with an NiN₆chromophore,

whereas the λ_maxvalues of the two bis(complexes) [Ni(cis-

dap)₂]^2ϩand [Ni(Hcis-dap)₂]^4ϩare both indicative of an

NiN₄chromophore.^[14]Clearly, deprotonation of the am-

monium groups is not followed by metal binding, and a

third ligand is required to achieve hexaamine formation.

Similarly, the bis(complexes) [CuH_xL₂]^(2ϩx)ϩ(0 Յ x Յ 2,

˚

of arbitrary size; selected bond lengths [A] and angles [°]: NiϪN1

1.928(3), NiϪN2 1.916(3), N2ϪNiϪN1 86.01(14); CuϪN11

2.027(4), CuϪN12 2.013(4), CuϪN21 2.027(4), CuϪN22 2.010(4),

N22ϪCuϪN12

N21ϪCuϪN11

174.4(2),

174.8(3),

N12ϪCuϪN11

N22ϪCuϪN11

95.91(17),

N12ϪCuϪN21 98.18(16), N22ϪCuϪN21 82.44(17)

CuN₄geometry with a very weak interaction with one of L ϭ cis-dap, trans-dap; and x ϭ 0, L ϭ cis-cptn) all show

the ClO₄^Ϫcounterions. The two endocyclic and the two ex- a characteristic band in their visible spectra with an absorp-

ocyclic nitrogen donors each have a trans orientation. Each tion maximum at 550Ϫ570 nm (Table 6). These values are

Eur. J. Inorg. Chem. 2001, 2525Ϫ2542

2533

K. Hegetschweiler et al.

FULL PAPER

of magnitude lower than those of [M(en)]^2ϩ. For the 1:2

complexes, this effect is doubled.

Table 5. Comparison of protonation (log K_i) and complex forma-

tion constants (log β₂) for some selected polyamines

(iii) As a consequence of the JahnϪTeller distortion, tria-

mine ligands that are restricted to a tripodal, facial coor-

dination usually show a characteristic inversion of the

IrvingϪWilliams series with higher stability for Ni^IIthan

for Cu^II. For cis-dap, however, no such inversion is ob-

served. As shown in Figure 9, the stabilities of the cis-dap

complexes fall in the expected range based on the linear

free-energy relationship of log β_Nivs. log β_Cufor ‘‘normal’’

open-chained polyamines.

H^ϩ

log K₂

Cu^II

log β₂

Ni^II

logβ₂

Polyamine^[a]

Triamines

log K₃

log K₁

[b]

[c]

cis-dap

trans-dap^[d]

trap

tmca

dapi

2.4

3.7

3.6

5.2

4.2

6.3

7.9

6.9

7.6

9.7

9.6

9.3

9.5

17.5

14.1

19.6

23.6

20.1

11.2

Ϫ

17.4

25.9

21.2

Diamines

cis-cptn

trans-cptn^[e]

cis-chxn

trans-chxn^[e]

ampy^[e]

6.1

7.2

6.3

6.6

6.8

7.1

9.7

9.9

9.7

9.8

10.4

9.9

19.8

15.8

20.0

21.1

15.0

19.6

12.7

Ϫ

13.8

14.3

8.1

en

13.4

^[a]Abbreviation of ligands and references: cis-dap ϭ cis-3,4-diamin-

opyrrolidine; trans-dap ϭ trans-diaminopyrrolidine; trap ϭ 1,2,3-

propanetriamine;^[27]tmca ϭ all-cis-2,4,6-trimethoxycyclohexane-

1,3,5-triamine;^[28]dapi ϭ cis-3,5-piperidinediamine;^[5]cis-cptn ϭ

cis-cyclopentane-1,2-diamine; trans-cptn ϭ trans-cyclopentane-1,2-

diamine; cis-chxn ϭ cis-cyclohexane-1,2-diamine;^[26]trans-chxn ϭ

trans-cyclohexane-1,2-diamine;^[26]ampy

ϭ 3-aminopyrrolidine;

en ϭ ethane-1,2-diamine^[26]. Ϫ ^[b]K_iϭ [H_iL] ϫ [H]^Ϫ1ϫ [H_iϪ1L]^Ϫ1

.

[c]

[d]

Ϫ

[e]

β₂ϭ [ML₂] ϫ [M]^Ϫ1ϫ [L]^Ϫ2. Ϫ

(3R,4R) enantiomer. Ϫ

Racemate.

Figure 9. Comparison of formation constants for Cu^II- and Ni^II(a-

mine) complexes; the linear free-energy relationship (squares) for

selected diamines (log β₂), triamines H₂NϪ(CH₂)_xϪNHϪ

CH₂)_yϪNH₂(log β₁), and tetraamines H₂NϪ(CH₂)_xϪNHϪ

(CH₂)_yϪNHϪ(CH₂)_zϪNH₂(log β₁) is log β_Niϭ 0.77 ϫ log β_CuϪ

1.84; this is compared with log β₂values of the potentially tripodal

ligands cis-dap, dapi, and tmca (circles); the data are from

refs.^[5,25,27]and from this work (Tables 2 and 3)

Table 6. UV/Vis data for Ni^II- [³A_2g-³T_1g(F) transition] and Cu^II-

(amine) complexes in aqueous solution (25 °C)

Complex

λ_max[nm]

ε

[Cu(Hcis-dap)]^3ϩ

[Cu(cis-dap)₂]^2ϩ

[Cu(cis-dap)(Hcis-dap)]^3ϩ

[Cu(Hcis-dap)₂]^4ϩ

[Cu(trans-dap)₂]^2ϩ

[Cu(ampy)]^2ϩ

688

552

566

561

560

668

550

545

571

577

553

40

86

69

78

140

64

142

92

8

(iv) All the cis-dap complexes show a pronounced tend-

ency to form protonated species [M^z(HL)]^(zϩ1)ϩ

,

[M^z(HL)₂]^(zϩ2)ϩ, and [M^z(HL)L]^(zϩ1)ϩ. [Pt(Hcis-dap)Cl₄]^ϩ,

[Pd(Hcis-dap)₂]^4ϩ, and [Cu(Hcis-dap)₂(OH₂)₂]^4ϩare repres-

entative examples that have been structurally characterized

by single-crystal X-ray diffraction analysis. For the divalent

metal cations (M ϭ Ni, Cu, Zn, Cd), the pK_avalues for

deprotonation of the partially protonated complexes fall in

the range 7.1Ϫ8.1, indicating only a slight decrease in the

[Cu(ampy)₂]^2ϩ

[Cu(cptn)₂]^2ϩ

[Ni(cis-dap)₂]^2ϩ

[Ni(Hcis-dap)₂]^4ϩ

[Ni(cis-dap)₃]^2ϩ

6

10

indicative of a trans (square-planar) CuN₄chromophore basicity of cis-dap upon coordination. This effect would be

with additional loosely bound water molecules at the api- incompatible with a significant interaction between the

ces.^[14]It is well known that the coordination of an addi- third amino group and the metal cation. It is also note-

tional nitrogen donor in an apical position would lead to worthy that the pK_avalues of [Cu(HL)₂]^4ϩare very similar

a significant red shift (pentaammine effect).^[16]The close for L ϭ cis-dap and L ϭ trans-dap (7.1, 7.7 versus 6.8, 7.5,

similarity of the spectral properties of the differently pro- respectively). The trans isomer is even slightly more acidic.

tonated species again indicates that deprotonation at the Since tridentate coordination of trans-dap can be ruled out

third nitrogen atom does not result in a significant re- on steric grounds, the even higher basicity of [Cu(cis-

arrangement of the chromophore.

dap)₂]^2ϩwould again be inconsistent with a significant in-

(ii) Compared to other tridentate cyclic triamines such as teraction between the third nitrogen atom and the metal

3,5-diaminopiperidine (dapi) or all-cis-2,4,6-trimethoxycy- cation.

clohexane-1,3,5-triamine (tmca), the formation constants

The arguments listed as (i)Ϫ(iv) clearly show that trident-

for [M(cis-dap)₂]^2ϩare remarkably low (Table 5). The ate coordination of cis-dap is not observed in labile solution

formation constants of [M(cis-dap)]^2ϩare about one order species. Additionally, monodentate coordination of any of

2534

Eur. J. Inorg. Chem. 2001, 2525Ϫ2542

Coordination Chemistry of cis-3,4-Diaminopyrrolidine and Related Polyamines

FULL PAPER

the studied ligands can also be excluded. Although some of metries: two exocyclic nitrogen atoms in a cis arrangement

the bis(complexes), particularly [M(ampy)₂]^2ϩ(M ϭ Ni, Ͼ two exocyclic nitrogen atoms in a trans arrangement Ͼ

Cu, Cd) and [Cu(trans-dap)₂]^2ϩ, show rather low stabilities, one endocyclic and one exocyclic nitrogen atom.

their formation constants are still considerably higher than

(b) Since all the ligands coordinate in a bidentate fashion,

those for monodentate amines such as NH₃or CH₃NH₂. the presence of an additional nitrogen atom reduces the nu-

Moreover, not only cis-dap but all the other ligands as well cleophilicity of the donor groups and, consequently, the tri-

give rise to a visible spectrum for [CuL₂]^2ϩthat is indicative amines are generally weaker ligands than the correspond-

of a trans-CuN₄chromophore.

ing diamines.

