A Structural and Dynamic Study of Cryptates
2r-l"q I
A Structural and Dynamlc Study of Cryptates
AMIRA ABOU-HAMDAN B.Sc. (American University of Beirut) B.Sc.(Hons.) (University of Adelaide)
This thesis is presented for the degree of Doctor of Philosophy
Department of Physical and Inorganic Chemistry University of Adelaide
July, 1990 Arnira Abou-Hamdan 1
C ontents
Summary lv
S tatem ent vll
Acknow ledge ments v1l1
Abbreviations IX
Chapter 1. General Introduction 1
Chapter 2. Structural Aspects of Cryptates 6
2.t Introduction 6
2.2 Experimental 13
2.2.1 Materials 13 2.2.2 NMR Spectroscopy 13 2.2.3 Crystallography l3 (a) X-ray Crystallography of ÍLl.C2lCslNCS l3 (b) X-ray Crystallography of [K.C2l(NCS)] 20 2.3 Results and Discussion 26 2.3.1 X-ray Crystallography 26 2.3.2 Cryptate Structure in Solution 31 (a) 13C Nun Specrroscopy 31 (b) 7ti NIr,tR Spectroscopy 38 Arnira Abou-Hamdan l1
Chapter 3. Cryprate Stabiliry 42
3.1 Introduction 42 3.2 Stability constant Determination 49 3.2.1 Determination of Cryptate Stability Constants by NMR Spectroscopy 49 3.2.2 Determination of Cryptate Stabiliry Constants by Potentiometric Titration 50 3.3 Results and Discussion 53
Chapter 4 Kinetic and Mechanistic Aspects of the Cryptates of C2ICs, C22CZ and Czll 65
4.1 Introduction 65 4.2 Kinetic Applications of NMR Spectroscopy 67 4.2.1 7Li NtvtR Spectroscopy 6l 4.2.2 23Na NMR Spectroscopy 69 4.2.3 Kinetic Applications 69 4.2.4 Lineshape Analysis 78 4.2.5 Calculation of Activation Parameters 80 4.3 Results and Discussion 82 4.3.t General Mechanistic Aspects of Cryptates 82 4.3.2 Exchange Kinetics of Li+ on [Li.C2lCs]* 84 4.3.3 Exchange Kinetics of Na* on [Na.C211]* and [Na.C2lC5]+ 94
4.3.4 Exchange Kinetics of Li+ on [Li.C22Cz]* 100 4.3.5 General Conclusions t07 Atnira Abou.-Hamdan ul
Chapter 5. Experimental 110
5.1 Materials 110
5.2 Synthesis 111
5.2.1 Preparation of C2lC5 111 5.2.2 Preparation of C22Cz rt2
5.3 Stability Constant Measurements 113 5.4 Preparation of NMR Samples lr5
5.5 In stru m entation 115
List of Publications 118 B ibl iography t20
Appendix 1 Lineshape Analysis Data 132 Appendix 2 Supplementary Crystallographic Data t43 Amira Abou-Hamdan lV
Summ ary
The crystal structures of the cryptate [Li.C2lCs]NCS and the diaza crown ether complex [K.C21]NCS have been determined by X-ray crystallography. The structures are compared with structures of related cryptates and diaza crown complexes to afford an assessment of the effeit of variation of the position and number of donor atoms in cryptands on the structure of alkali metal cryptates. The complexation of Li* by the cryptand C2lC5 has been studied in seven solvents by 7Li nmr spectroscopy and potentiometric titration. The stability constants tog{Ksldm3 mol-1} values at 298.2 K, for [Li.C21C5]* and [Ag.C21C5]. respectively are: in acetonitrile (4.15, 4.29), methanol (3.01 , J.69), dimethylformamide (1.80, 5.23), dimethylacetamide (1.85 , 4.45), and diethylformamide (1.72, 4.95). The Li+ exchange on the LLi.CzlC5l* is in the very slow regime of the 7fi nmr timescale in acetonitrile, propylene carbonate, and acetone, and within the 1Li nmr timescale in methanol, dimethylformamide, dimethylacetamide, and diethylformamide. Thus the respective decomplexation rate constants obtained from subsequent lineshape analysis are k¿ (298.2 K) = 2I.6 * 0.4, lL6 + 2,237 * 4, and 210 * 4 s-l The corresponding activation parameters are L,H. = 36.L + 0.9, 38.4 * 0.9, 49.0 * 2.1, and 27.8 + 1.5 kJ mol-l and Â^S* = -98.4 * 3.1, -76.5 * 3.0, -35.0 + 2.8, and -108 + 5 J K-l mol-l. The variation of the 13C and
7Li chemical shifts of ÍLi.CzlCsl* with solvent is employed in a structural investigation of this cryptate in solution. The equitibrium and kinetic data are discussed in conjunction with data from other related cryptates. Amira Abou-Hamdan V
complexation of Na* by the closely related ligands C2l, Czll and C2lC5 has been studied by 23Na nmr spectroscopy and potentiometric titration. The stability constants log{Ks/dm3 mol-1} values at 298.2 K for [Na.c2l]* in the three solvents dimethytformamide, dimethylacetamide and diethyformamide (of similar electron donating strength but different molecular size) are 2.10, 2.88 and 3.lg respectively, The log{Ksldm3 mol-l} values at 298.2 K, for [Na.C2rCs]* and [Na.c211]* respectively are: in dimethytacetamide (2.05, 4.74), and in diethylformamide (2.52, 5.10). The Na* exchange on the [Na.C2lCs]* is in the very fast regime of the 23Na nmr timescale in dimetylacetamide and diethylformamide. The Na* exchange on the [Na.C211]* is within the 23Na nmr timescale in diethylformamide. Thus the decomplexation rate consranr obtained is k¿ (298.2 K) = l8.z + 2.0 s-l and the corresponding activation parameters are L,H* = 67.1 + 1.9 kJ mol-l and Âs* = 4.4 * 5.0 J K-l mol-I. These data are compared
with each other and with similar complexes in the light of tigand and solvent molecular characteristics.
complexation of Li+ and Ag* by the clam-like cryptand c2zcz has
been studied in seven solvents by 1Li nmr spectroscopy and potentiometric titration. The stability constants log{Ksldm3 mol-l} values at 298.2 K, for lLi.c22c2l. and l{g.c22c2l* respectively are: in acetonitrile (7.8, 9.4), acetone (8.9, 13.1), water (<2, 6.0), methanol
(4.0, r0.2), dimethylformamide (3.5, 9.4), diethylformamide (3.1, B.z),
and pyridine (4.0, 5.0). The Li* exchange on the lLi.C22C2l+ is in rhe very slow regime of the 7Li nmr timescale in acetonitrile, acetone and pyridine, in the very fast regime in water, and within the 7Li nmr timescale in methanol, dimethylformamide and diethylformamide.
Thus the respective decomplexation rate constants obtained are k¿ Amira Abou.-Hamdan vl
(298.2 K) :971 + 42,240 + 7, and 916 + 28 s-1. The corresponding activation parameters are LH* :31.0 + 0.4,22.5 + r.2 and 26.7 * 0.6 kJ mol-l and Â,S* : -84.0 + 2.6, -I24 + 5 and -98.6 * 2.3 J K-l mol-l. These data aÍe discussed in the context of the effects of cryptand structure and solvent characteristics on cryptate lability and stability.
Amira Abou.-Hamd.an vltl
Acknowledgements
I sincerely wish to thank my supervisor, Dr s.F. Lincoln for his guidance, encouragement, and support throughout the course of this work.
I wish to thank my associates for their helpful discussion and suggestions, and my colleagues for their invaluable support .I especially wish to thank Andrea Hounslow for her advice on the operation of the spectrometers, and particularly for her encouragement and friendship. Many sincere thanks go to my relatives and friends for their endless support.
I gratefully acknowledge the financial support of a Commonwealth Postgraduate Research Award for the period of my candidature.
Finally, I wish to dedicate this thesis to my family whose support, interest, and encouragement throughout my academic years made the achievement of this work possible. Amira Abou-Hamdan 1X
Abbreviation s
The following abbreviations have been used in this thesis:
Czl 4,7 ,13 -trioxa- I , 10-diazacyclopentadecane C22 4,J,13,Ií-tetraoxa-1,I0-diazacyclooctadecane
C2lC5 4,7 ,l3 -trioxa- I ,L}-diazabicyclo t8.5.51eicosane C22C2 4;7,13,16-tetraoxa-1,I0-diazabicyclo[8.8.2]eicosane
CZI I 4,7 ,13, 1 8-tetraoxa- I ,|}-diazabicy clo t8.5.51eicosane C221 4,7 ,13 ,16,21 -penraox a-I ,I0-diazabi cy clo t8.8.51tricosane c222 4,7,13,16,21,24-hexaoxa-1,r}-diazabicyclotS.S.Sloctacosane
MeOH m ethan ol DMF N,N-dimethylformamide DEF N,N-diethylformamide DMA N,N-dimethylacetamide MeCN acetonitrile PC propylene carbonate TEAP tetraethylammonium perchlorate
CDCI3 the D represents the deuterium 12¡1;, also in the cases of D2O and (CD3)2CO.
d n - n-deuterated- DNI Gutmann donor number n m r nuclear magnetic resonance
F.I.D. free induction decay r.f. radio frequency ò chemical shift Hz Hertz p p m parts per million Arnira Abou-Hamdan
Chapter 1 General Introduction
The study of the complexation of metal ions and other species by multidentate macrocyclic ligands has been rapidly developing since the discovery of crown-ethers (macrocylic polyethers) in 1967 by
Pedersen 11 ,2,3), the introduction of cryptands (polyoxadiazabicycloalkane ligands) in 1968 by Lehn [4], and the more recent contribution by Cram in the synthesis of a new class of molecules, the spherands [5,6]. All three scientists have been the recipients of the 1987 Nobel prize in chemistry. This developing field is described by Lehn as "supramolecular chemistry' (and is also called "inclusion chemistry"), where the new supermolecules, of the types described above, or "hosts" bind the substrates or "guestn species by bonding interactions varying from those with significant covalent character to those which are of a secondary bonding character. Pedersen discovered the first crown-ether, dibenzo-18-crown-6 (Figure I .1), in the process of improving a method for olefin polymerization, catalyzed by vanadium complexes. Crown-ethers are cyclic molecules which form complexes with alkali metal ions. r^oì l^oì 00 0
0 0 0 0 \-o'---J \-oJ
Figure 1.1 Dibenzo-18-crown-6 and its potassium complex. Atnira Abou.-Hamdan 2
Concurrently, the increasing research on the complexation of alkali cations by naturally occurring macrocyclic antibiotics lead to the determination of the structure and synthesis of valinomycin, and its ability to mediate K+ transport in mitochondria Í71. Furthermore, the crystal structure of the K+-nonactin complex t8l (Figure 1.2) showed:
(a) a complete desolvation of the cation, and
(b) an appropriate folding of the 32 membered ring to afford an octacoordination of the potassium ion.
o o ao
a o
a a o o o CH: o
Figure 1.2 The nonactin molecule and the schematic representation of its K*-complex showing the wrapping of the macrocycle around the potassium ion. The carbonyl groups are represented by O, and the ether oxygen atoms by O.
The design and preparation of ligands which completely encapsulate metal ions and thereby imitate antibiotics, was successfully achieved by Lehn with the introduction of the cryptands Arnira Abou-Hatndan J
[9,10]. The general structure and naming of the cryptands are shown in Table 1.1.
Table 1.1 Cryptand Nomenclature
a b c cry ptan d
0 0 0 c111
I 0 0 CzTI 0 o 1 I 0 C22T
1 I 1 c222 N 0 3rd 1 0 C2lC5 bridge 0 csHro U9 C 3rd 1 I C22Cs bridge csHro
Cryptands are ideal cation receptors, they form highly stable complexes (cryptates) with metal cations which are of an appropriate size for the cryptand cavity (or crypt). The stability constants of cryptates are several orders of magnitude larger than the complex formation constants of naturally occurring or synthetic monocyclic ligands with alkali metal ions [10,] l]. A cryptand can be transformed from a cation receptor into a cation carrier by simple structural changes which may be accomplished by keeping the size of the intramolecular cavity constant and replacing some of the binding sites by non-binding groups ll2l. Amira Abou-Hamdan 4
The implications óf supramolecular chemistry, of which the cryptands are part, in biological and industrial fields are illustrated in the following applications: 1. transport processes and carrier design which are potentially important for biological and industrial applications [4,12-17]. Recently, the first complexation study of amino acids by cryptands has been reported t18l; 2. molecular catalysis [4,12];
3 . ionophores for ion-selective electrodes t 191 ; 4. extraction of alkali metal salts from water I20); 5. anion activation, which is useful industrially in some
polymerization processes 1,4; 6. anion receptors l4,2ll, which mimic proteins and enzymes;
7 . binding of toxic metal ions, such as Cd2*, Pb?*, and Hg2* t22-25); 8. light conversion devices Í26-321, realized by absorption of ultraviolet light, energy transfer and emission of visible light (A-
ET-E) as shown in Figure 1.3, with possible applications as luininescent probes for monoclonal antibodies in medical diagnostics. h u A ET *
N N
E
hu' Figure 1.3 Illustration of A-ET-E light conversion process performed by a Eu(III) cryptate. Amira Abou.-Hamdan 5
These existing and potential applications of cryptates stimulated this area of research. A number of important features, already established, aÍe discussed in the introductions of the following chapters. The aim of this work is to investigate structural, equilibrium and kinetic aspects of some monovalent cation cryptates and diaza crown ether complexes, which would complement and extend the existing information, thereby gaining further insight into supramolecular chemistry. Atnira Abou-Hamdan 6
Chapter 2 Structural Aspects of Cryptates
2.1 Introduction
The complexation of alkali metal ions by cryptands t33-381 and crown ethers 12,34,39,401 in solution and in the solid state has attracted much attention as a consequence of: (a) similarities between these complexations and those occurring with antibiotic molecules and other carrier molecules which appear to be important in biological membrane transport; and (b) their intrinsic interest as a route by which alkali metal ion coordination chemistry may be explored. The polyoxadiazabicycloalkanes, or cryptands, complex alkali metal ions (M*) to mainly form (i) inclusive cryptates in which M+ is enclosed within the cryptand cavity and (ii) exclusive cryptates in which M+ resides on the outside of a face of the cryptand. (Some possible external cryptates have been reported where the metal cation interacts with only one of the nitrogen bridgeheads [11,41].) The inclusive cryptands are exemplified by [Li.C21l]I Í421 (shown in Figure 2.1) and [Na.C22l]NCS Í431, and the exclusive cryptates are exemplified by [Na.C21Cs(NCS)] Í441 (shown in Figure 2.2), lNa.C211(NCS)I 1441, and lK.Czzl (NCS)I [43], whose strucrures have been determined by single-crystal X-ray diffraction methods. Amira Abou-Hamdan 7
Figure 2.1 (a) ORTEP piot of the structure tl-i.C2l lll l4Ll-
(b) Structural diagram generated from X-ray diffraction data using program SCHAKAL [45], ali atoms are shown as spheres of appropriate radii.
(a)
(b) Amira Abou-Hamdan 8
Figure 2.2Perspective viewof the structure [Na.C21C5(NCS)] showing the atomic numbering t441.
(b) Structural diagram generated from X-ray diffraction data using program SCHAKAL [45], all atoms are shown as spheres of appropriate radii.
c( 8) c(9 o 1) c(10) c(7) C(s) C b) c(2) N(3) c( 16 N(2) S c(1) c(4 No(1) ) N( 1) c(3) c( 1s) c(11)
o(3 )
(2) cfi2 )
c(13) c(14 ) (a)
(b) A¡nira Abou-Hamdan 9
In the inclusíve cryptates, the Li* and the Na* first coordination spheres are occupied solely by binding groups of the cryptand and the metal ion is centrally sited in the cryptand cavity, whilst in the exclusive cryptate the anion is within bonding distance of the metal ion and the metal ion is sited outside the cryptand cavity. Accordingly the înclusive and exclusive terminology is adopted for cryptates. Cryptands can assume any of three conformations (Scheme 2.1) according to the configuration of the two nitrogen bridgeheads, namely endo-endo, endo-exo and exo-exo.
Scheme 2.1
o-o o-o o ( I N N N o-o / o
endo-e ndo endo-exo exo-exo
The endo-endo form is the most probable conformer for the cryptate 136,42,44,46,47) where the electron density of both nitrogen atoms is directed toward the inside of the intramolecular cavity and appreciably complexed with the electrophitic metal ion. The cation-
site interactions are largely of electrostatic nature (charge-dipole and charge-induced dipole type) t481. The sequence of the two cryptands
4,7 ,13, 1 8-tetraoxa- 1 , 1 O-diazabicyclo [8.5.5] eicosane, or CZll, and 4,7,L3-tnoxa-1,10-diazabicyclo[8.5.5]eicosane, or C2IC5, and the dtaza Amira Abou-Hamdan 10
crown ether 4,7,13-trioxa-I,I0-diazacyclopentadecane, or Czl, (IUPAC rule B-1.53 requires the name 1,4,10-trioxa-7,I3- diazacyclopentadecane. However the name used in this thesis more conveniently shows the relationship of the crown ether to the cryptands C2ll and C2IC) is based on the substitution of the two
amine hydrogens of C21 by a bridging -CHzCHZOCHZCHz- moiety ro give Czll, and by a bridging -(CHz)S- moiety, to give C2IC5. The structures of these ligands, and others which are used throughout this thesis, are shown in Chart 2.1.
In the case of CZll, the cavity is delineated by four oxygen and two nitrogen donor atoms, whereas in the case of C2lC5, the fourth oxygen is replaced by the third methylene group in the -(CHz)s- moiety, but the three oxygen and two nitrogen atoms aÍe contained in a single ring as is also the case for monocyclic CzI. This systematic variation in principle facilitates an assessment of the effect of variation of the position and number of donor atoms in ligands on the structure, stability and lability of alkali metal cryptates. Generally, the structure of metal ion complexes in the solid state cannot be assumed to be retained in solution, since various solvational and conformational changes may occur on dissolution. However, in the case of alkali metal cryptates, nmr studies [44,49-5ll indicate that such an assumption may be made. These observations have major implications in the interpretation of the mechanistic aspects of cryptate equilibria in solution. As part of the assessment of such a systematic structural variation of C2lI, C2lC5 and Czl, the following X-ray diffraction studies examine (i) the structural effects of replacing an oxygen of C2ll by a methylene group to give the C21C5 on the complexation of Atnira Abou-Hatndan 11
Li* in the solid state, and (ii) the first reported structure for a 4,7,13- trioxa-I,I0-diazacyclopentadecane complex (although it was desirable to study several alkali metal complexes of this ligand the potassium complex was the only one for which crystals suitable for crystallographic studies were obtained). Amira Abou-Hamdan 12
Chart 2.1 Structures of various diaza-crown ethers and cryptands.
N N N 0 0 0 ) 0 0 0 0 ) ) ) 0 0 0 ) N N N
c21 c211 C?1C 5
N N N 0 0 0 0 0 0 ) ( ( 0 ) )( ) 0 0 0 0 0 0 ) N N N
c22 c221 C22C 2 Amira Abou.-Hamdan. 13
2.2 Experimental
2.2.t Materials
The lithium and potassium thiocyanate salts M(NCS) were dried
at 353-363K under high vacuum for 48 h and were stored over P2O5 under vacuum. The lLi.CzlCslNCS and [K.C21]NCS crystals were prepared by evaporating a methanolic solution equimolar in M(NCS) and the appropriate ligand to dryness, redissolving in acetone, and equilibrating over several weeks with light petroleum (b.p. 313-333 K) whereupon crystals slowly formed.
2.2.2 NMR Spectroscopy
A detailed description of lLi nmr spectroscopy is given in chapter 4, and the preparation and running of the solutions for 7Li and l3C nmr studies are described in chapter 5.
2.2.3 Crystallography
(a) X-ray Crystallography of [Li.C21Cs]NCS
Crvstal data. C1OHSOLiN3O3S, M=35I.44, monoclinic, space group F21/n, a = 8.348(2), b = 24.798(7), c : 9.l6IQ) Ä, þ : 90.74(2)0, U - 1896.2Ã3, Z = 4, Dc ]23l g cm-3, I.49 À(Mo-Ko) = ¡r(Mo-K) = ".-1, = 0.7107 Ä, ^r'1000) = 752. Amira Abou-Hamdan l4
Data collection. A plate-shaped crystal of dimensions 0.30 x 0.25 x 0.08 mm was mounted on a glass fibre with cyanoacrylate glue. Lattice parameters at 294 K \ilere determined by a least-squares fit to the setting parameters of 25 independent reflections, measured and refined by scans on an Enraf-Nonius CAD4 four-circle diffractometer employing graphite-monochromated Mo-Kq radiation. Intensity data were collected in the range I < 0 < width and horizontal apertures employed were (1.00 + 0.35 tanO)' and (2.40 + 0.05 tan0) mm respectively. Data reduction and application of Lorentz, polarization, and decomposition corrections were performed using the program SUSCAD I52l.Of the 1910 independent reflections collected l10l with I > 2.5o(1") were considered observed and used in the calculations.
Sfrncfrrre snlrtion anrT refine.menf The structure was solved by direct methods (MULTAN t531) which revealed all non-hydrogen atoms except lithium. The thiocyanate ion is disordered over the two sites with an occupancy ratio 80:20. Hydrogen atoms were included at calculäted sites (C-H 0.91 Al. All non-hydrogen atoms were modelled with anisotropic thermal parameters while hydrogen atoms were assigned isotropic group thermal parameters. Block-matrix least- squares refinement of all positional and thermal parameters and an overall scale factor converged, all shifts <
0.073, and w - 5.241[o2çf.¡ - 0.00009 FoZ].Maximum excursions in a final difference map were + 0.25 e Ä-3. Reflections with k + I odd were on average five times greater than those with k + I even, reflecting an approximate two-fold axis through the structure ut [, l, Atnira Abou-Hamdan 15
0. The pseudo-symmetry complicated both the solution and the refinement of the structure, and presumably contributed to the relatively high final R index. Scattering factors and anomalous dispersion terms were taken from the literature t541. All calculations were canied out using SHELX 76 l55l and ORTEP t561. The atom numbering scheme is given in Figure 2.3. Final atomic coordinates, bond distances, and bond angles aÍe given in Tables 2.1- 2.3 respectively. ,$nira Abou-Hamdan 16
Figure 2.3 (a) ORTEP plot of the structure lLi.CzlCslNCS showing the atomic numbering.
(b) Structural diagram generated from atomic coordinates of the crystal structure using program SCHAKAL [45], all atoms are shown as spheres of appropriate radii
(7 t9 NB C6
c10 05
t11 t19
tlB
t13 017
c16 t15 c20
(a)
(b) Amira Abou-Hamd,an 17
Table 2.1 Final Atomic Co-ordinates (x 164) for [Li.C21Cs]NCS*.
