The Species of Fehling's Solution

Article abstract of DOI:10.1002/ejic.201600168

A species model for the copper(II)/L-(+)-tartaric acid (LH₂, tartH₂) aqueous system between pH 1.9 and 12.3 has been established by potentiometric, UV/Vis spectroscopic, solubility and crystallographic studies. Eight species were detected, including the oligonuclear major species Cu₈L₆H_-10 in the neutral region and the mononuclear major species CuL₂H_-4 in the alkaline region, the latter being associated with Fehling's solution. The octanuclear complex was isolated as its lithium salt Li₇[Cu₈(L-tartH_-2)₄(L-tartH_-1)₂(H₂O)₆]NO₃·19H₂O (1). The "Fehling species" CuL₂H_-4 was crystallized as its sodium, potassium and caesium salts, namely K₂Na₄[Cu(L-tartH_-2)₂]·12H₂O (2), Na₆[Cu(L-tartH_-2)₂]·9H₂O (3), Na₆[Cu(L-tartH_-2)₂]·14H₂O (4), and Cs₆[Cu(L-tartH_-2)₂]·8H₂O (5). Each of them shows a distorted square-planar environment of copper(II), similar to the related non-chiral cuprate [Cu(rac-tartH_-2-κ²O²,O³)₂]^6- (S. Albrecht, P. Klüfers, Z. Anorg. Allg. Chem. 2013, 639, 280-284). By comparing the components of the UV/Vis spectra of Fehling's solution and the solids, we confirmed the identity of the major solution species.

Full text of DOI:10.1002/ejic.201600168

DOI: 10.1002/ejic.201600168
Full Paper
Speciation of Copper(II)
The Species of Fehling's Solution
Thomas G. Hörner[a] and Peter Klüfers*[a]
Abstract: A species model for the copper(II)/
(LH2, tartH2) aqueous system between pH 1.9 and 12.3 has been caesium salts, namely K2Na4[Cu(
established by potentiometric, UV/Vis spectroscopic, solubility tartH–2)2]·9H2O (3), Na6[Cu( -tartH–2)2]·14H2O (4), and Cs6[Cu(L-
L-(+)-tartaric acid species” CuL2H–4 was crystallized as its sodium, potassium and
L-tartH–2)2]·12H2O (2), Na6[Cu(L-
L
and crystallographic studies. Eight species were detected, in- tartH–2)2]·8H2O (5). Each of them shows a distorted square-pla-
cluding the oligonuclear major species Cu8L6H–10 in the neutral nar environment of copper(II), similar to the related non-chiral
region and the mononuclear major species CuL2H–4 in the alka- cuprate [Cu(rac-tartH–2-κ2O2,O3)2]6– (S. Albrecht, P. Klüfers, Z. An-
line region, the latter being associated with Fehling's solution. org. Allg. Chem. 2013, 639, 280–284). By comparing the compo-
The octanuclear complex was isolated as its lithium salt nents of the UV/Vis spectra of Fehling's solution and the solids,
Li7[Cu8(L-tartH–2)4(L-tartH–1)2(H2O)6]NO3·19H2O (1). The “Fehling we confirmed the identity of the major solution species.
Introduction
known. Strangely enough, textbooks such as that of Cotton et
al. assume that meso-tartaric acid is a component of Fehling's
solution,[3] although recipes from both the European and the
Fehling's solution, an alkaline solution containing copper(II) and
L-(+)-tartaric acid (LH2 = tartH2; the different isomeric forms of
U.S. Pharmacopeia require
higher stability of the cuprates formed. According to these pro-
cedures, copper sulfate, potassium sodium -(+)-tartrate (Sei-
L-(+)-tartaric acid because of the
tartrate are illustrated in Scheme 1), is a well-known standard
probe for detecting reducing substances. Originally invented by
Hermann Christian von Fehling in 1848 to determine the
amount of glucose in urine samples,[1] it is nowadays used to
distinguish between reducing and non-reducing sugars.[2]
Moreover, it is one of the first detection probes students are
faced with in undergraduate courses. Although the usage of
the deep-blue solution reaches back almost 170 years, the
structure of the obvious tartratocuprate(II) species that prevents
precipitation of cupric hydroxide in solution has remained un-
L
gnette salt) and sodium hydroxide are mixed in a molar ratio
of 1:3.8:5.8 in water (cCu = 0.217 mol L–1) to obtain deep-blue
solutions (pH 13.5–14).[4,5] Samples are checked for reducing
properties by adding a few drops of this solution; after heating
gently, the formation of a red precipitate of Cu2O indicates the
presence of reducing species.
At the beginning of the 20th century, several research
groups started to reveal the composition of the active species.
The first studies concerning alkalized solutions of copper(II) and
enantiopure tartaric acid were reported by Bullnheimer and
Seitz in 1899 and 1900, and by Traube in 1921. Both groups
concluded that the anion [Cu(L
-tartH–2)2]6– was the species in
Fehling's solution.[6–8] Owing to the methodological limitations
at that time, the formulae of some alkali salts of the cuprate
were resolved by elemental analysis, but the structure of the
species remained unclear. Since then only a few synthetic stud-
ies on compounds obtained in the alkaline region have been
published, most of them, such as the work of Pfeiffer et al. in
1948, confirmed earlier results.[9] In fact, until now, no com-
pound with copper and L-(+)-tartaric acid has been isolated
from solutions above pH 9 and characterized by single-crystal
X-ray analysis, unlike the case of racemic tartaric acid. With the
racemate, three compounds were isolated and fully character-
ized: Na4[Cu2(rac-tartH–2-κ2O1,O2:κ2O3O4)2]·10H2O,[10–12] Li4[Cu2-
(rac-tartH–2-κ2O1,O2:κ2O3O4)2]·11.75H2O and Na6[Cu(rac-tartH–2-
κ2O2,O3)2]·14H2O,[2] the last being considered a Fehling-analo-
gous complex. In terms of stoichiometry, these species resem-
ble the formulae Cu2L2H–4 and CuL2H–4.
