Thermodynamic studies of dynamic metal ligands with copper(II), cobalt(II), zinc(II) and nickel(II)

Article abstract of DOI:10.1080/00958972.2016.1262949

The self-assembly of unsymmetrical metal ligands from simpler components circumvents complex synthetic approaches. We show herein how a recently developed three-component assembly to create tripicolylamine-like ligands can be extended to the use of azides rather than pyridines. Substituent effects, metal dependence, and reversibility of the assembly are explored. Further, we show that the extent of assembly directly correlates with the differential binding of the templating metals between dipicolylamine-like and tripicolyamine-like ligands.

Full text of DOI:10.1080/00958972.2016.1262949

Journal of Coordination Chemistry
ISSN: 0095-8972 (Print) 1029-0389 (Online) Journal homepage: http://www.tandfonline.com/loi/gcoo20
Thermodynamic studies of dynamic metal ligands
with copper(II), cobalt(II), zinc(II) and nickel(II)
S. Reid Long, Chung-Yon Lin & Eric V. Anslyn
To cite this article: S. Reid Long, Chung-Yon Lin & Eric V. Anslyn (2016): Thermodynamic
studies of dynamic metal ligands with copper(II), cobalt(II), zinc(II) and nickel(II), Journal of
Coordination Chemistry, DOI: 10.1080/00958972.2016.1262949
To link to this article: http://dx.doi.org/10.1080/00958972.2016.1262949
Accepted author version posted online: 20
Nov 2016.
Published online: 04 Dec 2016.
Submit your article to this journal
Article views: 2
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=gcoo20
Download by: [University of Colorado at Boulder Libraries]
Date: 08 December 2016, At: 06:59

Journal of Coordination Chemistry, 2016
http://dx.doi.org/10.1080/00958972.2016.1262949
Thermodynamic studies of dynamic metal ligands with
copper(II), cobalt(II), zinc(II) and nickel(II)
S. Reid Long, Chung-Yon Lin and Eric V. Anslyn
department of Chemistry, the university of texas, austin, tX, usa
ABSTRACT
ARTICLE HISTORY
received 16 may 2016
accepted 6 october 2016
The self-assembly of unsymmetrical metal ligands from simpler
components circumvents complex synthetic approaches. We show
herein how a recently developed three-component assembly to
create tripicolylamine-like ligands can be extended to the use of
azides rather than pyridines. Substituent effects, metal dependence,
and reversibility of the assembly are explored. Further, we show
that the extent of assembly directly correlates with the differential
binding of the templating metals between dipicolylamine-like and
tripicolyamine-like ligands.
KEYWORDS
dynamic chemistry;
Click chemistry; itC;
thermodynamics
Aldehyde
Eq. Constant (M-1
)
Aldehyde
Eq. Constant (M-1
)
1500
140
610
30
660
710
520
320
190
67
CONTACT eric V. anslyn
anslyn@austin.utexas.edu
© 2016 informa uK limited, trading as taylor & francis Group

2
S. R. LONG ET AL.
1. Introduction
Tripodal ligands, such as tripicolylamine, have been used for chelating metals, but the syn-
thesis of analogous complexes can be complex when different chelating groups are desired
to generate unsymmetrical ligands [1–3]. Recent advances by our group demonstrated that
a dynamic tripodal ligand, scheme 1(b), can be created using dipicolylamine (2), a β-N-het-
erocyclic aromatic aldehyde, such as pyridine-2-carbaldehyde (1), and a Lewis acidic metal
ion, optimally zinc(II) [4–7]. This procedure creates the tripicolylamine analog in situ from
these simpler building blocks.
While there are numerous commercially available β-N-heterocyclic aldehydes, secondary
amines with two appropriate heterocyclic “arms” are not as widely available. The ability to
modify and rapidly generate such heterocycles could lead to a variety of potential tris-chelat-
ing amines that could form dynamic tripodal ligands in reactions similar to that of scheme
1 [8, 9]. This diversity of complexes could lead to a variety of potential applications for these
assemblies. For example, the system shown in scheme 1(b) has been utilized as a method
for measuring the enantiomeric excess of secondary alcohols, and other potential applica-
tions are currently under investigation [5–7].
In recent years, the Huigsten-Sharpless cycloaddition reaction, colloquially referred to as
the“click”reaction, has exponentially grown in popularity because the reaction is fast, quick,
and high yielding [10, 11]. The product of the click reaction is a triazole formed from the
cycloaddition of an azide and an alkyne, allowing nearly any potential functional group to
be included in the triazole by using the appropriate azide or alkyne (scheme 2). Furthermore,
many azides are commercially available or readily accessible through one or two steps of
synthesis [12]. Given the diversity of azides, if triazoles were able to participate in the assem-
bly shown in scheme 1, a large number of different multicomponent four-coordinate ligand
assemblies could be rapidly generated. Thus, using the precursor dipropargylamine, a
(a)
(b)
Scheme 1. (a) tripicolylamine with zinc(ii) forming a tripodal metal complex. (b) model dynamic
hemiaminal assembly (3) with pyridine ligands [4].
Scheme 2. dynamic hemiaminal assembly 5 formed with ligand 4, 1, and zinc(ii).