¡{ ¡f ¡{ Atom 1 0 10 v 0
s(1) 2272(9) 4150(1 ) 10010(9) s(1 ') 3043 1 93 I 4088 ( 29 ) 9603(50) c(2) 2435(29) 346q(10) 979T Qo) c(2') 1972(60) 351 4(20) 10337 $5) N(3) 2236(22) 3086(5) 996U(2o) N(3' ) 2669(62) 3089 (23 ) 10558(64) Lr(4) 7q99 (30 ) 4002(6) 5022(22) 0(5) 8053(7) 4658(3) 6199(B) c(6) 777o(12) 4566(5) 7689(13) c(7 ) 8428(11) 3989(5) 8005(9) N(8) 7767 Q ) 3577 ß) 7007 (7 ) c(e) 6225(9) 3409(q) 7522(B) c(10) 5232(10) 3145(4) 6366(10) c(11) !761 (9) 3535(4) 52u4(9) c(12) u321 (9) 3295(4) 3836(10) u\ 13 ) 5764(10) 3139(3) 2994(10) N( 14 ) 69r{9(7) 361 4(3¡ 2822(6) C( 15 ) 631 4(1 1 ) q023(q) 1852(10) C( 16 ) 7147(11) 4536(5) 2071 (1 o) o( 17 ) 69\7 Q) 4658(3) 3579 Q ) c( 18 ) 7 6z8(t z) 5157 (\) .r{090(13) C( 19 ) 7433(12) 5156(5) 566tr(28) c( 20 ) 8513(9) 3368(q) 222u(10) c( 21 ) 9495(10) 3116(4) 3484(10) 0( 22 ) 9909(7) 3538(3) 4564(7) c( 23 ) 10360(10) 3307 ( rr ) 6020(1 1 ) c( 2u ) 8896(10) 3156(q) 6929(10)
* The thiocyanate ion is disordered over two sites with an occupancy ratio of 80:20. The more populated site is occupied by atoms s(l), C(2), and N(3). A¡nira Abou-Hamdan 18
Tabte 2.2 Bond Lengths 1Å¡ for lLi.C2ICslNCS.
c(2) ---s(1 ) 1 .717 (29) 0(5) ---Li(q) 2.001 (21) c(11) ...1i(4) 2.573(25) 0(17) ---Li(4) 2:1 42(1 6) c(6) ---0(5) 1 :407 (z) c(7) ---c(6) 1 .557 ( 6) c(9) ---N(8) 1.438(9) c(10) ---c(9) l;488(11) c(12) ---c(11) 1.463(12) N(14) --c(13) 1.548(11) c(20) ---N(14) 1:5r{B(10) 0(17) ---c(16) 1;426(10) c(19) ---c(18) 1.453(1q) o(22) ---c(21 ) 1.478(11) c(24) ( --c 23 ) 1 :534(12) N(3) ---c(2) .966(zu) N(B) 2..111(19) N(14) ---LÍ(4)--Li(4) 2.275(23) j. o(22) ---L ( 4 ) 2 .360 (27 ) c(19) ---0(5) 1.42q(13) N(B) ---c(7) 1:\74(12) c(zu) --N(8) 1.r{08(12) c(1 1 ) ---c(10) 1:u62(12) c(13) ---c(12) 1.490(11) c(15) --N(14) 1.4q5(11) c( 1 6 ) --c( 15 ) 1.46¡{(1rr) c(18) --0(17) 1.438(12) c(21 ) 1.5q0(12) c(23) --C(20) --o(22) 1.495(1 1 ) Atnira Abou-Hamdan T9
Table 2.3 Bond Angles (') for [Li.C2lCs]NCS
N(3 ) -c(2) -s(1 ) 158.5(29) N(14) -r,i. ( 4 ) -0(5) i \9 :T (9) 0(17) -Li(4) -0(5) 76.3(6) 0(17) -Li(4) -N(14) 74.6(7) o(22) -L1( rl ) -N(8) 80.1(7) o(22) -Li(4) -o(17) 115.9(11) c(19) -0(5) -Li(4) 116:0(8) c(7 ) -c( 6) -o(5) 105.5(B ) c(7 ) -N( B) -Li(4) loz.8(8 ) c(9) -N( B) -c(7) 109:0(8 ) c(2u) -¡¡ ( 8) -c(7) 107 : )t (7 ) c(10) -c( 9) -N(B) 112.7 Q ) c(12) -c( 11) -c(10) 114.4(9 ) N(1lr) -c( 13) -c(12) 112 .1 (7 ) c(15) -t¡ ( 14) -Li(4) 108:4(7 ) c(20) -t't ( 1q) -Li(4) 108.6(8 ) c(20) -N( 1q) -c(15) 111 :2(7 ) 0(17) -c( 16) -c(15) 104:B(g ) c(18) -o( 17) -Li(q) 111i8(B ) c(19) -c( 18) -o(17) 105j9(1 r) c(21) -c( 20) -N(1 4 ) 109:7 0 ) c(21) -o( 22) -Li(\) 105:7 0 ) c(23) -o( 22) -c(21) 112;3(8 ) c(23) -c( 2u) -N(8) 112:7 (9) N(B) -L1 (q) -0(5) 85.5(9) N(1q) -Li(t{) -N(8) 124.7(8) 0(17) -1,1 ( ¡l ) -N(B) 158.5(13) o(22) -Li(rl) -0(5) 107j5(11) o(22) -L1( r{ ) -N(14) '78;q(8) c(6) -0(5) -r,i(4) 110.5(8) c(19) -0(5) -c(6) 1trl.2(10) N(8) -c(7) -c(6) 113.2(8) c(9) -N(8) -Li(q) 110i0(9) c(2q) -N(8) -Li(rt) 112:9(9) cQU) -N(8 ) -c(9) 113.9(8) c(11) -c(10) -c(9) r1o:6(8) c(13) -C ( 12 ) -c(11 ) 111 i5(7) c(13) -N ( 1lt ) -Li(4) l o.8(Z) c(15) -N ( 14 ) -c(13) 111.5(7) _N c(20) ( 1ll ) -c(13) 106.3(7) c(16) -(- ( 15 ) -N(14) 110.9(8) c(16) -o ( 1T ) -Li(4) 114.3(9) c(18) -0 ( 1T ) -c(16) 116.6(9) c(18) -c ( 19 ) -0(5) 107:3(10) oQ2) ( 21 ) -c(20) 109.rr(8) c(23) -0 ( 22 ) -Li(4) 103;5(7) cQU) ( 2? ) -o(22) ttz:6(T) Anira Abou-Harndan 20
(b) X-ray Crystallography of lK.C21(NCS)I
Crvstal data CttHzzKN¡OgS, M=351.5, monoclinic, space group P21ln, ¿ = 8.015(4), b = 14.755(4), c : 13.744(2) Å, B = 95.31(3)0, U - 1618.3Ä3, z = 4, Dx = 1.295 g cm-3, u(Mo-Ka) = 4.27 crn-I, r(000) = 668 e, max./min. transmission factors 0.939 and 0.894 respectively.
Data collection. Intensity data for a weakly diffracting crystal, 0.40 by 0.22 by 0.22mm, were measured at room temperature on an Enraf-Nonius CAD4 diffractometer fitted with Mo Ka (graphite monochromator) radiation, À 0.7107 Ä. A total of 4426 reflections were measured (a:20 scan technique; 0 max 22.5") of which 2II5 were unique (Ramat 0.026) and 981 satisfied the I > 2.5o(I ) criterion of observability. Corrections were applied for Lorentz and polarization effects Í571 and for absorption with the use of an analytical procedure tssl.
Structure solution and refinement. The structure was solved by direct methods with SHELX 76 and refined by a full-matrix least- squares procedure based on F t55]. Non-hydrogen atoms were refined with anisotropic thermal parameters and methylene hydrogen atoms were included at their tetrahedral estimates (C-H l.0SA); the amine- bound hydrogen atoms were not included. Owing to the appearance of non-sensible bond distances within the l5-membered ring the model was refined with constrained O-C, N-C and C-C bond distances of 1.42, I.46 and 1.50 Å respectively. After the inclusion of a weighing scheme, a = kl{oz1f ¡+çg lF l\\, the refinement converged to R = 0.089, Ro :0.099 for k = 3.45 and I = 0.001. The analysis of variance Amira Abou-Hamdan 2l indicated no special features and the maximum residual electron density peak in the final difference map was 0.53 e Ä-3.
The scattering factors used were those incorporated in SHELX 76 t55l and all calculations were performed on the University of Adelaide's VAX 1l-780 computer system. Fractional atomic coordinates, interatomic distances and angles are given in Tables 2.4,, 2.5, and 2.6, and the numbering scheme employed is shown in Figure 2.4 drawn at 15% probability ellipsoids with ORTEP II [58]. The high thermal motion associated with some of the ethylene carbon atoms is noteworthy. Disorder is quite common in these systems t44l and as a consequence disorder was looked for, but \/as not found in this system. Amira Abou.-Hamdan 22
Figure 2.4 ORTEP plot of the structure lK.C21(NCS)I showing the atomic numbering.
S1
c11
N3
o3 o1' c / o2 K
c6 c9 cs N1 N2 c10 o1 c4 c1 c3 c2 Arnira Abou-Hamdan 23
Table 2.4 Final Atomic Co-ordinares (x 194) for [K.C2l(NCS)]
Atom x v z
K 1ss9 (3) 4109 (2) L397 (2) s (1) 6s28 (s) 2143 (3) 1019 (3) o(1) -13?1 (11) s69s (6) 646 (6) o(2) 3ss9 (12) s692 (7',t 2803 (7) o (3) 1313 ( 12 ) 4328 ('71 33s3 (7) N (1) -1607 (13) 4180 (7) 2r00 (8) N (2) l-85? (14) 6664 (9) 10 17 (e) N (3) 3983 (14) 3360 (9) 1,2\3 (s) c (1) 198 (13) 6984 (9) 638 (12) c (2') -145s (14) 6626 (6t 896 ( 11) c(3) -2844 (L2' 5277 (L0l 936 (8) c (4) -27 64 (].6) 4944 (8\ 197 0 (e) c (s) -1269 (18) 38s3 (10) 3 102 (8) c (6) -L96 (]-2) 4432 (11) 3806 1t0 ) c(?) 26s9 (19) 4693 (10) 39'17 (L2',, c(8) 3163 (21) 56s1 (11) 37 87 (8) c (9) 3686 (1?) 6632 ('t') 2600 (13) c (10) 2288 (t9', 7076 (10) 1 973 (9) c (11) s008 (14) 28?0 (8) 11 13 (8) Atnira Abou-Hamdan 24
Table 2.5 Selected Bond Lengths (Ä) for tK.C21(NCS)].
K(1)-o(1) 2.87 r(7) K(1)-o(3) 2.772(t) K(1)-N(l) 2.e05(8) K( 1)-N(3) 2.80e(8)
c(11)-S(1) 1.64( 1) K(1)-o(2) 2.798(7) K( l)-o(1') 2.861(7) K( 1)-N(2) 2.e4s(8) c(1 1)-N(3) 1.11(r) Amira Abou-Hamdan. 25
Table 2.6 Selected Bond Angles (") for [K.C2l(NCS)].
c(2)-o( 1)-c(3) 107 ( 1) c(6) -o(3 )-c(7) l0e(1)
c( 1)-N(2)-c ( I 0) 10e ( 1)
c( 1) -C(2)-o ( 1) 102.8 (e)
c(3) -C (4)-N (1) 1 10( 1)
c(5) -C(6)-o (3) e7 (t)
c(7)-c(8) -o(2) 107 ( 1)
C (e)- c( l0)-N(2) 1 16(1) C (8) -o(2)-c(e) r05(1) c(4)-N(r )-c(5) 116(l) N(2)-c( 1)-C(2) 127 (t) o(1)-c(3)-c (4) I l6( 1) N(l )-c(5)-c (6) 118(l) o(3)-c(7)-c (8) r 17( 1)
o(2)-c(e)-c (10 ) ll8(l)
s(l)-c(11)-N(3 ) l l7(l) funira Abou.-Hamdan 26
2.3 Results and Discussion
2.3.1 X-ray Crystallography
The structural analysis shows ÍLi.C2lCslNCS to be an inclusive cryptate in which lithium resides within the C2ICS cavity and is within bonding distance of the two nitrogens and the three oxygen atoms of the cryptand (Figures 2.3a and 2.3b). The thiocyanate anion is not within bonding distance of lithium and there are no contacts between it and the cryptate which are significantly less than the sum of the relevant van der Waals radii. This is in contrast to the exclusive cryptate [Na.C2lCs(NCS)] in which the thiocyanate nitrogen is within bonding distance of sodium Í441 (Figures 2.2a and 2.2b). The bond distances to lithium are within the range 2.00(2)-2.36(3) Å which compares with the ranges of 2.081(6)-2.288(7) Å and 2.151(6)- 2.255(6) Ä respectively observed for the inclusive cryptates ll-i.C2llU l42l (Figures 2.la and z.Ib) and [Li.C22Cz]NCS t59l (Figures 2.5a and 2.5b). The structural differences between [Li.C21Cs]NCS and [Na.C21CS(NCS)] are a consequence of Na* being too large to be accommodated by the C2lC5 cavity, and a similar explanation holds for the structural differences between [Li.C21l]I and [Na.C211(NCS)]. (The effective six-coordinate ionic radii t60l of Li* and Na*, and the internal diameters of the CZLI and C2lCs cavities [33,11] are 0.76, I.02, 1.6, and 1.6 Ä respectively). However, the fourth oxygen in C2l1 provides an inducement for Li* to enter the cryptand cavity in [Li.C211]* which is absent in [Li.C21C5].. Thus the inclusive form of the latter cryptate is in part at least attributable to the endo-endo conformation of C21C5 in the cryptate. This results in the lone Amira Abou-Hamdan. 21 electron pairs of both nitrogens being directed towards the centre of the cavity, and thereby providing a bonding interaction with Li* which is stronger than that anticipated if Li+ remained outside the cavity. The endo-endo conformation is also exemplified by !L|C2ZCz]NCS and [Na.C22Cz(NCS)] cryptates Í471 which are shown in Figures 2.5 and 2.6. The crystal structure of [K.C2l(NCS)] is comprised of molecules of 4,J,13-trioxa-1,10-diazacyclopentadecane, potassium(I) cations and thiocyanate anions. Potassium(I) is l.4l Å above the mean plane of O(1), O(2) and O(3) (defined by 0.471x - 0.658y - 0.5882 = 6.604) of the 4,7 ,13-trioxa-1,10-diazacyclopentadecane ligand, which has a puckered planar conformation. The seven-coordinated potassium(I) is 2.811(1), 2.198(l) and 2.772(7) Å from O(1), O(2) and O(3) respectively, and 2.905(8) and 2.945(8) Ä from N(l ) and N(2) respectively of 4,'7,13-trioxa-1,I0-diazacyclopentadecane; it is 2.861(7) A from O(1') (which is related by the symmetry operation (l-x, -y, -z) of a second such ligand), and 2.809(8) Ä from N(3) of the thiocyanate as shown in Figure 2.4. The interaction with potassium(l) causes no major distortion of the 15-membered 4,J ,13-trioxa- I ,10- diazacyclopentadecane ring from the puckered planar structure expected for this ligand in the unperturbed state. In contrast the more flexible l8-membered ring of 1,4,7,10,13,I6- hexaoxacyclooctadecane distorts markedly from its planar puckered form in the free state to partially encapsulate sodium(l) in I,4,7,10,13,16-hexaoxacyclooctadecanesodium(I) thiocyanate in which seven-coordinated sodium(I) is bound by the six oxygens of the cyclic ligand and by one water molecule t39l.The l.4l Å distance between potassium and the trioxa plane of the 4,7,13-trioxa-1,10- Amira Abou-Hamdan 28
diazacyclopentadecane ring is much greater than the 0.14 and 0.37 Ä respectively observed for the sodium(I) to the trioxa plane distances of the analogous rings in 4,J ,L 3,1 8-tetraoxa- 1,10- diazabtcyclo[8.5.5]eicosanesodium(I) thiocyanate and 4,J,13-trioxa- 1,10-diazabicyclotS.5.5leicosanesodium(I) thiocyanate 1441. The variation in these distances in the two cryptates is largely attributed to the interaction between sodium(I) and the fourth oxygen in
4,7 ,13, I 8 -tetraoxa- 1 , 10-diazabicyclo [8.5.5 ] eicos anesodium(I) thiocyanate, and the absence of this interaction in 4,7 ,13-trioxa- 1 ,10- diazabicyclo[8.5.5]eicosanesodium(I) thiocyanate. The greater distance of potassium(I) above the trioxa plane of the 4,7,13-trioxa plane of the 4,J,13-trioxa-1,10-diazacyclopentadecane ring, by comparison to that observed for sodium(I) in the two cryptates, may be partly due to the larger size of potassium(I), but the dominant cause of this greater distance is the interaction with O(l') of the second 4,7,I3- trioxa- L,l0 - diazacy cl op en tadec an e. A further comparison of the IK.C2I(NCS)I and ÍK.C22CzNCS)] t6ll structures (both with a coordination number of seven) confirms the above conclusion. In the latter case, the potassium ion (which is within bonding distance of the two nitrogen atoms, four oxygen atoms and the anion) is 0.584 Ä above the tetraoxa plane (compared to the l.4l Ä K*-trioxa plane distance in the former), the difference is again attributed to the interaction of the K+ with O(1') in the case of the tK.c21(NCS)]. Amira Abou-Ha¡ndan 29
Figure 2.5 (a) Perspective drawing of the structure fLi.Cz2czlSCN showing the atomic numbering t591. (b) Structural diagram generated from X-ray diffraction dara t59l using program SCHAKAL Í451, all atoms are shown as spheres of appropriate radii. The structure is viewed almost perpendicular to the N-N axis. c'l 0 c9
c11
CB
c'l
C1
N2 N1
6
c2
C1 c4 c5 c3 (a)
0 -Lr
0
(b) Amira Abou.-Hamdan 30
Figure 2.6 (a) Perspective drawing of the structure [Na.C22Cz(NCS)] showing the atomic numbering 1621.
(b) Structural diagram generated from X-ray diffraction data t62l using program SCHAKAL [45], all atoms are shown as spheres of appropriate radii. The structure is viewed almost perpendicular to the N-N axis.
c
N
c
c6À
c5B c5À
cqe'
c3À c2B c4A
3g' clA N1' N1 c2A clB c7s' c7A (a)
Þ No
0
I
(b) Atnira Abou-Handnn 31
2.3.2 Cryptate Structure in Solution
(a) 13C NMR Spectroscopy
The resonances arising from C2lCs are in three distinct groups (the carbon numbering corresponds to that shown in the stylised C2ICS structure shown in Figure 2.7): those of carbon atoms adjacent to oxygen [C(9), C(Iz), and C(13)], of carbon atoms adjacent to nitrogen lC(2), C(10), and C(l1)1, and of carbon atoms adjacent to carbon only tC(3) and C(4)1. The assignments of C(9), C(12), and C(13) are made by comparison to those made for C2ll [44,49,511, achieved by selective decoupling of the C(3) protons which distinguishes the C(2) resonance from those of C(10) and C(11) and similar decoupling of C(9) and C(12) which distinguishes the resonances of C(10) and C(11) from that of C(2)), and the assignments of C(3) and C(4) are obvious. Also the assignments of C(10) and C(11) have been made by comparison to the spectra of C2l1 and its Na* and Li* cryptates î441. Only the assignment of C(2) and C(11) is a little uncertain for C2lC5 where proton coupled resonances overlap Í441, but in the cases of ÍLi.Czl Csl * in the studied solvents, the assignment of the superimposing C(2) and C(11) resonances is unambiguous. (In this thesis, the terms 'upfield' and 'downfield' refer to shifts to lower frequency and higher frequency respectively.) It is generally observed that the complexation of C222 by K*, Na* and Li+ Í491, C22l by K*, Na* and Li* [50,49], and C211 by Na* and Li* [5I,491 produces an upfietd shift of the cryptand 13C resonances with the exception of the C(11) resonance of [Na.C211]*, [Na.C21Cs]* and the equivalent resonance of îK.CZLI\*. This distinguishes these exclusíve cryptates Arnira Abou-Handan 32 from the others which exist in the inclusive form. This difference in shift direction is observed in the cases of lLi.C2lC5l* and [Na.C21C5]+ in deuterio-chloroform (CDCI¡) (Figure 2.9 and Table 2.8) confirming their existence as inclusive and exclusive cryptates respectively. Also the chemical shifts of C(3) and C(4) of [Li.C2lCs]* and [Na.C21C5]* in CDCI3, differ substantially, with those of the former differing most from those of C2lC5. This is consistent with the fact that in the case of the inclusive Li* cryptate, C(3) and C(4) being in close proximity to the metal cation consequently experience a major change in environment, whilst their environment is more similar to that in the cryptand in the exclusíve cryptate. The study of [Li.C21C5]+ in two other solvents, deuterium oxide (DzO) (Figure 2.7 and Table 2.7) and deuterio-acetone ((CD3)2CO) (Figure 2.8 and Table 2.7) show similar results to those observed in CDCI3. This l3C study shows the presence of an inclusive [Li.C21C5]. in the three solvents studied. The possibility of the existence an inclusivelexclusive equilibrium moving increasingly towards the exclusive cryptate as the solvent coordinating power increases is further investigated in a 7Li nmr study, where a larger range of solverits can be studied. Arnira Abou-Handan 33
TabLe 2.7 13C NtvtR Chemical Shiftsa for C21C5 and ÍLi.CZlCsl* in D2Qb and (CD3)zCOc at 297 K.
DO Carbon C2lCs [Li.C2lCsl*
t3 71.657 7 0.27 L l2 66.266 68.873 9 67.954 67.381
l1 55.229 53.660 2 55.229 53.660 10 s 4 .200 52.930
3 24.134 24.080 4 20.980 23 .7 00 (Cq¡tç6 l3 72.307 69.417 TZ 70.728 67.690 9 7 t.3t4 60.307
11 s7.336 52.842 2 s7.669 52.842 10 s3.388 52.131
3 29.095 23.417 4 21.931 22.839 a The cryptand and cryptate concentrations are 0.02 mol dm-3. The numbering system is that used in the structures shown in Figures 2.7 and 2.8. b Chemicat shifts are referenced to t-butanol external reference (31.6 ppm). c Chemical shifts are referenced to (CD¡)zCO (30.2 ppm). Anira Abou-HaruJan 34
Table 2.8 13C NMR Chemical Shiftsa for C2IC5, [Li.C21C5]+ and lNa.C2lCsl* in CDCl3b at 305.2 K.
Carbon C2lCs [Li.C21Csl* [Na.C21Cs'l* l3 7 r.298 68.383 67.897 t2 70.053 66.6 8 3 67 .594 9 7 0.539 65.377 68.596
ll 56.541 51.986 58.180 2 56.146 51.864 58.180 10 54.354 5t.287 s2.229
3 27.693 22.409 25.84t 4 20.98r 21.893 21.013
a The cryptand and cryptate concentrations are 0.21 mol dm-3. The numbering system is that used in the structure shown in Figure 2.9, reference [44]. b Chemical shifts are referenced to CDC1¡ (77.25 ppm). Atnira Ahou-Hamdan 35
Figure 2.7 13C-{1H} nmr spectra (75.47 MHz) of [Li.C2lCs]* in DzO solution at 297 K. The resonance numbering corresponds to that shown in the C2lC5 structure.
¡J 12
11 0 10
2 4 3
11 13 t2
I 12 3 10
4
C2lC5 (free)
11,2
13 12 10 3 9 4
[Li.C21C5].
EO 70 60 50 40 50 20 PPM Atnira Abou-Hamdan 36
Figure 2.8 13c-{1H} n-r specrra (75.47 MHz) of [Li.czlcs]* in (cD:)zco solution at 297 K. The upper field (multiplet) arises from (CD¡)zCo. The resonance numbering corresponds to that shown in the C2ICs structure and is identical to that shown in Figure 2.7.
13
11
t 4 J (cD3Þco
11 2 5
13 I 10 1
4
[Li.C21C5].
3
2 11 10 2 4
CZLCs (free)
80 70 60 50 40 50 20 PPIl Amira Abou-Hamdan JI
13C-{lH} Figure 2.9 nmr specrra (20.1 MHz) of [Li.C21C5]. and
[Na.C2lCs]* in CDCI3 solution at 305.2 K. The resonance numbering is identical to that shown in Figures 2.1 and 2.8, t441.
13
11
a 4
13 I '11,2 10 12
4 CDCI3
CZlCs (free)
139 11 t2 3 12 10
4
[Li.C2lC5].