Scheme 1. Various isomers of tartarate. Top: Fischer projections; bottom:
stereochemical drawings. The atomic numbering is used in the κ2O2,O3-style
specification of the copper-chelating ligand atoms .
[a] Fakultät für Chemie und Pharmazie,
Butenandtstraße 5–13, 81377 Munich, Germany
E-mail: kluef@cup.uni-muenchen.de
www.cup.lmu.de/ac/kluefers/homepage
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/ejic.201600168.
To reveal the species in solution, potentiometric,[13–18]
polarographic,[19–22] electrophoretic,[23] UV/Vis and CD spectro-
Eur. J. Inorg. Chem. 2016, 1798–1807
1798
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
scopic[24–28] as well as extraction studies[29] of the copper(II) kali ion error.[34] Thus, a reliable analysis was hampered in the
and L-(+)-tartaric system have been reported. However, only a pH regime that is typical of Fehling's solution. We therefore
few of them focused on the alkaline regime (pH > 9) or pre- supplemented potentiometric data with a UV/Vis spectroscopic
sented a model for a wide pH region. Upon a closer look, most analysis and solubility studies.
of them showed a similar model for the acidic and neutral re-
gions. Whereas mono- and dinuclear species such as CuL, CuLH
Combined Potentiometric Titration and UV/Vis
Spectroscopy
and Cu2L2 were formed at low pH values, oligonuclear species
such as Cu8L6H–10 or Cu6L4H–7 predominated in solution at neu-
tral pH. One of the first comprehensive studies of this system
was performed by Lefebvre in 1957. By using potentiometric
titration and UV/Vis spectroscopy, he suggested a model includ-
ing CuL in the acidic region and the octanuclear Cu8L6H–10 at
pH 6 and higher, with the Cu8 species being favoured over the
tetranuclear Cu4L3H–5 species with the same overall stoichio-
metry. He also drew conclusions about the alkaline region and
postulated species such as [Cu3L3(OH)6]6–, [CuL2(OH)2]2– and
[CuL2(OH)4]6–. The last was assigned as Fehling's complex.[13]
Later research groups confirmed the existence of mono- and
dinuclear complexes in the acidic region, especially the species
CuL, which is in equilibrium with its dimer Cu2L2. Still later,
By recording both the pH value and the UV/Vis spectrum of the
titration solution (Cu/L
-tart = 1:3, ccu = 0.01 mol L–1) at each
added volume increment of sodium hydroxide we obtained a
titration curve as well as a series of absorption spectra without
isosbestic points between pH 1.9 and 12.3. With increasing pH,
the spectra show a shift of λmax for the crystal-field transition
to lower wavelengths, from 816 (pH 1.9) to 676 nm (pH 12.3; see
Figure S4 in the Supporting Information). According to Martell's
definition of critical stability constants, the most reliable pub-
lished species, specifically, the species from refs.[15–17], were
chosen and their stability constants were considered for model-
ling.[35] Furthermore, the CuL2H–4 species, which is supposed to
be the predominating species in Fehling's solution, was taken
Johansson also examined the neutral region and confirmed the
[16]
presence of oligonuclear species Cu8L6H–10
.
However, the al-
into account. The protonation constants of L-(+)-tartaric acid
used in the refinement were determined by a separate potenti-
ometric titration (pKa1 = 2.58, pKa2 = 3.69; ionic strength I =
kaline environment was investigated through potentiometric ti-
tration by just one group after Lefebvre's studies: Blomqvist and
Still found the additional species Cu2L2H–3 and Cu2L2H–4 above
pH 9, which are formed by stepwise deprotonation of the
Cu2L2H–2 species.[15] The most recent potentiometric study of
0.5
M, NaClO4). Under these premises, a satisfying model con-
taining eight species resulted (Figure 1). Fewer species led to a
markedly worse fit, whereas more species did not improve the
goodness of fit (see the Exp. Sect.) but caused noisy molar ex-
tinction curves or refinement errors. Moreover, factor analysis
(see the Exp. Sect.) showed at least six species that contributed
significantly to the UV/Vis spectra, further compounds were
therefore not considered. The mixed stability constants (see
Table 1) are in good agreement with the species of the latest
published data, which have now been completed by the previ-
ously unconsidered CuL2H–4 species.[15–17] In addition to stabil-
ity constants, the spectral molar absorption for each absorbing
the copper(II)/L-(+)-tartaric acid aqueous system was performed
by Piispanen and Lajunen in 1995. By considering all the species
formulated in the past, the authors reported a model with
seven species in the pH region between 2.4 and 7.5: CuL, Cu2L2,
CuHL and Cu2L2H–1 for pH < 4.5, and Cu2L2H–2, Cu6L4H–7 and
Cu8L6H–10 in the neutral region. It was also shown that, despite
a three-fold molar excess of ligand over copper, the last two
oligonuclear species were the predominating complexes for pH
> 5. Nevertheless, as with several other studies, the alkaline
region was missing from this work.[17] Most recently, progress
has been made regarding the structures of copper–tartrate spe-
cies in the crystalline state. In their 2013 work, Liu et al. pre-
pared the cupric
L-(+)-tartrates [{Cu2(L-tart)2(H2O)2}n/n]·≈3.5H2O
and [{Cu2( -tart)2}n] as two- and three-dimensional coordination
L
polymers in a hydrothermal approach.[30] The structure of the
hydrated polymer has been described by other groups previ-
ously.[18,31–33] To the best of our knowledge, these structural
analyses have provided the only available crystallographic data
on copper complexes of enantiopure tartrate.