JOURNAL OF COORDINATION CHEMISTRY
3
Scheme 3. synthesis of a bis-triazole based analog 4 of dipicolylamine using copper catalyzed click
reaction with dipropargylamine and benzylazide.
Scheme 4. tri-triazole analog of dipicolylamine 2 was synthesized following similar procedure as analog 4.
bis-triazole-based analog of dipicolylamine was generated using benzylazide via the click
reaction (scheme 3, 4). We postulated that such bis-triazole amine structures could be used
to generate analogous dynamic tripodal structures (scheme 2, 5) as previously reported.
The results of these studies are reported herein.
2. Experimental
The bistriazole 4 was synthesized from dipropargylamine and benzyl azide using literature
procedures [9]. All aldehydes and metal salts were purchased from commercial vendors and
used as received. The assembly of the bis-triazole with aldehyde in the presence of the metal
was carried out in deuterated acetonitrile obtained from Cambridge Isotope Laboratory.
Equal parts aldehyde, metal, and 4 were dissolved in deuterated acetonitrile to form the
complex. 1H NMR spectroscopy were used to monitor the complex formation. Equilibrium
constants were measured by finding the ratio of the hemiaminal hydrogen to aldehydic
hydrogen as previously reported [1]. Exchange experiments were conducted using an equil-
ibrated complex solution to which a second equivalent of an aldehyde was added. The
sample was analyzed after 30 min and repeatedly for up to 5 days.
Similar experiments were carried out for metal exchange wherein a second equivalent
of metal was added to an equilibrated complex solution and the formation of new peaks
corresponding to the complex with the new metal were observed in mass spectra.
All ITC experiments were carried out in acetonitrile at concentrations of 10 μM metal ion
(as the perchlorate salt) in the cell with 250 μM of the three-coordinate ligands and 150 μM
of the four-coordinate ligands in the syringe. All solutions were degassed by purging with
nitrogen gas for 15 min. In a typical ITC run, 7.5 μL of the ligand was injected into the metal
ion over the course of 10 s with mild stirring (300 rpm). The injections were spaced every
10 min with a total of 45 injections.
3. Results and discussion
Aldehyde, metal perchlorate salts, and bistriazole 4 were added in a 1 : 1 : 1 ratio to an ace-
tonitrile solution. The formation of new peaks was observed in the 1H NMR spectra from 3.5
to 6.1 ppm (figure 1). These new resonances correspond to the formation of 5. The pattern

4
S. R. LONG ET AL.

JOURNAL OF COORDINATION CHEMISTRY
5
Table 1. equilibrium constants for various aldehydes for forming a four-coordinate complex (i.e. 5). all
equilibrium constants were determined from 1h-nmr spectra collected on a 400 mhz instrument with
approximately 60 mm of each assembly component dissolved in deuterated acetonitrile.
Eq. Constant (M−1
)
Aldehyde
1500
610
30
520
320
140
660
710
190
67
of peaks is similar to that seen for the complex when dipicolylamine is used as the secondary
amine [4]. The four peaks from 3.5 to 4.8 ppm correspond to the hydrogens on the meth-
ylenes of the bis(triazole)amine. These four hydrogens are nonequivalent in the complex,
because they are diastereotopic due to the newly created stereocenter in the ligand. The
remaining two peaks correspond to the hemiaminal center, with one peak at approximately
5.4 ppm for the hydrogen on the methine carbon and one broader peak at 6.1 ppm for the
hydrogen on the hydroxyl.
An equilibrium constant for 5 with the bistriazole 4 and zinc and 2-pyridinecarboxalde-
hyde was calculated from the integrals in the 1H NMR spectra. By measuring the integrals of
the aldehyde peak and the methine peak, a ratio between the components in solution was
obtained. Using the concentration of each of the components and this ratio, an equilibrium
constant was calculated using the formula published from our previous study (table 1, 1st
entry) that takes into account the fact that the metals also bind the bidentate ligand [4].
Thus, speciation of the metal between the bidentate and tridentate ligand is not necessary
in determination of the equilibrium constant. Further, because we are operating in