11,2
9rs ,12 3 10
4
lNa.C2lCsl*
EO 76 t2 66 6a 60 56 52 .! )¿ 26 21 20 ppm Atnira Abou-Hamdan 38
0) zfi NMR Spectroscopy
In the solid state state both [Li.C211]* and [Li.C21CS]* exist as inclusive cryptates, as discussed previously. If these structures are retained in solution, and the complexed Li+ does not interact with the solvent, the 7ti chemical shift of the cryptates should be independent of the nature of the solvent (since the solvent effect upon chemical shift occurs primarily in the first coordination layer of the metal cation) in contrast to the chemical shift of the solvated Li* ion which directly reflects the influence of the solvent on the magnetic field seen at the 7fi nucleus t631. In a range of solvents the 7ti chemical shift of [Li.C211]. is substantially independent of the nature of the solvent indicating that the interaction of the complexed Li* with solvent is minimal as anticipated for an inclusive cryptate t631. In contrast the 7Li chemical shifts of [Li.C2lCs]*, õçqrr, vâry over a 1.83 ppm range for the solvents studied, which compares with a range of 2.62 ppm for solvated Li* (Table 2.9). This suggests that direct interaction between Li* in lLi.CzlC5J+ and solvent occurs despite the inclusive structure observed in the solid state. It is seen from the space filling representation of inclusive ÍLi.Czl0sl* in Figure 2.3, that Li* is not exposed at the exterior surface of the cryptate, and that interaction with solvent can only occur if Li+ moves from the centre of the cryptand cavity towards the fifteen membered diazatrioxa ring (this ring is chosen as it contains five donor groups as opposed to four in the fifteen membered diazadioxa ring) to produce an exclusive structure for [Li.C2lCs]*. As there is no electron donating group in the -(CHz)s- moiety of C2lC5, there is unlikely ro be a substanrial hindrance to this movement which allows Li+ to increase its Amira Abou-Hamdan 39
coordination number to six or more through interaction with solvent. Thus a minimum of two equilibria are required to describe the complexation of Li* by C2lC5 as shown in Scheme 2.2, where the equilibrium between exclusive and inclusíve [Li.C21Cs]* moves increasingly towards the exclusive cryptate as the ability of the solvent to coordinate Li+ increases. Scheme 2.2
N N
)J \- 0 ) \-- 0 )î N N
exclusive tnclusive
The Gutmann donor number, D¡¡ [64,65], reflects the solvent coordinating ability (it is given by the heat of formation of the complex between SbCl5 and the solvent in 1,2-dichloroethane), and it is seen from Table 2.8 that for the oxygen donor solvents õ"o.. for [Li.C2lCs]* tends to increase with increase in DN. The nitrogen donor solvents produce very different õco., values for ÍLi.CzlCSl*. (These observations, however, do not indicate the same correlation with the donor number as for 23Na chemical shifts. The following possibte explanation for this difference in behavior is suggested by Popov et al. [65]: in the case of 23Na, the paramagnetic shielding term oo is - 30 times larger than the diamagnetic term od, and thus the former Table 2.9 Solution Compositions and 7Li Chemical Shifts (2S0.1 K) for rhe lLi.C2lCsl* Sysrem
Solvent DNa [LiClO¿] [C2lCs] õohrb òc..,rr' òobrb ò"c,.rc correctl on N (Li* solvated) (tl-i.C2l Csl *) appl i ed I mol dm-3 mol dm-3 PPm ppm ppm ppm ppm t ¡-i MeCN 14 .l 0.0201 0.01 l0 0.52 -0.26 -1.37 -2.t5 -0.78 s PC 15. I 0.0197 0.01 r 7 -0.29 -0.48 -0.04 -0.23 -0. 19 Acetone 17 .0 0.0196 0.01 r 6 2.10 1.02 0.81 -0.27 - 1.08 HzO I 8.0 0.0 l 89 0.0095 0.00 0.00 -0.216 -0.216 0 (3¡)¿ MeOH I 9 .0 0.0201 0.0108 0.04 -0.15 0.44 -0.35 -0 .7 9 Q3 's¡a DMF 26.6 0.0 l 99 0.0099 0.89 0.r 9 0.22 -0.48 -0.70 Pyridine 33.1 0.0207 0.0103 2.31 1.86 -0.56 -1.0r - 0.45 a Gutmann donor numbers from reference t64]. b Chemicals shifts referenced to 0.005 mol dm-3 solution of LiCIO¿ in water as external reference. cBulk diamagnetic susceptibility corrections made as described in Chapter 5. d These alternative D¡ values have been proposed for aqueous and methanol solutions. respectively. in references t65l and t66]. The D¡ values derived by Gutmann refer to dilute solutions of the solvents in 1,2-dichloroethane.
A Amira Abou-Hamdan 4T
dominates the chemical shifts. Whereas in the case of 7Lí, the values of the two terms ate much closer, which may account for the variation in the chemical shifts.) Unfortunately the inclusivelexclusive equilibria characterising ILi.C2lCsl* cannot be quantified as the òcog data do not permit the determination of the specific chemical shifts of exclusive and inclusive [Li.C2lCs]+ required to calculate equilibrium constants. The 39K chemical shift of inclusive lK.C222l* is also solvent independent while that of exclusive [K.C22l]* is solvent dependent, and an equilibrium between exclusive and inclusive forms of ÍCs.C222l* (moving increasingly towards exclusive cryptate as temperature increases) has been reported 167,681. Arnira Abou-Harndan 42
Chapter 3 Cryptate Stability
3. 1 Introduction
The solution chemistry of the complexes, or cryptates, formed between alkali metal ions and cryptands has been the subject of
extensive study [ 10, 1 1,33,44,50,51 ,63 ,69-7 5]. The most stable complexes known to date with the alkali and alkaline-earth metal cations are obtained with these ligands 17 6-781. The stability constants are several orders of magnitude larger than the complex formation constants of naturally occurring or synthetic monocyclic ligands with alkali metal ions t9-111. The stability of these cryptates is due to the macrobicyclic topology of the cryptands where the metal cation is contained in a three-dimensional intramolecular cavity of matching size (inclusive cryptates) 136,42,46,791. The lK.C222l+ cryptate is more stable by a factor of 105 than the K+ complex of its macrocyclic counterpart; this macrobicyclic cryptate effect is even larger than the related macrocyclic effect (Figure 3.1). Some of these bicyclic ligands display very high selectivities for one alkali or alkaline-earth metal cation where 'peak' selectivity is displayed by the smaller, less flexible cryptands and 'plateau' selectivity observed for the larger, more flexible cryptands (Figure 3.2. and Table 3.1). With the possibility, in principle, of the modification of these synthetic ligands at will, cryptates cover the whole spectrum from cation receptors to cation carriers [11,12,80]. funira Abou-Hamdan 43
Figure 3.1 The macrocyclic and macrobicyclic effects (^) on complex stability; values given are the stability constants (log Kr) of the K+ complex in methanol (top) and methanol in warer (95%) (botrom) t33].
Macrocyclic Effect A ol o O1 ( 2.2 6.1 3.9 oJo/
Cryptate Effect
o oì-o\,,,or eo^ol _N N\r4\-o\N 9 75 4.95 4'8 Lo,.-, o -) Lo-, o-) Atnira Abou-Hamdan 44
Table 3.1 Ionic Radii, Hydration Numbers of Metal Cations, Approximate Cavity Radii and Number of Binding Sites of Cryptands.
Ionic Hydra- Cavity No. of Cation Radius tion Ligand Radius Binding r(Å) a No.b (Å) c Sites
Li+ 0.7 8 6 cztl 0.8 6
Na* 0.98 6 c22t 1.1 7
K+ 1.33 6 c222 t.4 8
Rb* t.49 6 c322 1.8 9
Cs* 1.65 6 c332 2.1 l0
Mg2* 0.78 6 c333 2.4 ll
Ca2* 1.06 8 * Sr2 1.27 8
Ba2* 1.43 8
a Reference t8ll. b References [80,82]. c Measured on Corey-Pauling- Koltun molecular models, from references [10,83]. funira Abou-Hamdan 45
10.0 c222
9.0 c22l
8.0 c21 I c322
7.0
6.0 c333 log Ks
5.0
4.0
3.0
2.0 \.a-----a 0 0.75 1.0 1.25 't.50 '1.75 I Li. No. K Rb' .C¡' Metal lon Radius (Ä)
Figure 3.2 Selectivity of cryptands: log Ks values for reaction of several cryptands with alkali metal cations at 298.2 K versus cation radius. Data points: (a) in 95% methanol; (b) in merhanot t331. ,Atnira Abou-Hamdan 46
The bicyclic topology of the cryptand renders contraction and expansion of the cavity more difficult than in macrocyclic tigands. This explains the high stability of the cryptates formed with the cations whose size most closely fits the intramolecular cavity. Inclusion of a larger cation decreases the stability of the complex due to conformational strain imposed on the ligand, and inclusion of a smaller cation also brings the ligand out of its own equilibrium conformation to afford contact between cation and binding sites and leads to a decreased complex stability, this is exemplified in the case of c22l (cavity radius of ca. 1.1 Al which is of oprimal size ro form inclusive [Na.C221]*, but forms exclusive lK.Cz2ll* with the larger K*. These size correlations aÍe reflected in the stabilities of cryptates. Thus the stability of [M.C22l]+ in a range of solvents varies with M+ in the sequence Li* < Na* ) K*, which reflects that Li* easily enters the C22l cavity but is too small to estabtish optimal bonding distances, K* is too large to enter the cavity, and Na* is of optimal size. The number of cryptand binding sites also affects the stability constants of the cryptate. In the cases of C222 and C22Cg cryptands, the stability of Na* and K+ cryptates decreases by a factor of about 104 - 105 (in methanol-water) on replacement of two oxygen binding sites by two CH2 groups, that is, a factor of approximately 102 per site tl I l. It should also be noted that the stability constants decrease at low pH, due to protonation of the bridgehead nitrogen sites and their consequent unavailability for binding cations, this factor allows linkage of proton levels and free cation levels, a phenomenon of interest in membrane transport studies t101. The third factor affecting the cryptate stability is the ability of the solvent to compete with the cryptand for coordination with the Amira Abou-Hamdan. 47 metal ion since the complex formation involves the substitution of one or more solvent molecules from the inner coordination sphere of the metal ion. The difference of the two binding energies due to competition between the cryptand and solvent molecules for the cation should constitute a major contribution to the overall stability of the cryptate [69,84,85]. A major objective of the studies of the complexation of alkali metal ions by cryptands, has been the determination of the above factors controlling complex stability, which are reflected in the magnitudes of the stability constant, Kr, (given in equation 3.1, where kc and kd are the complexation and decomplexation rate constants respectively) in the generalised reaction
Scheme 3.1 kc M++l=-[M.L]* kd
t, kc [M.L.] where, n'= (3. 1) kd = rruruu for the complexation of lithium(I), sodium(I) or silver(I), M*, cryptare or crown ether complex where L represents a cryptand or a crown ether respectively. Such investigations will eventually elucidate the macrocyclic effect in general, and especially the cryptate effect and will improve the ability to predict the extent of ligand-cation interaction s. In this study L represents the cryptands 4,7,I3,18-tetraoxa-1,10- diazabicyclo[8.5.5]eicosane (C2ll), 4,7,13-trioxa-1,10- diazabicy clo t8.5.51eicos ane (C2lC5'¡, 4,7 ,13 ,16-tetaoxa- 1 ,1 0- diazabicyclo[8.8.2]eicosane (C22C), and the diaza crown ether 4,7,I3- Atnira Abou-Hamdan 48
trioxa-1,tO-diazacyclopentadecane (C2l) shown in Chart 2.1. The ligands CzlL, C2lC5 and C2l all possess the same fifteen-membered
ring containing three ether oxygen and two amine nitrogen atoms.
However, in the case of C?ll and C2ICS the two nitrogen atoms are
linked by -CHzCHzOCHzCHz- and -(CHz)s- respectively to form a second ring, while monocyclic Czl has no such linkage. These differences facilitate an assessment of the effect of a systematic variation of ligand structure on Na(I) complex stability and lability. The study of [Li.C2lCsl* and its Ag(I) analog in a range of solvents affords comparisons with the complexation of Li(I), Na(I) and Ag(I) by C2II and C2lC5, and an improved understanding of the effect of the cryptand donor atom variation in these cryptates. The novel ligand C22Cz t61l is also characterised by a bicyclic cryptand structure in which one of the arms possesses no donor atoms, and thus represents an intermediate structural stage between the diaza- crown ethers, Czl and C22, and the cryptands Czll, C221, and C222
in which all arms possess donor atoms. In addition to this, C22CZ has a clam-like structure resulting from the shortness of the Cz arm, as is shown by X-ray diffraction studies of [Li.C22CZl+ (Figure 2.5) and lNa.C22C2l* @igure 2.6). It is seen that Li+ in lLi.C2ZC2l* is sited deep in the throat of the cryptand, which contrasts with solid state [Li.C21C5]* and ÍLi.Czlll+ in which Li* is centrally sited in the cavity of the cryptand (as was shown in Chapter 2). The consequences of these differences on the equilibrium characteristics of lLi.C22C2l* and [Ag.CZ2Cz]* in a range of solvents are also examined in this study. þnira Abou-Hamdan 49
3.2 Stability Constant Determination
Stability constants of cryptates have been measured by a. number of methods including calorimetry [86], optical properties [14], electrochemical method based on anodic oxidation of cryptands [87], pH-metric titration [11,61], and may be measured by fast atom bombardment mass spectrometry t881. Predominantly, cryptate stability constants have been measured by nmr spectroscopy and potentiometric titration.
3.2.1 Determination of Cryptate Stability Constants by NMR Spectroscopy.
Analysis of cryptate stability constants by nmr spectroscopy has been carried out by two main methods: (a) through the chemical shifts: in the case of fast intermolecular metal ion exchange between solvated and complexed sites, the values of the stability constants are calculated from the variation of the chemical shift with the cryptand/metal ion mole ratio t89l (the upper limit for this direct nmr technique is 105 t90l). (b) through the resonance integrals: in the case of significantly different chemical shifts of the solvated and complexed metal ions, (i.e. slow rate of exchange of the metal ion between the two environments, on the nmr timescale), the stability constant is determined by (i) integrating the solvated and complexed cation bands to provide direct measurements of the relative concentrations of the solvated and complexed species, or (ii) a competitive nmr technique apptied to two cations that form stable complexes which Atnira Abou-Hatndan 50 measures the ratio of the stability constants (as long as the known and unknown stability constants differ by a factor of 103-104), this method permits the determination of stability contant values higher than the upper limit for the direct nmr technique t901.
3 .2.2 Determination of Cryptate Stability Constants by Potentiometric Titration.
The potentiometric titration with alkali-cation specific electrodes [11,71] is a suitable method for measurements in most solvents where 1 < log Ks < technique allowing indirect measurement of alkali metal cryptate stability constants greater than 106 where the very high stability of a complex results in free ion concentration being too low to be reliably determined by selective ion electrode t9ll.Due to the lower sensitivity of the nmr measurements, this technique was considered to be too expensive in cryptand and as the stability was to be studied in a range of solvents, it appeared that the potentiometric method would be most convenient for our purposes. Direct potentiometric measurement of metal ion concentration may be made using the appropriate cation selective electrode. The electrode potential is related to the concentration of the solvated metal ion:
E=Eo+clog[M*] (3.2) where the coefficient c is determined by calibration measurements of solutions containing known metal ion concentration and may vary Amira Abou-Hamdan 51
between 45 and 65 (where the emf is in mV). Having determined c, the concentration of free metal ion in solution containing cryptand may be obtained directly, from which the stability constant may be calculated. Two points should be noted at this stage. The thermodynamic stability constant Ktn is defined âs,
f ([M.Crypl *) [M.Cryp+] Kn1 *) *] (3.3) - f ([Cryp] ) [Cryp] ,f ( tMl [M where .,f (tM.Crypl*), f (tcrypl) and f ([M].) are the activity coeff,rcients of the cryptate, cryptand and metal ion respectively. Since these quantities are generally unknown, the stability constants reported here are concentration stability constants ,K. defined as
f (tcrvpl) [M.Cryp*] ¡(. f(tMl.l) (3.4)
In principle Ks represents the overall equilibrium (Scheme 3.1) which includes cryptand conformational changes (as discussed in chapter 2) and metal cation, cryptand, and cryptate solvational equil ibria. The indirect method employed consists of the determination of the stability constant for the formation of the Ag* cryptate (K¡g) by direct titration, then monitoring the competitive equilibrium established between the Ag* cryptate in the presence of Li* in solution as shown below, Arnira Abou-Hamdan 52
Kr' [Li.Cryp]* + Ag* : [Ag.Cryp]* + Li+ (3.5)
lAg.Cryp+l[Li*] where, Kr' (3.6), - [Li.Cryp*] [Ag*] from which K¡i is obtained since,
KLi= K¡gl Ks' (3.7)
Thus a solution of standard Ag* is titrated with a solution of [Li.Cryp*] where the initial concentrations of these species are known, the equilibrium concentration of Ag* is determined from the potential of the Ag/Ag+ electrode, Kr' may be evaluated from the equilibrium concentrations of the required species (equation (3.6)) directly determined from the stoichiometry of the equilibrium. The total metal ion concentrations are always higher than the cryptand concentrations. Under these conditions, the concentrations of uncomplexed cryptand at equilibrium aÍe assumed to be negligible Íe21. funira Abou-Hamdan 53
3.3 Results and Discussion
The results are summarized in Tables 3.2, 3.3, and 3.4, and values for other cryptates in some solvents are included in Tables 3.5,3.6 and 3.7 for comparison. The stability constants for [M.C211]+ are substantially greater than those for [M.C21C5]*, as seen from the data presented in Tables 3.5, 3.6 and 3.J, this is consistent with the presence of the extra oxygen donor atom in CZII generating a greater stability in its cryptates when M* = Ag*, Li* and Na+. This is probably a consequence of the greater electrostatic attraction of CZll for metal ions by comparison with C2lCs, which increases the competition between the cryptand and the solvent for M+-binding sites. In the common * solvents studied, the relative stabilities [Li.C21 l ] > indicate that the optimum fit of Li* into the C2ll cavity to produce an inclusive cryptate confers an increased stability over that of exclusive [Na.C211]*, while the relative stabilities [Li.C21C5]* < suggest that ÍLi.C2lCSl* has lost this stabilising feature. This is consistent with the variation of the 7Li shift of [Li.C2lCs]+ with solvent and supports the suggestion that this cryptate exists substantially in the exclusive form. In the oxygen donor solvents both Ag* (r = 1.15 Al cryptates are considerably more stable than their Li+ (r = 0.8 Ä) and Na+ (r = 1.02 Ä) analogs despite the larger ionic radius (r) t60l of Ag+ rendering it a poorer fit for the cryptand cavities. This suggests that there is a difference in the interaction between the cryptands and non-directional bonding Li* and Na*, and directional bonding Ag* which has a tendency to form two strong coaxial bonds t931. In acetonitrile [Ag.C21C5]+ is of a similar stability to irs Li+ Arnira Abou-Hamdan 54 analogue and less stable than its Na+ analogues and [Ag.C2ll]+ is less stable than its Li+ and Na+ analogues (Tables 3.2 and 3.7). This change probably reflects the tendency of Ag* to bond less effectively with oxygen donor atoms than with nitrogen donor atoms (the mean value for a single Ag-N bond is 23 kJ mol-l, that of Ag-O bond is 6 kJ mol-l [94]), with the consequence that a nitrogen donor solvent can compete more effectively with a cryptand for Ag*, than can an oxygen donor solvent [93,95]. The opposite results found for the Li+ and Na* analogues are possibly interpreted as due to the fact that nitrogen donor solvents classified as "soft" bases do not strongly solvate nhard" acids such as alkali metal ions 189,96,971 and thus compete less effectively with the cryptands.
It is seen that Ks observed for [Li.C21C5]+ and [Na.C21C5]* increases as the Gutmann donor number (Dr.r) decreases for the oxygen donor solvents (Table 3.2). A similar increase in Ks for the [Na.C21Cs]* (Table 3.6) is seen for the nitrogen donor solvents pyridine (DN=33.1) and acetonitrile (DN=14.1), but whereas rKs fits roughly into the sequence observed for the oxygen donor solvents, the ,K, observed in pyridine is greater than expected on the basis of its high D¡. This is probably a consequence of the position of the nitrogen-donor atom in the pyridine ring causing substantial steric hindrance towards bonding with Na*. In general, it is expected that the electron-donating abilities indicated for the solvents by DN will be modified by their steric characteristics tg5l.Similar variations of stability with the nature of the solvent have been observed for other cryptates Í72). AmiraAbou-Hamdan 55
Table 3.2 Stability Constants for Cryptates of Monovalent Metal Ions Formed with the Cryptand CZlCs in a Range of Solvents.
log (Kr/dm3 mol-l) at 298.2 K
* Solvent DN lAg.C2l Csl [Li.C21Csì* [Na.C21Csì* MeCN r4.l 4.29a 4.15a 5.09d MeOH 19.0 7.69a 3.01b 3.76d (23.5c) DMF 26.6 5.23a 1.80 b 2.97d
DMA 27.8 4.45 b 1.85b 2.05 a
EF 30.9 4.95 a r.72b 2.52 a
a Error is + 0.1 log units. b Error is + 0.2 log units. c Proposed alternative D¡ value [65,66]. The D¡ derived by Gutmann refer to dilute solutions of the solvent in 1,2-dichloroethane, where in the case of MeOH, the hydrogen bonding structure is disrupted. d
Reference t951. funira Abou-Hamdan 56
Table 3.3 Stability Constants for Sodium Complexes Formed with the Ligands CZl, CZICs and CZII in Solvents of Similar Type: Dimethylformamide, Dimethylacetamide and Diethylformamide.
log (Kr/dm3 mol-l) at 298.2 K
Solvent ¿N lNa.C2ll* [Na.C21C5]* lNa.C211l*
DMF 26.6 2.10 a 2.97 b 5.20 b
DMA 27.8 2.88 a 2.05 a 4.74 a
EF 30.9 3.19 a 2.52 a 5. r0 a
â Error is + 0.1 log units. b Reference [95]. Amira Abou-Hamdan 57
Table 3.4 Stability Constants for Cryptates of Monovalent Metal Ions Formed with the Cryptand C22C2 in a Range of Solvents.
log (Krldm3 mol-l) at 298.2 K
Solvent DN l\g.C22C2l*b lLi.C22C2l+b lNa.C22C2l*d
MeCN 14.l 9.4 7.8 9.5 c
Acetone t7 .0 I J 1 8.9
Hzo 18.0 6.0 <2 3.0 b (3:)a
MeOH r 9.0 r0.2 4.0 7.2 c (23.5\a
DMF 26.6 9.4 3.5 5.9 b
EF 3 0.9 8.23 3.1
Pyridine 33.1 4.97 4.0 8.2 c
a Proposed alternative D¡ values for H2O and MeOH rather than 18.0 and 19.0 respectively, obtained in 1,2-dichloroethane solution where the hydrogen bonding structure of H2O and MeOH is disrupted [65,66]. b Error is + 0.1 log units. c Error is + 0.2. d Reference [98]. funira Abou-Hamdan 58
The study of the stability constants for the [Na.C21]+, [Na.C211]* and [Na.C21CS] * has been carried out in three solvents (dimethylformamide, diethylformamide and dimethylacetamide) which have similar electron donating abilities as expressed through their Gutmann donor numbers DN of 26.6, 30.9 and 27.8 respectively, but which have significantly different molecular shapes and sizes. The stability constants characterising [Na.C21 1] * are greater than those characterising [Na.C2lCs]* and [Na.C21]* by two orders of magnitude or more (Table 3.3). Clearly this greater stability is a consequence of the fourth ether oxygen in Czll and its dipolar interaction with Na* which is still within bonding distance of Na*, eventhough [Na.C2ll]* exists in the exclusive form t951. The similarity of the stabilities of lNa.C2lCsl* and [Na.C21]* indicates that the presence of -(CHz)s- moiety of C2ICs has only a minor influence on the stability of sodium(I) complex. The lower stability constants observed for the cryptates [Na.C211]* and [Na.C21C5]* in dimethylacetamide compared with dimethylformamide and diethylformamide must be due to the differences in solvating ability caused by the additional steric bulk of the methyl group adjacent to the coordinating oxygen in that solvent. No similar effects are observed in the cases of the diethylformamide and dimethylformamide where the additional steric bulk of the ethyl groups in the former solvent are further a\r/ay from the coordinating oxygen. Table 3.5 Stability Constants for Some Cryptates and Diaza-Crown Ether Complexes of Li(I) in a Range of Solvents.