In this work, which was focused on the alkaline regime and
includes Fehling's solution, we present a species distribution
model that is complemented by crystallographic evidence of
the identified species. In particular, we provide structural data
on the major species in this part of the pH range, Cu8L6H–10
and the Fehling species CuL2H–4.
Results and Discussion
Figure 1. Distribution of species in the copper(II)/
L
-(+)-tartaric acid system
The strongly alkaline region is difficult to investigate by potenti-
ometric methods due to a marked increase in the so-called al-
[Cu/LH2 = 1:3, cCu = 0.01
M in 0.5 M NaClO4, LH2 = L-(+)-tartaric acid]. Charges
have been omitted.[36,37]
Eur. J. Inorg. Chem. 2016, 1798–1807
www.eurjic.org
1799
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
species was calculated (Figure 2).[36,37] As a result of these analy- The Neutral Region
ses, the dominating species of three pH regions are discussed.
The predominant species between pH 5.3 and pH 9.6 is
Cu8L6H–10, which represents almost 100 % of the Cu species
present at pH 8.1. Cu2L2H–4 plays a minor rule in this pH region.
Notably, all past solution models include such an octanuclear
species in preference to the formula Cu4L3H–5. In fact, the tetra-
nuclear formula is used in work based on elemental analysis,
such as that of Masson and Steele in 1899 and of Packer and
Wark in 1921, who may have had the octanuclear entity in their
hands.[38,39] Specifically, both research groups isolated a cop-
In the sense of a consensus model, the species of our solution
study are common to most of the published analyses with some
more rarely considered species skipped, namely Cu2L2H–3 and
Cu6L4H–7. Moreover, the frequently addressed CuL species was
also omitted for the reasons discussed in the next paragraph.
Table 1. Log ꢀ, λmax and εmax values of the different species observed in
HypSpec2014. Charges have been omitted.[36,37]
Species
log ꢀ
λmax(H2O) [nm]
εmax [L mol–1 cm–1
]
per(II) complex with
L-(+)-tartaric acid in the neutral region and
Cu
Cu2L2
–
9.22
4.38
–0.22
–17.32
–27.68
6.22
817
839
744
756
694
680
813
697
6.13
assigned to it the formula M3[Cu4(
L
-tartH–2)2( -tartH–1)]·4H2O
L
27.22 0.16
32.66 0.89
36.93 0.48
36.17 0.18
28.79 0.02
20.62 0.12
(38.76 0.32
(M = Na, K, Ag). Masson and Steele obtained the potassium salt
as a blue powder by mixing copper tartrate with an excess of
aqueous KOH and adding ethanol. (They falsely claimed this
compound as Fehling's complex, although it reacted neutrally
with litmus.[38]) In 1954, Ablov and Popovich reported the
Cu2L2H–1
Cu2L2H–2
Cu2L2H–4
CuL2H–4
CuLH
Cu8L6H–10
–7.36
isolation of the salts MIII[Cu4(
L-tartH–2)2(L-tartH–1)] with large
cations {M = [Co(NH3)6]3+, [Co(NH3)5(H2O)]3+, [Co(en)3]3+ and
[Cr(urea)6]3+} at pH 7.[40] In an attempt to use their procedure
for the preparation of crystals, we obtained the [Co(NH3)6]3+ salt
as small, dark-green crystals, which were reasonably suitable for
X-ray diffraction analysis. However, although the anion structure
was fairly well resolved and showed the expected octanuclear
structure, which results from doubling the net formula, the en-
tire analysis did not meet the quality criteria for publication due
to cation disorder (see Figure S24 in the Supporting Informa-
tion).
Crystals of improved quality were eventually obtained from
Cu(NO3)2·3H2O,
of 4:3:11 in water (pH ≈ 9). From these solutions, blue blocks of
Li7[Cu8( -tartH–2)4( -tartH–1)2(H2O)6]NO3·19H2O (1) were grown
L-(+)-tartaric acid and LiOH·H2O in a molar ratio
L
L
by slow evaporation over the course of several months. The
structure of the cuprate ion resembles the octanuclear anion of
the [Co(NH3)6]3+ salt and confirms the existence of an oligonu-
clear Cu8L6H–10 species in the addressed pH region. The assign-
ment by past research groups of the major species in the neu-
tral regime was, thus, correct.
Figure 2. Molar absorptivity spectra of the different species observed in Hyp-
Spec2014. Charges have been omitted.[36,37]
The Acidic Region
The Alkaline Region
As concluded by past researchers, mono- and dinuclear species
are formed between pH 1.9 and 5.0, that is, aquated copper(II) Owing to the fundamental problems of analysing strongly
ions Cu, CuLH, Cu2L2, Cu2L2H–1 and Cu2L2H–2. In contrast to alkaline solutions by a potentiometric approach, establishing a
most of the preceding studies, the inclusion of the species CuL model needed care. We found two species above pH 9.6, the
in the model did not result in a stable refinement. It should be minor Cu2L2H–4 and major CuL2H–4 species, the latter predomi-
noted that both CuL and Cu2L2 are cupric tartrate. The presence nating above pH 9.9 and being the only one above pH 11.7.
