A Carel N Kwamen1, Judith Jenniches2, Iris M Oppel2, Markus Albrecht1. 1. Institut für Organische Chemie, RWTH Aachen University, Landoltweg 1, Aachen, 52074, Germany. 2. Institut für Anorganische Chemie, RWTH Aachen University, Landoltweg 1, Aachen, 52074, Germany.
Abstract
Hierarchical helicates based on ketone-substituted titanium(IV)triscatecholates show different monomer-dimer behavior depending on different solvents. The dimerization constants of a whole series of differently alkyl-substituted complexes is analyzed to show that the solvent has a very strong influence on the dimerization. Hereby, effects like solvophobicity/philicity, sterics, electronics of the substituents and weak side-chain-side-chain interactions seem to act in concert.
Hierarchical helicates based on ketone-substituted titanium(IV)triscatecholates show different monomer-dimer behavior depending on different solvents. The dimerization constants of a whole series of differently alkyl-substituted complexes is analyzed to show that the solvent has a very strong influence on the dimerization. Hereby, effects like solvophobicity/philicity, sterics, electronics of the substituents and weak side-chain-side-chain interactions seem to act in concert.
Over the last 50 years supramolecular chemistry has evolved to an important independent branch of chemistry combining principles of the traditional disciplines (inorganic, organic, physical chemistry) and connecting those to biochemistry, material science or nanotechnology.1More than 30 years ago Lehn introduced the helicates as coordination compounds in which two or more linear ligand strands wrap around two or more metal ions.2 If the helicating ligands are not covalently linked but contain a non‐covalent connecting point (e.g. a metal ion or a hydrogen bond), helicate type coordination compounds may be formed in a hierarchical process (Scheme 1).3 Several “hierarchical” helicates as well as closely related cluster helicates have been described in the literature.4
Scheme 1
Hierarchical formation of helicate type complexes by incorporation of a metal ion into the spacer of the helicating ligand.
Hierarchical formation of helicate type complexes by incorporation of a metal ion into the spacer of the helicating ligand.In 2005 we described a hierarchical helicate based on 3‐carbonyl‐substituted catecholate ligands forming initially a mononuclear triscatecholate titanium(IV) complex which in the presence of lithium counter cations dimerizes to a triple‐lithium bridged coordination compound. The carbonyl may be an aldehyde, ketone, thioester or ester.5 For several reasons no dimer formation has been observed for amide derivatives yet: with secondary amides an NH⋅⋅⋅Ocatecholate hydrogen bond is blocking the lithium binding site, while tertiary amides are sterically too demanding for dimer formation (Scheme 2).
Scheme 2
The lithium dependent monomer‐dimer equilibrium based on carbonyl‐substituted titanium(IV) triscatecholates.
The lithium dependent monomer‐dimer equilibrium based on carbonyl‐substituted titanium(IV) triscatecholates.The hierarchically formed triscatecholate titanium(IV) helicates are exceptional in comparison to other hierarchical helicates. In the solid, the dimeric helicates are present while in solution the lithium bridged systems slowly reach the thermodynamic equilibrium between monomer and dimer.6The equilibrium ratio between monomer and dimer depends on the strength of lithium binding in the dimer or the ease of lithium removal, respectively. Thus, the kinds of carbonyl donors as well as of the solvents are highly influential on the equilibrium state. In addition, weak side‐chain interactions significantly can contribute to dimer stabilization or destabilization.6To illustrate the solvent dependence: the complexes of 2,3‐dihydroxybenzaldehyde as ligand at room temperature show dimerization constants of K
dim=10 ([D4]MeOH), 950 ([D8]THF) or 1330 ([D6]acetone). In [D6]DMSO or D2O only monomer and in [D3]acetonitrile only dimer is observed.5 Thus depending on the solvent, the whole spectrum from monomer to dimer can be detected by NMR spectroscopy at ambient conditions (Figure 1).
Figure 1
Stability domains in which the lithium dependent monomer dimer equilibrium of carbonyl substituted triscatecholate titanium(IV) complexes can be observed depending on the carbonyl moiety as well as on the solvent. The results of the present study on ketone derivatives are highlighted in the box.