6
S. R. LONG ET AL.
acetonitrile, there is no protonation of the ligands by the solvents that need to be accounted
for in the equilibrium constant determination.
Once the assembly using 4 was observed, we expanded the scope to determine whether
other heterocyclic aldehydes would also form the complex. Several different heterocyclic
aldehydes were screened with zinc and 4 in a 1 : 1 : 1 ratio at 60 mM. The equilibrium constant
for each heterocyclic aldehyde is shown in table 1 and as expected each heterocyclic alde-
hyde showed a different equilibrium for complex formation.
While all heterocyclic aldehydes formed complexes with 4 and zinc, 3,5-dibromopyridi-
necarboxaldehyde, isoquinoline-2-carboxyaldehyde, thiazole-4-carboxaldehyde and tri-
azole-2-carboxyaldehyde only formed a small amount of complex. Others, like 3,
5-bromopyridinecarboxaldehyde, imidazole-2-carboxaldehyde and thiazole-2-carboxalde-
hyde, almost exclusively formed the corresponding tripodal complexes (5). The formation
of complexes with the heterocyclic aldehydes is driven by a combination of the electrophilic-
ity of the aldehyde and the coordination ability of the heterocyclic nitrogen. The coordination
of the heterocyclic nitrogen of the aldehydes leads to a thermodynamically stable complex,
thereby forming an energy sink for the formation of the tripodal complexes. On the other
hand, the electrophilicity of the aldehyde influences the ability of the secondary amine to
add to the aldehyde.
In order to determine whether complex formation was dynamic and the heterocyclic
aldehydes could exchange, a second equivalent of a different aldehyde with a larger equi-
librium constant was added after the original solution had reached equilibrium. As could
be seen in the 1H NMR spectra, the original aldehyde peak grows in as the complex disas-
sembles and reassembles with the second aldehyde. For example, the weaker complex based
upon thiazole carbaldehyde disassembles in the presence of pyridine carbaldehyde to give
the more thermodynamically stable complex. Given this fact, in a mixture of aldehydes, the
complex that forms is dictated by the most stable metal complex (figure 2).
Additionally, the metal controls the extent of assembly. Metals that form stronger donor
coordinate bonds from the tetradentate assembly form more of the complex. Studies using
mass spectrometry were conducted with various metals to determine which metals would
lead to the most complex formation (5) under inert atmosphere and in degassed solvents.
The extent of complex formation was measured via mass spectrometry by a comparison of
the area of the peak corresponding to the tetradentate complex to the area of the peak
corresponding to metalated 5. Based on these studies, the largest amount of complex for-
mation occurred when using Zn(II). Co(II) led to approximately 80% of the complex formation
found for zinc. Mass spectroscopy studies with iron(II) and copper(II) showed little formation;
less than 10% of the amount formed with zinc.
Additionally, studies were carried out to evaluate the change in the thermodynamic driv-
ing force for the formation of the complex using each different metal ion to see whether the
thermodynamic driving forces could be correlated with the extent of complex formation.
Because the major driving force for assembly is the formation of the four-coordinate complex,
as such, the differences in energy between the metallation of tridentate 4 to the tetradentate
6 should directly correlate with the extent of complex formation for that metal. Isothermal
titration calorimetry was used to measure the thermodynamic parameters of the metal
binding to these two different ligands in acetonitrile. The goal was to generate relative trends
of binding constants to see how these correlate with the extent of assembly, and thus the