of a Cu2L2 species in the refinement points to the fact that in Attempts to crystallize the minor dinuclear species, the
the solid dinuclear building blocks of the respective polymers crystallization of which is straightforward in the analogous cop-
are recognised. Thus, a non-polymeric Cu2L2 solution species per(II)/rac-tart system,[2] were unsuccessful. As a starting point,
maybe derived from a dinuclear cutout of the polymer by satu- we had chosen the work of Bullnheimer and Seitz;[7] we inter-
rating the vacant bonding sites by additional aqua ligands. preted their Na2[Cu(
L
-tartH–2)]·4H2O as Na4[Cu2( -tartH–2)2]·
L
CuLH and Cu2L2H–1 seem to play a minor role in solution in 8H2O, in close resemblance to its racemic variant.[2,10–12] We
contrast to Cu2L2; Cu2L2H–2 reaches its maximum (46 %) at agree with the statement of the former authors that the mother
pH 4.9. None of these species has been isolated and character- liquors of this compound were unstable (the lithium analogue
ized, which may be explained by the tendency of the Cu2L2 showed still increased instability). However, we were not able
species to form polymeric precipitates.
to isolate the solid sodium salt.
Eur. J. Inorg. Chem. 2016, 1798–1807
www.eurjic.org
1800 © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
To clarify the marked difference between rac- and L-tart-
based dicuprates, we modelled both anions by a DFT approach.
For steric reasons, the centrosymmetric [Cu2(rac-tartH–2)2]4– tet-
raanion with its five-membered chelate ring cannot be assem-
bled with a single enantiomer. Instead, a reasonable dinuclear
analogue needed six-membered chelate rings with [Cu2(L-
tartH–2-κ2O1,O3-κ2O2,O4)2]4– bonding. Based on five different
functionals, Ahlrich's def2-tzvp basis set and the COSMO sol-
vent model, the racemic compound is 22 (TPSS) to 33 kJ mol–1
(B3LYP) more stable than the enantiopure compound, clearly
due to lower ring strain (see Figure 3 and Table 2). Because of
this instability, we assume that Cu2L2H–2 is not a species truly
competing with CuL2H–4 in Fehling's solution. Moreover, taking
into consideration the conditions of the standard recipe for the
preparation of Fehling's solution with respect to the species-
distribution curve obtained here,[4,5] and, extrapolating from
pH 12.3, we see that CuL2H–4 is the dominating species be-
tween pH 13.5 and pH 14 (Figure 4).
Figure 4. Species distribution diagram c(Cu) vs. pH. Here, the conditions of
the standard recipe for the preparation of Fehling's solution were considered
(cCu,tot = 0.217 mol L–1). Values at pH > 12.3 were extrapolated by using a
local variant of Motekaitis's SPE.[35]
of the CuL2H–4 species, resulted in a clear deep-blue solution
with no remaining solid (both experiments: cNaOH = 1 mol L–1).
This indicates that [Cu(OH)4]2– (≡ CuH–4) did not play a role in
Figure 3. Optimized structures (def2-tzvp, B3LYP, cosmo, d3) of [Cu2(rac-
L)2H–4] (left, rac = racemic) and [Cu2L2H–4] (right).
the titration solution. In fact, Cu(OH)2 is soluble in >5
M NaOH
(λmax = 641 nm, ε = 31.3 dm3 mol–1 cm–1; in 10
M
NaOH) only.
Typical synthetic procedures for Na2[Cu(OH)4] require 45–50 %
aqueous NaOH.[41,42] The situation is less clear for a CuLH–4 spe-
cies, which may be present as a minor species. Thus, attempts
were made to isolate crystals from high-pH solutions.
Table 2. Energy differences (ΔE) between Cu2L2H–4 (racemic, Erac
Cu2L2H–4 (enantiopure, Eena).
) and
Functional
ΔE = Erac – Eena [kJ mol–1
]
B3LYP
PBE0
TPSSh
BP86
TPSS
33.0
31.9
25.8
25.3
22.2
CuL2H–4 Salts
To confirm the CuL2H–4 species and analyse its structure, we
reproduced the synthetic procedure of Bullnheimer and Seitz
for Na6[Cu(
L-tartH–2)2]·13H2O and K2Na4[Cu(
L-tartH–2)2]·11H2O;
the latter is reported to co-crystallize with K3Na3[Cu(
L-tartH–2)2]·
11H2O, which, however, did not precipitate in any of our crystal-
lization batches. Moreover, powder X-ray diffraction patterns of
the batches always resembled the calculated patterns of the
The Dilemma of the Alkaline Regime: CuL2H–4 versus
CuLH–4 versus CuH–4
Owing to the known problems of potentiometric analyses in K2Na4 salt.[6] By mixing Cu(OH)2, potassium sodium
L-(+)-tartrate
the high-pH region, both the species CuH–4 and the hypotheti- tetrahydrate (Seignette salt), KOH and NaOH according to their
cal CuLH–4 instead of CuL2H–4 gave a similar fit with similar recipe, we obtained blue rods in 36 % yield by concentrating
stability constants, that is, measuring pH and UV/Vis spectra did the dark-blue solutions. Crystal-structure analysis revealed
not suffice to establish a reliable model. Thus, solubility studies K2Na4[Cu(
were performed in addition, which proved the solubility of composition of Fehling's complex. Similarly, Na6[Cu(
Cu(OH)2 in 1 NaOH (note: the final hydroxide concentration 9H2O (3) and Na6[Cu( -tartH–2)2]·14H2O (4) were isolated as blue
of a titration solution was approximately 0.1 mol L–1). By mixing crystals. In addition to the work of Bullnheimer and Seitz, we
Cu(OH)2 and NaOH in a molar ratio of 1:2 at cNaOH = 1 mol L–1, also prepared the caesium salt Cs6[Cu(
-tartH–2)2]·8H2O (5),
practically all the Cu(OH)2 was recovered and the resulting solu- which is a hygroscopic solid that is stable for a few hours only.