Þ
Log (Kr/¿mg mol-l) at 298.2 K for d' Þ
I + Solvent 4.I [Li.C21Cs]+ ll-i.c2l ll* lLi.c22ll* lLi.c222l+ lLi.c22l+ ñ MeCN I 4 I 4 .15 7.8 > lOc 10.33c 6.97 c 4.39f Acetone l7 .0 8.9 2.13f l 8.0 H2o <2 5.5e 2.50e 0.98c 33 a
1 9.0 MeOH 3.01 4.0b 8.04d 5.38d 2.6è t.0te 23.5 a
DMF 26.6 1.80 3.5 6.99c 3.58c -0.0f
DEF 30.9 I .12 J 1
P ridine 33.1 4.0 0.43 f a Alternative b D¡ values, references [65,66]. The value is 3.44 in gO% MeOH, reference t611. c Error + 0.1, reference a t721. Error + 0.1, 0.1 mol dm-3 EI+NCIO¿ supporting electrolyte, reference [92]. e Error 1 +0.2,0.10 mol N(CH3)aBr 0..' supporting electrolyte, reference tl ll. f Values determined through nmr techniques. Error + 0.0g the cases in of Acetone and Pyridine, and * 0.41 in the case of MeCN, reference t991. g in 95% MeOH, reference t1001. \oUr Table 3.6 Stability Constants for Some Cryptates and Diaza-Crown Ether Complexes of Na(l) in a Range of Solven ts.
Èr
I X. (Kr/¿m¡ èl Log mol-l) at 298.2 K for s- È
Solvent DN lNa.C22C2l+t [Na.C221]* lNa.C2l11*c [Na.C21Cs]*'
MeCN l4 .l 9.5 t2.4d 4 s.08 MeOH 19.0 93d 7.2 6 1 3.16 (23.5¡a 8.65c l 8.0 H2o 3.0 5.4c 3.2 (33)a 2.gd DMF 26.6 5.9 7.93c 5.2 2.87 Pyridine 23 .0 8.2 3.72 a Alternative D¡ values, references [65,66]. b Reference t981. c Reference tlz).d 0.10 mol dm-3 Bu+NClo+ supporting electrolyte, reference tl0ll. e Error + 0.1, 0.05 mol dm-3 EeNClOa supporting electrolyte, reference te5l.
o\ Table 3.7 Stability Constants for Some Cryptates and' Diaza-Crown Ether Complexes of Ag(l) in a Range of S olvents.
ê¡
I Log (Ks/dm3 mol-l) 298.2 I at K for Ër
S ol vent . Al [Ag.C2l Cs]* Ag,C22C2)*c IAg.C2l l ] [Ag.C22l]+ f{g.C222l+ l{g,C22)+
MeCN t4.1 4.29c 9.4 1.7d 1l .24d 8.99d
Acetone 17 .0 8.5 8b I 3 18.0 Hzo 6.0 8.5d 10.6e g.6h 33 a I 1.gf 19.0 MeOH 7 .69c 10.2 10.6d 14.64s 12.2e 7.901 23.s a
DMF 26.6 5.23c 9.4 g.6d t2.4lf t 0.07f 10.02j
DEF 30.9 4.95c 8.2
ridine 33.1 <2b 5.0 a Alternative Dlv values, references [65,66]. b Reference c d t951. This st¡dy. Error + 0.1, referenc e 1721, e Error + 0'1' reference 1102)' f Error 0.1, reference + [91]. I Error + 0.r. 0.r0 mol dm-3 EeNCloa supporting electrolyte, n reference l92l' Error < * 0.2,0.10 mol dm-3 N(cH¡)+Br supporting electrolyre, reference tlll. i Reference [103]. j 0.05 mol dm-3 o\ EeNCroa supporting erecrroryte, reference [g41. Atnira Abou-Ham"dan 62
The complexation of Li(I) and Ag(I) by C22Cz was studied in seven selected solvents, (Table 3.4), (which afford a range of electron donating abilities as indicated by their Gutmann donor numbers (DN) Í64,65,661), to provide a contrast between the effects of oxygen and nitrogen donor solvents. It is seen that in these solvents the stability of lLi.C2ZCzl* is substantially less than that of lAg.C22C2l*, and that the difference in stability is less in the nitrogen donor solvents acetonitrile and pyridine. This is consistent with a combination of the tendency for non-directional bonding Li+ to show a preference for oxygen donor atoms, and that of directional bonding Ag* to show a preference for nitrogen donor atoms t931. Consequently, a nitrogen donor solvent can compete more strongly with a cryptand for Ag*, than can an oxygen donor solvent. This is also reflected in the variation of stability with the nature of the solvent. Thus for lLi.C22Czl+ stability tends to decrease as D¡ increases. The apparently anomalous position of water in the sequence is resolved when a D¡ value = 33.0 is employed [65,66], which is probably more appropriate to aqueous solutions than D¡ = 18.0. On this basis the stability of lLi.C22Czl+ is lower than expected in pyridine for which DN = 33.1, and may indicate that the incorporation of the nitrogen donor atom in the ring structure causes steric hindrance which decreases the solvating power of pyridine. While the stability sequence for IAg.CZZCz]+ in the oxygen donor solvents is similar to those for lLLCZ2Czl+ and [Na.C22C21+, lTable 3.4), the stability of lAg.C22C2]* is decreased in acetonitrile and pyridine as a consequence of the greater ability of these nitrogen donor solvents to compete with C22Cz for Ag*. Anira Abou-Hamdan 63
A study of the effects of solvent variation oû the stabilities of cryptates suggests that the free cryptands are considerably more strongly solvated in aqueous solution than in non-aqueous solvents
191,921. Thus the increase in stability (102.1-104 times), in the cases of C22Cz, C2l1 and C22I complexes with Li+, Na+ and Ag* in methanol than in water (Tables 3.5,3.6 and 3.7) cannot be solely attributable to a decrease in the solvation of the cations [92,104] but also to a decrease in the solvation of the organic surface of the cryptate in the former solvent. Generally stabilities increase in the sequence: ÍLi.C2lCsl* [Li.c22]+ < ÍLi.c222l* < lLi.cL\czl+ < Þi.czzrl+ < Þi.c2l1l+ consisrent with the the six donor atoms of C2l1 and the optimal fit of Li* (r = 0.76 Äl into the Czll cavity (r 0.8 Äl (Table 3.1) in inclusive tl-i.C2lll+ generating the greatest stability. The increase in stability ÍLi.c222l+ < electrostatic interaction with Li* resulting from the decrease in the number of donor atoms from eight in C222 to seven ín C22l is offset by the smaller cavity size of C22l with the consequence that lLi.C22\+ is the more stable cryptate. Similarly the increase in stability: [Li.C22]+ < lLi.C22Czl* < tLi.CZlll+ indicates the influence of ligand stereochemistry on ligands with the same number and type of donor atoms whereby the flexible partial cavity formed when monocyclic C22 binds Li* confers less stability than the preformed cavities of the two cryptands. The number and type of donor atoms is more influential in determining the relative stabilities of the Ag* species as exemplified by the sequence: tAg.C2ll* - [Ag.CZICs]+ < [Ag.CZ2l+ lAg.C2Il]* - lAg.C22C2l+ < Atnira Abou-Hamdan 64
[Ag.C222]+ < effects as Ag* (r =1.15 Ä t60l) fits rhe C22l cavity ( r 1.1 Ål optimatly while it fits the C222 cavity (r 1.4 Äl loosely, and in consequence lAg.CZ2Il+ should gain an enhancement of stability over f{g.C222l+.) This stability sequence probably arises from rhe preference of Ag* for bonding with the two nitrogen donor atoms of the diaza-crown ethers and cryptands. It is also consistent with the stabilities of the Ag* complexes decreasing, both in absolute magnitude and relative to the analogous Li+ complexes, in acetonitrile as a consequence of the ability of this nitrogen donor solvent to compete with the ligands for Ag* to a greatt.l' extent than do oxygen donor solvents. In the common oxygen donor solvents studied, methanol and dimethylformamide, the Li* cryptates are less stable than their Ag* analogues, but the difference in stabitity is less for IM.C21 1] *. This probably reflects the increased srability characterising [Li.C2l l]. because of its inclusíve structure, whereas the larger Ag* is unlikely to form an inclusive cryptate with Czll and gain this enhanced stability. A combination of the inclusíve nature of [Li.c2l1]+ and the greater ability of a nitrogen donor solvenr to compete for Ag* accounts for the reversal of stability: [Li.C211]+ > [Ag.C2l1]+ found in acetonitrile. Atnira Abou-Hamdan 65
Chapter 4
Kin tic and Mechanistic Aspects of the Ctyptates of C21 Cs, C22Cz and CzTI
4.I Introduction
The kinetic and mechanistic studies of the complexation of alkali m etal i on s by cryptands hav e rec eived a great interes t [10,11,33,44,71,84,105-107] as a consequence of: (i) similarities between these complexations and those occurring with antibiotic molecules and other carrier molecules which appear to be important in biological membrane transport; and (ii) their intrinsic interest as a route by which alkali metal ion coordination chemistry may be explored. The function of the cryptands as either a metal ion receptor or a carrier is largely determined by the stability and cation exchange rates of the cryptates. The alkali metal cryptates have consequently been the subject of a number of kinetic studies which included stopped-flow studies 141,50,71,92,108-l l4l, temperature-jump relaxation studies [14,115], electrochemistry [116], spectrophotometry tll7l, pulse radiolysis î221, ultrasonic relaxation studies [118,119], and nmr lineshape analysis Í120,1211. Amongst these, nuclear magnetic resonance of alkali metal elements was shown to be a particularly useful technique. It is a very sensitive probe of the immediate chemical environment of an alkali metal ion Arnira Abou-Hamdan 66 in solutions and can be used to study ion-ion, ion-solvent and ion- ligand interactions 169,70,73,84,122,1231.
The kinetic and mechanistic studies discussed in this chapter are summarized by (A) Investigations of the effect of metal ion size and cryptand cavity stze on lability of cryptates have produced a substantial understanding of the mechanism of cryptate formation Í4,10,33,42,50,63,69-711, but there is a paucity of systematic data on the effect of the cryptand donor atoms on these characteristics.
Recently attention turned to the investigation of this effect as exemplified by the studies of [Na.C21Cs]* and the closely related [Na.C211]* cryptates [44,51,J2,73,84,95], where the cryptand in the former differ from that in the latter by the replacement of an oxygen by a methylene group. It is found that [Na.C21C5]* is less stable than [Na.C2l1]* in a range of solvents, largety as a consequence of its greater rate of decomplexation t841. This study seeks insight into this aspect of the cryptates through a kinetic investigation of Li+ exchange on the cryptates formed by C2lC5 in various solvents and comparisons with [Li.Czll]*, [Na.C2lCs]* and [Na.C2l 1]*.
(B) The determination of the factors controlling the lability of the complexation of sodium(I) by CZI, CZll and C2lCs in three solvents of different molecular shapes and sizes, and similar electron donating strength, namely dimethylformamide (DN=26.6), diethylformamide (DN=30.9) and dimetylacetamide (DN=27.8) has been carried out. It has previously been shown that these differences produce marked variations in the solvent lability and stoichiometries of solvated cobalt(Il) and nickel(Il) U241. It is therefore of interest to examine Atnira Abou-Hamdan 67 the variation of these solvent characteristics on the complexation of lNa.C2ll*, [Na.C21 1]+ and [Na.C21Cs]*.
(C) A kinetic study of the complexation of lithium(I) by C22Cz Í47,621, which possesses an unusual clam-like structure, and its silver(I) analogue in a range of solvents affords a systematic comparison with the complexation of lithium(I) and silver(I) by related cryptands and diaza-crown ethers shown in Chart 2.1. A considerable variation in the [Li.C22C2]* stability constants occurs with solvent variation as shown by the log (Ks/dm3mel-l¡ = 7.8, 8.9, <2,4.0,3.5, 3.1, and 4.0 values determined at 298.2 K in acetonitrile, acetone, water, methanol, dimethylformamide, diethylformamide, and pyridine, respectively (from chapter 3), which are intermediate values between those of [Li.C22]+ and [Li.Czll]*. A prime objective of such a study is to determine the kinetic and mechanistic origins of these differences in stability arising from the differences of the cryptand structure and from the variation in the nature of the solvent.
4.2 Kinetic Applications of NMR Spectroscopy
4.2.1 7ti wtr¡tR Spectroscopy
Nuclear magnetic resonance of the alkali metal elements is a very sensitive probe of the immediate chemical environment of an alkali metal ion in solutions, thus particularly useful in kinetic studies of Amira Abou.-Hamdan 68
alkali metal cryptates U3,84,1221. 7Lí nmr spectroscopy is a convenient and direct method for the investigation of the exchange 'lfi kinetics in lithium cryptates [63,69). has a natural abundance of 92.57 % and an intrinsic nmr sensitivity of approximately one third of the proton |25f, or near 900 times that of 13C, allowing its recent increased use in chemical exchange studies, as well as in biological investigations in living systems, where the lithium ion is of importance. The 7fi nucleus is quadrupolar with a spin number of 312. The resonance lines of lithium ion in solutions are exceptionally narrow (natural linewidth less than I Hz) and chemical shift can be measured with considerable accuracy 170,1261. The usual chemical shift range for 7Li cations is about 10 ppm which is relatively small compared to the other alkali metal ions. The frequency of the 7ti resonance is quite sensitive to the environment, but no correlation of the chemical shift with the solvent Gutmann donor number is observed, possibly due to the fact that the diamagnetic and the paramagnetic screening constants are of the same order of magnitude and, therefore, ring currents and neighbour anisotropy effects influence the magnitude and direction of the chemical shifts [27]. As long âs the rate of lithium ion exchange between the cryptate bound site and the free solution site is substantially less than the chemical shift separation, individual resonances can be observed for the two sites, thereby making the lithium cryptate kinetics agreeable to variable temperature 1Li nmr studies and lineshape analysis. Amira Abou-Hamdan 69
4.2.2 23Na NMR Spectroscopy
23Na nmr spectroscopy is a convenient and direct method for the investigation of the exchange kinetics in sodium cryptates
184,128,1291. 23Na has a natural abundance of 100% and an intrinsic nmr sensitivity of approximately one tenth of the proton [125], or near 500 times that of 13C, allowing its extensive use in studies of interactions with organic molecules and investigations of biological systems. The 23Na nucleus is quadrupolar with a spin number of 312. However, in the extreme narrowing conditions of rapid reorientational motion, the 23Na relaxation is characterised by a single exponential rate and gives rise to true Lorentzian lineshapes. The usual chemical shift range for 23Na cations is about 40 ppm which is relatively small for a heavy nucleus. The chemical shift is strongly solvent dependent and shows a linear correlation with the solvent Gutmann donor number t70].Complexation by cryptands usually results in a considerable chemical shift change and if the rate of sodium ion exchange between the cryptate bound site and the free solution site is substantially less than the chemical shift separation, then the individual resonances can be observed for the two sites. These characteristics allow an investigation of the sodium cryptate kinetics by variable temperature 23Na nmr techniques and lineshape analysis.
4.2.3 Kinetic Applications
The theory of the nmr techniques for the study of rate processes in general have been reviewed by Lincoln t1301. Rates of exchange, Amira Abou-Hamdan 70
activation parameters and mechanisms of exchange may all be derived (under favourable conditions) through the variation of the nmr resonance lineshape with temperature. It is appropriate to describe the physical and mathematical aspects of these methods which have been used throughout this study. \ilhen a sample is placed in a magneric fietd Ër the induced magnetization may be described by the net macroscopic magnetization vector, ü, in the z direction (of magnitude Mr"q). The application of an additional radio frequency field, H*1, of frequency û) provides magnetic field components along the x and y axes:
H = (Ht cos rDt, -H1sin cot, He) (4.1)
and ü is tipped away from the z axis, thereby decreasing the z component, Mz. The time required for Mz to regain its equilibrium value, Mr"q, is characterized by Tl, the spin-lattice or longitudinal relaxation time. The magnetization in the xy plane takes a period T2, the spin-spin or transverse relaxation time, to equilibrate to zero after .a perturbation. Bloch's phenomelogical equarions tl3ll describe the time dependence of ü in a stationary frame:
dM* . rrrY . ., Mx ìt = Y(MvHo + MzHlsin cot) ,, (4.2)
y(-M*Ho + MzHlcos tor) - # (4.3)
dM, (M.-M,ao) -y(M*H1sin <¡t + MyHlcos tot) - (4.4) dt T1 Atnira Abou-Harulan 7I
where y is the nuclear gyromagnetic ratio.
In the rotating frame of reference (rotating about z axis, where z is aligned with the Ho field) the time dependence of the transverse
magnetization ¡4xy may be reformulated as [130]:
Y--oMxy-iyH1M,sq (4.5)
where, u: l[Tz - i(ros - to).