Stability domains in which the lithium dependent monomer dimer equilibrium of carbonyl substituted triscatecholate titanium(IV) complexes can be observed depending on the carbonyl moiety as well as on the solvent. The results of the present study on ketone derivatives are highlighted in the box.Intense recent studies were focusing on the ester or thioester derivatives in [D6]DMSO or [D4]MeOH solution, respectively. Those solvents provide dimer stability windows which make a comparative investigation within the oxo‐6 or thioester5 series depending on different side‐chains possible. However, it would be of major interest to perform related studies which simultaneously allow the evaluation of the influence of the side chain as well as of different solvents.Therefore, our focus now was shifted back to the ketone based catecholate ligands 1‐H2 and found out that they are ideal candidates for the systematic investigation of the dimer stability with variation of the substituents as well as of the solvents. Some of the ketone derivatives were already studied in [D4]MeOH.5a Some new complexes are added in here and significantly different dimerization behavior is observed depending on the solvents [D6]DMSO, [D4]MeOH, [D3]acetonitrile, [D8]THF and [D6]acetone.
Results and Discussion
The required ligands 1‐H2 were prepared by Grignard addition of alkyl Grignard reagents to dimethoxybenzaldehyde 2 followed by Jones oxidation of the alcohols 3 and final ether cleavage of the protecting groups at 4. The obtained ligands 1‐H2 (3 equiv) were coordinated to titanoylbis(acetylacetonate) in the presence of lithium carbonate to obtain the hierarchical helicates Li[Li3(1)6Ti2] which in solution are in equilibrium with the monomeric species Li2[(1)3Ti] (Scheme 3).5
Scheme 3
Preparation of the ligands discussed in this study.
Preparation of the ligands discussed in this study.The catechol ligands bear ketone substituents of the n‐alkane series from methyl to dodecyl (1 a–l‐H2), β‐branched substituents (1 m–o‐H2), secondary substituents (1 p–s‐H2), and substituents with phenyl groups (1 t,u‐H2).7The crystal structure of the ethyl ketone Li[Li3(1 b)6Ti2] has been described earlier.5a In addition, the structure of the more sterically demanding cyclohexyl methyl K[Li3(1 o)6Ti2] and cyclohexyl‐substituted complex Li[Li3(1 s)6Ti2] have been obtained now. Figure 2 a shows the side view of the anion [Li3(1 o)6Ti2]− revealing the connecting bis‐titaniumtris‐lithium center while the top view (Figure 2 b) shows the relative orientation of the cyclohexylmethyl substituents. Hereby the cyclohexyl rings adopt roughly an alternating position with the “plane” of the six‐membered ring orientated parallel or orthogonal to the Ti⋅⋅⋅Ti axis. This allows a close packing with H⋅⋅⋅H distances of 2.29–2.98 Å between neighboring T‐shaped cyclohexyl rings. All cyclohexyl moieties adopt the chair conformation with an equatorial position of the methylene unit.8
Figure 2
Structure of the anion [Li3(1 o)6Ti2]− as observed in the crystal of K[Li3(1 o)6Ti2] (side (a) and top view (b)) and of [Li3(1 s)6Ti2]− (side (c) and top view (d)). Grey: C, white: H, red: O, blue: Li, yellow: Ti, the cyclohexylmethyl substituents are shown in black.
Structure of the anion [Li3(1 o)6Ti2]− as observed in the crystal of K[Li3(1 o)6Ti2] (side (a) and top view (b)) and of [Li3(1 s)6Ti2]− (side (c) and top view (d)). Grey: C, white: H, red: O, blue: Li, yellow: Ti, the cyclohexylmethyl substituents are shown in black.Li[Li3(1 s)6Ti2] is the sterically most crowded hierarchically formed dimer of this kind which has been structurally characterized so far. Due to the limited space around the central core the cyclohexyl planes have to orientate parallel to the Ti⋅⋅⋅Ti axis. In addition, the dimer has to “stretch” resulting in a somewhat longer Ti⋅⋅⋅Ti distance of 5.562(1) Å in [Li3(1 s)6Ti2]− compared to 5.444(1) Å in [Li3(1 o)6Ti2]−.Dimerization constants of the complexes Li[Li3(1 a–u)6Ti2] were determined by proton NMR at 295 K in [D6]DMSO, [D4]MeOH, [D3]acetonitrile, [D8]THF and [D6]acetone (Table 1).
