Shedding Light on the Interactions of Hydrocarbon Ester Substituents upon Formation of Dimeric Titanium(IV) Triscatecholates in DMSO Solution

Article abstract of DOI:10.1002/chem.201904639

The dissociation of hierarchically formed dimeric triple lithium bridged triscatecholate titanium(IV) helicates with hydrocarbyl esters as side groups is systematically investigated in DMSO. Primary alkyl, alkenyl, alkynyl as well as benzyl esters are stu

Full text of DOI:10.1002/chem.201904639

A Journal of
Accepted Article
Title: Shedding light on the interaction of hydrocarbon ester
substituents upon formation of dimeric titanium(IV)
triscatecholates in DMSO solution
Authors: Carel Kwamen, Marcel Schlottmann, David VanCraen,
Elisabeth Isaak, Julia Baums, Li Shen, Ali Massomi, Christoph
Räuber, Benjamin Joseph, Gerhard Raabe, Christian Göb, Iris
Oppel, Rakesh Puttreddy, Jas Ward, Kari Rissanen, Roland
Fröhlich, and Markus Albrecht
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201904639
Link to VoR: http://dx.doi.org/10.1002/chem.201904639
Supported by

10.1002/chem.201904639
Chemistry - A European Journal
FULL PAPER
Shedding light on the interaction of hydrocarbon ester
substituents upon formation of dimeric titanium(IV)
triscatecholates in DMSO solution
A. Carel N. Kwamen,[a] Marcel Schlottmann,[a] David Van Craen,[a] Elisabeth Isaak,[a] Julia Baums,[a] Li
Shen,[a] Ali Massomi,[a] Christoph Räuber,[a] Benjamin P. Joseph,[a] Gerhard Raabe,[a] Christian Göb,[b]
Iris M. Oppel,[b] Rakesh Puttreddy,[c] Jas S. Ward,[c] Kari Rissanen,[c] Roland Fröhlich,[d] and Markus
Albrecht*[a]
Dedicated to Prof. Dr. Carsten Bolm on the occasion of his 60th birthday
Abstract: The dissociation of hierarchically formed dimeric triple- Introduction
lithium bridged triscatecholate titanium(IV) helicates with hydrocarbyl
esters as side groups is systematically investigated in DMSO.
Primary alkyl, alkenyl, alkynyl as well as benzyl esters are studied in
order to minimize steric effects close to the helicate core. The 1H
NMR dimerization constants for the monomer/dimer equilibrium
show some solvent dependent influence of the side chains on the
dimer stability. In the dimer, the ability of the hydrocarbyl ester
groups to aggregate minimizes their contacts with the solvent
molecules. Due to this, most solvophobic alkyl groups show the
highest dimerization tendency followed by alkenyls, alkynyls and
finally benzyls. Furthermore, trends within the different groups of
compounds can be observed. E.g., the dimer is destabilized by
internal double or triple bonds due to  repulsion. A strong
indication for solvent supported London dispersion interaction
between the ester side groups is found by observation of an
even/odd alternation of dimerization constants within the series of n-
alkyls, n--alkenyls or n--alkynyls. This corresponds to the
interaction of the parent hydrocarbons as documented by an
even/odd melting point alternation.
In solids, weak interactions between molecules are most
important for the properties of the respective bulk materials.
They direct the orientation of molecules towards each other and
thus control structures and their properties.1 In biological
systems weak interactions are crucial for the spatial
arrangement of molecular entities and thus the function of e.g.
biopolymers like polynucleotides or proteins.2 Weak non-
covalent contacts between molecules can be easily observed in
the crystal by using X-ray diffraction methods.3 However, in
solution the most dominating interaction a molecule undergoes
is the one with the surrounding solvent.4 Furthermore, although
all weak interactions are based in principle on polarity-effects
there is a huge conceptual difference in the nature of e.g.
hydrogen bonds, -based interactions, dispersion interactions or
solvophobicity/-philicity as well as steric effects.5
[a]
A. C. N. Kwamen, M. Schlottmann, Dr. D. Van Craen, Dr. Elisabeth
Isaak, J. Baums, Li Shen, A. Massomi, Dr. C. Räuber, B. P. Joseph,
Prof. Dr. G. Raabe, Prof. Dr. M. Albrecht
Institut für Organische Chemie
Figure 1. Wilcox molecular torsion balance.
RWTH Aachen University
Landoltweg 1, Aachen 52074, Germany
E-mail: markus.albrecht@oc.rwth-aachen.de
Dr. C. Göb, Prof. Dr. I. M. Oppel
Institut für Anorganische Chemie
RWTH Aachen University
Landoltweg 1, Aachen 52074, Germany
Dr. R. Puttreddy, Dr. J. S. Ward, Prof. Dr. K. Rissanen
University of Jyväskylä,
Department of Chemistry,
P.O. Box 35, Jyväskylä 40014, Finland
Dr. R. Fröhlich
One approach for the study of weak interactions in solution
follows a concept which has been promoted in the 1990s by
Wilcox. He introduced a simple “molecular torsion balance” to
be evaluated by NMR spectroscopy (Fig. 1).6 Since this
pioneering work the “Wilcox system” has been frequently used
in order to study different aspects of weak intramolecular
interactions.7 In addition several other torsion balances
following the same basic principle have been described.8
Molecular torsion balances in general are molecules that can
adopt two different conformational states with only one having
two groups in close proximity for interactions (Fig. 2a).
However, it has been pointed out, that in solution the weak
interactions are dominated/enforced by cohesive solvent
effects. This means that the differences in the energy of the
[b]
[c]
[d]
Organisch-Chemisches Institut
Universität Münster,
Corrensstrasse 40, 48149 Münster, Germany
Supporting information for this article is given via a link at the end of
the document
This article is protected by copyright. All rights reserved.