Nuclear magnetic resonance may be observed through: (i) the continuous wave slow passage experiment which consists of slowly sweeping the r.f. field applied to a sample in a fixed magnetic field (or vice versa) and concurrently observing the spectrum; and (ii) the pulse methods where short, high energy magnetic pulses are applied at a discrete frequency and, generally, observing the nuclear spin system after the r.f. pulse is complete. The former method, which represents an easily visualized model of the nmr experiment will be considered in detail, and it will be shown at the end of this section that, the lineshape obtained from the pulse Fourier transform nmr experiment is equivalent to that derived from a continuous wave experiment. Under the continuous wave slow passage conditions, -yHlMrsqT2 v= (4.6) Generally H1 is small thus the absorption mode lineshape becomes Lorentzian: -yHlMrsqT2 (4.7) The classical Bloch equations which describe the transverse components of magnetization in the rotating frame require modification under conditions of chemical exchange Í132,1331 through the mean site lifetimes of the exchanging species. In the case of a nucleus exchanging between two sites A and B at a rate ftape - rA-1 pA = frgpg = rB-lpB , where p¡ and pg are the relative populations, and rA and Tg aÍe the mean lifetimes of the nucleus in sites A and B respectively. The time required for a nuclear spin to transfer from site A to site B is too small for nuclear spin precession to occur, thus the nucleus reaches site B with its phase memory of site A intact, and vice versa. This causes dephasing at site B and results in an increase in M*yB, the transverse magnetization at site B, at a rate ftn M*yA = tR-l M*yA and a decrease in M*yA at the same rate. Similarly, transfer of a nuclear spin from site B to site A causes dephasing at site A and an increase in MxyA at a rate kg MxyB - rg-l M*yg whilst decreasing M*yB at the same rate. That is: dM*raldt = M*yglrg - M*r¡lr4 (4.8) dM*rg/dt = M,çy4/r4 - M*rg/rg , (4.e) Amira Abou-Hamdan 73 due to exchange induced dephasing effects only. Incorporation of these effects yields to a modification of the Bloch equations to: dM*ta/dt = -üAMxyA - iyHlMrsq4 + M¡yg/rg M1y4/t4 (4. 10) dM*tg/dt = -ûBMxyB - iyHlMrsqg + M¡¡n4/r4 M¡ys/rs (4.1 1) where y - gyromagnetic ratio of the observed nucleus Ht = magnitude of the applied r.f. field I aA = Tr^ i(ooe - t¡) 1 qB = T^ i(@6s - o) TZi the transverse relaxation time, T2, of site j in the absence of exchange. co : frequency of the r.f. field rlloj : Larmour (resonant) frequency of the observed nucleus in the absence of exchange The chemical exchange of an uncoupled spin system is studied in the case of a continuous wave experiment, then compared with the same study obtained from a pulse Fourier transform nmr experiment. I. Under continuous h,ave slow passage conditions Under these conditions MzA=MzeqA=pAMzeq (4.r2) Amira Abou.-Hatndan 74 MzB:MzeqB:pBMzeq (4.13) dM*ya _ dM^ys and dt dt :0 (4.t4) The total transverse magnetization, ¡{xy M*yA + M*yB may now be expressed in terms of r¡ and rs ll32,l34l : -iyH 1M2 rA + rB + rArB(cApB + ûBpA)) Mxy (4.15) (1 + anrn) (l + ûBrB) - 1 The nmr lineshape is proportional to the imaginary part of ¡4xy 1130,135,1361 and is given by: -yHlMreo{Ytl + a+(ps/T24 + pe/Tzs)l + QR) v Y2+R2 (4.16) where T= PBTA=PATB Âo= úJoA - O0B 8t¡ = l/2 (lcoea - @osl) - o 4o Y= t (+rr- õr,¡2 * ,. ffi. {"r; Âto a= r (õc,r - 7(Ve - pB)) llArollÂro R= õor [l + r (TzA*To)l * z t(7r,..^) *7@e,- pB) In the general case of an uncoupled spin system undergoing chemical exchange the lineshape obtained from a pulse Fourier transform nmr experiment is equivalent to that derived from a continuous wave experiment tl37l. Although the methods discussed Atnira Abou-Hamdan 75 aÍe not applicable when there is coupling between exchanging nuclear spins, coupling within an exchanging group may be accomodated by separately applying equation (4.16) to each component of the doublet and summing the two to produce the total lineshape. More complex systems may be analysed utilising a more sophisticated treatment [138,139]. The variation in the nmr lineshape may be studied using equation (4. 16) as the exchange rate is varied from "very slow" to "uery fast", for example, by increasing the temperature. (i) Very Slow Exchange Limit tR-l In this limit equation (4.16) approximares ro: -YHTPAM'"'TIA-I -v@ u=ru*Tru Gl7) which contains no chemical exchange parameters and describes two Lorentzian lineshapes centered at tos¡ and oos with line widths at half height of: wA = 2fT¿e.t" = 2lT2¡ and wB : 2lT2gs6s = 2lTZg , ignoring inhomogeneous field effects. (ii) Slow Exchange t4-1, ag-l << The absorption mode lineshape: funira Abou-Hamdan 76 _._ -yHrpaM'"q(Tza-1 + ta-1) , -yHlpsMrrr(Tzs-1 + re-1) u=(r *{r , (4.18) describes two Lorentzian lineshapes centered at tooA and co08 which are "exchange broadened". The line widths at half height are now: wA = 2fl2¡¡o1s = 2fT2t + 2/tn and wg = 2f|zp,obs = 2lTZg + 2ltB,. (iii) Fast Exchange r¡-1, 19-l > A single broad Lorentzian lineshape centered at (om¡ pA + o0B pB) is observed. The line width observed is w = 2lT2q6s = 2p¡'flze, + 2ps[Tzø + 2p¡2ps2 (ron - cÐ0B)2 (tn + rs) (4. 1e) (iv) Very Fast Exchange Limit The absorption signal is now single ^ Lorentzian lineshape centered at (c,lon pe + o0B ps) described by -yHlMzeo(paTzl-l + Þ sTzs- l) v= (4.20) (pnTze-l * psTzB-\2 + (pRooe + pBoos - r)2 The line width observed is w = Zlrzoa, = 2p¡.frzn + zpslrzs. Amira Abou-Hamdan 77 The rate of exchange is so fast that the exchanging nuclear spin experiences the weighted average of the environments A and B and no chemical exchange information is contained in the lineshape. 2. Under pulsed nmr conditions A short (10-o - 10-4 s) r.f. pulse of high energy, H-1, is directed along the x' axis (of the rotating frame X', y', z') applying a torque to the magnetization vector, ü, (initially aligned with the z axis) which causes it to rotate towards the y' axis, generating x and y components of ü. Immediately after the pulse, ú begins to its equilibrium position along the z axis. The resultant decay of the M*, signal to zero is sampled at regular time intervals and stored as a free induction decay (F.I.D) signal. An F.I.D. is obtained by setting Ht = 0 and solving the differential equations (4.10) and (4.11) to give [140]: ¡4xy Cle-O*t + Cze-O-t (4.21) where C1, CZ are integration constants and 2ó* = (qa + rA-l + GB + rg-l) * [(qe + rA-1 - ûB - ts-l¡Z + 4rA-lrB-\ln The Fourier transform, S, of the F.I.D. is given by: oo $= Jtuo.-ttr-r¡r)t dt 0 i Mo(tA + rB + rRrs(ûApB + qBpA)) (1+qepe)(laûBpB)-l (4.22) where, Atnira Abou-Hamdan. 78 ûA:TZtl+i(ooA-o) qB:TZg-l+i(<¡oe-o) o = the variable frequency of nucleus of interest tol = the fixed pulse carrier frequency. The absorption mode lineshape is derived from the imaginary part of equation (4.22) and is the same as that obtained in the continuous rwave experiment. 4.2.4 Lineshape Analysis A theoretical spectrum proportional to the experimental spectrum may be calculated for a particular T value according to equation (4.16) using the following input parameters: vs, Vc the chemical shift of a coalescing pair of peaks in the absence of exchange ws, wç the width at half height of each peak in the absence of exchange ps, pç the relative populations of each exchanging site R the estimated rate of exchange, R = (Tcps)-l = (rrp.)-l where subscripts c and s refer to the complexed and solvated metal cations respectively in this study, but in general, may refer to any two exchanging environments. The theoretical spectrum is generated by solving l(l + âÍcÞs + arspc)K + a(f + p. - l)Ll G (4.23) (K2 + L2) (lv. - v*l) where G = the signal intensity at frequency v Atnira Abou-Hamdan 79 2î'lvs - v5l a R v - vc f- Vs - vc Ws fs: 2lvs - vsl Wg tC - 2lv" - vrl L = f(l a âfc + arf) - (ar" + ps) K : r"p. + rsps + a(rcrs - f(f - 1)). Spectra for a variety of T values are generated and compared to the experimental one to obtain a best fit R value, where R = l/r"ps = l/r.p. (r" and Ts are the mean lifetimes, pc and ps are the relative populations of the complexed and solvated metal ions respectively) and thus the decomplexation rate constants, ftd (= Rp, = r.-l) are obtained for a series of temperatures. As described above, the input parameters required for the generation of the theoretical spectra are the chemical shifts, linewidths and populations of the site resonances in the absence of exchange. The 1l-i and 23Na linewidths and chemical shifts (and their temperature dependences) employed in the lineshape analysis were obtained through a combination of extrapolation from low temperatures, where no exchange induced modification occured, and the determination of the linewidths and chemical shifts in solutions containing either solvated metal ion or complexed metal ion alone in the coalescence temperature range observed for the solutions which contained both species. The ÍLi.C21CSl* and [Na.C21l]* systems were analysed, by generating the theoretical lineshapes and the subsequent Atnira Abou-Hamdan 80 determination of the rates of exchange, using an interactive program LINSHP. This program is based on the methods of Nakagawa [141], and Siddall, Stewart and Knight tl42l and adapted for the BNC-12 computer of the HX-90E spectrometer by Williams [143]. It provided a convenient visual display on a cathode ray oscilloscope of the experimental lineshape with a calculated theoretical spectrum generated from the input parameters. In the case of the lLi.C2ZCfl+ system, a computer program, LINPFT Í1441, a modified version of LINSHP written for the VAX 11- 780 mainframe computer was used for the determination of the rates of exchange. The analysis involved determining the best fit R values which minimised the residuals of fit between the theoretical and experimental spectra. 4.2.5 Calculation of Activation Parameters The Eyring equation [145,1461 from the transition state theory expresses the dependence of the decomplexation rate constant, k¿ on the activation parameters and temperature: kd = GyTD exp(-Â,FI.lRT + (4.24) ^S./R) where ks is Boltzmann's constant It is Planck's constant R is the gas constant Afl* and Â,9 are the activation enthalpy and entropy and T is the temperature (K) Amira Abou-Hamdan 8l Kinetic data may be fitted to the Eyring relationship in this exponential form or in the more convenient linear form below (where r.-l is substituted for k¿): L,H* h AJF ln (r.Z) + (4.2s). RT lln ks rl The non-linear, weighted least squares program DATAFIT tT5l was used on a Vax 11-780 computer to derive k¿, L,H* and using ^,S* the above relationship. DATAFIT minimises the residual differences in an n-dimensional sum of squares space between a calculated and an experimental surface (in this study, defined by k¿ and temperature). The errors quoted for the activation parameters provided by DATAFIT are the standard deviations for each parameter in the sum of squares space. These values only take into account the errors between the input parameters and not of other possible systematic errors. Atnira Abou-Hamdan 82 4.3 Results and Discusslon 4.3 .I General Mechanistic Aspects of Cryptates The formation of a. cryptate involves a complex combination of effects involving solvational, conformational and coordination changes in the metal ion and cryptand in solution 1341. Two possible mechanisms for the exchange of the alkali metal cation between the solvated and cryptate sites have been proposed ll20,l47l: (i) the first order dissociative mechanism (monomolecular) in which the alkali metal cation is completely decomplexed before another alkali metal cation complexes with the cryptand: ka M+ + CryP lM.crypl* (4.26), k¿ in this case, Ilt" = k¿ independent of both the solvated cation and the cryptand concentrations, where ts is the mean lifetime of the complexed cation. (ii) the second order exchange process (bimolecular) in which decomplexation and solvation of the complexed alkali cation takes place simultaneously with complexation and desolvation of a solvated cation (the asterisk is a typographical notation only to distinguish between the alkali metal cations): *M* + [M.Cryp]* [*M.Cryp¡* + M+ (4.27), in this case, llt" = k proportional to the solvated cation concentration. funÌra Abou.-Hamdan 83 The former mechanism is exemplified by [Na.C222]* tI22l and ÍLi.C222l* [120] in various solvents. The latter is exemplified by lLi.C2ZLl* in solvents of high dielectric constants t1481. The Eigen-Winkler mechanism Í149,1501, a simplified mechanistic scheme based on the mechanisms proposed for the interactions of alkali metal ions with antibiotics, represents a possible pathway for the predominant mononuclear mechanism. The overall complexation of the Li+ ion to the cryptand involves the initial diffusion controlled formation of the encounter complex in which both the metal ion and cryptand retain their complete solvational shell, characterized by kt. This is followed by the first Li*-C21C5 bond formation which involves desolvation of the solvated Li* ion, and possible conformational and solvational changes in the cryptand, characterized by kZ. All subsequent steps are incorporated in the final step and are represented by k3 which include further displacement of the solvent bound to the Li* ion (a total displacement in the case of inclusive [Li.C21C5]* whereas the Li+ ion is probably still partially solvated in the exclusive form of this cryptate), leading to the formation of the lithium cryptate in which, for CZICS a total displabement of the solvent occurs in the inclusive form of this cryptate, whereas the Li* ion is probably still partially solvated in the exclusive form, and for C22CZ the Li* ion is probably still partially solvated as can be seen from the space filling representation in Figure 2.5. The mechanism may be schematically represented for [Li.C21C5]+ in Scheme 4.1, noting that this scheme is almost certainly over simplified. Amira Abou-Hamdan, 84 Scheme 4.1 kt Li+ + C2lC5 Li+.... C2lC5 k-t kz Li*. C2IC5 k-z kz ll-i.C2l Csl * k-z The identification of the rate determining step for the cryptate formation is intricate because of conformational equilibrium in the cryptand, and desolvation of both the metal ion and the cryptand will contribute to the energetics of each step [10,151], leading to the cryptate formation changing to an earlier or later step in the above scheme. In the nmr studies, only two rate constants (k" and kO) are derived for the overall equilibrium shown in Scheme 3.1. Thus, these rate constants are closely related to the rate determining steps for the complexation and decomplexation processes of cryptates. 4.3.2 Exchange Kinetics of Li+ on [Li.CzlCs]* In acetonitrile, propylene carbonate, and acetone the rate of exchange of Li+ between the solvated and [Li.C2lCs]+ environments is too slow to cause any significant broadening of the separate 1Li resonances characterising these environments up to the solvent boiling point temperature. However, conservative lower limits for Ts at 298.2 K 50, 100, and 200 ms are estimated for these systems respectively. These limits may be obtained through equation 4.28 Atnira Abou-Hamdan 85 where wsSs is the observed width (Hz) of the [Li.C21C5]* resonance at half amplitude, l.5wsbs is the width which would be observed if the rate of Li+ exchange was sufficient to increase the natural linewidth by half Í1521. 1.5nwq6s - Trwsþs : l/t" : ps/(rspc) - kd (4.28) In methanol, dimethylformamide, diethylformamide, and dimethylacetamide a temperature dependent coalescence of the 1Li resonances arising from solvated Li+ and tLi.CzlCsl* (Figure 4.1) yields the kinetic parameters for the decomplexation of ÍLi.CzlCsl* (equation (4.2g)) shown in Table 4.1. These parameters are derived from the temperature variation of the mean lifetime of lLi.C2lC5l+, rç, through equation (4.30). The rc values are derived using complete lineshape analysis (as described in the previous section) of the coalescing 7Lí resonances observed for solutions i - x, as exemplified by Figure 4.1. The decomplexation rate constant values, kd, are given (i) in the midst of the coalescence region where the modification to the spectra due to the chemical exchange is at a maximum and hence the most reliable values for lt¿ are obtained, and (ii) at 298.2 K for the purpose of relevant comparisons with other systems. Li+ + C2lCs ll-i.C2lCsl+ (4.2e) k¡ = llrc : @BTlh) exp(-L,H*lRT + AsF/R) (4.30) Atnira Abou-Hatndan. 86 Figure 4.1 Typical exchange modified 116.59 MHz 1Li nmr spectra of a diethylformamide solution of LiCIO q (0.0261 mol dm-3) and C2lC5 (0.0091 mol dm-3). Experimental temperatures and spectra appear to the left of the figure, and the best fit calculated lineshapes and corresponding Ts values appear to the right. The resonance of [Li.C21C5]* appears upfield from the solvated Li+. The solid curves represent the simultaneous best fit of the data derived from each set of solutions to equation (4.30). -þl: ?00 Hz T (K) ï (ms) C 304 6 40 289.t. 77 28t*.8 9.8 280 1 13.6 275.5 16.t* TXPT IA LC tunira Abou-Hamdan 87 Figure 4.2 The temperature variation of Tç for Li* exchange on tLi.C2lCsl*.Data points for the methanol solutions (i) and (ii) are represented by circles and triangles respectively; for dimethylformamide solutions (iii)-(vi) by triangles, squares, circles, and inverted triangles respectively; for diethylformamide solutions (vii) and (viii) by circles and triangles respectively; and for dimethylacetamide solutions (ix) .and (x) by triangles and circles respectively (these values are muttiplied by four). í 100 T t c 50 (s K) metho nol 20 10 5 dimethytocetq mide (x4) dimethytfo rmqmide 2 1 diethytfo rmo mide 30 32 34 3b tohtt tKll Table 4.1 Solution Conrposition and Kinetic Parameters for Li* Exchange on lLi.C2lCSl* in Various Solventsa Solution Solvent (solvated)] k¿ (298.2K) LH,t* aSd* ILi* ILi.C2lC5.] ! mol dm-3 dm-3 s-l I mol kJ mol- J K-l mol-l d MeOH 0.05 3 0 0 .0 420 20.4 + 0.4 37.2 + 0.8 -95.0 + 2.8 I ii 0.03 60 0.05 90 22.7 + 0.5 34.9 + 0.9 -101.6 + 3.3 Ë' l-ll 21.6 + 0.4 36.1 + 0.9 -98.4 + 3.1 s DMF 0.0150 0.0074 120+3 38.9 + 1.3 -74.8 + 4.7 It IV 0.0120 0.0079 ll0 + 5 38.0 + 2.3 -77 .5 + 3.4 il V 0.01 l0 0.01 10 120+3 34.0 + l.t -73.1 + 4.3 r vi 0.005 8 0.0 r 40 120+3 39.3 + 1.0 -78.7 + 3.6 il iii-vi 116+2 38.4 + 0.9 -16.5 + 3.0 vii DEF 0.0170 0.0091 215+5 27.7 + 1.6 -110.3 + 5.3 il ,103.5 viii 0.0091 0.0170 196+3 29.0 + 1.2 , + 4.2 vll-vlll 210+4 21.8 + 1.6 -101 .6 + 5.3 IX DMA 0.0140 0.0071 241+6 49.0 + 1.3 -34.0 + 4.3 X 0.00 8 4 0.0126 235+5 48.9 + l.l -35.6 + 3.8 lx-x 231+4 49.0 + 2.1 -35.0 + 2.8 a Kinetic parameters were determined from DATAFIT analysis. Errors quoted represent one standard deviation. oo oo Table 4.2 Kinetic Parametersa for Li* and Na* Exchange on Cryptates in a Range of Solvents Cryptate Solvent DNb 10-s kc Q98.2 K) k6 Q98.2 K) L,H6n ÂSd* dm3 mol-l s-l s-1 kJ mol-1 J K-l mol-l ll-i.C2l Csl + MeCN I 4 I slow [Li.C21Csl. PC I 5 1 slow [Li.C2lCsl* Acetone t7 .0 slow * 19.0 [Li.C21Cs] MeOH 0.221 21.6 + 0.4 36.1 + 0.9 -98.4 * 3.1 (23.5)c + [Li.C2l Cs] DMF 26.6 0.073 116+2 38.4 + 0.9 '76.5 + 3.0 * ll-i.C2l Csl DEF 30.9 0.149 210+4 27.8 + 1.5 -108 + 5 [Li.C21Cs]* DMA 27 .8 0.124 237 +4 49.0 ,+ 2.1 -35.0 + 2.8 [Li.c2t 1l* DMFd 26.6 1.27 1.0130 + 0.0033 64.4 * 2.5 -64.8 + 5.9 [Na.C2lCsl* MeCNe I 4 I 100 84.8 + 1.6 57.9 + 0.7 -13.8 + 2.1 lNa.C2lCsl* PCe 1 5 1 25.s 19.4 + 0.5 10.3 * 0.5 r5.3 + 1.4 * [Na.C21Csl Acetonee l7 .0 84 878 + 6 54.4 * 0.4 -6.1 É 1.2 ! 