The solid-state structures of [Pt(Hcis-dap)Cl₄]^ϩ,

Although in terms of absolute stability, the metal binding

[Pd(Hcis-dap)₂]^4ϩ, and [Cu(Hcis-dap)₂(OH₂)₂]^4ϩreveal ex- ability of the triamines is lower, the opposite behavior is

clusive coordination of the metal cation to the two primary observed in neutral and acidic solutions. This is based upon

amino groups. Although the potentiometric measurements the ability of the dap complexes to bind additional protons,

do not of course allow a direct structural assignment, com- which is, of course, not possible for the diamines. In terms

parison of cis-dap with the model diamines provided ample of conditional stability,^[29]cis-dap is a rather efficient com-

evidence for a similar binding to the exocyclic nitrogen plexing agent for Cu^IIin acidic solution and is clearly super-

atoms in solution: Metal complexes of ampy are generally ior to any of the known aliphatic or alicyclic diamines. It is

less stable than the corresponding cis-cptn complexes. Cle- in fact a remarkable property of this ligand that it is capable

arly, coordination to the two exocyclic donors is preferred. of significant complexation even under acidic conditions,

Molecular mechanics calculations on [Cu(cis-dap)- such as at pH ϭ 2.5 (Figure 3).

(OH₂)₄]^2ϩalso showed that coordination to the two prim-

ary amino groups would be favored by about 9.3 kJ mol^Ϫ1

.

Structural Aspects of the Metal Complexes

For [Cu(cis-cptn)]^2ϩand [Cu(ampy)]^2ϩthe difference in

Although the coordination behavior of cis-dap in labile

stability (∆log β₁) is 2.4, corresponding to a difference in complexes must be regarded as being mainly bidentate, in

free enthalpy of 13.5 kJ mol^Ϫ1

principal a facial tridentate coordination is possible, as ex-

Comparison of the metal complex stabilities of the differ- emplified by the inert Co^IIIcomplex [Co(cis-dap)(tach)]^3ϩ

.

ent ligands investigated in this study reveals the order cis- However, the structural parameters for this complex, par-

cptn Ͼ cis-dap Ͼ trans-cptn Ͼ ampy Ͼ trans-dap. Regard- ticularly the long CoϪN bonds, clearly reveal a severely

ing [Cu(HL)]^3ϩformation, the formation constant of the strained structure. We were unable to model this particular

trans-dap complex is lowered by about two orders of magni- Co^IIIN₆geometry with a commercially available MM pro-

tude compared with that of the cis-dap derivative, and for gram.^[30,31]It seems that a nonharmonic description of the

the bis(complexes) [Cu(HL)₂]^4ϩ

,

[Cu(HL)L]^3ϩ

,

and Co^IIIϪN bond would be required, which was not imple-

[CuL₂]^2ϩthe decrease in stability is about four orders of mented by the force field used. In a complex with bidentate

magnitude. Similar effects are observed for trans-cptn and coordination, the noncoordinating endocyclic nitrogen

cis-cptn. The cis-cyclopentane-1,2-diamine structure is cle- atom can either accept a proton or bind to an additional

arly favorable with regard to metal binding. This is in con- metal center. The first possibility has been discussed extens-

trast to the cyclohexane-1,2-diamine structure, for which ively for solid-state and solution species in the previous sec-

the trans isomer has been reported to be more stable.^[26]The tion. Although we do not have any experimental evidence

formation constants of the cis-cptn complexes are in close that the latter possibility is realized in aqueous solution, the

agreement with those reported for the corresponding en crystal structure of [Cu(Hcis-dap)₂(OH₂)₂](SO₄)₂·3.5H₂O Ϫ

complexes. Evidently, the cis-cyclopentane-1,2-diamine 2x H^ϩϩ x Cu^2ϩ, with its nonstoichiometric composition

framework does not provide any stabilization in terms of (0.01 Յ x Յ 0.11) shows that multiple binding of metal

pre-organization of the donor groups. Again, this is in con- cations to a cis-dap molecule is at least possible in the solid

trast to the cyclohexanediamines, where significantly in- state. It is noteworthy that cis-dap is structurally related to

creased stability is observed for the trans-cyclohexane-1,2- the open-chained propane-1,2,3-triamine (trap): The trap

diamine. Since cis-cptn and cis-dap have related structures, molecule can be regarded as a substructure of cis-dap.

the reduced ability of the latter to bind metal cations must However, the open-chained trap is conformationally flexible

be interpreted in terms of a lower nucleophilicity of the whereas cis-dap is rigid, thus representing a model of the

donor groups, caused by the electron-withdrawing effect of triamine with a locked conformation. Interestingly, the co-

the endocyclic nitrogen atom. Consequently, the higher li- ordination behavior of these two ligands towards Cu^IIis

gand field of cis-cptn is able to generate a low-spin electron similar. The ligand trap, by analogy with cis-dap, forms

configuration for [Ni(cis-cptn)₂]^2ϩ, whereas [Ni(cis-dap)₂]^2ϩsimilar protonated Cu complexes in acidic solution, and a

is high spin. Similar considerations are applicable for the bridging mode with bidentate and monodentate coordina-

trans-cptn and trans-dap systems. The observed decrease in tion has been established from the crystal structure of [Cu-

metal complex stability in the above series can thus be ra- (trap)₂]Cl₂.^[27]However, it appears that the locked conforma-

tionalized in terms of a combination of steric (a) and elec- tion is not optimal for metal binding: The trap complexes

tronic (b) effects:

are generally more stable than the corresponding cis-dap

(a) For the five-membered ring frame, a steady decrease analogues. Moreover, with other metal cations such as Ni^II,

in metal complex stability is observed for the following geo- the flexibility of trap allows tridentate coordination, which

Eur. J. Inorg. Chem. 2001, 2525Ϫ2542

2535

K. Hegetschweiler et al.

FULL PAPER

500 MHz NMR spectrometer (resonance frequencies: 500.13 MHz

for ¹H and 125.9 MHz for ¹³C{¹H}). Chemical shifts are given rel-

ative to [D₄]sodium (trimethylsilyl)propionate (D₂O) or tetra-

methylsilane (all other solvents) as internal standards (δ ϭ 0). pD

values (corrected)^[5]were determined with a glass electrode. The ¹H

NMR titration experiment on cis-dap was performed and evaluated

as reported previously for dapi.^[5]Ϫ UV/Vis spectra of the Ni^II

complexes were measured with a Uvikon 941 spectrophotometer

(H₂O; 25 Ϯ 3 °C). Ϫ Cyclic voltammograms were recorded as de-

scribed previously (BAS C2 cell, BAS 100B/W2 potentiostat, Au

is not accessible for cis-dap. Analogously, the structure of

trans-dap represents another conformation of trap that is

even more disfavored for metal binding.

Another interesting aspect is the comparison of the three

cyclic triamine ligands tach, dapi, and cis-dap (Scheme 1).

The ligand tach shows a very pronounced preference for

facial tridentate coordination, giving an adamantane-like

structure with all six-membered chelate rings. Attempts to

enforce a bidentate coordination mode failed. Reaction

with Pt^IIresulted in oxidation to Pt^IVwith a concomitant working electrode, Pt counter electrode, Ag/AgCl reference elec-

trode, ambient temperature).^[5]Ϫ C,H,N analyses were performed

by H. Feuerhake (Universität des Saarlandes).

facial coordination of tach.^[32]On the other hand, cis-dap

shows a strong preference for bidentate coordination. A fa-

cial, tridentate coordination mode comprising exclusively

five-membered chelate rings is severely strained. Such an

[M(cis-dap)] structure can formally be derived from the

[M(tach)] moiety by the removal of one CH₂and one CH

Materials: Unless otherwise stated, all chemicals used for the syn-

thetic work were commercially available products of reagent-grade

quality. For the potentiometric and spectrophotometric titrations,

metal salts of highest available quality (Ͼ 99.95%) were used. 3-

group. The [M(dapi)] structure contains one six-membered Pyrroline^[33]and (3S,4S)-1-benzyl-3,4-pyrrolidinediyl bis(methane-

sulfonate)^[34]were prepared as described in the literature. [Co-

and two five-membered chelate rings, in analogy to nor-

(tach)(NO₃)₃] was obtained from [Co(tach)Cl₃].^[35]Dowex 50 W-X2

adamantane. As a true intermediate in this series, dapi pre-

dominantly coordinates in a tridentate mode. However, bi-

(100Ϫ200 mesh, H^ϩform) and Dowex 2-X8 (50Ϫ100 mesh, Cl^Ϫ

form) were purchased from Fluka. The anion resin was converted

dentate coordination is possible, as has been observed for

into the OH^Ϫform using dilute aqueous NaOH.^[5]

Pd^II.^[4]