10.1002/chem.201904639
Chemistry - A European Journal
FULL PAPER
balance’s two states are mainly due to the minimization or
maximization of the contact surface with the solvent.9
The homo- or hetero-dimerization of appropriate species may
serve as a tool to develop alternative kinds of molecular
balances (Fig. 2b/c). If the energetics of the central dimerizing
unit are well understood, it is possible to study the contribution of
side-chain interactions on the dimerization. This has been
utilized by Schneider,10 Cockroft11 or Schreiner12 in order to
evaluate interactions even as weak as dispersion forces in
solution. 13
a strong indication of the influence of London dispersion
interactions as an important driving force to stabilize the dimer. 17
In the present study we aim to use hierarchical helicates for the
estimation of solvophobic/-philic effects of hydrocarbyl
substituents in DMSO in order to be able to determine the
relative strength of hydrocarbyl solvatization depending on small
structural changes. Figure 3 schematically summarizes effects
which are important during the dimerization process.
In principle, the stability of the main core of the hierarchical
helicates depends more on the binding strength of lithium in the
interior of the central complex moiety than compared to the
solvatization of lithium cation. Hereby some main factors play an
important role:
1) The carbonyl moiety attached to the catecholate is highly
influential. Esters are the best donors for lithium coordination
methanol-d4
(Kdim
(methyl ester) = 32.000 L/mol) compared to
methanol-d4
ketones (Kdim
(methyl ketone) = 3.700 L/mol) and the
even weaker aldehyde (Kdimmethanol-d4 = 10 L/mol).16
2) The ability of the solvent to solvate lithium cations is crucial to
destabilize the dimers. Solvents that strongly bind lithium like
DMSO easily remove the lithium from the dimer and lead to a
high amount of monomer in solution (KdimDMSO-d6 (methyl ester) =
175 L/mol)17 while less well coordinating solvents like methanol-
d4 (Kdim (methyl ester) = 32.000 L/mol), acetonitrile-d3 (Kdim
(methyl ester) = 47.000 L/mol) or acetone-d6 (Kdim (methyl ester)
= 65.000 L/mol) populate the dimer.
Figure 2. Different concepts for the development of “molecular balances”.
Since 2005 we study the chemistry of hierarchically formed
helicates14 which are essentially dimeric triple-lithium bridged
bis(titanium(IV) triscatecholates).15 In solution, a monomer-dimer
equilibrium can be observed and its energetics can be
accurately determined by NMR spectroscopy (Scheme 1). 16
Figure 3. Different effects influencing the dimerization behavior to form
hierarchically assembled helicates with hydrocarbyl ester substituted ligands in
solution.
Scheme 1. Hierarchical formation of dimeric triple-lithium bridged titanium(IV)
catecholate helicates and the equilibrium between monomer and dimer
structures observed in solution.
3) The kind of central metal ion controls the charge of the
triscatecholate complexes. In case of titanium(IV) ions the
triscatecholate complex possesses a double negative charge
while in case of gallium(III) the complex is triple negative. Due to
the strong electrostatic attraction of the lithium cations
In a recent study we could show that alkyl ester substituents
have a distinct influence on the dimerization behaviour in DMSO
which in many cases can be explained by steric, electronic and
solvation effects of the side chains. For a special example with
“space filling” isopropyl ester groups it even was possible to get
methanol-d4
gallium(III) forms much more stable dimers (Kdim
(aldehyde) = 200.000 L/mol) in solution compared to titanium(IV)
This article is protected by copyright. All rights reserved.

10.1002/chem.201904639
Chemistry - A European Journal
FULL PAPER
methanol-d4
(Kdim
(aldehyde) = 10 L/mol).16 A study with related 8-
In the present study the focus will be on the influence of primary
hydrocarbyl substituents at ester catecholate ligands on the
dimerization in DMSO solution as described under 5-7 and the
influence of the “external” cation will be briefly discussed as well.
The influences of steric as well as inductive effects are
minimized by choosing primary ester substituents with at least
three carbon atoms in order to focus on the solvent influence. As
shown in Figure 3, solvent interactions become strong with
groups that are not buried in the groove of the helicate. At the
same time steric and inductive effects are minimized in this case.
The focus will be on alkyl-, alkenyl- and alkynyl- as well as
benzyl-substituted catechol esters as ligands (Fig. 4).
oxoquinolinate ligands reveals that negatively charged
monomeric complexes are required in order to form the dimers.
Charge neutral monomers do not form dimers. 18
4) Substituents at the aromatic units of the catechol may act as
electron donors or acceptors and thus alter the donor ability of
the catecholate as well as the carbonyl oxygen atoms towards
lithium cations. 19
5) Substituents at the carbonyl of the ligand may destabilize the
dimer due to their bulkiness. 20 However, one borderline example
has been found in which the dimer is highly favoured by a bulky
ester side chain due to destabilization of the monomer. 21
6) The carbonyl substituents have different inductive donor or
acceptor abilities and thus modulate the electron density at the
carbonyl oxygen. E.g., due to the stronger inductive donor effect
Results and Discussion
the ethyl ester substituted complex shows a higher dimer
DMSO-d6
stability (Kdim
= 830 L/mol) compared to the methyl ester
Computational considerations. It has been mentioned above
that in this study, in order to focus on the solvent influence,
steric as well as inductive effects are minimized by choosing
primary OCH2R ester substituents.
(KdimDMSO-d6 = 175 L/mol).17
7) In the dimer the contact surface of side chains with the
solvent is minimized and thus solvophobic effects will favour the
dimer.17
8) Weak attractive “through space” interactions between the
substituents (e.g. dispersion forces) may be a stabilizing factor
in favour of the dimer. However, such weak interactions in most
cases are hidden by the stronger ones described before. 17
Figure 5. Minimized structures (RHF 6-31G(d)) of Li[Li3(1Hex3Ti)2],
Li[Li3(2Z2Hex3Ti)2], Li[Li3(2E2Hex3Ti)2] and Li[Li3(32Hex3Ti)2] as well as the
computed intramolecular dispersion energies for the dimeric complexes EDisp B
(green) as the interaction between neighboring aromatic catecholates and
ester side chains of the two complex units, for interaction between the ester
substituents of the first and aromatic catecholates of the second complex unit
and the interaction between the aromatic catecholates of the two units EDisp C
(red) and between neighboring ester side chains of the units EDisp A (blue).
Figure 4. Ester catechol derivatives for hierarchically assembled helicates
Li[Li3(L3Ti)2]. The derivatives shown in grey have already been described
earlier.16,17
This article is protected by copyright. All rights reserved.