L-tartH–2)2]·12H2O (2), quite close to the reported
L-tartH–2)2]·
M
L
L
tion was clear and colourless. A mixture of Cu(OH)2, L-(+)-tar- In general, crystals were isolated only from relatively concen-
taric acid and NaOH in a molar ratio of 1:1:2, which resembles trated solutions (cCu ≈ 1.1 mol L–1). The instability of the com-
the stoichiometry of a CuLH–4 species, in water yielded 26.8 % plexes, which was evidenced by red or black discoloration due
of dissolved Cu(OH)2, whereas a mixture of Cu(OH)2,
L-(+)-tar- to the formation of copper oxides, increased in the series
taric acid and NaOH in a molar ratio of 1:2:6, the stoichiometry 3 < 4 ≈ 2 << 5.
Eur. J. Inorg. Chem. 2016, 1798–1807
www.eurjic.org
1801
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Solid-State versus Solution UV/Vis Spectra
starting point to construct the Cu8 cluster. It is based around
the Cu2 and Cu6 centres. To build the entire cluster, one should
add two more copper atoms to each of the two tartH–2 ligands
of the two bis(diolato) cores. The bonding sites for each of
these two copper atoms are the deprotonated hydroxy-carboxy
functions, which form five-membered chelate rings on κ2O1,O2
and κ2O3,O4 coordination in addition to the κ2O2,O3 diolato
mode. One should now double this motif by the application of
a two-fold axis close to Cu4 and Cu5. Finally, two terminating
tartH–1 ligands are added to the Cu1/Cu3 and Cu7/Cu8 couples.
As a result, the six copper atoms that are not bonded in a bis(di-
olato) core share a common bonding pattern: a cis-bis(alkoxido)
and cis-bis(carboxylato) pattern. It should be noted that the
octanuclear cluster is not formed from two Cu4 parts by Jahn–
Teller-elongated, weaker contacts and is thus not prone to
cleavage in solution. Accordingly, the Cu8 species is a stable
and persistent solution species.
As additional proof to demonstrate that the crystals' anion
[Cu(L
-tartH–2)2]6– resembles the high-pH solution species, the
compounds were studied by UV/Vis spectroscopy in aqueous
solution and in the solid state (see the Supporting Information).
The observed λmax values for the d–d transitions of 2, 3 and 5
are presented in Table 3. For comparison, λmax,tit = 680 nm. In
the case of compound 4, only a few crystals were obtained and
therefore it was not possible to record its UV/Vis spectrum. The
absorption spectra of each compound both in aqueous solution
and in the solid state give good agreement with the absorption
spectra of the species CuL2H–4 obtained from the analysis of
the potentiometric data. Thus, the major species, practically the
only one, in Fehling's solution is CuL2H–4 {≡ [Cu(L
-tartH–2)2]6–}.
Table 3. λmax of the d–d transitions of compounds 2, 3 and 5.
Compound
λmax(H2O) [nm]
λmax(solid) [nm]
The Cu8(tart)6 assembly in Figure 5 is complemented in the
actual crystal structure by aqua ligands and bridging carboxyl-
ate O atoms, which bind along the Jahn–Teller-elongated axes.
The coordination number of the cupric centres thus is either
five (Cu2, Cu4, Cu6, Cu7) or six (Cu1, Cu3, Cu5, Cu8). In the case
of Cu3, a disordered nitrate ion completes the coordination
sphere. (There is some uncertainty whether or not the nitrate-
assigned electron density is caused by a hydrogen carbonate
ion because the reaction solution was left in air for crystalliza-
tion.)
2
3
5
673
671
671
656
661
669
Structures
Li7[Cu8(
L
-tartH–2-κ4O1,O2,O3,O4)4(
-tartH–1-
L
κ3O1,O2,O4)2(H2O)6]NO3·19H2O
The octanuclear anions are linked by Cu8 and Cu1 bonding
to neighbouring carboxylate ligands and build up a two-dimen-
sional network. We assumed that these links are replaced by
Following common rules, species of low nuclearity dominated
the acidic as well as the alkaline region, but an oligonuclear
aggregate makes up the neutral solution. In fact, the repeatedly
formulated Cu8L6H–10 species was precisely confirmed by
an X-ray study of crystals of 1. The octanuclear cuprate exhibits
aqua ligands in aqueous solution, so that isolated [Cu8(
L-
tartH–2)4(
L
-tartH–1)2(H2O)x]6– (x ≥ 8) exist in aqueous media. The
a
saddle-shaped
[Cu8(
L
-tartH–2-κ4O1,O2,O3,O4)4(
L-tartH–1-
coordination sphere of the copper atoms can be interpreted in
greater depth by using continuous shape measures (CShM), as
established by Alvarez et al.[43] Details are presented in the Sup-
porting Information.
κ3O1,O2,O4)2(H2O)6]6– ion. Figure 5 shows the Cu–tart frame-
work of the structure (ORTEP plot is provided in Figure S16 in
the Supporting Information). The whole anion lies in the asym-
metric unit, the apparent C2 symmetry is non-crystallographic.
The cupric centres show typical strongly Jahn–Teller-elongated
octahedra with tartrate O atoms forming the equatorial planes
of four closer ligand atoms.