Table 1
Solvent dependent dimerization constants K
dim for the equilibrium between two monomers Li2[(1)3Ti] and one dimer Li[Li3(1)6Ti2] as obtained at 295 K by proton NMR integration at a concentration of 2×10−3 mol L−1.
Ligand (R)
[D6]DMSO
CD3OD
[D3]MeCN
[D8]THF
(D3C)2C=O
1 a (Me)
monomer
3890±505
715±84
1430±177
1240±152
1 b (Et)
monomer
785[a]
3260±420
1960±247
28 590±3913
1 c (Pr)
35±3
1110[a]
7460±990
3170±408
54 150±7979
1 d (Bu)
55±5
1500[a]
5400±710
4000±521
30 780±4219
1 e (Pent)
25±2
1015[a]
5520±727
6215±822
36 980±5082
1 f (Hex)
180±19
965[a]
3120±402
6075±802
16 560±2245
1 g (Hept)
90±9
725[a]
3640±472
3260±420
18 030±2448
1 h (Oct)
115±12
1425[a]
3890±505
4960±651
12 320±1659
1 i (Non)
110±11
1125±137
5515±677
2570±328
19 250±2618
1 j (Dec)
90±9
665[a]
5250±70
5480±721
15 900±2884
1 k (Undec)
85±8
740±88
5825±83
2780±356
12 350±1665
1 l (Dodec)
80±8
1200[a]
5155±677
5270±692
8560±1142
1 m (iBu)
175±18
175±18
600±70
7570±1007
1090±132
1 n (CH2cyBu)
135±14
100±10
700±83
3150±406
2950±378
1 o (CH2cyHex)
95±9
55±5
540±63
2180±276
6690±887
1 p (iPr)
25±2
7±1
1730±216
9720±1301
1040±126
1 q (3‐Pent)
monomer
4±1
505±58
6390±845
2000±252
1 r (cyPent)
160±17
40±4
375±42
7970±1061
4700±615
1 s (cyHex)
monomer
30±3
965±116
6390±845
2560±326
1 t (Ph)
160±17
70±7
130±13
950±114
455±52
1 u (Bz)
50±5
146±15
290±32
675±79
985±119
[a] From reference [5a].
Solvent dependent dimerization constants K
dim for the equilibrium between two monomers Li2[(1)3Ti] and one dimer Li[Li3(1)6Ti2] as obtained at 295 K by proton NMR integration at a concentration of 2×10−3 mol L−1.Ligand (R)[D6]DMSOCD3OD[D3]MeCN[D8]THF(D3C)2C=O1 a (Me)monomer3890±505715±841430±1771240±1521 b (Et)monomer785[a]3260±4201960±24728 590±39131 c (Pr)35±31110[a]7460±9903170±40854 150±79791 d (Bu)55±51500[a]5400±7104000±52130 780±42191 e (Pent)25±21015[a]5520±7276215±82236 980±50821 f (Hex)180±19965[a]3120±4026075±80216 560±22451 g (Hept)90±9725[a]3640±4723260±42018 030±24481 h (Oct)115±121425[a]3890±5054960±65112 320±16591 i (Non)110±111125±1375515±6772570±32819 250±26181 j (Dec)90±9665[a]5250±705480±72115 900±28841 k (Undec)85±8740±885825±832780±35612 350±16651 l (Dodec)80±81200[a]5155±6775270±6928560±11421 m (iBu)175±18175±18600±707570±10071090±1321 n (CH2cyBu)135±14100±10700±833150±4062950±3781 o (CH2cyHex)95±955±5540±632180±2766690±8871 p (iPr)25±27±11730±2169720±13011040±1261 q (3‐Pent)monomer4±1505±586390±8452000±2521 r (cyPent)160±1740±4375±427970±10614700±6151 s (cyHex)monomer30±3965±1166390±8452560±3261 t (Ph)160±1770±7130±13950±114455±521 u (Bz)50±5146±15290±32675±79985±119[a] From reference [5a].Table 1 and Figure 3, and Figure 4 summarize the obtained dimer stabilities. It is obvious that the dimerization constants rise in the order [D6]DMSO<[D4]MeOH<[D3]acetonitrile≤[D8]THF<[D6]acetone as it is roughly summarized in the highlighted box in Figure 1. The observed trend in dimerization constants can neither be correlated with the polarity of the solvent (Reichardt polarity parameters: ET=45.1 (DMSO), 55.4 (MeOH), 45.6 (MeCN), 37.4 (THF) and 42.2 (acetone))9 nor with the ability to dissolve lithium cations.10 However, the dimerization tendency roughly follows the lipophobicity/philicity of the respective solvent. Alkanes are not soluble in DMSO, they show low solubility in methanol and acetonitrile and they are more or less miscible with THF and acetone. Thus, more hydrophilic solvents stabilize the highly charged monomer with two “free” lithium counter cations in which oxygen atoms are exposed to the surface of the complex while lipophilic solvents prefer the less charged dimer with the oxygen atoms buried within the complex.