10.1002/chem.201904639
Chemistry - A European Journal
FULL PAPER
Initially some computational investigations on the selected
listed in Figure 4 have been determined in DMSO-d6 under the
standardized conditions. As a rough measure of the “size” of the
ester substituents the number of carbon atoms in the ester side
chain has been selected. In Figure 6 the obtained dimerization
constants of the monomer dimer equilibria are listed against the
number of the carbon atoms of the ligand ester side groups (see
also Table 1). On the first view, the distribution of data points
seems to be more or less chaotic. However, clear trends are
observed upon a more detailed look. For example, it is found
that the possible maximum of Kdim increases with the “size” of
the side chain.
Figure 6 shows the Kdim of the n-alkane derivatives as orange
line.17 The regions in which Kdim of the branched primary alkyl
(green), n-alkenyl (red), n-alkynyl (blue) and benzyl derivatives
(magenta) are located reveal that the n-alkyls possess the
highest dimerization tendency closely followed by the branched
primary alkyls with decreasing Kdim for n-alkenes > n-alkynes >
benzyl esters.
dimeric
complexes
Li[Li3(1Hex3Ti)2],
Li[Li3(2Z2Hex3Ti)2],
Li[Li3(2E2Hex3Ti)2] and Li[Li3(32Hex3Ti)2] have been performed.22
The starting geometries were built using the crystal structure of
Li[Li3(1Me3Ti)2] as framework and subsequently optimized (RHF
3-21G*). The resulting structures were then further refined
employing a larger basis set (RHF 6-31G*). Three dispersion
interaction energies EDisp A, EDisp B and EDisp C were calculated.23
EDisp
(blue squares in figure 5) measures only the weak
A
interactions between the hydrocarbon substituents of the ester
side chains of the two complex units in the dimer. EDisp B (green
squares in figure 5) includes the ring fragments of the aromatic
catecholates in addition to the hydrocarbon substituents
included in the former calculation. EDisp C (red squares in figure 5)
is defined as the difference between EDisp A and EDisp B and thus
measures the weak interactions between the ring fragments of
one complex unit and the ring fragments as well as the side
chains of the other. It was found that in the gas phase as
expected no significant dispersion interactions occur between
the hydrocarbon side chains of the ester substituents in case of
primary esters. However, strong London dispersion interactions
between the substituents and aromatic catecholate units
significantly contribute to the stability of the dimer (Fig. 5).
Experimental studies. The catechol ester ligands as well as
their complexes Li[Li3(L3Ti)2] were prepared as described
before.16,17,24 Monomers as well as dimers are observed by
proton NMR spectroscopy in DMSO-d6 at a concentration of
2•10-3 mol L-1 and the corresponding dimerization constants are
obtained by integration. For the determination of reliable data it
is essential that the quality of the DMSO-d6 used in all studies is
the same. The water content of the used samples was
determined to be 0.12 ± 0.04 mol L-1.17 In this context it also
would be of interest to study heteroleptic complexes to evaluate
interactions between different kinds of substituents. However,
mixed ligand complexes only can be prepared as mixtures with
statistical composition, so that no reliable data can be obtained
on inter-substituent interactions.16d Attempts to specifically
obtain heterodimers by mixing of “complementary” titanium(IV)
triscatecholates were not successful yet.
Figure 6. Distribution of the dimerization constants of primary hydrocarbyl
substituted ester complexes Li[Li3(L3Ti)2] in DMSO-d6 depending on the
number of C-atoms of the substituent with branched alkyl (green), n-alkenyl
(red), n-alkynyl (blue) and benzyl derivatives (magenta). The already reported
data of the n-alkanes are used as standard and are shown as orange line.17
The role of the “external” cation. Prior to a systematic study of
the substituent effects, the influence of the “external cation” on
the dimerization constant has been determined.
Complexes Li[Li3(13Ti)2] with alkyl substituted esters. The n-
alkyl ester derivatives of Li[Li3(13Ti)2] and some crystal
structures thereof have been described before.17 The trend of
the dimer stability was explained by an electronic inductive effect
increasing from the methyl Li[Li3(1Me3Ti)2] to ethyl Li[Li3(1Et3Ti)2]
and propyl ester Li[Li3(1Pr3Ti)2]. With longer chain length this
effect can be neglected. However, with longer alkyls the chain
reaches out into the solvent leading to strong solvophobic
effects favouring aggregation with a maximum of Kdim for the
heptyl ester Li[Li3(1Hept3Ti)2]. With even longer chain length Kdim
drops again due to an unfavourable entropic contribution.17
Following this, branched primary alkyl esters have been
investigated (Fig. 7). Hereby, the isobutyl substituted
Li[Li3(1iBu3Ti)2] (Kdim = 1050 L/mol) behaves very similar in its
stability in DMSO-d6 to the corresponding n-propyl Li[Li3(1Pr3Ti)2]
(Kdim = 1100 L/mol)17 and close to the n-butyl derivative
Li[Li3(1Bu3Ti)2] (Kdim = 1200 L/mol)17. The dimerization trend
Therefore a series of complex salts M[Li3(1Me3Ti)2] (M = Li, Na, K,
Rb, Cs) with the methyl ester catecholates was prepared and
the dimerization constants have been measured in DMSO-d6 to
be Kdim = 175 L/mol (M = Li), 40 L/mol (M = Na, K, Cs) and 50
L/mol (M = Rb). All dimerization constants with a cation M
different to lithium are very similar but lower than the one
obtained for the all-lithium salt. Thus, the fourth cation does not
have a direct influence on the dimerization process. In case of
an external lithium cation, the effective concentration of lithium is
higher and the formation of the dimer is favored.
The influence of the ester substituents. All complexes
Li[Li3(L3Ti)2] with ligands 1-4 shown in Figure 4 were prepared
and spectroscopically characterized.
Thermodynamic
investigations
and
structural
considerations. The dimerization constants of all compounds
This article is protected by copyright. All rights reserved.