19.0 Ar [Na.C2l C5]* MeOHe 104 1800 + 50 44.9 + 0.1 -31.9 + 0.4 (23.5\c I [Na.C2lCsl* DMFC 26.6 2t4 28800 + 300 40.0 + 0.1 -25,3 r 0.5 è lNa.C21ll* PCf 1 5 I 2r0 0.03 6 r 9.0 [Na.C211]* MeOHf 31 .0 2.5 (23.5)c lNa.C21 I I. DMFg 26.6 19.2 I2.l + 0.2 83.5 + 0.5 55.8 + r.2 a Quoted errors represent one standard deviation. b Gutmann donor number 164). c ,l e References [65,66]. Reference [69]. Reference [84]. f Reference 1721. B Reference [51]. oo Aøtn Abou-ÌI¿uadan 90 It is seen from Figure 4.2 that the temperature variations of rç for each of the solutions studied for a given solvent are indistinguishable. Thus the mean lifetime of lLi.CzlCsl*, rc, is independent of the concentration of solvated Li+ (Table 4.1) consistent with the non-participation of solvated Li+ in the rate determining step of the predominant pathway for Li+ exchange on ÍLi.CZlCsl*, and the operation of a monomolecular mechanism for the decomplexation of Li+ from the cryptand. Monomolecular decomplexation mechanisms also operate for [Li.C211]+, [Na.C211]*, and [Na.C2lC5]+ 169,51,841. However, it is not invariably rhe case thar the monomolecular mechanism operates in cryptate systems. This is exemplified by [Li.C2zl]+ for which the Li+ exchange process changes from a monomolecular mechanism in methanol, which has a dielectric constant of 32.7, to a bimolecular mechanism (in which both the leaving and entering Li+ are bound by C22l in the transition state) as the solvent dielectric constant increases in acetonitrile and propylene carbonate which have dielectric constants of 38.0 and 69.0 respectively tl20l. Whereas in the case of the highly flexible C222, the Li* exchange on tLi.CzzZl* also follows a dissociative mechanism pathway. A suggested explanation is that this highly flexible C222 wraps itself around the small Li* such as the only possible exchange of the cation between the solvated and complexed sites would be the dis sociative pathway t I 201. Other factors than the geometrical structure of the complex such as the nature of the solvent [48], the concentration of the cation [92,113,153], and the counterion t1541 cannot be ignored. Variation in the nature of the solvent has a substantial effect on the magnitude of k¿ characterising ÍLi.CZlCsl*. Of the four solvents for Atnira Abou-Hamdan 9I which quantitative data are available, the weaker electron donating solvent, methanol , yields the smaller kl value while dimethylformamide, diethylformamide and dimethylacetamide, which aÍe of similar structure and DN, yield /c¿ values grouped within a factor of two of each other over the experimental temperature range. (Nevertheless, this similarity of k¿ is a consequence of differing, b:, compensating, LHd* and ÂSd* contributions for the latter three solvents.) The kc values exhibit a '.smaller variation than k¿ consistent with the variation of K, for ÍLi.CZlCsl* in these solvents being predominantly determined by k¿ variations as is also observed for [Na.C2lC5]+ and other cryptates (Tab\e 4.2). In dimethylformamide the greater stability of [Li.C21l]+, by comparison with [Li.C21C5]+, is predominantly a consequence of its 9000 fold smaller kd, with a modest contribution arising from its l7 fold greater kc. The smaller k¿ for [Li.C211]+ is attributable to the greater electrostatic interaction between Li+ and the six donor atoms of C2ll, and the inclusive nature of the cryptate. In contrast C2lC5 has five donor atoms and ÍLi.CzlCSl* exists, in part at least, in the exclusive form and is consequently expected to be more labile. The increase of k¿ in the sequence lLi.Czlll* . [Na.C211]* . îLi.CzlCsl* . [Na.C21Cs]* in dimethylformamide (Table 4.2) suggesrs rhat rhe greater electrostatic attraction of CZII for Li+ and Na+ is a particularly important factor determining the decomplexation rate. The greater stability of [Na.C21Cs]*, by comparison ro thar of ÍLi.CzlCsl*, in dimethylformamide is a consequence of the first cryptate possessing a 250 fold greater k¿ and a 2900 fold greater kc (Table 4.2). A similar relationship exists for methanol. The lower lability of lLi.CzlCsl* by comparison with [Na.C21Cs]* may be Amira Abou-Hamdan. 92 attributable to the smaller size of Li+ and its consequently stronger electrostatic interaction with ligands. Thus the magnitude of kc partially reflects the sequential desolvation of L7*, which is more strongly solvated than Na*, and in consequence kc for [Li.C2lCs]+ is significantly smaller than that for [Na.C21Cs]*. Simitarly the smaller value of k¿ for ILi.CzlCsl* probably reflects the greater electrostatic interaction between Li+ and C2lC5, and the existence of this cryptate in an inclusive/exclusive equilibrium compared with the exclusive lNa.C21C5l+ cryptate. * The free energies of solvated Li* , C2lC5, lLi.CZlCsl and the transition states will be determined to a substantial extent by the electron-donating ability of the solvent (expressed by DN), the number of sovent molecules bound to Li+ in the fully solvated state and in the cryptate, the steric interactions of solvent molecules in these environments, and to a lesser degree the secondary solvation of Li*, C2lC5, lLi.CzlCsl*. These factors cannot be quantified, bur a general trend of an increase in the A¿ values characterising [Li.C21CS]+ and [Na.C21C5l+ to increase as solvent DN increases (Table 4.2) indicates that the influence of the solvent is similar for both cryptates. There is no apparent correlation between the solvent dielectric constant and k¿ (Table 4.2). The decomplexation mechanism involves sequential resolvation of the metal ion and accompanying stereochemical changes. However, the gross aspects of the variation of kd may be explained on the basis of equation (4.31), where ÂGd* for decomplexation is largely the difference between LGr., the free energy change arising from structural rearangements in tM.C21Csl* to reach its transition state stereochemistry in the absence of solvent, and ÂGr', the free energy change arising from involvement of solvent Arnira Abou-Hamdan 93 [Li.C2lCs]** + S [Li.C2lC5]** + g -Á- 1 ÂC5' I ILi.C2lCs.S].+ ÂG.- t [Li.C2l Cs.S]-" I r LCc- I LCr' ^Cr' ÁOc' I ^cd d C2lCs ^G [Li.SJ*-t [Li.SJ' + C2lCs l + [Li.C2lC5]. + S [Li.C2lCs]. S Strong Donor Solvent Weak Donor Solvent Figure 4.3 Simplified reaction profile for the complexation and decomplexation of lLi.C2lCSl* in solvents in which kc and ft¿ show little and substantial variation with the nature of the solvent (S). The free energies of the lLi.CzlCSl* + S ground states in strong and weak donor solvents are normalized to the same value in both profiles. The solvent molecule shown bound in the transition state lLi.CZlCS.Sl** is in addition to those already bound in the [Li.C2lC5]* ground stares. The transition state lLi.CzICs.Sl** only exists in the absence of solvent participation in the activation process. The [Li.Sn]* species represent the ground state solvated ions. Amira Abou-Hamdan 94 in the activation process which causes an increase of M+ solvation in tM.C2lCsl* in the transition state. AGcl* - LGr* - AGs* (4.3 1 ) As the magnitude of ÂG¡* is defined to be independent of solvent, it is evident that as AGs* increases with the D¡ of the solvent AG¿* decreases and kd increases [84], as shown qualitatively in Figure 4.3. Similar increases in h and solvent DN have been observed for other cryptates L51,721. 4.3.3 Exchange Kinetics of Na+ on [Na.CzlI]+ and [Na.C21C5]+ In diethylformamide and dimethylacetamide the rate of exchange of Na+ between the solvated and [Na.C2lCs]* environments is in the fast exchange limit of the 23Na nmr timescale, in other words the exchange of the sodium ion between the two sites, solvated ion and the cryptate, is greater than ,tiht , where Âv is the difference between the chemical shift (in Hz) of each site resulting in only one populâtion-average resonance observed t631. Thus the lower limits for the respective ft¿ values at 298.2 K - 680, and 175 s-1. (This is also the case with the diaza-crown ether, Czl, in the three solvents dimethylformamide, diethylformamide and dimethylacetamide which are of similar electron donating strength but different molecular size tl55t). In diethylformamide a temperature dependent coalescence of the 23Na resonances arising from solvated Na+ and [Na.C2ll]+ yietds the + kinetic parameters for the decomplexation of [Na.C21 1] (equation Atnira Abou-Hamdan 95 (4.32)) shown in Table 4.3. These parameters are derived from the temperature variation of the mean lifetime of [Na.C211]+, Tc, through equation 4.33. The Tç values are derived using complete lineshape analysis (as described in section 4.2) of the coalescing 23Na resonances observed for solutions i - iii. The decomplexation rate constant values, kd, are given (i) in the midst of the coalescence region whe¡e the modification to the spectra due to the chemical exchange is at a maximum and hence the most reliable values for k¿ are obtained, and (ii) at 298.2 K for the purpose of relevant comparisons with other systems. k Na+ + CzIl [Na.C211]+ (4.32) kd /k,l : llr" : (bBTD exp(-ÅFI"lRT + (4.3 3 ) ^.y/R) It is seen from Figure 4.4 that the temperature variations of r, for each of the solutions studied for this system are indistinguishable. Thus the mean lifetime of [Na.C21l]+, rc, is independent of the concentration of solvated Na+ (Table 4.4) consistent with the predominant decomplexation mechanism involving a monomolecular reaction of [Na.Czlll+ in the rate determining step, as is also the case in a range of other solvents including dimethylformamide t841. The kinetic parameters derived for the three solutions studied are given in Table 4.3 as are the kinetic parameters derived from the t. data of the three solutions fitted simultaneously to equation (4.33). The increase of solvent electron-donating power (indicated by an increase in DN) has been observed to cause an increase in the Anira Abou-Hamdan 96 Figure 4.4 The temperature variation of Tc for Na* exchange on [Na.C211]*. Data points for the diethylformamide solutions (i)-(iii) are represented by triangles, squares and circles respectively. The solid curve represents the simultaneous best fit of the data derived from the three solutions to equation (4.33). 2.0 diethytformomide 1 0 tJ1 :¿ 0.5 .(JP 0.2 0 1 0 0 27 2B 29 30 31 lo4l t tt<-11 ! Table 4'3 Solution Composition and Kinetic Parameters for Na* Exchange on [Na.C2ll]* s 1n Diethy form am ide.a I x: :s Solution *] [Na*(solvated)] [Na.C2l I k¿ L,H¿' ÂS¿* mol dm-3 l mol dm-3 s- kJ mol-l J K-l mol-l 0.06 8 0.037 (3s0.0 K) (298.2 K) ll50 * 40 14+2 65.0 + 2.8 -1.8 + 7.9 ll 0.05 I 0.040 1220 + 40 14+2 69.8+2.8 12.3+7.8 lll 0.024 0.069 1190 + 40 14 + 2 66j * 3:l 3.1 + g.5 (i-iii) 1180 + 30 14 + 2 67.l + t.g 4.4 + 5.0 a The quoted errors represent two standard deviations \ì\o Atnira Abou-Hamdan 98 Table 4.4 Kinetic Parameters for Na* Exchange on [Na.C21]*, [Na.C2lCs]+ and [Na.C211]* Systems in Dimethylformamide, Diethylformamide and Dimethylacetamide. Sy stem kc Q98.2 K) k¿ (298.2 K) LH¡ ÂSd* dm3 mol-l s-l s-l kJ mol- I J K-l mol-l Dimethylformamide [Na.C211]*a 1.92 x 106 t2.r 83.5 55.8 lNa.C2lCsl*b 2.14 x 107 2.88 x 104 40.0 -25.3 [Na.C21]+c fast fast Diethylformamide [Na.C211]* 2.29 x 106 18.2 67.1 4.4 [Na.C2lC5]* fast fas t [Na.C21]+c fas t fast Dimethylacetamide lNa.C211l+c 2.49 x 106 45.2 64.8 4.3 * [Na.C21Cs] fas t fast * lNa.C2U fast fast a Reference 1731. b Reference [84]. c Reference [155] Atnira Abou-Hamdan 99 decomplexation rate constant values, k4, over several orders of magnitude for both [Na.Czll]+ and [Na.C21Cs]*, but has little effect on the complexation rate constants, k", such that the substantial variations observed in the overall stability constant values, Kr, are predominantly a consequence of the variation in k¿ t841. This clearly indicates the involvement of the solvent in the rate determining step characterised by kd. However, in the case of dimethylformamide (dmÐ, diethylformamide (deÐ and dimethylacetamide (dma), which aÍe of similar electron-donating power, their differences in molecular size cause only minor variations in k¿ (Table 4.4) and Ks (as seen in chapter 3) for [Na.C21l]+. In dmf, kd characterising [Na.C21l]* is 2000 times less than k¿ characterising [Na.C21Cs]*. The small increase in ka observed for [Na.C211]* in changing from dmf to def and dma solvents, if proportionately reflected in the kd values for [Na.C21C5]*, will place this system in the fast exchange limit of the nmr timescale; and it is reasonable to conclude that the variation in Ks observed for [Na.C21Cs]* in def and dma (from chapter 3) are largely a consequence of variations in kd, rather than in k", by analogy to the [Na.C2l l]* system. The small variation in the lability of [Na.C2ll]* in dmf, def and dma (quatified by k¿) contrasts with the several orders of magnitude increase in the lability of the solvent exchange process observed with increase in solvent size in the lCo(dmÐ 6]2*, [Co(defl612* and [Co(dma)O]2* systems and the analogous nickel(Il) systems Í1241. This increase in lability is attributed to increased steric crowding in a dissociatively activated solvent exchange mechanism. It is also reported that increasing solvent size in the [Mn(solvent)612+ system causes a change from an associatively to a dissociatively activated funira Abou-Hamdan 100 solvent exchange mechanism 1128,1561. However, in the cases of [Na.C211]* and [Na.C21Cs]+ it appears that the variations in the molecular size of dmf, def and dma, are insufficient to cause substantial variations in k¿ through changes in steric interactions. Nevertheless, major variations in the stability and tability of the cryptates [Na.C211]* and [Na.C21Cs]* have been observed when significant differences in D¡ occur [84,1571. 4.3.4 Exchange Kinetics of Li+ on [Li.C2ZCZ]+ In methanol, dimethylformamide, and diethylformamide a complete lineshape analysis of the temperature dependent coalescence of the 7Li resonances arising from solvated Li+ and [Li.C22Cz]+ (Figure 4.5) yields rc, the mean lifetime of Li* in ÍLi.C22Czl+, for solutions whose compositions are given in Table 4.5. It is seen from Figure 4.6 that the temperature variations of tç for each of the solutions studied for a given solvent are indistinguishable. Thus the mean lifetime of ÍLí.CZ2Czl+, rç (= llk¿, where ka is the decomplexation rate constant of this cryptate) is independent of the concentration of solvated Li+ (Table 4.5) consistent with the non- participation of solvated Li+ in the rate determining step of the predominant pathway for Li+ exchange on [Lí.C22C2\+, and the operation of a monomolecular mechanism for the decomplexation of Li+ from the cryptand as shown in equation (4.34). Li+ + C22Cz [Li.c22Cù+ (4.34) frd = Ilr" = &sTD exp(-ÂI1-lRT + ASþ/R) (4.3s) t) f: ) Amira Abou-Hamdan 1 0 1 Figure 4.5 Typical exchange modified 116.59 M'Ii'z lLi nmr spectra of a diethylformamide solution of LiCIO q (0.0206 mol dm-3) and C22Cz (0.0068 mol ¿m-3). Experimental temperatures and spectra appear to the left of the figure, and the best fit calculated lineshapes and corresponding Tç values appear to the right. The resonance of lLi.CZ2C2l* appears upfield from the solvated Li*. -H-, 200 Hz T (K) l, (ms) 295 9 IJ 28tr.B 19 211 1 32 263 8 51 256.tr 13 248.0 109 EXPT CALI Amira Abou-Hamdan r02 Figure 4.6 The temperature variation of Ts for Li* exchange on [Li.c22c)*. Data points for the methanol solurions (i) -(iii); for rhe dimethylformamide solutions (iv)-(vi); and for rhe diethylformamide solutions (vii)-(ix) aÍe all represented by circles, triangles, and squares respectively. The solid lines represent the best fits of the combined data for each trio of solutions to equation (4.35). 50 lT sK) 20 methonoI 10 5 dimethytformsmíde 2 1 0.5 diethytformqmide 0.2 31 33 35 37 39 r+1 43 45 to4tt (t<-1) Table 4.5 Solution Composition and Kinetic Parameters for Li+ Exchange on lLi.C22Cflt in Various Solvents.a Solution Solvent ILi*(solvated)] ÍLi.CZZCz*l k¿ LHrt* ÀS¿* mol dm-3 mol dm-3 s-l kJ mol- I J K-l mol-l N\ (240 K) (298.2 K) I N MeOH 0.0101 0.0927 37.7 + 0.6 971 .6 + 39.2 31.0 0.4 -84.0 2.6 + + \ ll 0.007 3 0.01 l3 31.4 + 0.3 860.2 + 42.4 29.9 + 0.5 -88.6 + 3.1 I 111 0.0128 0.0075 37.9 + 0.4 1021.6 + 61.2 31.4 + 0.4 -80.1 + 1.6 l -11I 31.9 + 0.4 970.8 + 4l.l 31.0 + 0.4 -83.8 + 1.8 (2e8.2 K) iv DMF 0.0 r 17 0.0rt2 235 + 6.5 2l.9 + 1.5 -126 + 4.5 ?r v 0.0071 0.0132 246 + 9.0 22.5 + L2 -124 + 4.2 n vi 0.0143 0.0060 240 + 6.7 23.1 + 1.5 -122 + 5.1 il iv-vi 240 + 6.8 225 '+ 1.3 -124 + 4.8 (265 K) (298.2 K) vii DEF 0.01 3 8 0.0091 216 + 3.5 938 + 20 26.7 + 0.6 -98.3 + 2.1 vl11 0.01 13 0.0 r 70 232 + 8.7 913 + 62 24.8 + 1.3 -105.1 + 5.3 n ix 0.0070 0.0136 202 + 2.2 879 + 22 26.8 + 0.4 -98.8 + 1.6 tl vii-ix 216 + 3.5 917 + 28 26.7 + 0.6 -98.6 + 2.3 a (/.) Kinetic parameters were deternlinecl from DATAFIT analysis. Errors quotecl represent one standard deviation. Table 4.6 Kinetic Parameters for Li* Exchange on ÍLi.C22Czl+ and Other Cryptates in a Range of Solvents. Cryptan d Solvent AI I 0-5 kc Q98.2 K) k¿ (298.2 K) LHd* À,sd" dm3 mol- I s- I s-l kJ mol- I J K-l mol-l lLi.C22C2T MeCNa l4 .l slow ÍLi.C22Cz| Acetonea t7 .0 sl ow 18.0 ÍLi.c22Cùr H2Qa fast (3¡ )t' l 9.0 lLi.C22C2l. MeOHa 97 .t 971 + 42 3l .0 0.4 -84.0 + 2.6 (23.5)b * lLi.CZ2Czl+ DMFA 26.6 7.60 240+7 22.5 + 1.2 -124 * 5 lLi.C2ZCzl* DEFã 3 0.9 I 1.5 916 + 28 26.7 * 0.6 -98.6 + 2.3 ÍLi.C22Cz| Pyridinea 33.1 slow 19.0 [Li.C21C5]- MeOHa 0.221 2t.6 36 -98.4 (23.5)b ll-i.C2lCsl* DMFA 26.6 0.07 3 I l6 38.4 -7 6.5 ll-i.C2lCsl- DEFå 30.9 0 .149 210 27.8 -108 d . r 9.0 [Li.c2l I ] MeOHd 4.8 4.4 x l0-3 (23.s)t, I s' ll-i.c21|* DMFE 26.6 t.27 0.0130 64.4 -64.8 \ I 9.0 îLi.c22t). MeOHf rg2r 78.4 23.8 - 129 (23 .5 )b lLi.c222l+ MeCNf I 4 I 39200 420 11.6 - 157 1 9.0 lLi.c222)+ MeOHd >300 (23.Ðu a This study, quoted errors represent one standard deviation. b Gutmann donor number 1641. c References [65,66] d Reference 1921. e Reference [69].f Reference il581. à Amira Abou-Hamdan 105 The kinetic parameters for the decomplexation of lLi.CZ2c2l+ (Table 4.5) are derived from the temperature variation of Tc through equation (4.35). In acetonitrile, acetone, and pyridine the rate of exchange of Li+ between the solvated and lLi.CZ2Czl+ environments is too slow to cause any significant broadening of the separate 1Li resonances characterising these environments close to the solvent boiling point temperature. However, conservative lower limits for Tc Q98.2 K) of - 140, 100, and 125 ffiS, are estirnated in acetonitrile, acetone, and pyridine, respectively, by calculating rç which would broaden the [L|.C22C2]+ resonance width observed ar 298.2 K by a factor 1.5 (equation (a.28)). In water, at temperatures close to the freezing point, a single narrow resonance is observed for solutions containing [L|.C22C2]+ and solvated Li* consistent with exchange being in the very fast limit of the nmr timescale. This variation in lability with the nature of the solvent is similar to that observed for other cryptates for which it is generally found that the magnitude of k¿ increases with solvent DN [71,84] and is much more dependent on the nature of the solvent than is that of kc. This is consistent with the transition state being more similar to the solvated metal ion and the free cryptand than the cryptate. Two observations may be made: (i) a solvent with a high electron-donating ability is expected to compete effectively with the cryprand for binding with the cation, thus causes an increase in solvation in the decomplexation activation process of the cryptate which stabilizes the transition state. A considerably smaller decomplexation rate constant than predicted characterizes lLiCZZC2l+ exchange in pyridine. This can be interpreted by the fact that the donor atom in pyridine in Amira Abou-Hamdan 106 situated within the aromatic ring thus sterically prevented from reaching the Li+ in the cryptate, and (ii) the li¿ values for [LiC22Cfl* in rhe methanol and dimethylformamide are in agreement with the general trend of an increase in the magnitude of ft¿ with increasing D¡ of solvents over a wide range of solvents, but not over their narower Du range. However, occasionally some juxtaposition may occur for solvents of similar DN values t69l and' this probably results from a combination of superimposition of steric effects and solvent structure effects on DN. The lLi.C22Czl+ data may be compared with those characterising ÍLi.C21Csl*, ILi.C21 1l*, ÍLi.C22\+, and lLi.CZZTI+ for which monomolecular decomplexation mechanisms also operate in the solvents listed in Table 4.6 U20,691. The more open and flexible structure of C22C2 probably accounts for the greater frç and k¿ values characterising lLi.C22C2l+ by comparison to those pertaining to the relatively rigid inclusive [Li.C211]+ in which Li* is bound by the same number and type of donor atoms. Although C2lCs has one less donor atom than C22C2, its relative rigidity also confers a lower lability on tLi.CzlCsl* (which exists in an inclusive/exclusive equilibrium in solution) than that of lLi.C2ZCzl+. As size increases cryptand flexibility also increases and the optimum fit of Li+ to the cryptand cavity is lost with the consequence that the labilities of lLi.C22ll+and [Li.Czzz)+ approach that of [LL.C22C2]+. It is of interest to extend these studies to other alkali metal cations and preliminary 79.39 MHz 23Na nmr investigations of the kinetics of ÍNa.CZ2Czl+ have been carried out t981. The k¿ value obtained for dimethylformamide was 13 s-l at 298.2 K, lower than for Amira Abou-Hamdan rc] lLi.CZ2Czl+ (240 s-1, Table 4.5) and lic values are similar (k":1x 107 dm3 mol-l s-l). Thus the higher selectivity of c22cz to Na+ compared to Li+ (from chapter 3) is a consequence of a smaller ft¿ value, at least for this solvent. This must be the result of stronger interaction of Na* with the cryptand binding sites where the Na* reaches its optimal coordination number and the C22C2 experiences the least conformational strain t611. The determination of the kinetic data in other solvents and with other metal cations, 'is of interest for further understanding of the complexation of cations by this novel cryptand. These studies are part of the continuing work in our laboratories. 