1-Acetyl-3-pyrroline: To an ice-cooled solution of 3-pyrroline

(4.00 g, 57.88 mmol) in MeOH (50 mL), acetic anhydride (23.64 g,

231.56 mmol) was added dropwise. The clear solution was then

stirred for 30 min at room temperature, cooled in an ice bath once

Conclusion

The basic coordination mode of the cyclic triamine cis-

3,4-diaminopyrrolidine is a bidentate coordination through

the two exocyclic amino groups. This mode has been estab-

lished in a variety of solid-state structures. The potentio-

metric data and the spectroscopic properties of cis-dap

complexes have also provided evidence for this coordination

mode in aqueous solution. The noncoordinated endocyclic

nitrogen atom represents a comparatively strong basic site,

and as such it is readily protonated. A facial, tridentate co-

ordination mode results in a highly strained structure, as

shown for the inert [Co(cis-dap)(tach)]^3ϩ. The formation of

such strained species is not trivial and requires specific syn-

thetic strategies. Alternatively, coordination of cis-dap to

two different metal cations is possible. In this mode, one

metal cation is bound through the primary amino groups,

whereas the other is coordinated to the endocyclic nitrogen

donor. The coordination behavior of cis-dap is in marked

contrast to that of other cyclic triamine ligands such as cis-

3,5-diaminopiperidine or all-cis-cyclohexane-1,3,5-triamine,

where bidentate coordination or a bridging mode are of mi-

nor importance or even unknown.

more, and made alkaline with 2  aq. NaOH. After extraction with

dichloromethane (3 ϫ 100 mL), the combined organic phases were

dried with Na₂SO₄and concentrated to dryness. The oily residue

was sublimed (10^Ϫ2mbar, 40 °C) to yield 1-acetyl-3-pyrroline as

white crystals (6.07 g, 94%). Ϫ C₆H₉NO (111.1): calcd. C 64.84, H

8.16, N 12.60; found C 64.92, H 8.41, N 12.51. Ϫ ¹H NMR

(CDCl₃): δ ϭ 2.08 (s, 3 H), 4.25 (m, 4 H), 5.81 (m, 1 H), 5.88 (m,

1 H). Ϫ ¹³C{¹H} NMR (CDCl₃): δ ϭ 22.1 (CH₃), 52.8 (CH₂), 54.2

(CH₂), 125.0 (CH), 126.5 (CH), 169.1 (C).

1-Acetyl-cis-3,4-pyrrolidinediol:

1-Acetyl-3-pyrroline

(8.93 g,

80.35 mmol) and 4-methylmorpholine 4-oxide monohydrate

(12.99 g, 96.11 mmol) were dissolved in acetone/H₂O (1:1, v/v;

160 mL). A solution of OsO₄(2.5% in tert-butyl alcohol; 1 mL)

was added and the pale-yellow solution was stirred for 24 h.

Na₂S₂O₅(1.5 g) was then added and the mixture was stirred for a

further 1 h. It was then acidified to pH ϭ 2 with 50% H₂SO₄and

extracted with CH₂Cl₂(3 ϫ 100 mL). The aqueous layer was di-

luted to a total volume of 1000 mL with H₂O and was sorbed

onto Dowex 50 W-X2. The resin was washed with H₂O until neut-

ral and the collected eluent was sorbed onto Dowex 2-X8. This

second column was rinsed with H₂O and the neutral eluent was

concentrated to dryness yielding the diol as a colorless oil, which

solidified after drying in vacuo (10.55 g, 90%). Ϫ C₆H₁₁NO₃

(145.2): calcd. C 49.65, H 7.64, N 9.65; found C 49.65, H 7.84, N

9.64. Ϫ ¹H NMR (D₂O): δ ϭ 2.05 (s, 3 H), 3.36 (dd, J ϭ 4.5,

12.5 Hz, 1 H), 3.47 (dd, J ϭ 5.5, 11.5 Hz, 1 H), 3.61 (dd, J ϭ 5.5,

12.5 Hz, 1 H), 3.78 (dd, J ϭ 6.0, 11.0 Hz, 1 H), 4.31 (m, 1 H), 4.35

(m, 1 H). Ϫ ¹³C{¹H} NMR (D₂O): δ ϭ 23.8 (CH₃), 52.6 (CH₂),

54.2 (CH₂), 72.7 (CH), 73.5 (CH), 176.1 (C).

Experimental Section

Safety Notes: Organic polyazides and perchlorate salts of metal

complexes with organic ligands are potentially explosive. The or-

ganic polyazides should only be handled in dilute solutions and

should not be isolated as pure substances. The perchlorate salts

should only be isolated in small quantities and should not be dried

at elevated temperatures.

(3S*,4R*)-1-Acetyl-3,4-pyrrolidinediyl Bis(methanesulfonate): 1-

Acetyl-cis-3,4-pyrrolidinediol (4.00 g, 27.56 mmol) was dissolved in

dry pyridine (150 mL) and the resulting solution was cooled to Ϫ10

1

Instrumentation: H and ¹³C{¹H} NMR spectra were measured in °C (ice/NaCl). Methanesulfonyl chloride (4.67 mL, 60.10 mmol)

D₂O, CDCl₃, CD₃CN, or [D₆]acetone at 28 °C with a Bruker DRX

was added dropwise with stirring. The ice bath was removed and

2536

Eur. J. Inorg. Chem. 2001, 2525Ϫ2542

Coordination Chemistry of cis-3,4-Diaminopyrrolidine and Related Polyamines

the mixture was stirred for 1 h at room temperature. The suspen- 56.2 (CH). Ϫ Single crystals were grown from an aqueous solution

FULL PAPER

sion was then poured into Et₂O (1000 mL) and the resulting mix-

ture was left to stand for 2 d at 4 °C. The Et₂O was decanted and

the remaining white solid was redissolved in hot acetone (150 mL).

This solution was subjected to flash chromatography (silica gel/

by slow evaporation of the water; cis-dap·3HNO₃·H₂O was pre-

pared by deprotonation of cis-dap·3HCl·H₂O (1.00 g, 4.38 mmol)

on Dowex 2-X8 (OH^Ϫform). The alkaline fraction was concen-

trated to dryness and the residual oil was redissolved in MeOH

acetone) and the eluate was concentrated to dryness. Yield 7.50 g (50 mL). The resulting solution was acidified with 3 equiv. of

(90%) of a white solid. Ϫ C₈H₁₅NO₇S₂(301.3): calcd. C 31.89, H HNO₃(3  in H₂O) and left to stand at 4 °C for 14 h. The precipit-

5.02, N 4.65; found C 31.80, H 5.01, N 4.60. Ϫ ¹H NMR ([D₆]ace-

ate obtained was collected by filtration and dried in air. Yield 1.12 g

tone): δ ϭ 2.01 (s, 3 H), 3.26 (s, 3 H), 3.27 (s, 3 H), 3.62 (dd, J ϭ (83%). Ϫ C₄H₁₆N₆O₁₀(308.2): calcd. C 15.59, H 5.23, N 27.27;

5.5, 10.0 Hz, 1 H), 3.74 (dd, J ϭ 6.0, 11.0 Hz, 1 H), 3.83 (dd. J ϭ found C 15.71, H 5.26, N 26.95.

5.0, 10.0 Hz), 4.07 (dd, J ϭ 6.0, 11.0 Hz, 1 H), 5.38 (m, 1 H), 5.43

[Pd(Hcis-dap)₂](ClO₄)₄·H₂O:

A mixture of cis-dap·3HCl·H₂O

(m, 1 H). Ϫ ¹³C{¹H} NMR ([D₆]acetone): δ ϭ 21.8 (CH₃), 38.5

(CH₃), 38.6 (CH₃), 48.8 (CH₂), 49.9 (CH₃), 77.1 (CH), 77.3 (CH),

169.4 (C).

(0.40 g, 1.75 mmol) and PdCl₂(0.16 g, 0.87 mmol) in H₂O (50 mL)

was refluxed for 3 h and the clear yellow solution was sorbed

onto Dowex 2-X8 (OH^Ϫform). The column was washed with H₂O

until a neutral eluent was obtained. The combined alkaline frac-

tions were acidified with 70% HClO₄and the solvents were care-

fully evaporated under reduced pressure until a volume of 5 mL

remained. Conc. HClO₄(5 mL) was added and the solution was

set aside at room temperature for 2 d. Slightly yellow crystals pre-

cipitated. These were collected by filtration and washed with EtOH

and Et₂O (0.52 g, 82%). Ϫ C₈H₂₆Cl₄N₆O₁₇Pd (726.6): calcd. C

13.22, H 3.61, N 11.57; found C 13.33, H 3.51, N 11.54. Ϫ ¹H

NMR (D₂O; pD Ͻ 2): δ ϭ 3.64 (m, 4 H), 3.82 (m, 4 H), 3.95 (m,

4 H); (D₂O; pD Ͼ 12): δ ϭ 3.02 (m, 4 H), 3.28 (m, 4 H), 3.60 (m,

4 H). Ϫ ¹³C{¹H} NMR (D₂O; pD Ͻ 2): δ ϭ 50.8 (CH₂), 50.9

(CH₂), 62.4 (CH), 62.5 (CH); (D₂O; pD Ͼ 12): δ ϭ 52.5 (CH₂),

52.6 (CH₂), 63.9 (CH), 64.0 (CH). Ϫ UV/Vis: λ_max(ε) ϭ 290 (347).

Ϫ Crystals were grown by slow evaporation of the water from an

aqueous solution of [Pd(Hcis-dap)₂](ClO₄)₄that had been acidified

with a few drops of HClO₄.