10.1002/chem.201904639
Chemistry - A European Journal
FULL PAPER
increases switching to the neopentyl Li[Li3(1Neo3Ti)2] (Kdim = 1380
series of cycloalkyl methyl derivatives is observed for the
cyclohexyl methyl complex Li[Li3(1CH2cyHex3Ti)2] (Kdim = 1380
L/mol). The same decrease of Kdim has already been observed
for the analogous secondary ester of cyclohexanol within the
series of cycloalkanol esters.17 This is attributed to the special
conformational properties of the cyclohexyl group related to ring
inversion.
L/mol) and then to the 2-ethylbutyl ester Li[Li3(12EtBu3Ti)2] (Kdim
=
1530 L/mol). The increase of the dimerization constants is
attributed to a growing contact surface with the solvent resulting
in a stronger influence of the solvophobicity of alkyl groups
towards DMSO enforcing alkyl aggregation.
Table 1. Dimerization constants Kdim [L/mol] of the complexes Li[Li3(L3Ti)2] in
DMSO-d6 at room temperature.
L =
Kdim
L =
Kdim
[L/mol]
[L/mol]
1Me
175±20 a
1095±135 a
1920±240 a
2690±345 a
2410±300 a
810±90 a
1380±170
830±100
2040±260
2600±330
40±5
1Et
830±100 a
1195±145 a
1530±190 a
2330±300 a
1080±125 a
1050±130
1530±190
1310±160
1380±170
4960±650
540±65
1Pr
1Bu
1Pent
1Hex
1Hept
1Oct
1Non
1Dec
1Undec
1iBu
1Neo
12EtBu
1CH2cyBu
1CH2cyHex
1CH2cyOct
23Bu
1CH2cyPr
1CH2cyPent
1CH2cyHept
2All
2Z2Pent
24Pent
2E2Hex
2E3Hex
2E4Hex
2E2,4Hex
27Oct
700±80
930±110
710±85
1450±180
1530±190
480±40b
1290±160
1380±170
10±1
175±20
230±25
20±2
455±50
580±65
640±75
90±8
2E2Pent
2Z2Hex
2Z3Hex
2Z4Hex
25Hex
26Hept
28Non
210Undec
32Bu
330±35
970±115
1610±200
1700±215
1810±225
370±40
1200±145
2020±255
20±2
30±3
420±50
360±40
950±115
750±90
Figure 7. Dimerization constants of primary branched alkyl derivatives of
Li[Li3(L3Ti)2] in comparison to the corresponding n-alkyl esters (dotted blue
line).
29Dec
3Prop
Complexes Li[Li3(23Ti)2] with alkenyl substituted esters. A
series of alkenyl derivatives Li[Li3(23Ti)2] has been prepared and
characterized. It was possible to obtain a crystal structure of
Na[Li3(2All3Ti)2] (Fig. 8).
33Bu
32Pent
34Pent
33Hex
36Hept
38Non
310Undec
42Me
33Pent
32Hex
35Hex
37Oct
The core of the dimer is similar as described before for
analogous hierarchically formed helicates.16-21 It is noticed that
there is no direct contact between the allyl groups (closest
distance: 5.3 Å). The methylene units are still buried in the
groove of the helicate showing short distances to the
neighboring catecholates (CH…Carom = 2.9-3.3 Å), while the
terminal alkene slightly sticks out of the vicinity of the dinuclear
coordination compound. However, close distances between the
internal sp2 hybridized Calkene and Ccatechol atoms as low as 3.5 Å
are observed and should result in some  repulsion. 25
39Dec
4Bn
1060±130
20±2
170±20
140±15
390±45
43Me
270±30
50±4
80±8
90±9
110±10
44Me
42,6Me
42,5Me
42,3,5,6Me
4Me5
42,4Me
43,5Me
42,4,6Me
44iPr
20±2
130±15
[a] Ref 17; [b] Ref 16d
A similar trend is also observed with the cycloalkyl methyl ester
substituted complexes. Kdim raises dramatically from the
cyclopropyl methyl Li[Li3(1CH2cyPr3Ti)2] (Kdim = 830 L/mol) to the
cyclooctyl methyl derivative Li[Li3(1CH2cyOct3Ti)2] (Kdim = 4960
L/mol). The extraordinarily high dimer stability of the latter is
attributed to the large “volume” of the cyclooctyl group opening
up the possibility of strong substituent-substituent interactions,
thus dramatically reducing the contact with solvent molecules
upon dimerization. Kdim of Li[Li3(1CH2cyBu3Ti)2] (Kdim = 1310 L/mol)
is in the same region as the one observed for the neopentyl
ester Li[Li3(1Neo3Ti)2] indicating a similar “size and shape” of the
two different substituents. In addition a drop of Kdim within the
This article is protected by copyright. All rights reserved.