4.3.5 General Conclusions The study of the exchange of Li* on [Li.C2lCs]* in conjunction with the observations made in the cases of [Li.C2ll]*, [Na.C21C5]*, and * [Na.C21 I ] demonstrated the effects of: (i) the two structural conformations, inclusive or exclusive, otr the energetics of cryptate formation, and (ii) the dominant effect of the fourth oxygen donor atom in CzII in greatly increasing the stability and decreasing the lability of lLi.C2lll* by comparison with rhose of [Li.C21Cs]*. The results of the study of the [Na.C2l]*, [Na.C211]*, and [Na.C2lCs]* also demonstrated the dominant effect of the fourth oxygen donor atom in CZll, where the stability of [Na.C21 1] * is greatly increased and its lability decreased by comparison with those of [Na.C2lCì* and [Na.C21]*. The variation in the molecular size of dimethylformamide, diethylformamide and dimethylacetamide, which aÍe of similar electron donating strength, had little effect on the stability and lability of the sodium(I) complexes studied. Amira Abou-Hcmdan 108 The open and flexible structure of C22C2 resulted in fLi.C22Cù* being less stable and more labile than inclusive [Li.C2ll]* which has the same number and type of donor atoms, but has a more rigid structure. However, as a consequence of its clam-like structure, C22CZ is more able to approach optimum bonding distances with Li* than is C222, despite the latter cryptand possessing more donor atoms, with the consequence that [LL.C22C2]* is more stable and of a similar or lesser lability than ÍLi.C222l+ depending on the solvenr. These comparisons illustrated the interplay of the effects of the number of donor atoms and cryptand flexibility which produce variations in cryptate stability and lability. Thus, in general terms, the following conclusions aÍe summarized: (a) The metal ion exchange on cryptates proceeds through one of two possible pathways: associative or dissociative. Although the latter is the more prevailing exchange mechanism for cryptates as found for the lLi.C2l C5l *, and lLi.C22C2l+ in various solvents, the former operates in some cases depending upon factors such as the geometrical structure of the complex, the nature of the solvent, the concentration of the cation, and the type of the counterion. (b) Generally, the decrease in cryptate stability observed with increasing solvent DN, is reflected in an increase in the cryptate lability (quantified by /'.¿) rather than to any grear difference in kc. Ir is seen that [Li.C21C5]* and lLi.C2ZC2l. are encompassed within rhis trend. This suggests that the transition states for their cation exchange resemble the products more closely than the cryptates. The increase of the decomplexation rate constant, kd, with increasing solvent DN, is consistent with a resulting stronger solvation of the Arnira Abou-Harndan 109 transition state which leads to a decrease of the free energy change for decomplexation of the cryptate. (c) The decomplexation process for a cryptate is generally characteúzed by large negative aSd* values, which is probably due to the coordination of the solvent molecules to the metal ion while still bound to the cryptand. The decomplexation therefore involves substantial .opening up of the cryptate and the relatively large LHd* arise from a number of contributions: (i) disruption of the solvation of the cryptate, (ii) opening of the cryptate, and (iii) breaking of the metal-ion bonds. Thus a decrease in k¿ is observed with a closer fit of the cation in the cryptand cavity and the increase of the cation- cryptand electrostatic interactions resulting from the increase in number of donor atoms of the cryptand. Amira Abou-Hamdan. 110 Chapter 5 Experimen tal 5.1 Materials Lithium, sodium and silver perchlorate .salts (Fluka) were dried under high vacuum for 48 h and were stored over Pzos under nitrogen. Cryptand precursor CZl (Merck) and Silver nitrate (Matthey-Garrett) were used as provided. Cryptand Czll (Merck) was distilled then dried under vacuum. Linde 3Å and 4^ molecular sieves (BDH) were activated by heating in a furnace for 8 h at 450 K. Acetonitrile, methanol, propylene carbonate, acetone, pyridine, dimethylsulphoxide, dimethylacetamide, dimethylformamide and diethylformamide were purified and dried by literature methods [159,1601 and were stored under nitrogen over Linde 3Å molecular sieves in the case of acetonitrile and methanol, and over Linde 4Å molecular sieves in the case of the other solvents. The water content of these solvents was below the Karl-Fischer detection level of ca. 50 ppm. Tetraethylammonium perchlorate (TEAP) was prepared by acidification of tetraethylammonium hydroxide (40% in water, Fluka) with perchloric acid resulting in the precipitation of TEAP. The crude material was recrystallized from water until free from acid (no precipitate was observed upon addition of AgNO3 to an aqueous solution) and dried under high vacuum at 335 K for 24 h and stored over PZOS under vacuum. Amira Abou-Hamdan 111 5.2 Synthesis 5.2.1 Preparation of CZICs The cryptand C2lC5 has been discussed in the literature Í12,1601 but no full experimental details of its preparation was found and accordingly the preparative method employed in this study is a general modification tl6ll of the general method of Lehn and co- workers ll62l. A solution of 7,4,10-trioxa-7,13- diazacyclopentadecane (Kryptofix-2l, Merck) (2.0 E, 9.2 mmol) and triethylamine (2.5 E, 25.0 mmol) in dry benzene (100 cm3), and a solution of pentanedioyl dichloride (1.41 g, 8.34 mmol) in dry benzene (100 crn3) added simultaneously to d.y benzene (1200 ctn3) over 8h at 295 K ,with vigorous stirring under d.y nitrogen using Perfusor motor-driven syringes. After filtration and removal of the solvent under vacuum the residue was chromatographed on 'flash' silica t1631 [Merck, 230-400 mesh, methanol-dichloromerhane (4:96), Rf = 0.301, removal of the solvent and drying under vacuum afforded the cryptand diamide as a white solid (2.29, 84%), m.p. 388-390 K; 1; lH Vmax. 1630, 1 130 .r- n.m.r. (CDCI¡), õ I .8-5.2, a complex spectrum with large peaks at 3.55 and 3.60 p.p.m.; m/e 314(M+), 286, 271, 246, 239, 28 (100). The diamide was reduced with borane- dimethyl sulphide tl64l as follows. Diamide (1.3 g, 4.1 mmol) was dissolved in dry tetrahydrofuran (30 cm3) and treated with boron trifluoride etherate (1.0 cffi3, 8.2 mmol) at 325 K under dry nirogen. The reaction mixture was heated to reflux and borane-dimethyl sulphide (Fluka, I0% Me2S, 1.2 ch3, 10.9 mmol) was slowly added with a syringe. Heating was continued for 3 h and diethyl ether and Me2S distilled off as they formed. After allowing to cool to room Atnira Abou-Hamdan TT2 temperature, solvent was removed under vacuum and the white residue was treated with 6 mol dm-3 hydrochloric acid (25 cm3). The resultant solution was heated to reflux for 12 h and then evaporated to dryness. The crude cryptand was obtained from the hydrochloride salt after ion exchanging an aqueous solution on Dowex 1-8 x (oH- form, 50-100 mesh), concentration of the basic eluent, extraction with chloroform.(4 x 50 cm3), and evaporation of the combined extracts. Distillation yielded C2lC5 (1.1 g, 92%) as a colourless oil, b.p. ca.3B3 K I 0.025 mm (sublimation block), m/e 286 (M*,100), 269, 255; Vmax. 1460m, 1350m, ll25s, 1070m c--l; lH n.m.r. (CDCI3), ò 1.35 (6 H, m, aliphatic -CHz-), 2.50 (12 H, m, -NCH2-), 3.40 ppm. (12 H, n, -OCHz-). 5.2.2 Preparation of C22Cz The ligand 4,7 ,r3, I 6-tetraoxa- r ,10-diazabicy cto[8. s.2] eicosane, C22Cz, was prepared according to a method described by Dale and co- workers t611. One solution of 1,4,10,13-tetraoxa-7,16- diazacyclooctadecane (Kryptofix 22, Merck) (2.98 g, Il.2 mmol) in acetonitrile ( 100 .rn3), and another solution of etyleneglycol ditosylate t1651 (4.r7 E, 11.3 mmol) in acetonirrile (100 cm3) were added synchronously over 32 h using Perfusor motor-driven syringes, to a stirred suspension of dry, finely powdered Nazco¡ (5 g) in refluxing acetonitrile (100 .rn3). Stirring and refluxing continued for 7 days, the solid salts were removed by centrifugation and washed with acetonitrile (2 x 100 cm3) and the combined solvents removed under vacuum. The residue was taken up in CHCI¡ (50 cm3) and water (5 cm3), and the aqueous phase further extracted with CHCI3 (3 x 50 "*3). After evaporation of CHCI3, the residue was Atnira Abou-Hamtlan 113 refluxed with NaOCH3 / CH3OH to destroy any unreacted tosyl functions, and the methanol was evaporated. The residue was taken up in water (50 cm3) and the aqueous solution repeatedly extracted with small portions of CHCI3 (3 x 10 cm3). The aqueous layer was then concentrated and further extracted with larger portions (4 x 10 .*3) of CHC13 yielded yielded the sodium tosylate complex of C22C2 which on pyrolysis in Kugelrhor at 473 K/0.01 mmHg gave the free ligand C22Cz (2.03 g, 63%), m.p. 333-338 K. 1H nmr (CD3OD) õ 2.68 (8H, r, NCÉlzCHzO),2.93 (4H, s, NCHzCHzN), 3.64 (8H, s, OCHzCHzO), 3.70 (8H, r, NCHzCË1zO); in CDCI3: ò 2.60, 2.J5,3.65,3.70. l3C nmr (CD:OD) ò 52.8 (NCH2CH2N), 57.6 (NCÉ12CH2O), 71.6, 71.7 (CHzO). Ethylene glycol ditosylate was prepared according to the literature t 1651 and was dried under vacuum and stored under dry nitrogen prior to use. 5.3 Stability Constant Measurements The stability constants of Ag* cryptates were measured by direct potentiometric titration. Cell potentials were measured using an Orion Research 54720. In a typical rirrarion, 10-3 mol dm-3 AgNo3 or AgCloa (in the case of methanol where insoluble precipitations formed when using AgNO¡) solution (20 cm3) was ritrared with lO-2 mol dm-3 ligand (Czr, CZlCs or c22Cz) solurion (approximarely 5 cm3) and the Ag* concentration was monitored with a silver wire electrode. The reference electrode was a Ag wire in l0-2 mol dm-3 AgNo3 solution (AgClO+ in the case of methanol). All solutions were 0.05 mol dm-3 in TEAP to keep the Ag* activity coefficient constant. The sample cell was connected to the reference cell by a salt bridge containing the Arnira Abou-Hamrlan tr4 TEAP background electrolyte solution. The titrations were carried out with a steady stream of dry nitrogen bubbled through the solution preventing any ingress of air or moisture, this also served to stir it. The sample solution was contained in a Princeton Applied Research polarographic cell with water jacket for thermostatting at 298.2 + 0.01 K. To avoid the possibility of systematic errors resulting from incorrect potential measurements, the electrode response was calibrated in each solvent by measuring the potential of known Ag* concentration solutions (usually by adding l0-2 mol dm-3 AgNO3 solution to 20 cm3 TEAP solution). Plots of porenrial versus log [Ag*] were linear with slopes varying from 55 to 65 (emf in mV). Readings were taken (up to 5 min equilibration time) when no further change in emf was observed. The direct potentiometric titrations of NaClO4 solutions with Czl were carried out in a similar manner to the Ag* titrations described above, with the free Na* concentration monitored with a Radiometer G502 sodium ion selectrode which was also calibrated in solutions of known Na* concentration. Typically, l5 points after the equivalence point where the error in Ks is smallest, were used in determining the stability constants t1661. The indirect determination l92l of Li* cryptate stability constants using competitive titration between Ag* and Li* cryptate, was carried out by titration of 10-3 mol dm-3 AgNO3 (20 cm3; with a solurion I0-2 mol dm-3 in both cryptand and LiCIO+, monitoring rhe [Ag.] with a Ag wire electrode. The stability constants were calculated as described in chapter 3. Amira Abou-Hamdan 115 5.4 Preparation of NMR Samples Samples of perchlorate salts and cryptands for study were prepared as follows. A stock solution of the anhydrous salt was prepared by weight in a volumetric flask and made up with the solvent of interest. The cryptand was weighed into a 1 or 2 cm3 volumetric flask and then made up with the stock solution according to the required ratios of free to coordinated metal ion. About 0.7 cm3 of each solution was degassed and sealed under vacuum in 5 mm o.d. nmr tubes (507-PP, \ù/ilmad Glass Co., modified by placing a glass joint at the open end to allow sealing under vacuum). For variable temperature 1Li and 23Na nmr studies, these tubes were coaxially mounted in 10-mm nmr tubes (513-PP, Wilmad Glass Co.) containing either Dzo, d6-acetone or d6-dimethyl sulfoxide (depending on the temperature under investigation), which provided the deuterium lock signal. For studies of the variation of 1Li chemical shifts with the nature of the solvent the sealed 5-mm nmr tubes were coaxially mounted in 10-mm nmr tubes containing 0.005 mol dm-3 LiCIO+ in ll9 vlv DzOtHzO solvent, and the 1fi resonance of this solution was used as an external reference. All preparations and transfers were carried out in a dry nitrogen flushed glove-box. 5.5 Instrumentation l3C nmr measurements were recorded on a Bruker CXP-300 nmr spectrometer operating at 15.47 MHz. An average of 50 000 Amira Abou-Hamdan 116 transients were accumulated into a 8192-point data base over a 20 000 Hz spectral width. 1Li nmr spectra were recorded on the CXP-300 nmr spectrometer operating at 116.64 M}lz or on a modified Bruker HX-90E nmr spectrometer at 34.977 MHz (in the case of ILi.CzlC5l* in MeOH). An average of 1000 transients were accumulated in a 2048-point data base over a¡ 8000 Hz spectral width for each solution prior to Fourier transformation. 23Na nmr spectra were .run on the CXP-300 spectrometer operating at 19.39 MHz. For each temperature studied, an average of 6000 transients were accumulated in a 2048-point data base over an 8000 Hz spectral width prior to Fourier transformation. In the variable temperature studies, data were collected at temperature intervals of ca. 5 K. Sample temperature was controlled to within * 0.3 K with a Bruker B-VT 1000 variable temperature unit calibrated with a copper-constantan thermocouple. The temperature unit was checked using the temperature dependence of the l¡¡ resonances of methanol and ethylene glycol [167-169]. Data obtained on the CXP-300 was transferred from the Aspect 2000 computer to the Nicolet BNC- l2 computer of the HX-90E using rhe Bruker program, SPECNET, for lineshape analysis t1301 using LINSHP [143]. In the case of the C22Cz systems, data obtained on the CXP-300 was transferred to a VAX I I -780 mainframe computer and complete lineshape analysis was performed using a program similar to LINSHP u44l Bulk diamagnetic susceptibility corrections of the 7f¡ chemical shifts were made using the expression: òcorr : òobs - 4nl3(X¡ef - Ð Atnira Abou.-Hamdan I17 where òcorr and òsþs are the corrected and observed chemical shifts, and X¡ef and X are the volume diamagnetic susceptibilities of the reference and sample solutions respectively t1701. A Johnson Matthey magnetic susceptibility balance was used to determine X¡ef and X. Amira Abou-Hantdan 118 List of Publications I Amira Abou-Hamdan, Trevor W. Hambley, Andrea M. Hounslow, and Stephen F. Lincoln, "A Structural Study of the Complexation of thq Lithium Ion by the Cryptand 4,7,13-Trioxa-1,10- diazabicyclo[8.5.5]eicosane (C2lC)", J.'Chem. Soc., Dalton Trens., t981, 489-492. 2. Amira Abou-Hamdan, Ian M. Brereton, Andrea M. Hounslow, Stephen F. Lincoln, and Thomas M. Spotswood, "An Equilibrium and Kinetic Study of the Complexation of Lithium and Sodium Ions by the C.yptand 4,7,13-Trioxa-1,10-diaza- bicyclo[8.5.5]eicosane (C21C5)", J. Inclusion Phenom., 1987, 5, 137 -141. 3. Amira Abou-Hamdan, Stephen F. Lincoln, Michael R. Snow, and Edward R. T. Tiekink, "The Crystal and Molecular Structure of 4,7 ,13 -Trioxa- I , l0-dia zacy clopentadecanepotassium(I) Thiocyanate", Aust. J. Chem., 1988, 41, 1363-1367. 4 Philip Clarke, Amira Abou-Hamdan, Andrea M. Hounslow, and Stephen F. Lincoln, "Kinetic and Equilibrium Studies of the Sodium(I) Cryptates [Na.C211]* and [Na.C21Cs]*, and rhe Sodium(I) Diaza Crown Ether Complex [Na.C21]* in Non-aqueous Solution", Inorg. Chím. Acta, 1988, 154, 83-87. Atnira Abou-Hamdan rt9 5 stephen F. Lincoln and Amira Abou-Hamdan, "complexation of Lithium Ion by the Cryptand 4,7 ,|3-Trioxa- 1,1O-diaza- bicyclo[8.5.5]eicosane (C21Cs) in a Range of Solvenrs. A 7Li Nuclear Magnetic Resonance and Potentiometric Titration Study", Inorg. Chem., accepted for publication. 6. Amira Abou-Hamdan and Stephen F. Lincoln, "Complexation of Lithium(I) and Silver (I) by 4,7 ,13,16-Tetraoxa- l,I}-diaza- bicyclo[8.8.2]eicosane in a Range of Solvents. A 1Li Nuclear Magnetic Resonance and Potentiometric Titration Study", J. Phys. Chem., submitted for publication. Amira Abou-Hamdan 120 B ibliography tll C.J. Pedersen, J. Am. Chem. Soc., 1961, 89, 7017. fzl C.J. Pedersen and H.K. Frensdorf, Angew. Chem., Ittt. Ed. Engl., 19J2, ll, 16. t3l C.J. Pedersen, J. Inclusion Phenom., 1988, 6, 337. t4l J.-M. Lehn, J. Inclusion Phenom., 1988, 6,351. tsl D.J. Cram, J. Inclusion Phetlom., 1988, 6,397. t6l D.J. Cram, T. Kaneda, R. Helgeson, and G.M. Lein, J. Am. Chem. Soc., 1919, 101, 6752. [7] C. Moore and B.C. Pressman, Biochem. Biophys. Res. Commun., 1964, 15, 562. t8l B.T. Kilbourn, J.D. Dunitz, L.A. Pioda, and W. Simon, J. Mol. Biol., 1967, 30, 559. tel B. Dietrich, J.-M. Lehn, and J.P. Sauvage, Tetrahedron Lett., 1969,2885. t10l J.-M. Lehn, Struct. Bonding (Berl.irt), 1973, 16, 1. t11l J.-M. Lehn and J.P. Sauvage. J. Am. Chem. Soc., 1975, 97, 6700. f,t2l J.-M. Lehn, Pure Appl. Clrcm., 1979, 51, 979. t13l M.C. Espanol and D.M. de Freitas, Inorg. Chem., 1987, 26, 4356. t 141 B. Tümmler, G. Maas, E. Weber, W. V/ehner, and F. Vögtle, ,L Am. Chem. Soc., 1971, 99, 4683. Amira Abou.-Ham¿lan t2r t15l S. Lindebaum, J.H. Rytting, and L.A. Sternson, "Progress in Macrocyclic Chemistry", Vol. 2, R.M. Izatt and J.J. Christensen (eds.), Wiley, New York, 1979. t16l B.F. Ehrlich and J.M. Diamond, J. Membr. Biol., 1980, 52, t87 . ltTl R. Margalit and G. Eisenman, J. Membr. Biol., 1981, 61,209. t18l A.F. Danil de Namor, M.C. Ritt, M.-J. Schwing-Weill, F. Arnaud-Neu, and D.F.V. Lewis,. ./. Chem. Soc., Chem. Commun., 1990, 116. tl el J.C. Lockhart, "/. Chem. Soc., Faraday Trans. I, 1986, 82, 1161. t20l T.M. Fyles, Can. J. Chem., 1981, 65, 884. 12tl P. Quintela, R.C.S. Reno, and A.E. Kaifer, J. Phys. Chem., 1987, 91, 3582. l22l P. Lardinois, I. Rebena, and B. Hickel, New J. Chem., 1988, 12, 21. t23l H.-J. Buschmann, Inorg. Chim. Acta, 1985, 98, 43. Í241 F. Arnaud-Neu, B. Spiess, and M.J. Schwing-Weill, J. Am. Chem. Soc., 1982, 104, 5641. 12sl I.M. Kolthoff, Anal. Chem., 1979, 51, 1. t26l M. Gubelmanû, A. Harriman, J.-M. Lehn, and J.L. Sessler, ,/. Chem. Soc., Chem. Commun., 1988, 77. t27l B. Alpha, J.-M. Lehn, and G. Mathis, Angew Chem., Int. Ed. Engl., 1987, 26, 266. t28l B. Alpha, V. Balzani, J.-M. Lehn, S. Perathoner, and N. Sabbatini, Angew Chem., Int. Ed. Engl., 1987, 26, 1266. tzel N. Sabbatini, S. Dellonte, A. Bonazzi, M. Ciano, and V. Balzani, Inorg. Chem., 1986, 25, 1738. Arnira Abou-Hamdan 122 t30I R.S. Bann'wart, S.A. Solin, M.G. De Baker, and J.L. Dye, J. hn. Chem. Soc., 1989, 111, 5552. t3 1l G. Blasse, L.H. Brixner, and N. Sabbatini, Chem. Phys. Letters, 1999,159,504. t32l N. Sabbatini, S. Perathoner, G. Lattanzi, S. Dellonte, and V. Balzani, J. Phys. Chem., 1987, 91, 6136. t33l J.-M. Lehn, Acc. Chem. Res., 1978, ll, 49. Í341 R.M. Izatt, J.S. Bradshaw, S.A.' Nielsen, J.D. Lamb, J.J. Christensen, and D. Sen., Chem. Rev., 1985, 85,211. t35ì B. Dietrich, J.-M. Lehn, and J.P. Sauvage, ,I. Cheru. Soc., Chem. Commun., 1973, 15. t36l D. Moras, B. Metz, and R. Weiss, Acta Crystallogr., Sect. B, 19'13, 29, 393. t37l B. Metz, D. Moras, and R. Weiss, J. Am. Chem. Soc., 1977, 99, 6532. t3 8l B. Metz, D. Moras, and R. Weiss, Chem. Commun., 1970, 217. t3el M. Dobler, J.D. Dunitz, and P. Seiler, Acta Crystallogr., Sect. B, 1974, 30, 2741. t40l S.F. Lincoln, A. White, and A.M. Hounslow, ,I. Chem. Soc., Faraday Trans. I, 1981, 83, 2459. t4ll B.G. Cox, J. Garcia-Rosas, and H. Schneider, Notw. J. Chim., 1982, 6, 397. t42l D. Moras and R. Weiss, Áct¿ Crystallogr., Sect. B, 1973,29, 400. t43l F. Mathieu, B. Metz, D. Moras, and R. Weiss, ,/. Am. Chem. Soc., 1978, 100, 4412. Amira Abou-Hamdan r23 Í441 S.F. Lincoln, E. Horn, T.W. Harnbley, M.R. Snow, I.M. Brereton, and T.M. Spotswood, J. Chem- Soc., Dalton Trants., 1986, t075. t45l Program SCHAKAL, E. Keller, Institut fur Anorganische Chemie der Universitat Freiburg, Freiburg, FGR. 146l F. Mathieu and R. Weiss, J. Chem. Soc., Chem. Comrnun., 1913, 816. t41l T. Alfheim, S. Buøen, J. Dale, and K.D. Krautwurst, Acta Chem. Scand., 840, 1986, 40. t48l Y.M. Cahen and A.I. Popov, "/. Solutiott Chem., I9J5, 4, 599. t4el L. Echegoyen, A. Kaifer, H.D. Durst, and G.W. Gokel, "/. Org. Chem., 1984, 49, 688. ts0t E. Schmidt, J.