1-Acetyl-cis-3,4-diaminopyrrolidine: A stirred mixture of (3S*,4R*)-

1-acetyl-3,4-pyrrolidinediyl bis(methanesulfonate) (7.50 g, 24.89

mmol) and NaN₃(8.10 g, 124.60 mmol) in dry DMF (200 mL) was

heated at 100 °C for 4 h. The white solid formed was filtered off

and washed with dry DMF (2 ϫ 50 mL). To the dark-red solution

was added 10% Pd/C (500 mg) and the suspension was immediately

transferred to an autoclave for hydrogenation. A small sample was

extracted from the reaction mixture with Et₂O for NMR character-

ization of the diazide. This sample was set aside, the Et₂O was

1

evaporated, and the oily residue was redissolved in CD₃CN. Ϫ H

NMR (CD₃CN): δ ϭ 1.91 (s, 3 H), 3.28 (dd, J ϭ 5.0, 12.5 Hz, 1

H), 3.38 (dd, J ϭ 5.5, 11.0 Hz, 1 H), 3.56 (dd, J ϭ 5.0, 12.5 Hz, 1

H), 3.77 (dd, J ϭ 6.0, 10.5 Hz, 1 H), 4.38 (m, 1 H), 4.43 (m, 1 H).

Ϫ

¹³C{¹H} NMR (CD₃CN): δ ϭ 22.2 (CH₃), 49.0 (CH₂), 50.3

(CH₂), 62.0 (CH), 63.0 (CH), 171.6 (C). Ϫ The main portion of

the reaction mixture was hydrogenated with vigorous stirring (room

temp., 4 bar) for 12 h and then the mixture was filtered. The clear

filtrate was acidified to pH ϭ 2 with 3  HCl, diluted to a total

volume of 1000 mL with H₂O, and sorbed onto Dowex 50 W-X2.

The column was washed with H₂O and 0.5  HCl. Elution with

3  HCl gave a yellow fraction, which was concentrated to dryness.

¹H NMR spectroscopy indicated mainly the formation of 1-acetyl-

cis-3,4-diaminopyrrolidine (ca. 95%) besides some cis-3,4-diamino-

pyrrolidine (5.00 g of crude product). Ϫ ¹H NMR of the main com-

ponent (D₂O; pD Ͻ 2): δ ϭ 2.12 (s, 3 H), 3.74 (dd, J ϭ 5.5,

13.0 Hz, 1 H), 3.87 (dd, J ϭ 6.0, 12.0 Hz, 1 H), 3.94 (dd, J ϭ 6.5,

13.0 Hz, 1 H), 4.16 (dd, J ϭ 6.5, 12.0 Hz, 1 H), 4.35 (m, 1 H), 4.40

(m, 1 H); (D₂O; pD Ͼ 12): δ ϭ 2.06 (s, 3 H), 3.25 (dd, J ϭ 5.0,

12.0 Hz, 1 H), 3.35 (dd, J ϭ 6.0, 11.0 Hz, 1 H), 3.47 (m, 1 H), 3.51

(m, 1 H), 3.59 (dd, J ϭ 6.5, 12.5 Hz, 1 H), 3.76 (dd, J ϭ 6.5,

11.0 Hz, 1 H). Ϫ ¹³C{¹H} NMR for 1-acetyl-cis-3,4-diaminopyrro-

lidine (D₂O; pD Ͻ 2): δ ϭ 24.1 (CH₃), 49.6 (CH₂), 51.0 (CH₂),

52.4 (CH), 53.3 (CH), 176.1 (C); (D₂O; pD Ͼ 12): δ ϭ 23.8 (CH₃),

53.5 (CH₂), 54.6 (CH), 55.2 (CH₂), 55.6 (CH), 175.8 (C).

[Pt(Hcis-dap)Cl₄]Cl·H₂O: cis-dap·3HCl·H₂O (250 mg, 1.09 mmol)

was dissolved in H₂O (10 mL) and deprotonated using Dowex 2-

X8 (OH^Ϫform). The resulting solution was then concentrated to

dryness under reduced pressure. The oily residue was redissolved

in dry EtOH (5 mL) and this solution was cooled to 0 °C. A solu-

tion of H₂PtCl₆·6H₂O (283 mg, 0.55 mmol) in dry EtOH (10 mL)

was then added dropwise, which led to the deposition of a yellow

precipitate. The suspension was stirred for 2 h at 65 °C. The solid

was then collected by filtration, recrystallized from hot H₂O (3 mL)

and conc. HCl (20 mL), and washed with EtOH and Et₂O. Yellow

powder (150 mg, 55%). Ϫ C₄H₁₄Cl₅N₃OPt (492.5): calcd. C 9.75,

H 2.87, N 8.53; found C 10.16, H 2.89, N 8.68. Ϫ ¹H NMR (D₂O;

pD Ͻ 2): δ ϭ 3.96 (m, 4 H), 4.38 (m, 2 H). Ϫ ¹³C{¹H} NMR

(D₂O; pD Ͻ 2): δ ϭ 48.9 (CH₂), 66.2 (CH).

[Cu(Hcis-dap)₂(OH₂)₂](SO₄)·3.5H₂O: cis-dap·3HCl·H₂O (250 mg,

1.09 mmol) was dissolved in H₂O (10 mL) and deprotonated using

Dowex 2-X8 (OH^Ϫform). The resulting solution was concentrated

to dryness under reduced pressure and the oily residue was redis-

solved in MeOH (50 mL). A solution of CuSO₄·5H₂O (272 mg,

1.09 mmol) in H₂O (5 mL) was added, which led to the deposition

of a blue precipitate. The suspension was diluted with H₂O (10 mL)

and refluxed for 1 h. The insoluble solid was then removed by fil-

tration. MeOH was added dropwise to the deep-blue solution until

it became turbid. Single crystals were grown by heating the suspen-

sion until the precipitate redissolved and then allowing it to slowly

cool to room temperature. Ϫ C₈H₃₅CuN₆O_13.5S₂(559.1): calcd. C

17.19, H 6.31, N 15.03; found C 17.00, H 6.19, N 14.89.

cis-3,4-Diaminopyrrolidine: The crude 1-acetyl-cis-3,4-diaminopyr-

rolidine (7.40 g) was dissolved in 3  HCl (150 mL) and the solu-

tion was heated to 100 °C for 1 h. It was then diluted to a total

volume of 2000 mL with H₂O and sorbed onto Dowex 50 W-X2.

The column was eluted with H₂O, 0.5  HCl, and 3  HCl. The

last fraction was collected and concentrated to dryness. The white

solid obtained was redissolved in H₂O (50 mL) and reprecipitated

with conc. HCl (100 mL). The product was washed with EtOH and

Et₂O and dried in air yielding cis-dap·3HCl·H₂O (5.37 g, 69%). Ϫ

C₄H₁₆Cl₃N₃O (228.5): calcd. C 21.02, H 7.06, N 18.39; found C

1

20.99, H 6.97, N 18.37. Ϫ H NMR (D₂O; pD Ͻ 2): δ ϭ 3.71 (m, [Co(cis-dap)(tach)]Cl₃·1.5H₂O: cis-dap·3HCl·H₂O (241 mg, 1.05

2 H), 3.99 (m, 2 H), 4.41 (m, 2 H); (D₂O; pD Ͼ 12): δ ϭ 2.53 (m,

mmol) was dissolved in H₂O (10 mL) and deprotonated using

2 H), 3.11 (m, 2 H), 3.23 (m, 2 H). Ϫ ¹³C{¹H} NMR (D₂O; pD Ͻ Dowex 2-X8 (OH^Ϫform). The resulting solution was concentrated

2): δ ϭ 49.4 (CH₂), 52.5 (CH); (D₂O; pD Ͼ 12): δ ϭ 53.9 (CH₂), to dryness under reduced pressure and the oily residue was redis-

Eur. J. Inorg. Chem. 2001, 2525Ϫ2542

2537

K. Hegetschweiler et al.

FULL PAPER

solved in dry MeOH (50 mL). This solution was added dropwise HCl (200 mL) in small portions whilst stirring. The product precip-

to a purple solution of [Co(tach)(NO₃)₃] (3.16 mmol) in dry

itated as a white solid, which was collected by filtration, washed

MeOH. The mixture was refluxed for 14 h and then concentrated with H₂O, and dried in vacuo (5.23 g, 83%). Ϫ C₇H₁₄O₆S₂(258.31):

to dryness. The brownish residue was suspended in H₂O (30 mL) calcd. C 32.55, H 5.46; found C 32.42, H 5.31. Ϫ ¹H NMR

and the suspension was acidified with 3  HCl and heated to boil-

ing for a few minutes. Insoluble components were removed by fil-

(CDCl₃): δ ϭ 1.67Ϫ1.73 (m, 1 H), 1.98Ϫ2.12 (m, 5 H), 3.10 (s, 6

H), 5.00 (m, 2 H). Ϫ ¹³C{¹H} NMR (CDCl₃): δ ϭ 18.3 (CH₂),

tration. The orange-red filtrate was diluted to a volume of 500 mL 28.4 (CH₂), 38.5 (CH₃), 80.5 (CH).

with H₂O and sorbed onto Dowex 50 W-X2. The column was

cis-1,2-Diaminocyclopentane: A mixture of (1S*,2R*)-1,2-cyclopen-

eluted with 0.5  HCl to give a pink fraction, which was discarded.