The formation of crystals of non-chiral copper–tartrate
assemblies seems to be less critical. By using aiding ligands,
supramolecular networks and ordered functional crystalline sol-
ids have been reported. Thus, the decanuclear compound
[Cu10(rac-tart)4(rac-tartH–1)4]·(apy)8·13H2O (apy = 2-aminopyr-
idine) and the pentadecanuclear cluster [Cu15(rac-tartH–2)6-
(OH)6(H2O)10]·20H2O were synthesized to examine the role of
hydrogen bonds and antiferromagnetic or ferromagnetic inter-
actions.[44,45] Moreover, we recently described the isolation and
molecular structure of K8[Cu10(rac-tartH–2)4(rac-tartH–1)4(H2O)4]·
18H2O.[2] The structure of the decanuclear anion in this cuprate
devoid of aiding nitrogen ligands is ruled by the same princi-
ples as derived for 1. As for L-(+)-tartrate, further cupric ions
become attached to two bis(diolato) nuclei resulting in carb-
oxylate-terminated Cu3(μ-O)3 cores. In the racemic compound,
the Cu3O3 hexagons are the only structural motif, in 1, they are
connected by a Cu4O4 octagon.
Figure 5. Octanuclear, almost C2-symmetrical copper(II)–tartrate assembly in
crystals of 1. Carbon atoms are shown in gray, copper in blue, oxygen in red
and hydrogen in white. Lithium ions, the nitrate ion and water molecules
have been omitted for clarity.
An analysis of the magnetic properties is pending. However,
antiferromagnetic spin-coupling paths are clearly recognisable
Below, the mononuclear bis(diolato) structure of the in terms of the Cu–μ-O–Cu angles. Considering the dx2–y2 orbital
CuL2H–4 core is described in detail. This core is also a suitable in each plane made up of the four short Cu–O contacts as the
Eur. J. Inorg. Chem. 2016, 1798–1807
www.eurjic.org
1802
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
magnetic orbital, two antiferromagnetic paths with obtuse Cu–
O–Cu angles become evident (Figure 6). We will report on the
results of the magnetic investigation in a separate work.
Figure 7. Molecular structure of the anion in 2. Carbon atoms are shown in
grey, copper in blue, oxygen in red and hydrogen in white. Sodium ions,
potassium ions and water molecules have been omitted for clarity. Ellipsoids
are plotted at the 75 % probability level.
Figure 6. Antiferromagnetic spin-coupling pathways in terms of Cu–μ-O–Cu
angles in 1. Note the two clearly resolved fields of angles: more obtuse angles
of about 130° that impart strong super-exchange in terms of the Gooden-
ough–Kanamori rules,[46] and angles of less than 110° that impart weaker
spin coupling. Note that an overall S = 0 ground state of the Cu8 cluster
results.
The Bis(diolato) Anions in M6[Cu(
2–5
L
-tartH–2-κ2O2,O3)2]·xH2O
The central copper(II) ions of 2–5 are part of two diolato che-
lates in a distorted square-planar environment. The bond
lengths between the copper(II) and alkoxo oxygen atoms are
similar, ranging from 1.890 (Cu3–O33 in 3) to 1.996 Å (Cu1–
O4 in 2). The distance between the next oxygen atoms of a
carboxylato group and the central copper atom in 3 and 5 is
rather large (dCu–O > 2.96 Å). The coordination in 2 and 4, how-
ever, can be regarded as a 4+2 environment due to Cu–O con-
tacts of about 2.5 Å along the Jahn–Teller-elongated axis. A
continuous shape measures analysis of the coordination envi-
ronment is given in the Supporting Information.[43,47] In sum-
mary, the tartratocuprate(II) ions show a slight distortion in the
square-planar base, caused by the incompatibility of the chiral
ligands and the intrinsic centrosymmetry of a square-planar
motif. As an illustration, the dihedral angle between the two
CuO2 triangles of the spiro motif around copper ranges from
3.7(7)° for the Cu1 atom in 3 to 18.2(2)° in the caesium salt.
Figure 8. Hydrogen bonding in 2 viewed along [001]. Carbon atoms are
shown in grey, copper in blue, oxygen in red, hydrogen in white, sodium in
green and potassium in magenta.
Conclusions
A species model for the copper(II)/L-(+)-tartaric acid (LH2, tartH2)
system in aqueous solution in the pH range from 1.9 to 12.3
has been established by potentiometric, UV/Vis spectroscopic
and solubility studies. Herein, seven tartrato–copper(II) species
The molecular structure of K2Na4[Cu(
-tartH–2-κ2O2,O3)2]· were identified and refined: CuLH, Cu2L2, Cu2L2H–1, Cu2L2H–2,
L
12H2O (2), which is also representative of 3–5, is depicted in Cu8L6H–10, Cu2L2H–4 and the major species in Fehling's solution,
Figure 7. Like 4 and 5, the crystals of 2 are stabilized by an CuL2H–4. As the focus of this work, structural information is pro-
extensive three-dimensional hydrogen-bonding system. A sec- vided for the major species of the neutral and alkaline regimes.
tion of the crystal structure is shown in Figure 8. (Owing to the From almost neutral solutions, the Cu8L6H–10 species, which was
poor crystallinity of 3, hydrogen atoms bonded to the solvent frequently formulated in previous solution studies, crystallized
oxygen atoms could not be located properly and therefore have as its lithium salt, Li7[Cu8(
been omitted.)
L
-tartH–2-κ4O1,O2,O3,O4)4(
L-tartH–1-
κ3O1,O2,O4)(H2O)6]NO3·19H2O.