10.1002/chem.201904639
Chemistry - A European Journal
FULL PAPER
The following trends are observed within the series of alkenyl
esters Li[Li3(23Ti)2]:
(i) The red squares in Figure 9 represent the dimerization
constants of the n-alkenyl substituted complexes Li[Li3(23Ti)2]
with terminal double bonds. The most simple, the allyl derivative
Li[Li3(2All3Ti)2] shows a very low Kdim = 40 L/mol (for comparison
Li[Li3(1Pr3Ti)2] Kdim = 1095 L/mol). This can not only be attributed
to the lower solvophobicity of the allyl compared to the propyl
group. It rather is expected that there is a strong repulsion
between the alkene unit and the neighbouring catechol of the
second complex moiety ( repulsion, 27 see also discussion of
crystal structure). The trend of relatively low dimerization
constants compared to the alkanes is also observed for the
other complexes with 2-alkenyl substituents (Li[Li3(2E2Pent3Ti)2]:
Kdim
= = 700 L/mol,
330 L/mol, Li[Li3(2Z2Pent3Ti)2]: Kdim
Li[Li3(2E2Hex3Ti)2]: Kdim = 710 L/mol, and Li[Li3(2Z2Hex3Ti)2]: Kdim
=
970 L/mol).
Figure 8. Structure of the anion [Li3(2All3Ti)2]- as found in the crystal. Top: “side
view” (orthogonal to the Ti-Ti axis), bottom: “top view” down the Ti-Ti axis, left:
ball and stick representation, right CPK representation. Yellow: Ti, blue: Li,
grey: C, white: H, red: O, the allyl groups are shown in purple.
(ii) Starting from the allyl derivative, the dimerization constants of
the helicates with terminal double bonds increase with the chain
length up to Kdim = 1810 L/mol for Li[Li3(25Hex3Ti)2]. Similar trends
are observed for shifting of the double bond “outwards” from 2-
pentenyl Li[Li3(2Z2Pent3Ti)2] and Li[Li3(2E2Pent3Ti)2] to 3-hexenyl
esters Li[Li3(2Z3Hex3Ti)2] and Li[Li3(2E3Hex3Ti)2] (Fig. 8).
The dimerization constants of the alkene complexes Li[Li3(23Ti)2]
are presented in Table 1/Figure 9.
In general alkanes are nearly insoluble in DMSO (e.g. n-
pentane: 3.5 g L-1 at 20-30 °C) while the corresponding alkenes
show a higher solubility (e.g. pentenes: 71 g L-1 at 20-30 °C).26
Li[Li3(26Hept3Ti)2] (Kdim = 370 L/mol) shows a dramatic decrease
of the dimerization tendency which rises again for
Li[Li3(27Oct3Ti)2] (Kdim
= 1290 L/mol). For the 8-nonene
Li[Li3(28Non3Ti)2] (Kdim = 1200 L/mol) as well as the 9-decene
complex Li[Li3(29Dec3Ti)2] (Kdim = 1380 L/mol) the constants are at
the same level while they rise to Kdim = 2020 L/mol for
Li[Li3(210Undec3Ti)2].
This already gives
a
rough indication of the relative
solvophobicity of alkanes versus alkenes in DMSO.
The difference can be observed in comparing the stability of
dimeric n-alkene substituted complexes Li[Li3(23Ti)2] with the
corresponding n-alkanes Li[Li3(13Ti)2]. As a trend the alkenyl
ester helicates possess lower dimerization constants as the
corresponding alkanyl groups in DMSO-d6. Due to the lower
solvophobicity of the alkenyl groups, the dimer is less stabilized
than the monomer or the other way round the monomer is better
solvated in case of alkenyls than in case of alkyls.
The high dimerization constant of Li[Li3(210Undec3Ti)2] may be
explained considering an enthalpy/entropy compensation.17 With
short alkane chains the dimerization constants rise due to their
solvophobicity. At the same time the repulsion between the
double bonds and the catecholates is reduced with every
methylene unit added in the substituent. With long chain length
(nonene, decene) the solvophobicity results in chain aggregation
and entropically originated reduction of Kdim. In the undecenyl
complex Li[Li3(210Undec3Ti)2] the double bond is far away from the
central helicate and can interact well with the solvent resulting in
an increase of the dimerization constant due to a lower influence
of the unfavourable entropic contribution at the terminus of the
ester chain.
The heptenyl derivative Li[Li3(26Hept3Ti)2] is exceptional in the
series of -n-alkenes. However, a similar drop of Kdim has been
observed with n-hexyl in the series of n-alkanes.
(iii) In case of internal double bonds, it is found that the E-
configurated derivatives show a lower Kdim as observed for the
Z-isomers (e.g. Li[Li3(2E2Pent3Ti)2]: Kdim
Li[Li3(2Z2Pent3Ti)2]: Kdim = 700 L/mol).
=
330 L/mol L,
Complexes Li[Li3(33Ti)2] with alkynyl substituted esters.
Alkynyl ester substituted complexes Li[Li3(33Ti)2] were prepared
and the crystal structures of [Li3(32Bu3Ti)2]- and [Li3(33Bu3Ti)2]-
have been determined (Fig. 10). Again the central triple lithium
bridged triscatecholate titanium(IV) complex units possess the
same structural features as observed for many related examples
before.16-21 However, the strict separation of neighbouring
Figure 9. Dimerization constants of n-alkenyl derivatives Li[Li3(23Ti)2] in
comparison to the corresponding n-alkyl esters (dotted blue line).
This article is protected by copyright. All rights reserved.

10.1002/chem.201904639
Chemistry - A European Journal
FULL PAPER
alkynyl substituents in the crystal structure is remarkable. Close
contacts of the internal methylene groups of the side chains with
the catecholate aromatic units force the outer part of the alkynyl
substituents away from each other. Closest H…H contacts are
observed for [Li3(32Bu3Ti)2]-: d(OCH2…H2CO)
=
5.2 and
d(OCH2…H3C) = 3.5 Å and for [Li3(33Bu3Ti)2]-: d(OCH2…H2CO) =
3.6 Å.
As described in the computational considerations, dispersion
effects between the ester substituents and the catechols seem
to stabilize the dimer while corresponding interactions between
the side chains are negligible.
However, the alkyl-aryl interactions are weakened in case of
neighbouring electron clouds at the ester (double or triple bonds).
The proximity of the -systems results in some degree of
repulsion between the  electrons.
Figure 11. Dimerization constants in dmso-d6 of n-alkynyl derivatives
Li[Li3(33Ti)2] in comparison to the corresponding n-alkyl esters (blue dotted
line).
The determination of the dimerization constants of the alkynes
reveals lower constants as observed for the alkenes
corresponding to the lower solvophobicity of alkynes in DMSO
(Fig. 11). For the alkynes with terminal alkyne moieties an
increase of the dimerization constant with the chain length is
observed. As observed for the corresponding alkenes, no drop
of the dimerization constants is found with longer chain length.
The destabilizing entropic contribution to G which is observed
for the n-alkanes17 seems not to be present in case of terminal
“solvophilic” groups.
It is remarkable that all complexes with triple bond in 2-position
Li[Li3(3Prop3Ti)2],
Li[Li3(32Bu3Ti)2],
Li[Li3(32Pent3Ti)2]
and
Li[Li3(32Hex3Ti)2] show about the same dimerization constant (Kdim
= 20-30 L/mol). The same is observed at higher values for the 3-
alkynes Li[Li3(33Bu3Ti)2], Li[Li3(33Pent3Ti)2] (Kdim = 230 L/mol) and
Li[Li3(33Hex3Ti)2] (Kdim = 360 L/mol). This indicates that here the
repulsion between alkynes or between alkyne and catechol
is a dominating interaction for the stability of the dimers.
However, this destabilizing effect decreases with the distance of
the triple bond to the central helicate moiety.
Complexes Li[Li3(43Ti)2] with benzyl substituted esters. The
benzyl ester complexes Li[Li3(43Ti)2] have been prepared and
the
derivatives
Li[Li3(4Bn3Ti)2],
Na[Li3(42Me3Ti)2]
and
Li[Li3(42,4,6Me3Ti)2] were structurally characterized (Fig. 12). In the
crystal of Li[Li3(42,4,6Me3Ti)2] the enantiomeric left () and right
handed forms () are severely disordered (see SI). For
[Li3(4Bn3Ti)2]- it is observed that the benzyl groups seem to avoid
contact to each other, while again close contact is found
between the benzylic position and catechol units of the other
…
complex moiety. Distances CH2 Ccat as low as d = 2.82 Å are
observed. In [Li3(42Me3Ti)2]- all six methylbenzyl units adopt
different orientations showing the that no interacttion occurs
between those groups which would induce some preferential
orientation. For [Li3(42,4,6Me3Ti)2]- all six aromatics are parallel to
the plane of the lithium atoms showing close CHMe…CHMe
contacts of 2.9 Å.
Figure 10. Structure of the anions [Li3(32Bu3Ti)2]- (a) and [Li3(33Bu3Ti)2]- (b) as
found in the crystal. Left: “side view”, right: “top view” down the Ti-Ti axis.
Yellow: Ti, blue: Li, grey: C, white: H, red: O, the allyl groups are shown in
purple.
This article is protected by copyright. All rights reserved.