Further elution with 3  HCl gave an orange fraction, which was

concentrated to dryness. Recrystallization from 6  HCl yielded

65 mg (14%) of the trichloride salt. Ϫ C₁₀H₂₉CoCl₃N₆O_1.5(422.7):

calcd. C 28.42, H 6.92, N 19.88; found C 28.41, H 6.68, N 19.75.

Ϫ ¹H NMR (D₂O): δ ϭ 1.88 (m, 3 H), 2.14 (m, 3 H), 2.60 (d, J ϭ

11.0 Hz, 2 H), 2.79 (d, J ϭ 11.0 Hz, 2 H), 2.84 (s, 1 H), 3.09 (s, 2

H), 3.78 (s, 2 H). Ϫ ¹³C{¹H} NMR: δ ϭ 33.6 (CH₂), 33.7 (CH₂),

42.4 (CH), 42.7 (CH), 59.3 (CH), 62.6 (CH₂). Ϫ UV/Vis: λ_max(ε) ϭ

351 (142), 475 (102). Ϫ CV: E ϭ Ϫ210 mV. Ϫ Crystals of the

chloride tetrachlorozincate salt suitable for X-ray diffraction ana-

lysis were grown by layering a solution of [Co(cis-dap)(tach)]Cl₃

and ZnCl₂in conc. HCl with EtOH at 4 °C.

tanediyl bis(methanesulfonate) (4.49 g, 17.38 mmol) and NaN₃

(3.40 g, 53.30 mmol) was heated at 100 °C whilst stirring in dry

DMF (100 mL) for 4 h and then stirred for an additional 12 h at

room temperature. H₂O (100 mL) was then added and the mixture

was extracted with Et₂O (4 ϫ 100 mL). The combined ethereal

phases were washed with H₂O (100 mL) and dried with Na₂SO₄.

A small sample (5 mL) was set aside until the ether evaporated,

and the oily residue was dissolved in CDCl₃for NMR characteriza-

tion of the diazide. Ϫ ¹H NMR (CDCl₃): δ ϭ 1.60Ϫ1.68 (m, 1 H),

1.76Ϫ1.94 (m, 5 H), 3.80 (m, 2 H). Ϫ ¹³C{¹H} NMR (CDCl₃): δ ϭ

19.9 (CH₂), 27.9 (CH₂), 64.4 (CH). Ϫ The main portion of the

ethereal solution was concentrated under reduced pressure to a vol-

ume of 200 mL and then diluted with EtOH (150 mL). This solu-

tion was transferred to an autoclave and hydrogenated for 26 h at

4 bar under vigorous stirring in the presence of 10% Pd/C (300 mg).

The resulting suspension was heated to boiling for a few minutes

and the catalyst was removed by filtration. The clear filtrate was

concentrated under reduced pressure to a volume of 150 mL. Conc.

HCl was added portionwise until the product precipitated (white

needles). The solid was collected by filtration, washed with EtOH,

and equilibrated in air, yielding cis-cptn·2HCl·0.5H₂O (1.57 g; 50%

based on the amount of methanesulfonate). Ϫ C₅H₁₅Cl₂N₂O_0.5

(182.1): calcd. C 32.98, H 8.30, N 15.38; found C 33.00, H 8.30, N

15.35. Ϫ ¹H NMR (D₂O; pD Ͻ 2): δ ϭ 1.80Ϫ1.93 (m, 3 H),

1.94Ϫ2.00 (m, 1 H), 2.23 (m, 2 H), 3.92 (m, 2 H); (D₂O; pD Ͼ 12):

δ ϭ 1.38 (m, 2 H), 1.52 (m, 1 H), 1.70 (m, 1 H), 1.86 (m, 2 H),

3.07 (m, 2 H). Ϫ ¹³C{¹H} NMR (D₂O; pD Ͻ 2): δ ϭ 21.7 (CH₂),

29.8 (CH₂), 55.6 (CH); (D₂O; pD Ͼ 12): δ ϭ 22.7 (CH₂), 33.7

(CH₂), 57.3 (CH).

trans-(3R,4R)-Diaminopyrrolidine: A mixture of (3S,4S)-1-benzyl-

3,4-pyrrolidinediyl bis(methanesulfonate) (11.9 g, 34.06 mmol) and

NaN₃(5.2 g, 80.00 mmol) in dry DMF (150 mL) was stirred at 100

°C for 4 h. After cooling to room temperature, the white solid was

filtered off and the resulting solution was subjected to hydrogena-

tion in a two-step procedure. (1) Reduction of the diazide: The

solution was transferred into an autoclave and hydrogenated for

12 h at 4 bar in the presence of 10% Pd/C (500 mg) under vigorous

stirring. The catalyst was removed by filtration. The filtrate was

acidified with 3  HCl and concentrated to dryness. The brownish

residue was redissolved in H₂O (250 mL) and sorbed onto Dowex

50 W-X2. The column was washed with H₂O and eluted with HCl

(0.5 , 3.0 , 6.0 ; 500 mL each). The latter two fractions were

combined and concentrated to dryness. (2) Cleavage of the benzyl

group: The product was taken up in a mixture of MeOH (100 mL),

H₂O (50 mL), and acetic acid (20 mL) and the suspension was

again hydrogenated in an autoclave for 12 h at 4 bar in the presence

of 10% Pd/C (400 mg). The catalyst was removed and the clear

solution was concentrated to dryness. The solid was then sus-

pended in refluxing conc. HCl (100 mL) and small portions of H₂O

were added until complete dissolution. After cooling to room tem-

perature, the product was precipitated by adding EtOH (50 mL),

collected by filtration, washed with EtOH and Et₂O, and dried in

vacuo. White crystals of the composition trans-(3R,4R)-

[Ni(cis-cptn)₂](ClO₄)₂: cis-cptn·2HCl·0.5H₂O (200 mg, 0.99 mmol)

was dissolved in H₂O (10 mL) and deprotonated using Dowex 2-

X8 (OH^Ϫform). The resulting solution was concentrated to dry-

ness under reduced pressure and the oily residue was redissolved

in dry MeOH (100 mL). A solution of Ni(ClO₄)₂·6H₂O (183 mg,

0.50 mmol) in dry MeOH (25 mL) was added. The resulting yellow

solution was stored at 4 °C for 14 h. Yellow crystals suitable for X-

ray diffraction studies precipitated. These were collected by filtra-

tion and dried in air (227 mg, 85%). Ϫ C₁₀H₂₄Cl₂N₄NiO₈(457.9):

calcd. C 26.23, H 5.28, N 12.24; found C 26.24, H 5.26, N 12.21.

Ϫ UV/Vis: λ_max(ε) ϭ 357 (8.8), 444 (8.3), 574 (5.3).

dap·3HCl·0.25H₂O

were

obtained

(3.32 g,

45%).

Ϫ

C₄H_14.5Cl₃N₃O_0.25(215.0): calcd. C 22.34, H 6.80, N 19.54; found

1

C 22.25, H 6.82, N 19.54. Ϫ H NMR (D₂O; pD Ͻ 2): δ ϭ 3.70

(dd, J ϭ 8.0, 10.0 Hz, 2 H), 4.12 (dd, J ϭ 7.0, 10.0 Hz, 2 H), 4.42

(m, 2 H); (D₂O; pD Ͼ 12): δ ϭ 2.53 (dd, J ϭ 5.0, 10.0 Hz, 2 H),

3.02 (m, 2 H), 3.15 (dd, J ϭ 6.0, 10.0 Hz, 2 H). Ϫ ¹³C{¹H} NMR

(D₂O; pD Ͻ 2): δ ϭ 50.0 (CH₂), 54.6 (CH); (D₂O; pD Ͼ 12):

δ ϭ 54.9 (CH₂), 62.1 (CH). Ϫ Single crystals of trans-(3R,4R)-

dap·3HCl·0.5H₂O were grown from an aqueous solution by slow

evaporation of the water.

(1R*,2R*)-1,2-Cyclopentanediyl Bis(methanesulfonate): Racemic

trans-1,2-cyclopentanediol (1.86 g, 18.21 mmol) was converted into

the bis(methanesulfonate) as described above for the cis isomer.

Yield 2.71 g (58%). Ϫ C₇H₁₄O₆S₂(258.3): calcd. C 32.55, H 5.46;

found C 32.46, H 5.19. Ϫ ¹H NMR (CDCl₃): δ ϭ 1.87 (quint, J ϭ

7.0 Hz, 2 H), 1.91Ϫ1.97 (m, 2 H), 2.22 (m, 2 H), 3.07 (s, 6 H), 5.08

(m, 2 H). Ϫ ¹³C{¹H} NMR (CDCl₃): δ ϭ 20.9 (CH₂), 30.5 (CH₂),

38.4 (CH₃), 84.8 (CH).