Eur. J. Inorg. Chem. 2016, 1798–1807
www.eurjic.org
1803 © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
stants were refined by minimizing the objective function, the sum
of squares U. U is defined by Equation (1):
The main aim of this work was to reveal the structure of
Fehling's complex, the tartratocuprate complex of the strongly
alkaline region, the tentative CuL2H–4 species. Unfortunately,
potentiometry is hampered in the strongly alkaline media that
are obtained when Fehling's solution is prepared following the
typical recipes of the various pharmacopeias. Thus, UV/Vis spec-
troscopic as well as solubility studies on Cu(OH)2 in various me-
dia were conducted to overcome the limited reliability of the
potentiometric results. As the most powerful support, four salts
of the anion in question were structurally resolved, namely
U = Σi = 1,npWiri
(1)
in which Wi is a (diagonal) weight and ri is a residual (i. e., the
difference between observed and calculated pH).[36,37] To estimate
the standard deviations of the absorbances, an absorbance-error
function was determined by scanning a slightly acidic solution of
K2Cr2O7 (Grüssing, 99 %) about 20 times in the region 235–310 nm.
HypSpec was then able to calculate the weights and yielded a sin-
gle scale, σ, through which the quality of fit could be estimated.
Moreover, HypSpec adjusted the molar absorptivities and defined
the minimum number of absorbing main species by performing a
factor analysis (= in HypSpec a singular value decomposition of the
absorption matrix. The singular values should be positive or zero in
the absence of experimental errors. The number of positive values
is the same as the number of absorbing species, that is, the number
of independent columns in the data).[36] To reduce the number of
K2Na4[Cu(
κ2O2,O3)2]·9H2O (3), Na6[Cu(
Cs6[Cu(
-tartH–2-κ2O2,O3)2]·8H2O (5). Each copper(II) centre
L
-tartH–2-κ2O2,O3)2]·12H2O
-tartH–2-κ2O2,O3)2]·14H2O (4) and
(2),
Na6[Cu(L-tartH–2-
L
L
shows a distorted square-planar coordination environment of
the tartrate's alkoxido groups. The absorption maxima of the d–
d bands of these solids are in good agreement with the spectra
calculated from the solution analysis.[36,37]
Thus, the active species in Fehling's solution, which is re-
refining parameters, a 0.01
M
Cu(ClO4)2·6H2O solution was scanned
sponsible for the characteristic deep-blue colour, is the [Cu(L-
and the corresponding species “Cu” was set as “known”.
tartH–2-κ2O2,O3)2]6– anion.
Synthesis and Characterization of Compounds in General: All
reactions were carried out in aqueous solution under ambient con-
ditions. Solid compounds formed during the reaction or that did
not dissolve after a certain time were separated from the reaction
solution by filtration through a glass filter frit (G3 or G4) and/or by
centrifuging for 12 min at 3000 rpm. All chemicals purchased for the
synthesis of the compounds were used without further purification.
Table S2 in the Supporting Information shows all the chemicals,
their purity and their manufacturers.
Experimental Section
Combined Potentiometric and UV/Vis Titration of the Copper(II)
and -(+)-Tartaric Acid System in General: The combined potenti-
L
ometric and UV/Vis titration to determine the protonation constants
of the ligand and the stability constants of the formed copper(II)
complexes were carried out at T = (25.00 0.02) °C under nitrogen
in aqueous solution (four-necked Schlenk flask). The reagents
Copper(II) Hydroxide: Copper(II) sulfate pentahydrate (99.9 g,
0.400 mmol) was dissolved in boiling water. The blue solution was
cooled to 50 °C and, within 10 min, a 25 % NH4OH solution
(69.6 mL, 0.930 mol) was added dropwise with stirring, until the
reaction mixture became blue-violet. The greenish-blue precipitate
was allowed to settle and the slightly alkaline solution was sepa-
rated from the solid component by decantation. The precipitate
was washed with water (2 × 800 mL), suspended in water (600 mL)
and cooled to 10 °C. Sodium hydroxide (83.00 g, 2.075 mol) was
dissolved in water (337 mL) and added to the blue suspension un-
der continuous stirring at 10 °C. The light-blue precipitate was de-
canted and washed with water (600 mL), until the wash solution
had a neutral pH and was free of sulfate. Washing with acetone,
decanting (3 × 100 mL) and drying for 1 d at 45 °C gave copper(II)
hydroxide (27.1 g, yield: 69 %) as a homogeneous light-blue pow-
der, which was stored in a brown flask to protect from light.
Cu(ClO4)2·6H2O (Fluka, >98 %),
and NaClO4 (Riedel-de-Haën, pure, >97.5 %) were used without fur-
ther purification. A 1.0 NaOH solution (Fischer Chemical, factor
L-(+)-tartaric acid (Fluka, ≥99.5 %)
M
limits 0.999–1.001), which was standardized against potassium
hydrogen phthalate (Aldrich, ≥99.95 %) as a primary standard, func-
tioned as titre. To keep carbonate impurity in the basic solution as
low as possible, nitrogen was bubbled through it before the titra-
tion was started. The NaOH solution was added in known volume
increments (see Table S1 in the Supporting Information) to the sam-
ple solution by an automatic titrator system, Titrando 809, whereas
the pH was measured after each added volume with a combined
glass electrode with a fixed ground-joint diaphragm (unitrode). The
electrode was calibrated with standard buffers at pH 4.01, 7.00 and
9.00 (Merck). When the pH was constant for a longer time, a UV/Vis
spectrum was recorded by using a Cary50 conc UV/Vis spectrometer
with an optical probe (Hellma Analytics 661.702-UV, stainless steel,
optical path length 10 mm). This procedure was performed in the
wavelength region of 350–900 nm and in the pH region of 1.890–
12.235. Deionized, oxygen-free and carbonate-free water served as
a baseline.
Characterization of the Synthesized Compounds: To determine
the amount of carbon and hydrogen in the bulk sample, elemental
analyses were performed with Vario Micro cube and Vario EI instru-
ments. UV/Vis spectra were recorded with Cary50 conc UV/Vis (solu-
tion) and Cary500 Scan UV/Vis/NR spectrometers (solid samples).