10.1002/chem.201904639
Chemistry - A European Journal
FULL PAPER
Figure 13. Dimerization constants in dmso-d6 of benzyl derivatives
Li[Li3(43Ti)2].
Conclusions
In here a systematic approach to study solvophobic effects in
DMSO as a weak interaction of the solvent with hydrocarbyl
ester groups of hierarchically formed helicates is described. In
order to minimize steric effects only primary alcohols are
introduced as esters.
One general trend is immediately obvious when investigating the
dimerization constants of the many examples prepared in this
study. Alkyl-esters in general possess higher dimerization
constants in DMSO-d6 than alkenes followed by alkynes and
finally benzyl groups. This strongly correlates with the solubility
Figure 12. Structure of the anions [Li3(4Bn3Ti)2]- (a), [Li3(42Me3Ti)2]- (b) and
[Li3(42,4,6Me3Ti)2]- (c) as found in the crystal. For [Li3(42,4,6Me3Ti)2]- only one of the
disordered enantiomers is shown. Left: “side view”, right: “top view” down the
Ti-Ti axis. Yellow: Ti, blue: Li, grey: C, white: H, red: O, two benzyl groups are
shown in purple.
of
the
respective
compound
classes
in
DMSO
(alkanes<alkenes<alkynes<aromatics, Figure 14).26
Simple aromatic compounds (benzene, mesitylene) are miscible
with DMSO.26 Aromatic units therefore do not show solvophobic
behaviour in this solvent. Consequently the dimerization
constants of the hierarchical helicates Li[Li3(43Ti)2] drop
dramatically showing that the side chains of the monomer are
well solvated and do not tend to aggregate (Fig. 13). Somewhat
surprisingly are the dimerization constants of the complexes with
methyl groups in 3-position of the aromatic units in
Li[Li3(43Me3Ti)2] and Li[Li3(43,5Me3Ti)2] which indicate that here
some weak attractions may occur between neighbouring
substituents.
Figure 14. Domains in which the dimerization constants of alkyl (green),
alkenyl (red), alkynyl (blue) and benzyl ester (magenta) substituted complexes
Li[Li3(L3Ti)2] are observed.
Within the different classes of compounds more subtle trends
can be observed like better dimer stabilization in case of Z-
alkenes compared to E-alkenes. It also is found, that in case of
internal double or triple bonds destabilization occurs which is
This article is protected by copyright. All rights reserved.