Figure 3
Dimerization constants K
dim (at 295 K) of the complexes Li[Li3(1 a–u)6Ti2 in [D6]DMSO, [D4]MeOH, [D3]MeCN, [D8]THF and [D6]acetone.
Figure 4
K
dim (at 295 K) of the complexes Li[Li3(1 a–u)6Ti2 as observed in different solvents.
Dimerization constants K
dim (at 295 K) of the complexes Li[Li3(1 a–u)6Ti2 in [D6]DMSO, [D4]MeOH, [D3]MeCN, [D8]THF and [D6]acetone.K
dim (at 295 K) of the complexes Li[Li3(1 a–u)6Ti2 as observed in different solvents.Different trends can be observed for the dimerization constants based on the different solvents (Figure 4): [D: The K
dim values in [D6]DMSO follow trends as observed earlier for the corresponding esters.6 For the methyl‐ as well as ethyl‐ketones only monomers are observed while with gradually increasing numbers of carbon atoms of the n‐alkyl substituent K
dim increases reaching a maximum for the hexyl compound. With longer n‐alkyls K
dim stepwise decreases again. Figure 5 shows a comparison of the dimerization constants of the n‐alkyl‐substituted ketone (blue) and ester derivatives (red) revealing similar shapes of the trend lines (dotted lines).
Figure 5
K
dim of the n‐alkyl (blue) and n‐alkoxy (red) substituted complexes plotted against the chain length (number of C or C+O atoms) of the respective side chain. The dotted lines represent the respective trend lines.
K
dim of the n‐alkyl (blue) and n‐alkoxy (red) substituted complexes plotted against the chain length (number of C or C+O atoms) of the respective side chain. The dotted lines represent the respective trend lines.The branched compounds show in most cases much lower dimerization tendencies than the linear ones with the exception of the isobutyl and cyclopentyl substituted complexes. The complexes with aromatic side chains prefer the formation of the monomer due to the high solvophilicity of aromatics in DMSO. The exceptionally high K
dims of the isobutyl and cyclopentyl ketones may be due to some interactions between the side chains in the dimeric helicates (e.g. dispersion11) in addition to solvophobic effects. The isobutyl complex hereby may be compared to the corresponding isopropyl ester in which stabilizing dispersive interactions have been verified.6[D:5a In deuterated methanol there seems to be virtually no difference in the electronic influence of different substituents. The observed dimerization constants lead to the impression that here only sterics are controlling the dimer stability. The methyl ketone as the sterically least demanding group results in the highest K
dim. The longer n‐alkyl derivatives show very similar dimer stabilities. However, it is reduced in the complexes with sterically more demanding β‐ and even more with α‐branched side chains.[D: In deuterated acetonitrile the dimerization constant gradually increases from the methyl to the n‐propyl ketone and after this reaches a plateau starting with butyl. A drop in K
dim is found for the hexyl to octyl substituted derivatives. The initial increase of K
dim can be attributed to the increasing donor ability of the substituents while later on mainly steric effects are important. This results in low dimerization constants of the complexes with branched side chains.[D: Starting with the methyl ketone the dimerization constant gradually increases until it reaches the pentyl derivative. The hexyl and higher n‐alkyl substituted complexes show a strong even–odd behavior with the even alkyl groups resulting in higher and the odd in lower dimerization constants. This even–odd behavior is an indication for a direct interaction between the alkyl chains.12 The K