(1S*,2R*)-1,2-Cyclopentanediyl Bis(methanesulfonate): A solution

of cis-1,2-cyclopentanediol (2.50 g, 24.48 mmol) in dry pyridine

(50 mL) was cooled to 0 °C in an ice/NaCl bath. Methanesulfonyl

chloride (3.89 mL, 50.00 mmol) was then added dropwise. A white

trans-1,2-Diaminocyclopentane: A mixture of (1R*,2R*)-1,2-cyclo-

pentanediyl bis(methanesulfonate) (2.71 g, 10.49 mmol) and NaN₃

precipitate appeared. The ice bath was removed and the solution (2.05 g, 31.53 mmol) in dry DMF (100 mL) was allowed to react

was stirred at room temperature for 1 h and then cooled to 0 °C

once more. Excess pyridine was neutralized by the addition of 3 

and then hydrogenated as described above for the cis isomer. The

catalyst was removed by filtration, and the filtrate was acidified

2538

Eur. J. Inorg. Chem. 2001, 2525Ϫ2542

Coordination Chemistry of cis-3,4-Diaminopyrrolidine and Related Polyamines

FULL PAPER

with 3  HCl and concentrated to dryness. The residue was redis-

solved in H₂O (1000 mL) and sorbed onto Dowex 50 W-X2. The tions prior to and after each measurement. All titrations were per-

column was eluted with H₂O and aqueous HCl (0.5  and 3.0 ;

formed at 25.0 °C under nitrogen (scrubbed with an aqueous solu-

500 mL each). The last fraction was concentrated to dryness and tion of the inert electrolyte). To determine the pK_avalues of the

the residual white solid was dried in vacuo (1.13 g, 62%). Ϫ ligands, several alkalimetric titrations were carried out with analyt-

and the stability of the electrode was checked by calibration titra-

C₅H₁₄Cl₂N₂(173.1): calcd. C 34.70, H 8.15, N 16.18; found C ically pure samples of the corresponding hydrochlorides. Test solu-

1

34.73, H 8.03, N 16.09. Ϫ H NMR (D₂O; pD Ͻ 2): δ ϭ 1.83 (m, tions for the titration experiments were prepared using the solid

2 H), 1.89 (quint, J ϭ 7.0 Hz, 2 H), 2.32 (m, 2 H), 3.81 (m, 2 H); hydrochlorides (cis-dap·3HCl, cis-cptn·2HCl, trans-cptn·2HCl) or

(D₂O; pD Ͼ 12): δ ϭ 1.32 (m, 2 H), 1.64 (quint, J ϭ 8.0 Hz, 2 H), a stock solution (ampy·2HCl), standardized aqueous stock solu-

1.95 (m, 2 H), 2.77 (m, 2 H). Ϫ ¹³C{¹H} NMR (D₂O; pD Ͻ 2): tions of the metal chlorides (Cu, Ni, Zn), or solid Cd(NO₃)₂·4H₂O.

δ ϭ 24.5 (CH₂), 32.5 (CH₂), 58.0 (CH); (D₂O; pD Ͼ 12): δ ϭ 22.8 Complete equilibration was ensured by performing acidimetric

(CH₂), 34.9 (CH₂), 62.4 (CH).

back titrations (with 0.1 mol dm^Ϫ3HCl or 0.1 mol dm^Ϫ3HNO₃).

The data were only considered reliable when no significant hyster-

esis was observed.

[Cu(3R-ampy)(3S-ampy)](ClO₄)₂: Racemic ampy·2HCl (1.75 g,

11.19 mmol) was dissolved in H₂O (10 mL) and deprotonated using

Dowex 2-X8 (OH^Ϫform). The resulting solution was concentrated

to dryness under reduced pressure and the oily residue was redis-

solved in dry EtOH (50 mL). A solution of Cu(ClO₄)₂·6H₂O

Spectrophotometric Titrations: The titrations were carried out as

described in the previous section (50 mL sample solution,

0.01Ϫ0.02 mol dm^Ϫ3total Cu, 1 mol dm^Ϫ3KNO₃, 25.0 Ϯ 0.1 °C).

(1.85 g, 4.99 mmol) in dry EtOH (25 mL) was added dropwise. The Additionally, the titration cell was equipped with an immersion

blue precipitate was collected by filtration, washed with Et₂O, and

probe (HELLMA), which was connected to a diode-array spectro-

dried in air (1.75 g, 81%). Ϫ C₈H₂₀Cl₂CuN₄O₈(434.7): calcd. C

photometer (J & M, TIDAS-UV/NIR/100Ϫ1). An in-house com-

22.10, H 4.64, N 12.89; found C 21.76, H 4.56, N 12.81. Ϫ UV/ puter program was used to control the addition of the titrant (1 mol

Vis: λ_max(ε) ϭ 554 (127). Ϫ Single crystals were grown by slow dm^Ϫ3KOH). This program allowed a sample spectrum to be re-

cooling of a hot saturated solution of [Cu(ampy)₂](ClO₄)₂in

MeOH.

corded prior to each addition of an aliquot of the titrant.

Calculation of the Equilibrium Constants: All equilibrium constants

were calculated as concentration constants and pH was defined as

Ϫlog [H^ϩ]. The pK_avalues of the ligands and the formation con-

Potentiometric Measurements: Potentiometric titrations were car-

ried out with a Metrohm 713 pH/mV meter and a Metrohm com-

bined glass electrode with an internal Ag/AgCl reference.^[5,14]The stants of the complexes were calculated using the computer pro-

sample solutions were titrated with 0.1  or 1.0  KOH using a gram HYPERQUAD.^[36]All equilibrium constants were evaluated

Metrohm 665 piston burette. Addition of the titrant and the pH

meter reading were controlled using an in-house computer program

with fixed values for the total concentrations of the reactants and

for pK_w(13.78 for µ ϭ 0.1 mol dm^Ϫ3, 13.71 for µ ϭ 1.0 mol

and a PC. The ionic strength of each 50-mL sample solution was dm^Ϫ3).^[26]For the refinement of the formation constants of the

0.1 mol dm^Ϫ3KCl, 0.1 mol dm^Ϫ3KNO₃, or 1.0 mol dm^Ϫ3KNO₃metal complexes, fixed values were used for the protonation con-

Table 7. Crystallographic data for H₃cis-dapCl₃·H₂O and H₃trans-dapCl₃·0.5H₂O

Compound

H₃cis-dapCl₃·H₂O

H₃trans-dapCl₃·0.5H₂O

Empirical formula

Molecular mass

Temperature [K]

Crystal system

Space group

C₄H₁₆Cl₃N₃O

228.55

293(2)

monoclinic

P2₁

6.3903(7)

11.4227(13)

7.4879(9)

114.969(7)

495.49(10)

2

C₄H₁₄Cl₃N₃O_0.5

218.53

293(2)

orthorhombic

P2₁2₁2

12.476(2)

12.783(3)

6.4860(10)

90.00

1034.4(3)

4

1.403

0.837

456

˚

a [A]

˚

b [A]

˚

c [A]

β [°]

V [A ]

3

˚

Z

D_calcd.[g cm^Ϫ3

]

1.532

0.880

240

µ [mm^Ϫ1

F[000]

]

Crystal size [mm]

θ min. and max.

Data set

Transmission, min. and max.

Total data, unique data

Reflections with I Ͼ 2σ(I)

Parameters/restraints

R₁, wR₂[I Ͼ 2σ(I)]

R₁, wR₂(all data)

0.30 ϫ 0.35 ϫ 0.20

5.51 to 27.50

Ϫ8/8; Ϫ14/14; Ϫ9/9

not measured

4510, 2261

2048

0.50 ϫ 0.45 ϫ 0.35

2.28 to 24.02

Ϫ14/14; Ϫ13/13; Ϫ7/7

not measured

6274, 1493

1483

149/9

152

0.0266, 0.0527

0.0332, 0.0542

0.224 and Ϫ0.204

0.006

0.0251, 0.0696

0.0253, 0.0698

0.454 and Ϫ0.149

0.005

3

˚

Largest peak/hole [e/A ]

Max. shift/error

Eur. J. Inorg. Chem. 2001, 2525Ϫ2542

2539

K. Hegetschweiler et al.

FULL PAPER

Table 8. Crystallographic data for [Pd(Hcis-dap)₂](ClO₄)₄·2H₂O, [Pt(Hcis-dap)Cl₄]Cl·H₂O, [Co(cis-dap)(tach)][ZnCl₄]Cl·C₂H₅OH, [Ni(cis-

cptn)₂](ClO₄)₂, and [Cu(ampy)₂](ClO₄)₂

Compound

[Pd(Hcis-dap)₂](ClO₄)₄· [Pt(Hcis-dap)Cl₄]Cl·

[Co(cis-dap)(tach)][ZnCl₄]Cl· [Ni(cis-cptn)₂](ClO₄)₂[Cu(ampy)₂](ClO₄)₂

2H₂O

H₂O

EtOH

Empirical formula

Molecular mass

Temperature [K]

Crystal system

Space group

C₈H₂₈Cl₄N₆O₁₈Pd

744.56

293(2)

monoclinic

P2₁/c

9.8381(15)

12.6413(19)

9.9609(15)

104.93(3)

1197.0(3)

2

C₄H₁₄Cl₅N₃OPt

492.52

298(2)

orthorhombic

Pc2₁b

7.984(3)

11.210(3)

14.271(5)

90.00

1277.3(7)

4

C₁₂H₃₂Cl₅CoN₆OZn

577.99

233(2)

monoclinic

P2₁/c

8.0147(12)

18.527(3)

15.440(2)

93.5(5)

2288.3(6)

4

1.678

2.372

1184

C₁₀H₂₄Cl₂N₄NiO₈

457.94

293(2)

monoclinic

C2/c

22.450(4)

7.605(2)

12.016(2)

116.68(3)

1833.1(7)

4

1.659

1.395

C₈H₂₀Cl₂CuN₄O₈

434.72

213(2)

monoclinic

P2₁

8.633(3)

10.789(3)

8.850(4)

110.41(4)

772.5(5)

2

˚

a [A]

˚

b [A]

˚

c [A]

β [°]

V [A ]

3

˚

Z

D_calcd.[g cm^Ϫ3

]

2.066

1.317

752

2.561

12.006

920

1.869

1.807

446

µ [mm^Ϫ1

F[000]

]

952

Crystal size [mm]

Transm. coeff.