10.1002/chem.201904639
Chemistry - A European Journal
FULL PAPER
due to  repulsion between the substituents and between the
substituents and catechol aromatics.
observation of a corresponding even/odd alternating behaviour
of the dimerization constants of the n-hydrocarbyl derivatives
(although in some cases opposite to the melting point behaviour
of the parent compounds) indicates some significant interactions
between the alkyl chains depending on the even versus odd
number of carbon atoms.
Although computational gas-phase models indicate virtually no
interaction between the substituents, in DMSO solution
solvophobicity compresses the hydrocarbyl chains together and
thus stabilizes the dimer. Thus, in the present case we directly
observe solvent supported London dispersion interactions
between neighboring hydrocarbyl groups.
Acknowledgements
We are grateful for funding by the international graduate school
Seleca (Deutsche Forschungsgemeinschaft) and for financial
support from the Academy of Finland (RP: grant no. 298817)
and the University of Jyväskylä.
Keywords: Coordination compounds • Helicate •
Thermodynamics • Weak interactions • Solvent effects
[1]
[2]
(a) G. Desiraju, Angew. Chem., Int. Ed. Engl., 1995, 34, 2311–2327; (b)
G. Desiraju, Angew. Chem., Int. Ed., 2007,46, 8342–8356; (c) A. Otero-
de-la-Roza, E. R. Johnson, J. Contreras-Garcia, Phys. Chem. Chem.
Phys., 2012,14, 12165–12172.
(a) A. R. Fersht, Proc. Natl. Acad. Sci. USA, 1995, 92, 10869-10873;
(b) I.-C. Lin, U. Rothlisberger, Phys. Chem. Chem. Phys., 2008, 10,
2730–2734; (c) R. Paulini, K. Müller, F. Diederich, Angew. Chem. Int.
Ed. 2005, 44, 1788 – 1805.
[3]
[4]
[5]
[6]
W. Massa, Crystal Structure Determination, Springer, 2004.
C. A. Hunter, Angew. Chem. Int. Ed. 2004, 43, 5310-5324.
F. Biedermann, H.-J. Schneider, Chem. Rev. 2016, 116, 5216-5300.
(a) S. Paliwal, S. Geib, C. S. Wilcox, J. Am. Chem. Soc. 1994, 116, 10,
4497-4489; (b) E. Kim, S. Paliwal, C. S. Wilcox, J. Am. Chem. Soc.
1998, 120, 11192-11193.
[7]
(a) H. Gardarsson, W. B. Schweizer, N. Trapp, F. Diederich, Chem. Eur.
J. 2014, 20, 4608-4616; (b) F. Hof, D. M. Scofield, W. B. Schweizer, F.
Diederich, Angew. Chem., Int. Ed. 2004, 43, 5056-5059; (c) B. Bhayana,
M. R. Ams, J. Org. Chem. 2011, 76, 3594-3596; (d) M. R. Ams, M.
Fields, T. Grabnic, B. G. Janesko, M. Zeller, R. Sheridan, A. Shay, J.
Org. Chem., 2015, 80, 7764–7769; (e) K. Nakamura, K. N. Houk, Org.
Lett. 1999, 1, 2049-2051.
Figure 15. Comparison of the progression of the dimerization constants of n-
alkyl (a), -n-alkenyl (b) and -n-alkynyl (c) substituted complexes
Li[Li3(L3Ti)2] (red lines) with the melting points of the corresponding
hydrocarbons (blue lines). The dashed lines connect neighbouring pairs of
even or odd numbered side chains.
[8]
(a) B. Bhayana, D. A. Dickie, Cryst. Growth & Design 2018, 18, 6404-
6410; (b) B. Bhayana, J. Org. Chem. 2013, 78, 6758-6762; (c) W. R.
Carroll, P. Pellechia, K. D. Shimizu, Org. Lett. 2008, 10, 3547-3550; (d)
Y. S. Chong, W. R. Carroll, W. G. Burns, M. D. Smith, K. D. Shimizu,
Chem. Eur. J. 2009, 15, 9117-9126; (e) W. R. Carroll, C. Zhao, M. D.
Smith, P. J. Pellechia, K. D. Shimizu, Org. Lett. 2011, 13, 4320-4323;
(f) J. Hwang, P. Li, M. D. Smith, K. D. Shimizu, Angew. Chem. Int. Ed.
2016, 55, 8086-8089; (g) J. M. Maier, P. Li, E. C. Vik, C. J. Yehl, K. D.
Shimizu, S. M. S. Strickland, J. Am. Chem. Soc. 2017, 139, 6550-6553;
(h) W. B. Jennings, B. M. Farrell, J. F. Malone, Acc. Chem. Res. 2001,
34, 885-894; (i) P. Cornago, R. M. Claramunt, L. Bouissane, J. Elguero,
Tetrahedron 2008, 64, 3667-3673; (j) I. Alkorta, F. Blanco, J. Elguero,
Tetrahedron Lett., 2008, 49, 7246-7249; (k) W. B. Jennings, N.
O`Cannell, J. F. Malone, D. R. Boyd, Org. Biomol. Chem. 2013, 11,
However, some additional weak interactions which are not
obvious seem to be hidden under the ones described above.
The melting points of n-alkanes as well as terminal n-alkenes
and n-alkynes show some even/odd alternating behaviour with
progression of the chain length. This is explained by different
molecular packing (chain-chain interaction) of the compounds in
the 3-D crystal lattice. 28 The shape of the melting point curves of
the alkanes, alkenes or alkynes are different to the ones of the
corresponding catechol ester complexes (Fig. 15). However,
This article is protected by copyright. All rights reserved.