JOURNAL OF COORDINATION CHEMISTRY
7

8
S. R. LONG ET AL.
Table 2. Binding constants of zinc, cobalt, nickel, and copper perchlorates with the di- or tri- triazole
amine ligand 4 and 6 in acetonitrile, respectively. the constants were determined using itC. the ratio
of binding constant was calculated by dividing the binding constant for 6 over 4. errors are estimated
at 15%.
4 (J)
6 (J)
Ratio of binding constant
Zinc(ii)
6.0e7
4.3e6
5.6e8
2.3e7
6.3e8
6.1e7
2.6e9
5.1e7
10.5
14.2
4.7
Cobalt(ii)
nickel(ii)
Copper(ii)
2.2
ITC data were not corrected for protolytic equilibrium that takes into account ligand pKa’s
that would have to have been determined and pH values in acetonitrile.
Previous studies have shown that the free energy of ligation can be calculated for each
of the ligands, but these studies were carried out in water [10, 11]. Metal ion perchlorate
salts in water do not have the same thermodynamic parameters as in acetonitrile, making
these values an ineffective comparison for our system. Using perchlorates allowed for all the
salts to be soluble in acetonitrile.
To correlate the differences in affinities of the tri- and tetradentate ligands with the extent
of assembly, the standardized ratio was used (table 2). The expectation is that when the
tetradentate ligand has the largest binding relative to the tridentate ligand, the maximum
amount of assembly will occur. Thus, irrespective of whether the number of amines differs
between the two ligands, the ratio of binding affinities should correlate with extent of assem-
bly. The trend in ratio reaffirmed the mass spectrometry results. The cobalt(II) and zinc(II)
show similar fold change (14.2 vs.10.5, respectively). Similarly, the mass spectroscopy results
showed similar extent of assembly with these two metals. Copper shows a favorable change
in binding constants, albeit smaller than zinc or cobalt, and a significantly lower extent of
assembly in the mass spectrometry results. On the other hand, nickel(II) shows a change less
than one fold between the two ligands and therefore does not lead to any complex formation
in the mass spectrum.
4. Conclusion
We have demonstrated that a bistriazole 4 can undergo reaction with heterocyclic aldehydes
to form four-coordinate ligands. The dynamic complexes form with components that are
the best metal and ligand combinations. The formation is reversible, such that the addition
of differing aldehydes with a higher affinity leads to preferential formation of the complex
containing that aldehyde. The complex also forms with several metals, and the extent of
formation correlates with the differences in chelation strength between the assembly and
the corresponding reactants.
Acknowledgements
The authors would like to thank Karin Keller for her help with the mass spectroscopic measurements.
We also would like to acknowledge the following agencies for their generous financial support of this
work: National Science Foundation Grant CHE-1212971 and Welch Regents Chair to EVA, Grant F-0046.

JOURNAL OF COORDINATION CHEMISTRY
9
Disclosure statement
No potential conflict of interest was reported by the authors.
Funding
This work was supported by the National Science Foundation [grant number CHE-1212971]; and Welch
Regents Chair to EVA [grant number F-0046].
References
[1] Z. Dai, J.W. Canary. New J. Chem., 31, 1708 (2007).
[2] A.G. Blackman. Polyhedron, 24, 1 (2005).
[3] R. Custelcean. Chem. Soc. Rev., 43, 1813 (2014).
[4] L. You, S.R. Long, V.M. Lynch, E.V. Anslyn. Chem.-Eur. J., 17, 11017 (2011).
[5] L. You, G. Pescitelli, E.V. Anslyn, L. Di Bari. J. Am. Chem. Soc., 134, 7117 (2012).
[6] L. You, J.S. Berman, E.V. Anslyn. Nat. Chem., 3, 943 (2011).
[7] L. You, J.S. Berman, A. Lucksanawichien, E.V. Anslyn. J. Am. Chem. Soc., 134, 7126 (2012).
[8] H.C. Kolb, M.G. Finn, K.B. Sharpless. Angew. Chem. Int. Ed., 40, 2004 (2001).
[9] Q. Wang, T.R. Chan, R. Hilgraf, V.V. Fokin, K.B. Sharpless, M.G. Finn. J. Am. Chem. Soc., 125, 3192
(2003).
[10] V.V. Rostovtsev, L.G. Green, V.V. Fokin, K.B. Sharpless. Angew. Chem. Int. Ed., 41, 2596 (2002).
[11] S. Giannattasio, S. Gagliardi, M. Samaja, E. Marra. Brain Res. Protoc., 10, 168 (2003).
[12] T.R. Chan, R. Hilgraf, K.B. Sharpless, V.V. Fokin. Org. Lett., 6, 2853 (2004).