θ min. and max.

Data set

0.76 ϫ 0.34 ϫ 0.31

0.4342, 0.6855

2.14 to 33.08

0.46 ϫ 0.46 ϫ 0.27

0.0139, 0.0578

1.81 to 30.00

0.46 ϫ 0.10 ϫ 0.10

0.4084, 0.7974

1.72 to 28.33

0.50 ϫ 0.45 ϫ 0.30

not measured

2.86 to 25.82

0.65 ϫ 0.63 ϫ 0.04

0.445, 1.000

2.84 to 25.00

Ϫ15/10; Ϫ18/15; Ϫ14/14 Ϫ11/11; Ϫ15/15; Ϫ20/20 Ϫ10/7; Ϫ23/24; Ϫ20/20

Ϫ27/27; Ϫ8/8; Ϫ14/14 Ϫ8/10; Ϫ7/12; Ϫ8/10

Total data, unique data 12170, 4107

4326, 2260

2043

146/2

0.0419, 0.1088

0.0465, 0.1127

1.998/Ϫ2.092

0.000

19108, 5691

3725

343/0

0.0450, 0.0980

0.0785, 0.1076

0.829/Ϫ1.233

0.008

6637, 1672

1161

194/0

0.0423, 0.0947

0.0652, 0.1023

0.430/Ϫ0.458

0.000

1880, 1587

1554

208/1

0.0375, 0.0933

0.0386, 0.0940

0.571/Ϫ0.836

0.000

Reflns. with I Ͼ 2σ(I)

Parameters/restraints

R₁, wR₂[I Ͼ 2σ(I)]

R₁, wR₂(all data)

3295

225/2

0.0389, 0.1078

0.0508, 0.1153

1.308/Ϫ0.914

0.014

3

˚

Max. peak/hole [e/A ]

Max. shift/error

Table 9. Crystallographic data for [Cu(Hcis-dap)₂(OH₂)₂](SO₄)₂·3.5H₂O Ϫ 2x H^ϩϩ x Cu^2ϩ

Compound

x Ͻ 0.01

x ϭ 0.08

x ϭ 0.11

Empirical formula

Molecular mass

Temperature [K]

Crystal system

Space group

C₈H₃₅CuN₆O_13.5S₂

559.08

100(2)

orthorhombic

Pccn

24.635(2)

13.1027(12)

13.4177(12)

4331.0(7)

C₈H_34.84Cu_1.08N₆O_13.5S₂

563.99

293(2)

orthorhombic

Pccn

24.670(5)

13.110(3)

13.417(3)

4339.4(16)

8

C₈H_34.78Cu_1.11N₆O_13.5S₂

565.54

293(2)

orthorhombic

Pccn

24.703(5)

13.115(3)

13.479(3)

4366.9(17)

8

1.720

˚

a [A]

˚

b [A]

˚

c [A]

3

˚

V [A ]

Z

8

D_calcd.[g cm^Ϫ3

]

1.715

1.277

2353

1.727

1.353

2369

µ [mm^Ϫ1

F[000]

]

1.366

2375

Crystal size [mm]

Transm. coeff.

θ min. and max.

Data set

Total data, unique data

Reflns. with I Ͼ 2σ(I)

Parameters/restraints

R₁, wR₂[I Ͼ 2σ(I)]

R₁, wR₂(all data)

Max. peak/hole [e/A ]

Max. shift/error

0.58 ϫ 0.48 ϫ 0.16

0.485, 0.753

1.76 to 33.00

Ϫ22/37; Ϫ20/19; Ϫ20/18

44850, 8148

6386

0.3 ϫ 0.25 ϫ 0.1

not measured

1.65 to 24.02

0/28; 0/15; 0/15

3407, 3407

2854

0.25 ϫ 0.20 ϫ 0.10

not measured

2.32 to 26.04

Ϫ28/29; Ϫ16/16; Ϫ16/16

32097, 4157

2484

323/61

284/0

0.0356, 0.0839

0.0535, 0.0918

1.578/Ϫ0.371

0.001

0.0546, 0.1446

0.0662, 0.1567

0.603/Ϫ1.508

0.000

0.0694, 0.1164

0.1325, 0.1514

0.810/Ϫ0.464

0.000

3

˚

2540

Eur. J. Inorg. Chem. 2001, 2525Ϫ2542

Coordination Chemistry of cis-3,4-Diaminopyrrolidine and Related Polyamines

stants of the ligands, as obtained from the pK_adeterminations. The

FULL PAPER

Acknowledgments

spectrophotometric measurements were evaluated using the com-

puter program SPECFIT.^[37,38]The spectrum of free Cu^2ϩwas

measured separately and was not refined. The free ligands and their

protonation products were considered as colorless.

We thank Anton Zaschka, Dr. Anja Zimmer, Dr. Volker Huch,

Prof. Dr. Michael Veith (all Saarbrücken) and Dr. Peter Osvath

(Melbourne) for valuable help and fruitful suggestions. The struc-

[18]

ture of [Cu(cis-cptn)₂](ClO₄)₂

was solved and refined by Man-

Crystal Structure Determination:^[39]X-ray diffraction data were col-

lected with the following diffractometers: STOE IPDS [protonated

ligands, Ni complex, Cu(cis-dap) complexes with x ϭ 0.08, 0.11],

STOE STADI-4 (Pt complex), Nonius Kappa-CCD [Cu(cis-dap)

complex with x Ͻ 0.01], Siemens SMART PLATFORM (Co com-

plex, Pd complex), Siemens P4 four-circle diffractometer [Cu-

(ampy) complex]. The crystallographic data are compiled in

Tables 7, 8 and 9. Graphite-monochromated Mo-K_αradiation (λ ϭ

uela Winter (Bochum). Financial support from the Studienstiftung

des Deutschen Volkes (D. K.) and the Fonds der Chemischen Indu-

strie (K. H.) is gratefully acknowledged.

[1]

A. E. Martell, R. D. Hancock, Metal Complexes in Aqueous

˚

Solutions, Plenum Press, New York, 1996.

0.71073 A) was used throughout. Low-temperature measurements

[2]

P. Chaudhuri, K. Wieghardt, Prog. Inorg. Chem. 1987, 35, 329.

at 233(2), 213(2), and 100(2) K were performed on the Co complex,

the Cu(ampy) complex, and the Cu(cis-dap) complex with x Ͻ

0.01, respectively. The data for all the other compounds were col-

lected at ambient temperature. Standard reflections were monitored

during the data collections. The standard reflections of the Pt com-

plex and [Cu(ampy)₂](ClO₄)₂decreased by 6.2% and 2%, respect-

ively, during the measurements, and decay corrections were applied

accordingly. All data sets were corrected for Lorentz and polariza-

tion effects. Absorption corrections were applied to the data for

the Cu(cis-dap) complex with x Ͻ 0.01 (face-indexed numerical),

the Pt complex (empirical with ψ-scans), the Cu(ampy) complex

(empirical with DIFABS), and the Co and Pd complexes (empirical

with SADABS). The structure of the Pt complex was solved by

the heavy atom method; all other structures were solved by direct

[3]

K. Hegetschweiler, Chem. Soc. Rev. 1999, 28, 239.

H. Manohar, D. Schwarzenbach, W. Iff, G. Schwarzenbach, J.

[4]

Coord. Chem. 1979, 8, 213.

J. W. Pauly, J. Sander, D. Kuppert, M. Winter, G. J. Reiß, F.

[5]

Zürcher, R. Hoffmann, T. F. Fässler, K. Hegetschweiler, Chem.

Eur. J. 2000, 6, 2830.

[6]

H. C. Wormser, J. Pharm. Sci. 1969, 58, 1038.

[7]

D. R. Reddy, E. R. Thornton, J. Chem. Soc., Chem. Commun.

1992, 172.

[8]

J. Skarzewski, A. Gupta, Tetrahedron: Asymmetry 1997, 8,

1861.

[9]

D. Cremer, J. A. Pople, J. Am. Chem. Soc. 1975, 97, 1354.

[10]

H. Toftlund, E. Pedersen, Acta Chem. Scand. 1972, 26, 4019.

[11]

G. W. H. Potter, M. W. Coleman, A. M. Monro, J. Heterocycl.

Chem. 1975, 12, 611.

methods.^[40]All structures were refined by full-matrix least-squares

[12]

S. Ongeri, D. J. Aitken, H.-P. Husson, Synth. Commun. 2000,

techniques^[41]against F². One of the SO₄

counterions of the

2Ϫ

30, 2593.