10.1002/chem.201904639
Chemistry - A European Journal
FULL PAPER
5278-5291; (l) S. Yamada, N. Yamamoto, E. Takamori, Org. Lett. 2015,
17, 4862-4865; (m) B. U. Emenike, S. N. Bey, R. A. Spinelle, J. T.
Jones, B. Yoo, M. Zeller, Phys. Chem. Chem. Phys. 2016, 18, 30940-
30945; (o) A. E. Aliev, W. B. Motherwell, Chem. Eur. J. 2019, 25, 10516
– 10530.
Shilov, G. N. Petrova, D. A. Kuznetsov, Eur, J. Inorg. Chem., 2016,
5215-5221.
[16]
(a) M. Albrecht, S. Mirtschin, M. de Groot, I. Janser, J. Runsink, G.
Raabe, M. Kogej, C. A. Schalley, R. Fröhlich, J. Am. Chem. Soc., 2005,
127, 10371-10387; see also: (b) M. Albrecht, M. Baumert, J.
Klankermayer, M. Kogej, C. A. Schalley, R. Fröhlich, Dalton Trans.,
2006, 4395-4400; (c) M. Albrecht, X. Chen, D. Van Craen, Chem. Eur.
J. 2019, 25, 4265-4273; (d) D. Van Craen, M. Albrecht, G. Raabe, F.
Pan, K. Rissanen, Chem. Eur. J., 2016, 22, 3255–3258.
[9]
(a) S. L. Cockroft, C. A. Hunter, Chem. Commun., 2006, 3806-3808; (b)
S. L. Cockroft, C. A. Hunter, Chem. Soc. Rev. 2007, 36, 172-188; (c) L.
Yang, C. Adam, G. S. Nichol, S. L. Cockroft, Nat. Chem. 2013, 5, 1006-
1010.
[10] (a) T. Liu, H.-J. Schneider, Angew. Chem. Int. Ed. 2002, 41, 1368-
1370; (b) H.-J. Schneider, Acc. Chem. Res. 2015, 48, 1815−1822.
[11] L. Yang, J. B. Brazier, T. A. Hubbard, D. M. Rogers, S. L. Cockroft,
Angew. Chem. Int Ed Engl. 2016, 55, 912-916.
[17] D. Van Craen, W. H. Rath, M. Huth, L. Kemp, C. Räuber, J.
Wollschläger, C. Schalley, A. Valkonen, K. Rissanen, M. Albrecht, J.
Am. Chem. Soc. 2017, 139, 16959–16966.
[18] M. Albrecht, M. Fiege, M. Baumert, M. de Groot, R. Fröhlich, L. Russo,
K. Rissanen, Eur. J. Inorg. Chem., 2007, 609-616.
[12] S. Rösel, C. Balestrier, P. R. Schreiner, Chem. Sci. 2017, 8, 405-410.
[13] M. A. Strauss, H. A. Wegner, Eur. J. Org. Chem. 2019, 295-302.
[14] (a) Review: M. Albrecht, Hierarchically assembled titanium(IV) helicates.
in Synergy in Supramolecular Chemistry. Nabeshima, T., CRC Press,
Boca Raton, 2015, chapter 2, 19-32; (b) M. H. W. Lam, S. T. C. Cheung,
K.-M. Fung, W.-T. Wong, Inorg. Chem. 1997, 36, 4618-4619; (c) T.
Konno, K. Takuda, T. Suzuki, K.-I. Okamoto, Bull. Chem. Soc. Jap.
1998, 71, 1049-1054; (d) X. Sun, D. W. Johnson, D. Caulder, R. E.
Powers, K. N. Raymond, E. H. Wong, Angew. Chem. Int. Ed., 1999, 38,
1303-1306; (e) Y. Sakata, S. Hiraoka, M. Shionoya, Chem. Eur. J.,
2010, 16, 3318-3325; (f) S. Hiraoka, Y. Sakata, M. Shionoya, J. Am.
Chem. Soc., 2008, 130, 10058–10059; (f) S. Cardona-Serra, E.
Coronado, P. Gavin, J. Ponce, S. Tatay, Chem. Commun. 2011, 47,
8235–8237; (g) M. Wang, V. Vajpayee, S. Shanmugaraju, Y.-R. Zheng,
Z. Zhao, H. Kim, P. S. Mukherjee, K.-W. Chi, P. J. Stang, Inorg. Chem.
2011, 50, 1506–1512; (h) K. Yoneda, K. Adachi, K. Nishio, M.
Yamasaki, A. Fuyuhiro, M. Katada, S. Kaizaki, S. Kawata, Angew.
Chem. Int. Ed. 2006, 45, 5459–5461; (i) A.-X. Zhu, J.-P. Zhang, Y.-Y.
Lin, X.-M. Chen, Inorg. Chem. 2008, 47, 7389-7395; (j) R. Ishikawa, M.
Nakano, A. Fuyuhiro, T. Takeuchi, S. Kimura, T. Kashiwagi, M.
Hagiwara, K. Kindo, S. Kaizaki, S.Kawata, Chem. Eur. J. 2010, 16,
11139 – 11144; (k) G. N. Newton, T. Onuki, T. Shiga, M. Noguchi, T.
Matsumoto, J. S. Mathieson, M. Nihei, M. Nakano, L. Cronin, H. Oshio,
Angew. Chem. Int. Ed. 2011, 50, 4844 –4848; (l) M. Okamura, M.
Kondo, R. Kuga, Y. Kurashige, T. Yanai, S. Hayami, V. K. K. Praneeth,
M. Yoshida, K. Yoneda, S. Kawata, S. Masaoka, Nature 2016, 530, 465
– 468; (m) T. Nakagawa, H. Danjo, M. Kawahata, K. Yamaguchi,
Tetrahedron 2019, 75, 315-323.
[19] D. Van Craen, M. de Groot, M. Albrecht, F. Pan, K. Rissanen, Z. Anorg.
Allg. Chem., 2015, 641, 2222-2227.
[20]
(a) M. Albrecht, E. Isaak, M. Baumert, V. Gossen, G. Raabe, R.
Fröhlich, Angew. Chem. Int. Ed., 2011, 50, 2850-2853; (b) M. Albrecht,
E. Isaak, H. Shigemitsu, V. Moha, G. Raabe, R. Fröhlich, Dalton Trans.
2014, 43, 14636-14643.
[21] M. Albrecht, E. Isaak, V. Moha, G. Raabe, R. Fröhlich, Chem. Eur. J.
2014, 20, 6650-6658.
[22] K. B. Lipkowitz, T. R. Cundari, D. B. Boyd, Reviews in Compututational
Chemistry 2009, Vol. 26, 43-44.
[23] See for comparison: S. Rösel, H. Quanz, C. Logemann, J. Becker, E.
Mossou, L. C. Delgado, E. Caldeweyher, S. Grimme, P. R. Schreiner, J.
Am. Chem. Soc. 2017, 139, 7428–7431.
[24] C. J. Gramer, K. N. Raymond, Org. Lett. 2001, 3, 2827-2830.
[25] D. Sorsche, M. Schaub, F. W. Heinemann, J. Habermehl, S. Kuhri, D.
Guldi, J. Guthmullere, S. Rau, Dalton Trans., 2016, 45, 12846-12853.
[26] Gaylord Chemical Company, Dimethyl sulfoxide solubility data, 2014,
Bulletin 102.
[27] H. Zhang, X. Jiang, W. Wu, Y. Mo, Phys. Chem. Chem. Phys., 2016, 18,
11821-11828.
[28]
(a) A. Baeyer, Ber. Dtsch. Chem. Ges. 1877, 10, 1286; (b) R. Boese,
H.-C. Weiss, D. Blaser, Angew. Chem. Int. Ed. 1999, 38, 988-992. (c)
For melting points of alkenes and alkynes see: J.-C. Bradley, A.
Williams, A. Lang Jean-Claude Bradley open melting point dataset,
https://figshare.com/articles/Jean_Claude_Bradley_Open_Melting_Poin
t_Datset/1031637.
https://doi.org/10.6084/m9.figshare.1031637.v2,
2014.
[15] (a) B. A. Borgias, S. R. Cooper, Y. B. Koh, K. N. Raymond, Inorg.
Chem. 1984, 23, 1009-1016; (b) T. A. Bazhenova, N. V. Kovaleva, G. V.
This article is protected by copyright. All rights reserved.

10.1002/chem.201904639
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
FULL PAPER
Obvious trends can be found in the
seemingly chaotic distribution of
dimerization constants of
hierarchically assembled helicates in
DMSO revealing the stabilizing effect
of solvent supported London
dispersion effects.
A. C. N. Kwamen, M.Schlottmann, D.
Van Craen, E. Isaak, J. Baums, L. Shen,
A. Massomi, C. Räuber, B. P. Joseph, G.
Raabe, C. Göb, I. M. Oppel, R.
Puttreddy, J. S. Ward, K. Rissanen, R.
Fröhlich, and M. Albrecht*
Page No. – Page No.
Shedding light on the interaction of
hydrocarbon ester substituents upon
formation of dimeric titanium(IV)
triscatecholates in DMSO solution
This article is protected by copyright. All rights reserved.