dim values of the β‐branched derivatives are related to the dimerization constants of the n‐alkyls while the bulky secondary ketones result in unusually high ones (even higher than the n‐alkyls).[D: The dimerization behavior of the n‐alkanes in [D6]acetone indicates a strong influence of the electron donating alkyl groups from methyl to ethyl to n‐propyl leading to increasing K
dim values. With longer alkyl chains the dimerization constants gradually decrease showing some even/odd alternating behavior. Due to higher steric demands the β‐branched systems possess somewhat lower and the α‐branched very low dimerization constants.Our investigations show that there is a strong solvent dependence of the monomer dimer equilibrium of Li[Li3(1 a–u)6Ti2] based on different effects in different solvents resulting in very different stability patterns of the set of compounds in the investigated solvents.13 [D6]DMSO, [D3]MeCN and [D6]acetone show more or less easy to explain patterns of K
dim: initially K
dim increases with growing chain length while it decreases with longer chains. This may be due to an entropy effect as observed for the corresponding n‐alkyl esters.The solvents [D4]MeOH and [D8]THF behave in an unexpected way: In case of [D4]MeOH only steric effects seem to be influential, leading to lower K
dim values with bulkier side chains. In [D8]THF higher dimer stability is observed with more bulky groups. This observation may be due to a direct attractive interaction between the side chains in this solvent. Bulkier groups are able to have direct contact to each other while less bulky groups have to adopt their conformation appropriately.14 This interpretation is supported by the observation of an even odd behavior of the dimerization constants in case of the long n‐alkyl chains.
Conclusions
The monomer dimer equilibrium of ketocatechol based hierarchical helicates Li2[(1)3Ti]/Li[Li3(1)6Ti2] is an ideal tool to investigate weak interactions of different side chains in different solvents.15 Thus it represents an interesting alternative to Wilcox molecular balance.16 Variation of the solvent leads to dramatically different patterns of the stability constants revealing the influence of effects such as sterics, electronics and side‐chain—side‐chain interactions. Often the concerted influence of all the effects is clear. However, in [D4]MeOH only sterics seem to be responsible for the dimer stability. Some observed dimerization constants are exceptionally high (Li[Li3(1 m,r)6Ti2] in [D6]DMSO, Li[Li3(1 p)6Ti2] in [D3]acetonitrile and Li[Li3(1 m,
‐s)6Ti2] in [D8]THF). In those cases, some attractive side chain‐side chain interactions between the bulky groups seem to become important, which may be based on London dispersions.17
Conflict of interest
The authors declare no conflict of interest.As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.SupplementaryClick here for additional data file.
Authors: Jean-François Ayme; Jonathon E Beves; David A Leigh; Roy T McBurney; Kari Rissanen; David Schultz Journal: Nat Chem Date: 2011-11-06 Impact factor: 24.427
Authors: Graham N Newton; Tatsuya Onuki; Takuya Shiga; Mao Noguchi; Takuto Matsumoto; Jennifer S Mathieson; Masayuki Nihei; Motohiro Nakano; Leroy Cronin; Hiroki Oshio Journal: Angew Chem Int Ed Engl Date: 2011-02-25 Impact factor: 15.336
Authors: Peter R Schreiner; Lesya V Chernish; Pavel A Gunchenko; Evgeniya Yu Tikhonchuk; Heike Hausmann; Michael Serafin; Sabine Schlecht; Jeremy E P Dahl; Robert M K Carlson; Andrey A Fokin Journal: Nature Date: 2011-09-14 Impact factor: 49.962
Authors: Iván Cortés; Estefanía Cordisco; Teodoro S Kaufman; Maximiliano A Sortino; Laura A Svetaz; Andrea B J Bracca Journal: RSC Adv Date: 2021-06-01 Impact factor: 4.036