[13]

Cu(cis-dap) complex with x Ͼ 0.01 proved to be disordered. These

anions exhibited two split positions for one of the oxygen atoms,

which were refined with occupancies of 50% each. Non-hydrogen

atoms were refined with anisotropic displacement parameters, ex-

cept for the disordered oxygen positions (O32A, O32B) of the

SO₄^2Ϫcounterions. In general, the positions of the hydrogen atoms

were calculated (riding model with U_isoϭ 1.2 ϫ U_eqof the bonded

heavy atom), with the exception of the H₂O hydrogen atoms of

H₃trans-dapCl₃·0.5H₂O, [Pt(Hcis-dap)Cl₄]Cl·H₂O, and the Cu(cis-

dap) complex with x ϭ 0.08 and x ϭ 0.11, which were not consid-

ered. All other hydrogen atoms of H₃trans-dapCl₃·0.5H₂O, all

H(ϪN) and H(ϪO) hydrogen atoms of H₃cis-dapCl₃·H₂O, all hy-

drogen atoms of the Ni and Pd complexes (including H₂O), and

the hydrogen atoms of the cation [Co(cis-dap)(tach)]^3ϩ, together

with the OH hydrogen atom of EtOH, were located and refined

with variable isotropic displacement parameters. The H(ϪN) atoms

of the coordinated amino groups of [Pt(Hcis-dap)Cl₄]Cl·H₂O were

placed in calculated positions and were refined with U_isoϭ 1.2 ϫ

U_eq. The positions of the hydrogen atoms of the noncoordinating

ammonium groups and of all the H₂O molecules of the Cu(cis-dap)

complex with x ϭ 0.005 were located and refined isotropically with

D. Kuppert, C. Roth, J. Sander, K. Hegetschweiler, Z. Kristal-

logr. NCS 2001, 216, 111.

[14]

A. Zimmer, D. Kuppert, T. Weyhermüller, I. Müller, K. Heg-

etschweiler, Chem. Eur. J. 2001, 7, 917.

E. Prenesti, P. G. Daniele, M. Prencipe, G. Ostacoli, Polyhedron

[15]

1999, 18, 3233.

[16]

J. Bjerrum, B. V. Agarwala, Acta Chem. Scand., Ser. A 1980,

34, 475.

[17]

B. J. Hathaway, J. Chem. Soc., Dalton Trans. 1972, 1196.

A low-quality single-crystal X-ray diffraction study of [Cu(cis-

[18]

cptn)₂](ClO₄)₂confirmed the square-planar (trans) CuN₄geo-

˚

¯

metry: P1, a ϭ 5.999(1) A, b ϭ 7.758(2) A, c ϭ 9.751(3) A,

α ϭ 81.67(2)°, β ϭ 86.83(3)°, γ ϭ 82.71(2)°, Z ϭ 1 for

C₁₀H₂₄Cl₂CuN₄O₈, R ϭ 0.089 for 1138 observed reflections [I

Ͼ 2σ(I)] and 116 parameters, wR₂ϭ 0.210 for all 1534 unique

˚

reflections. The average CuϪN bond length is 2.00 A.

A. F. Wells, Structural Inorganic Chemistry, 5th ed., Clarendon

Press, Oxford, 1984.

P. Comba, A. F. Sickmüller, Inorg. Chem. 1997, 36, 4500.

P. Hendry, A. Ludi, Adv. Inorg. Chem. 1990, 35, 117.

P. D. Newman, M. B. Hursthouse, K. M. Abdul Malik, J.

Chem. Soc., Dalton Trans. 1999, 599.

C. J. Andres, P. H. Lee, T. H. Nguyen, A. I. Meyers, J. Org.

Chem. 1995, 60, 3189.

A. Corruble, J.-Y. Valnot, J. Maddaluno, Y. Prigent, D. Dav-

oust, P. Duhamel, J. Am. Chem. Soc. 1997, 119, 10042.

L. Doms, L. Van den Enden, H. J. Geise, C. Van Alsenoy, J.

Am. Chem. Soc. 1983, 105, 158.

R. M. Smith, A. E. Martell, R. J. Motekaitis, Critically Se-

lected Stability Constants of Metal Complexes, NIST Standard

Reference Database 46, Version 5.0, USA, 1998.

A. Zimmer, I. Müller, G. J. Reiß, A. Caneschi, D. Gatteschi,

K. Hegetschweiler, Eur. J. Inorg. Chem. 1998, 2079.

M. Weber, D. Kuppert, K. Hegetschweiler, V. Gramlich, Inorg.

Chem. 1999, 38, 859.

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

˚

common bond lengths (0.775 Ϯ 0.005 A for HϪO and 0.855 Ϯ

˚

0.005 A for HϪN). The small amount of additional Cu in this par-

ticular specimen (1.5 eA^Ϫ3, x Ͻ 0.01) was not considered in the

˚

refinement. The structures of H₃cis-dapCl₃·H₂O, H₃trans-

dapCl₃·0.5H₂O, and [Pt(Hcis-dap)Cl₄]Cl·H₂O were solved and re-

fined in polar space groups. Their absolute structure parameters

were refined to values of Ϫ0.03(6), Ϫ0.01(9), and Ϫ0.041(14), re-

spectively.^[42]The structure of [Cu(ampy)₂](ClO₄)₂proved to be a

racemic twin.

[27]

[28]

Molecular Modeling Calculations: Molecular modeling calculations

were carried out using the commercially available program MO-

MEC97.^[30]

[29]

[30]

A. Ringbom, J. Chem. Educ. 1958, 35, 282.

P. Comba, T. W. Hambley, N. Okon, G. Lauer, MOMEC: A

Eur. J. Inorg. Chem. 2001, 2525Ϫ2542

2541

K. Hegetschweiler et al.

FULL PAPER

[39]

Molecular Modeling Package for Inorganic Compounds, Heidel-

berg, 1997.

Crystallographic data (excluding structure factors) for H₃cis-

dapCl₃[H₂O, H₃trans-dapCl₃·0.5H₂O, [Pt(Hcis-dap)Cl₄]Cl·

[31]

P. Comba, A. F. Sickmüller, Angew. Chem. 1997, 109, 2089;

H₂O,

[ZnCl₄]Cl·EtOH,

[Pd(Hcis-dap)₂](ClO₄)₄·2H₂O,

[Cu(ampy)₂](ClO₄)₂,

[Co(cis-dap)(tach)]-

and [Cu(Hcis-

Angew. Chem. Int. Ed. Engl. 1997, 36, 2006.

J. E. Sarneski, A. T. McPhail, K. D. Onan, L. E. Erickson, C.

dap)₂(OH₂)₂](SO₄)₂·3.5H₂O Ϫ 2x H^ϩϩ x Cu^2ϩ(x ϭ 0, 0.08,

0.11) have been deposited with the Cambridge Crystallographic

Data Centre as supplementary publication nos. CCDC-161100

to -161109. Copies of the data can be obtained free of charge

on application to the CCDC, 12 Union Road, Cambridge

CB2 1EZ, U.K. [Fax: (internat.) ϩ 44-1223/336-033; E-mail:

deposit@ccdc.cam.ac.uk].

[32]

N. Reilley, J. Am. Chem. Soc. 1977, 99, 7376.

[33]

J. S. Warmus, G. J. Dilley, A. I. Meyers, J. Org. Chem. 1993,

58, 270.

[34]

U. Nagel, E. Kinzel, J. Andrade, G. Perscher, Chem. Ber. 1986,

119, 3326.

L. Hausherr-Primo, K. Hegetschweiler, H. Rüegger, L. Odier,

[35]

[40]

[41]

[42]

R. D. Hancock, H. W. Schmalle, V. Gramlich, J. Chem. Soc.,

Dalton Trans. 1994, 1689.

G. M. Sheldrick, SHELXS-97, Program for Crystal Structure

Solution, Göttingen, 1990.

G. M. Sheldrick, SHELXL-97, Program for Crystal Structure

Refinement, Göttingen, 1997.

H. D. Flack, Acta Crystallogr., Sect. A 1983, 39, 876.

Received May 9, 2001

[36]

P. Gans, A. Sabatini, A. Vacca, Talanta 1996, 43, 1739.

[37]

R. A. Binstead, B. Jung, A. D. Zuberbühler, SPECFIT/32 Ver-

sion 3.0, Spectrum Software Associates, Marlborough, MA

01752, USA, 2000.

H. Gampp, M. Maeder, C. J. Meyer, A. D. Zuberbühler, Tal-

[38]

[I01165]

anta 1985, 32, 95.

2542

Eur. J. Inorg. Chem. 2001, 2525Ϫ2542

Article Doi

The coordination chemistry of cis-3,4-diaminopyrrolidine and related polyamines

DOI: 10.1002/1099-0682(200109)2001:10<2525::AID-EJIC2525>3.0.CO;2-Z

Source and publish data:

Authors:

Article abstract of DOI:10.1002/1099-0682(200109)2001:10<2525::AID-EJIC2525>3.0.CO;2-Z

Full text of DOI:10.1002/1099-0682(200109)2001:10<2525::AID-EJIC2525>3.0.CO;2-Z

Products guided by the article

R&D Labs maybe for 121187-63-1

Relevant to this article

Hot Product