-M. Tremillon, J.-P. Kintzinger, and A.I. popov, J. Am. Chem. Soc., 1983, 105, 7563. t51l S.F. Lincoln, I.M. Brereton, and T.M. Spotswood, J. Chem. Soc., Faraday Trans. I, 1985, 1623. ls2l J.M. Guss, SUSCAD, Data Reduction Program for the CAD4 Diffractometer, University of Sydney, 1976. t53l G. Germain, P. Main, and M.M. Woolfson, Acta Crystallogr., Sect. A, 1971, 27, 368. t54I 'International Tables for Crystallography', Kynoch Press, Bermingham, 1974, Vol. 4, 99. t55l G.M. Sheldrick, SHELX 76, Program for X-ray Crystal Structure Determination, University of Cambridge, England, t97 6. ts6l C.K. Johnson, ORTEP, Thermal Elipsoid Plotting Program, Oak Ridge National Laboratories, Oak Ridge, Tennessee, 1965. Amira Abou-Humdan t24 lsTl PREABS and PROCES-DaIa Reduction Programs for CAD4 Diffractometer, University of Melbourne, 1981. ts8l C.K. Johnson, ORTEP II, Report ORNL-3794, Oak Ridge National Laboratory, Tennessee, U.S.A., 197I. t5el P. Groth, Acta Chetn. Scand., Sect. A, 1985, 39,73. t60l R.D. Shannon, Acta Crystallogr., Sect. A, 1916, 751. ^32, t6 1l T. Alftreim, J. Dale, P. Groth, and K.D. Krautwurst, ../. Chem. Soc., Chem. Commun., 1984, 1502. î621 P. Groth, Acta Chem. Scand., Sect. A, 1985, 39, 68. t63l Y.M. Cahen, J.L. Dye, and A.I. Popov, "/. Phys. Chem., 1975, 79, 1289. t64l V. Gutmann, "Coordination Chemistry in Nonaqueous Solutions", Springer-Verlag, Wien, 1968. t65l R.H. Erlich, E. Roach, A.I. Popov, J. Am. Chem. Soc., 1970, 92, 4989. t66l rW.J. DeWitte and A.I. Popov, ,/. Soln. Chem., 1976, 5, 231. t67) J.-S. Shih and A.I. Popov, Inorg. Chem., 1980, 19, 1689. t68l E. Mei, A.I. Popov, and J.L. Dye, J. Am. Chem. Soc., 19J7, 99, 6532. t6el Y.M. Cahen, J.L. Dye, and A.I. Popov, -/. Phys. Chem., 19J5, 79, 1292. t70l A.I. Popov, Pure Appl. Chem., 1979, 51, 101. tT rl B.G. Cox, J. Garcia-Rosas, and H. Schneider, J. Am. Chem. Soc., 1981,103,1054. t72l B.G. Cox, J. Garcia-Rosas, and H. Schneider, J. Am. Chem. Soc., 1981,103,1394. 173l S.F. Lincoln, I.M. Brereton, and T.M. Spotswood, J. Chem. Soc., Faraday Trans. I, 1986, 82, 1999. Atnira Abou.-Hamdan t25 Í7 4l L.F. Lindoy, "The Chemistry of Macrocyclic Ligand Complexes", Cambridge University Press, 1989. t7 sl H.-J. Buschmann, Inorg. Chim. Acta, 1986, 125,31. Í7 6l G. Anderegg, Helv. Chím. Acta, 1975, 58, 1218. t77l E. Kauffmann, J.-M. Lehn, and J.P. Sauvage, ibid., 1976, 59, 1 099. t78l J.-M. Lehn and J. Simon, ibid., 1977, 60, l4l. Í7el F.J. Tehan, B.L. Barnett, and J.L. Dye, J. Am. Chem. Soc., 19J4, 96, 7203. t80l W. Simon, V/.E. Morf, and P.Ch. Meier, Struct. Bonding (Berlin), 1973, 16, 113. t8 1l L. Pauling, "The Nature of the Chemical Bond", 3rd ed., Cornell University Press, Ithaca, New York, 1960. t82l V/.E. Morf and W. Simon, Helv. Chim. Acta, 1973, 54, 2683. t83l J.-M. Lehn and J.P. Sauvage, Chem. Commun., 197I, 440. t84l S.F. Lincoln, I.M. Brereton, and T.M. Spotswood, J. Am. Chem. Soc., 1986, 108, 8134. t85l B.G. Cox, N.V. Truong, and Schneider, "I. Am. Chem. Soc., 1984, 106, 1213. t86l H.-J. Buschmann, ./. Soln. Chem., 1986, 15, 453. t87l G. Ritzler, F. Peter, and M. Gross, J. Electroanal. Chem., 1983, 146,285. t8 8l G. Boras, C. Bosso, and M.R. Vignon, J. Inclusion Phenom., 1989, 7, 637. tsel E.Mei, L. Liu, J.L. Dye, and A.I. Popov, J. Soln. Chem., 1977,6, 77 t. te0l R.D. Boss and A.I. Popov, Inorg. Chem., 1986,25, 1147. Atnira Abou-Hamdan t26 tell J. Gutknecht, H. Schneider, and J. Stroka, Inorg. Cheru., 1978, t7, 3326. îe2l B.G. Cox, H. Schneider, and J. Stroka, J. Am. Chem. Soc., 1978, 100, 4746. te3l F.A. Cotton and G. Wilkinson, "Advanced Inorganic Chemistry", Interscience, New York, 1980. te4) H.-J. Buschmann, Inorg. Chim. Aeta, 1985, 102, 95. te5l S.F. Lincoln, B.J. Steel, I.M. Br'ereton, T.M. Spotswood, Polyhedron, 1986, 5, 1597. te6l A. Hourdakis and A.I. Popov, J. Soln. Chem., 19J7, 6,299. Íe7l N. Ahmad and M.C. Day, J. Am. Chetn. Soc., 197J,99,941. tesl S.F. Lincoln and A. Stephens, unpublished results. teel M. Shamsipur and A.I. Popov, Inorg. Chim. Acta, 1980, 43, 243. t1001 G. Anderegg, Helv. Chim. Acta, 1981, 64, 1790. tl01l M.F. Lejaille, M.H. Livertoux, and C. Guidon, Bull. Soc. Chim. Fr., 1978, 1373. 11021 F. Arnaud-Neu, B. Spiess, and M. Schwing-Weill, Helv. Chim. Acta, 1977 , 60, 2633. tl03l F. Arnaud-Neu, E. Marques, M.-J. Schwing-Weill, C. Dietrich- Buchecker, J.-P. Sauvage, and J. Weiss, New J. Chem., 1988, 12, 15. t1041 D. Feakins and P.J. Voice, J. Chem. Soc., Faraday Trans. I, 1912, 69, 1390. t1051 H.-J. Buschmann, J. Inclusion Phenom., 1989, 7, 58I. t1061 B.G. Cox, J. Garcia-Rosas, and H. Schneider, Ber. Bunsenges. Phys. Chem., 1982, 86, 293. Amira Abou-Hamdan t27 t1071 B.G. Cox, D. Knop, and H. Schneider, J. Am. Chem. Soc., 19i8, 100, 6002. t1081 B.G. Cox and H. Schneider, J. Atn. Chem. Soc., Chem. Coruntun., 19J7, 99, 2809. tlOel B.G. Cox, N.V. Truong, J. Rzeszotarska, and H. Schneider, -/. Aru. Chem. Soc., 1984, 106, 5965. tl l0l V.M. Loyola, R.G. Wilkins, and R. Pizer, J. Am. Chem. Soc., 1975, 97, 1392. tl r ll B.G. Cox and H. Schneider, ,/. Am. Chem. Soc., 1980, I02, 3628. Írtzl V.M. Loyola, R. Pizer, and R.G. Wilkins, "/. Am. Chem. Soc., 1977, gg, 7195. tl13l B.G. Cox, J. Garcia-Rosas, and H. Schneider, ,/. Phys. Chem., 1980, 94, 3179. tl l4l J.-M. Lehn, Neurosci. Res. Prog. Bull., 19'76, 14, 133. tll5l K. Henco, B. Tümmler, and G. Maas, Angew. Chem., Int. Ed. Engl., 1977, 16, 538. tlr6l H. Tsukube, K. Takagi, T. Higashiyama, T. Iwachido, and N. Hayama, ./. Chent. Soc., Perkin Trans. I, 1987, 1697. tr l7l A.-M. Albrecht-Gary, C. Dietrich-Bucheker, Z. Saad, and J.p. Sauvage, -/. Am. Chem. Soc., 1988, 110, 1467. tll8l F. Eggers, T. Funck, K.H. Richmann, H. Schneider, E.M. Eyring, and S. Petrucci, -/. Phys. Chem., 1987, 91, 1961. tllel D.P. Cobranchi, G.R. Phillips, D.E. Johnson, R.M. Barton, D.J. Rose, E.M. Eyring, L.J. Rodriguez, and S. Petrucci, J. ptrys. Chem., 1989, 93, 1396. tl20l M. Shamsipur and A.I. Popov, J. Phys. Chem., 1986, 90, 5997. Ænira Abou-Hamdan r28 lr2rl J.M. Ceraso, J.L. Dye, J. Arn. Chem. Soc., 1973, 95, 4432. u22l J.M. Ceraso, P.B. Srnith, J.S. Landers, and J.L. Dye, "I. Phys. Chem., l9ll, 81, 760. u23l E. Schchori, J. Jagur-Grodzinski, and M. Shporer, J. Am. Chem. Soc., 1973, 95, 3842. It24l S.F. Lincoln, A.M. Hounslow, and A.N. Boffa, Inorg. Chem., 1986, 25, 1039. tr2sl B. Lindman and S. Forsen, "NMR.and the Periodic Table", R.K. Harris and B.E. Mann (eds), Academic Press, London, r97 8. tr26l Y.M. Cahen, P.R. Handy, E.T. Roach, and A.I. Popov, J. Phys. Chem., 1915, 79, 80. Írz7l G.E. Maciel, J.K. Hancock, L.F. Lafferty, P.A. Mueller, and W.K. Musker, Inorg. Chent., 1966, 5, 554. t1281 C. Cosy, L. Helm, and A.E. Merbach, Helv. Chim. Acta, 198'1, 70, 1516. tt2el P. Lazlo, Angew. Chem., Int. Ed. Eng., 1978, 17,254. tl30l S.F. Lincoln, Prog. React. Kinetícs, 1977, 9, l. t1311 F. Bloch, Plrys. Rev., 1946, 70, 460. tt32l H.S. Gutowsky, D.M. McCall, and C.P. Slichter, "/. Chem. Phys., 1953, 21, 219. tl33l H.M. McConnell, ./. Chem. Phys., 1958, 28, 430. u34l H.S. Gutowsky, A. Saika, ./. Chem. Phys., 1953, 21, 1688. tl3sl M.T. Rodgers and J.C. Woodbrey, "/. Phys. Chem., 1962, 66, 540. t1361 I.O. Sutherland, Ann. Rep. NMR Spect., 1971, 4,71. lt37l T.C. Farrar and E.D. Becker, "Pulse and Fourier Transform NMR", Academic Press, New York, 197I. Amira Abou-Hamdan r29 tl38l J.I. Kaplan, "I. Chem. Phys., 1972, 57, 5615. t13el J.I. Kaplan, .I. Chem. Phys., 1973, 59, 990. t1401 R.K. Gupta, T.P. Pitner, and R. rù/asylishen, J. Mag. Res., 1974, 13,393. t l41l T. Nakagawa, Bull. Chem. Soc. Jap., 1966, 39, 1006. Ír421 T.H. Siddall, W.E. Stewart, and F.D. Knight, "/. Phys. Chem., 1910, 74, 3590. t1431 E.H. Williams, Ph.D. Dissertatiorì,' University of Adelaide, r980. lr44l FORTRAN 77 program, by A. White and P. Clarke. t1451 W.F.K. Wynne-Jones, H. Eynng, J. Chem. Phys., 1935, 3, 492. t1461 S. Glasstone, K.J. Laidler, and H. Eyring, "Theory of Rate Processes", McGraw-Hill, New York, 1941. Í1471 E. Schchori, J. Jagur-Gordzinski, Z. Ltrz, and H. Shporer, ./. Am. Chem. Soc., l9ll, 93, 7133. t1481 E. Schmidt and A.l. Popov, J. Am. Chem. Soc., 1983, 105, 1873. Ít4el E. Grell, T. Funck, and F. Eggers, "Membranes-A Series of Advances", Vol. 3, G. Eisenman (ed.), Dekker, New York, t97 5. tl50l W. Burgermeister and R. V/inkler-Oswatitsch, Top. Curr. Chem., 1977, 69, 91. tr5ll J.-M. Lehn, J.P. Sauvage, and B. Dietrich, J. Am. Chem. Soc., 19'77, 92, 2916. Ítszl R.K. Harris, "Nuclear Magnetic Resonance Spectroscopy", Pitman, UK, 1983. t1531 A. Delville, H.D.M. Stover, and C. Detellier, J. Am. Chem. Soc., 1985, L07, 4172. Atnira Abou-Hamdan 130 t1541 B.O. Strasser, K. Hallenga, and A.I. Popov, J. Atn. Cheru. Soc., 1985, 107, 7gg. t1s5l P Clarke, A. Abou-Hamdan, A. M. Hounslow, and S.F. Lincoln, Inorg. Chim. Acta, 1988, 154, 83. trs6l L. Fielding and P. Moore, J. Chem. Soc., Chem. Commun., 1988,49. t1571 H.-J. Buschmann, Inorg. Chim. Acla, 1986, I20, 125. tl58l B.G. Cox, I. Schneider, and H. Schneider, Ber. Bunsenges. Phys. Chem., 1980, 84, 470. tl5el D.D. Perrin, V/.L.F. Aramaego, and D.R. Perrin, "purification of Laboratory Chemicals", Znd edn., Pergamon, Oxford, l 980. t1601 A.I. Vogel, "Textbook of Practical Organic Chemistry", Longman, London, 1910. tr 6ll I.M. Brereton, Ph.D. Dissertation, University of Adelaide, 1 985. Í1621 B. Dietrich, J.-M. Lehn, J.P. Sauvage, and J. BLanzat, Tetrahedron, 1973, 29, 1629. t1 631 W.C. Still, M. Kahn, and A. Mitra, .,/. Org. Chem., 1978, 43, 2923. t1641 H.C. Brown, S. Narasimhan, and Y.M. Choi, Synthesis, 1981, 996. t1651 J. Dale and P.O. Kristiansen, Acta Chem. Scand., 1972, 26, 147t. t1661 F.J. Rossotti and H. Rossotti, "Determination of Stability Constants", McGraw-Hill, New York, 1961. u67l D.S. Raiford, C.L. Fisk, and E.D. Becker, Anal. Chem., 1979, 51, 2050. Anira Abou-Hamdan 131 t1681 A.L. Van Geet, Anal. Chem., 1968. 40, 2227. tl6el A.L. Van Geet, Anal. Chem., 1970, 42, 619. t1701 D.H. Live and S.I. Chan, Anal. Chem., 1970, 42,791. Atnira Abou-Hamdan r32 A endix 1 Lineshape Analysis Data Table 1 Lineshape Analysis Data for [Li.C2lCs]* in MeOH, Tube (i) T¿ç¡/K k¿/s-l 277.40 6.5 283.4r 9.0 289.43 12.7 294.t2 l7 .4 297.40 20.7 302.54 25.9 307.68 34.5 3t2.83 40.8 317 .97 55.7 328.30 63 .6 Table 2 Lineshape Analysis Data for [Li.C2lCs]* in MeOH, Tube (ii). T KK /s-l 271.40 6.9 280.89 8.6 283.4r 9.9 289.43 13.1 297.40 20.5 3 00.49 27 .9 307.68 34.9 Amira Abou-Hamdan. r33 3t2.83 45.r 317 .97 56.2 Table 3 Lineshape Analysis Data for lLi.C2lCsl* in DMF, Tube (iii). T¿s1/K k¿ls - I 280. I 1 ls-e 284.7 5 54.5 289.40 70.1 294.04 93.1 304.60 180.2 309.40 224.2 3t4.r9 268.5 3 18.99 328.9 323.78 409.9 328.17 504.9 Table 4 Lineshape Analysis Data for [Li.C2lCs]* in DMF, Tube (iv). T¿ç1/K k¿/s- | 280. l 1 46.5 289.40 76.9 299 .40 129.5 309.40 221.0 3 1 8.99 309.7 328.58 43 8.1 338.17 472.0 Arnira Abou-Hatndan 134 Table 5 Lineshape Analysis Data for lLi.C2lCSl* in DMF, Tube (v). T¿s1/K k¿/ s- | 289.40 81.6 294.04 96.9 299.40 130.0 3 04 .60 182.0 309.40 222.6 314.t9 23t.4 3 18.99 319.8 323 .7 8 392.6 328.58 488.8 Table 6 Lineshape Analysis Data for [Li.C2lCs]* in DMF, Tube (vi) T¿ç1/K k¿ls-I 280.1 1 45.9 284.7 5 56.0 289 .40 77.0 294.05 104.7 299.40 132.9 3 04 .60 t7 5.5 309.40 222.3 3t4.r9 280.8 3 18.99 347.1 323.78 409.3 328.17 486.9 Amira Abou-Hamdan. 135 Table 7 Lineshape Analysis Data for lLi.CZlCsl* in DEF, Tube (vii). T¿ç1/K k¿ls - 1 2t 5 .46 6r.2 218 .25 71.0 280.1 I 73.7 282.89 97.6 284.15 r02.1 287 .54 1 10.6 289.40 t29.8 292.19 15 3.8 294.05 164.3 296.83 t83.2 298.69 203.2 302.68 240.2 3 04 .60 25t.7 Table 8 Lineshape Analysis Data for [Li.C21CS]* in DEF, Tube (viii) T¿s¡/K li¿ls - I 27 5 .46 67.2 218 .25 72.8 280.1 1 75.5 282.89 l0l .0 284.15 r07 .9 287 .54 116 .4 289 .40 13 1.5 292.19 153.8 Amira Abou.-Hamdan 136 294.05 t7 4.0 298.69 206.0 302.68 243.9 3 04 .60 256.4 Table 9 Lineshape Analysis Data for [Li.C2lCs]* in DMA, Tube (ix). T ¿'¿¡/K k¿ls-r 280. I 1 60.3 282.89 75.9 284.7 5 87.5 287 .54 I 06.5 292 . t9 17 0.4 294.05 184.6 296.83 220.1 298.69 250.6 Table l0 Lineshape Analysis Data for [Li.C2lCS]* in DMA, Tube (x). T¿ç1/K li¿ls- I 280.1 r 62.4 282.89 76.8 284.15 87.5 287 .54 114 .4 289.40 125.8 292.19 165.0 294.05 t7 8.2 29 6 .83 20r.6 Atnira Abou-Hamdan t37 298.69 23s.2 Table 11 Lineshape Analysis Data for [Na.CzIl]t in DEF, Tube (i) T2¿1/K k¿l s- I 328.58 21r .9 333.38 313.9 338.17 479.9 342.97 697.7 347 .77 955.4 352.56 1230.6 351 .36 1842.1 362.r6 2568.5 366.95 327 6.3 Table 12 Lineshape Analysis Data for [Na.czlll* in DEF, Tube (ii). Tr"t/K k¿ls-l 328.5 8 220.9 333.38 351.3 338.17 520.3 342.97 728.2 347 .77 1027.7 352.56 1519.7 357 .36 1925.4 362.16 2463.7 366.95 3335.9 Atnira Abou-Hamdan 138 Table 13 Lineshape Analysis Data for [Na.C2lI]* in DEF, Tube (iii) T¿s1/K fr¿l5 - I 328.58 2t4.8 333.38 314.5 338.17 504.7 342.97 699.9 352.56 14t0.2 357 .36 19 85.3 362.16 27 02.7 366.95 3413.5 Table 14 Lineshape Analysis Data for [Li.C22CZl* in MeOH, Tube (i). T¿s1/K k¿ls - I 22r.7 5 9.0 227 .00 t4.r 230.15 19.7 232.25 2I.I 235.40 28.5 239.60 36.9 242.7 5 46.2 248.13 66.9 253.16 8 7.0 Amira Abou.-Hamdan. r39 Table 15 Lineshape Analysis Data for lLi.CZ2Czf* in MoOH, Tube (ii) T¿s¡/K ft¿ls-l 221 .7 5 9.9 224.90 t2.4 227 .00 14.5 230.r5 20.t 232.25 22.1 235.40 27.7 239 .60 3 8.5 242.15 43.5 248.13 60.6 Table 16 Lineshape Analysis Data for [Li.C22Cz]* in MeoH, Tube (iii). T tKk /s- I 22t .7 5 9.1 224.90 12.0 227 .00 14.4 232.25 20.4 237.50 35.5 240.65 39.8 242.7 5 47 .l 245.90 57 .r 248.r3 64.3 25 t .20 7 6.6 Atnira Abou-Hamdan 140 Table 17 Lineshape Analysis Data for [Li.C22CZ]* in DMF, Tube (iv). T¿s1/K k¿ls- I 279.50 134.7 284.7 5 156.4 290.00 r92.5 29 4 .20 222.6 298.10 238.5 302.10 259 .0 3 05 .50 287.6 308.90 328.6 313.40 357 .l Table 18 Lineshape Analysis Data for [L|.CZ2CZ]* in DMF, Tube (v). T lKk /s-l 279.50 142.8 284.7 s t7 2.7 287.90 t92.9 290.00 200.8 29 6 .00 232.t 302.10 263.r 307 .7 0 322.6 313.40 357.2 Arnira Abou-Hamdan I4l Table 19 Lineshape Analysis Data for [Li.C22Cz]* in DMF, Tube (vi). T¿ç1/K k¿/s-1 279.50 135.1 284.7 5 15 8.7 290.00 196.7 296.00 232.6 299.84 256.4 302.r0 263.1 305.49 303.0 307.70 322.6 3tt.l4 348.4 Table 20 Lineshape Analysis Data for [Li.C22Cz]* in DEF, Tube (vii). T¿q1/K k¿ls - I 248.13 91.2 253.16 116.3 256.40 136.6 25 8.50 15 8.4 26t.65 177.6 263 .7 5 194.7 266.90 228.4 271.t0 3 08.6 27 4.25 361.7 279.50 444.4 284.7 5 53t.9 290.00 67 t.t funira Abou-Hatndan 142 295.85 806.5 Table 21 Lineshape Analysis Data for [Li.C22Cz]* in DEF, Tube (viii). T¿ç1/K k¿ls-l 248.13 90.6 25 8 .50 r63.6 263 .7 5 20r.3 268.82 263.5 27 4.25 360.9 279.50 45 0.8 284.7 5 547.6 295 .85 845.3 Table 22 Lineshape Analysis Data for [Li.C22Czl* in DEF, Tube (ix). T KK /s-l 248.13 91.9 253.16 115.1 258.39 156.6 261.78 t7 6.9 263.7 5 t95.4 268.82 246.6 27 4.25 356.2 279.s0 436.7 290.00 644.4 Amira Abou-Hamdan 143 A endix z Supplementary Ctystallographic Data Hydrege¡ atom positional (x103¡ and thermal (x102¡ parameters for lLi.CZtCslNCS. xy 7 u11 H(6A) 833(1 ) 483(t ) 829(:| ) 9(1) H(68) 663(1 ) 458(1 ) 788(1 ) 9(1) H(7A) 816(1) 389(1 ) 900(1 ) 9(1) H(78) 958( 1 r{oo(t ) ) 790(1 ) 9(1) H(9A) 639(i ) 315(o) 831 (1 H(98) ) 9(1) 566(1 ) 372(o) 7BB(1 H(10A) q2B(1 ) 9(1) ) 299rc) 68o(1 ) 9(1) H(108) 585(1 ) 2B6(o) 592(1 ) e(1) H(11A) 519Q) 390(1 ) 526 (7 ) 4(2) H(118) 373(4) 365(3) 559 (7 ) 5(2) H(t2A) 367(1) 298(o) qoo(1 H(128) ) 9(1) 371(11 356(0) 327 U) 9(1) H(13A) 631(1) 285(o) 350(1 ) 9(1) H(138) 5u2( 1 ) 3ozi 6; 203(1 H(15A) ) 9(1) 645(1 ) 391 (0) 85(1) 9(1) H( 1 5B) (0 518(1) 407 ) 20q(1 ) 9(1) H(16A) 667 () r{82(0) 1r{z(t H(168) ) 9(1) BzT() rr50 (0 ) 18q(1) H(1BA) 9(1) 875(1 ) 518(0) 385(1 H(188) ) 9(1) 706(1 ) 546(o) 366(1 H(19A) ('t ) 9(1) 631 ) 519(o) 590(2) 9(1) H(198) Bo2( 1 ) 545(o) 610(2) H(20A) 9(1) 914(1) 365(0) 177 U) 9(l) H(208) Bz5f ) 309(0) 151(1) H(21A) 9(1) 1oq7(1 ) 296(0) 310(1) H(218) 9(1) 887(1 ) 28q(o) 395(1 H(23A) ) 9(1) 1100(1) 299 (o) 587(1 H(238) ) 9(1) 1 099(1 ) 357 rc) 656(1 H(2lrA) ) 9(1) 926(1 ) 307(o) 791(1) 9(1) H(2rrB) 838(1 ) 2B5(o) 648(1 ) 9(t ) Atnira Abou-Hamdan. r44 Thermal parameters (x103) for [Li.C2lCs]NCS ul1 u22 u33 u23 u13 u12 s(1) 10r{(5) 52(3) 158(5) 10(3) -67(4) s(1') 252(23) -7(3) c(2) 35(B) 113(28) 27 (0) 23(9) 12(7 e()1\ 43(13) ) -2(10) N(3) 141 (13) 59(10) 99(12) 45(11 ) -47(10) -46(10) N(3') 87 (20\ Li('r) 226 (21-) 49(9) 72(12) 0(5) 52(1 3) -6(12) -7 Q6) eB(5) 55(6) 78(5) 14(tr) 13(4) -7(4) c(6) 77(7) 93(13) 84(e) c(7 -55(8) e(6) 2(7 ) ) 75rc) 1oB(14) 43(6) -11(7) 13(5) N(B) 50(rr) -14(Z) 56(6) 65ß) 0(5) -13(4) -3( 4) c(9) 66(5) 63(B) 47 6) rr5(6) 11(5) -20( 5) c(10) 5B(6) 5B(B) B1 (7) ql -5(6) 15 (5) -1 1 ( c(11) 42(5) 3B(7) 44(5) c(12) 3(6) 5(4) i/( 5) 42(5) 7e (9) 77 Q) 15(6) 13(5) c(13) -3( 5) 7u(6) 31 (6) 76(7) -4(5) 5l N(14) 3(6) -10( 53(4) 51(6) 32(u) 3(q) 20(3) 16( r{) c(15) 83(6) 67(10) 53(6) e(6) c(16) -6(5) 19( 6) e3(B) 71(11) 62(7 ) 1B(7) 0(17) 20(6) 27 Q) 84('i) 38(5) B1(5) 24(t{) 20(rr) c(18) 83(7) q1 -19(3) (9) e7 ß) -14(8) 2(6) 4(5) c(19) BB(B) rr6(9) 196(15) -64(10) -30(9) 12(6) c(20) 60(5) 80(9) 72(7 ) 8(7) 11(5) -g(5) c(21) 63(6) 68(9) 79(7) o(22) -29Q) 8(5) 20(5) 93(ri) 6e(6) 80(5) e(5) c(23) (6) 15(4) -6(q) 57 82(9) 99(8) 23(7) -1.q(6) 6(5) c(2q ) 54(6) 103(10) 58(6) 36(7) 2(5) 9(6) Amira Abou.-Hamdan r45 I{ydro gen atom p arameters (x 1 0a ; x 1 03 ) for 4,7,L3-Trioxa- 1, 1 O-diaz acyclo- pent adecanepot assium( I) thio cyanate Àtom r u (11) H (11) 184 (13) 68?4 (9) -140 (12) 21 t (3) Hl12) 83 (13) ??00 (e) 7 80 (L2l 27r (31 r{(21) -25't 9 (141 6935 (6) 5s0 (11) 271 (3) H(221 -147s (14) 66s5 (6) 1680 (11) 27I (s) H (31) -3065 (12) 468s (10) 480 (8) 2'11(31 H (32) -3916 (12) 5723 (10) 837 (B) 27r (3't H (41) -39s2 (16) 4728 (81 2202 (9') 27L (31 H(421 -22'18 (16l 5498 (8) 2425 (91 27I (3',r, H (s1) -75? (18) 3177 (10) 30s3 ( 8) 27r (31 H (s2) -2431 (18) 3841 (10) 3442 (81 27L (31 H (61) -Bs (12) 4229 (Ir',) 4s64 (10) 27t (31 H (62) -s76 (12) s134 (11) 3?4 9 (10) 27r (31 H (?1) 358s (19 ) 4189 (10) 42L7 (r2'l 2'11, (3',, H('t2l 2062 (tgl 4939 (10) 4s98 (12) 27t (31 H (81) 2639 (2Ll 6228 (Il',t 4133 (8) 27t (3',) H (82) 4438 (21) 5s45 (11) 4096 (8) 2'lL (3' H (91) 4788 (17) 6?16 (7) 2213 (13) 271 (3) H (92) 3822 (r'1',,t 7023 (7 | 326? (13) 271 (31 H (101) 2s00 (19) ?790 (10) 1859(9) 21L (31 H (102) 1184 (19) 6989 (10) 2357 (e) 2't r (31 { o oH pOÈ¡r q'P ñp8F ll ä' Èg j ('13 u (721 {O tt- oO Atom v u (11) u (221 u (33) u (23) u ) ,fì Fr c¡PrJ i'J :-a (1 tro K 3441(3) -29L (2) 3 603 (2) 62 lL sB (1) se (1) 6 (1) r0 (i) -3 l- à! (2 FlÞ Yä. e 1 34'12 (51 -2143 (3) -1or 9 (3) 9r 12 6s (2) 111 (2 ) -10 (2) 33 r(2 F,t l. -Pcì 3B (3 o 1 -13?1 (11) s69s (6) 646(6) 97 (3 118 (3) ?o (3) -13 (3) 1(3 rtfX'¡¡ (3) (3) (3) (3 -8 (3 RE. | ü o 2 3sse (12) 56e2 (7 | 2803 (?) 88 (3 142 116 -48 -B Pô (3 I 19p.Ë o 3 1313 (r2) 4328 ('t I 33s3 (?) 130 (3 124 (3) 8e (3) 31 (3) 30 (3 33 J ñÂ. uô (3) (3) 4 3 (3 (3 N 1 -160? (13) 4180 (? ) 2100 (8) ?3 (3 8? 139 (3) -2't -1s -;o b-<Ë- (3 (3 N 2 18s7 (14 ) 6664 (e) 101? (9) 2Lr 13 68 (3) 1?9 (3) -12 (3) 100 2r oX (lÉX (3 lo- 3 6017 (14 (9) (9) ?3 (3 e2 (3) 13? (3) 22 (31 s (3 1? N ) -3360 -1213 ¡J F (3 E-: 'o, é+ c 1 198 (13) 6984 (e) 638 (12) 14e (3 s7 (3) 236 (3) 3s (31 L26 46(3 P õ ¡*B (3) (3) (3) (3 108 (3 N¿+ 2 -14ss (14 ) ' 6626 (61 896 (11) 19s (3 1e8 113 -8s -?e ÊDO (3 (3 a) lt) -¡-,rtì JR c 3 -2844 (Lzl 527't (Lol e36 (8) s8 (3 186 (3) 70 (3) s (3) 14 20 tj ¡ (3 (3) (3) (3',, s9 (3 (3 t¡ c 4 -2'r 64 (t6l 4944 (8) t9?0(9) 103 9? 23s -32 -36 ùî R (3 (3 E*q ã' c -126e (18) 38s3 (10) 3102 (8) 1?0 (3 119 (3) 104 (3) -3 (3) 72 -2s FOots qs. (12) 4432 (11) 3806 (10) 18? (3 168 (3) 114 (3) 3? (3) 33 (3 64 (3 {Ì c(6 -196 ro'ìH- ôrtf c(7 26s9 (19) 4693 (10) 3s't't (t2I 307 (3 11? (3) 98 (3) -6(3) -83 (3 -2 (3 óo) (21) (11) 378? (8) 20s (3 588 (3) 80 (3) (31 (3 -182 (3 pb ñ.lc'ÉË c(8 3163 56s1 -27 -37 (31 (3 (3 a+ri c(e 3686 (1? ) 6632 (1 | 2600 (13) 106 (3 92 332 (3) -126 (3) -30 ? ßÊ- ìl .Ì< c (10) 2288 (19) ?o?6 (10) 19?3 (9) 4s8 (3 68 (3) 18? (3) -2e (3) 182 (3 -4 6 (3 U!es, (11) (8 (8) 56 (3 s8 (3) ?6 (3) 1r (3) 0 (,3) -8 (3 J (U õrf.o c 4992lr4',) -28?0 ) -1113 . ;1 H -I ÈË lH¡ +uco xr lL ôi ìt }Il -F òr