Because only the diffuse reflectance spectra could be determined
on the latter spectrometer, the measured data was converted into
absorption data by using the Kubelka–Munk Equation (2):[48]
Preparation of the Initial Solution: Cu(ClO4)2·6H2O (667.3 mg,
1.801 mmol, 1.000 equiv.) and
5.400 mmol, 2.999 equiv.) were mixed together in freshly prepared
0.5 NaClO4 solution (180 mL) prepared in distilled water. After
L-(+)-tartaric acid (810.5 mg,
(1 – R∞)2
M
K
S
f(R∞) =
=
(2)
stirring for 0.5 h under a constant stream of nitrogen, the first UV/
Vis spectrum of the sample solution was recorded.
2R∞
Evaluation of Data: The potentiometric data (95 points) were ana-
in which R∞ is the relative reflectivity of an infinitely thick layer of
lysed by using the HypSpec program[36] (Protonic software, 2014), a sample, K is the absorption coefficient of the sample and S is the
which uses non-linear least-squares methods. The stability con-
scattering coefficient of the sample.
Eur. J. Inorg. Chem. 2016, 1798–1807
www.eurjic.org
1804
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
For measurement, the samples were mixed with barium sulfate to
minimize the intensity of the colour so that absorption follows the
Kubelka–Munk function. Owing to an automatic detector change
at 800 nm by the Cary500 Scan UV/Vis-NR spectrometer, small ab-
sorption jumps at this wavelength could not be prevented.
product was homogeneous (see Figure S26 in the Supporting Infor-
mation).
Synthesis of Na6[Cu(
(99.00 mmol, 3.960 g, 11.00 equiv.) and
L
-tartH–2-κ2O2,O3)2]·9H2O (3): NaOH
-(+)-tartaric acid
L
(16.00 mmol, 2.402 g, 1.778 equiv.) were carefully mixed together
in distilled water (7 mL) at 4 °C. The reaction mixture was heated
at 60 °C, until all components dissolved. Then copper(II) hydroxide
(9.00 mmol, 878 mg, 1.00 equiv.) in water (1 mL) was added drop-
wise and the resulting deep-blue suspension was stirred at 62 °C
for 1 h. To separate the blue solution from remaining copper(II)
hydroxide, the mixture was filtered and centrifuged. The concentra-
tion of the viscous deep-blue solution with KOH yielded 3 as blue
crystals after several weeks (2.85 g, yield: 48.3 %). The product was
very soluble in water (pH 12.5) and drying in vacuo resulted
in the loss of crystallinity. UV/Vis: λmax (H2O) = 671 nm (ε =
22.28 dm3 mol–1 cm–1); λmax(crystal) = 661 nm. C8H4CuNa6O12·9H2O
(655.73): calcd. C 14.65, H 3.38; found C 14.46, H 3.33. The powder
X-ray diffraction pattern gave good agreement with that calculated
on the basis of the single-crystal diffraction studies and indicated
the product was homogeneous (see Figure S25).
X-ray Crystallography: Crystals suitable for single-crystal X-ray dif-
fraction were selected by using a microscope (Leica MZ6 with polar-
ization filters), covered with paraffin oil and mounted either on a
micro mount or a loop. The measurements were performed at 100,
173 or 300 K with Oxford XCalibur 3 and d8Venture diffractometers
with graphite-monochromated Mo-Kα radiation (λ = 0.71073 Å). The
structures were solved by direct methods (ShelxS2014) and refined
by full-matrix, least-squares calculations on F2 (ShelxL2014).[49] All
non-hydrogen atoms were initially located and later refined aniso-
tropically; hydrogen atoms were refined isotropically. Suitable intra-
and intermolecular hydrogen bonds to locate hydrogen atoms on
the difference Fourier map in compounds 1, 2, 4 and 5 were esti-
mated by using Ortep-3 2.02.[50] Owing to the poor crystallinity of
3, hydrogen atoms bonded to solvent oxygen atoms could not be
located properly and therefore were omitted. In the case of 1, some
oxygen atoms are disordered and therefore the corresponding
hydrogen atoms were omitted. C–H bonds and the corresponding
angles were idealized (a C–H bond length of 1.000 Å and angles of
approximately 108.9° were assumed). Shelxle (version 7.25)[51] was
used as a graphical user interface for . Powder X-ray diffraction was
performed with a Huber G670 diffractometer with Co-Kα1 radiation
(λ = 1.788965 pm) and a Ge-111 monochromator at room tempera-
ture. The measurement range was 5.0–99.98°. The STOE WinXPow
software package (version 2.21)[52] was used to visualize the powder
diffraction data (GRAPHIC) and to calculate the theoretical pattern
from single crystal data (THEO).
Synthesis of Na6[Cu(
(99.00 mmol, 3.960 g, 11.00 equiv.) and
L
-tartH–2-κ2O2,O3)2]·14H2O (4): NaOH
-(+)-tartaric acid
L
(16.00 mmol, 2.402 g, 1.778 equiv.) were carefully mixed together
in distilled water (7 mL) at 4 °C. The reaction mixture was heated
at 60 °C, until all components dissolved. Then copper(II) hydroxide
(9.00 mmol, 878 mg, 1.00 equiv.) in water (1 mL) was added drop-
wise and the resulting deep-blue suspension was stirred at 62 °C for
2.5 h. To separate the blue solution from the remaining copper(II)
hydroxide, the mixture was filtered and centrifuged. Large blue
crystals of 4 were isolated after 1 h following the addition of eth-
anol (15 mL) to the reaction solution (5 mL) and diluting with water
(5 mL).
CCDC 1423088 (for 1), 1423089 (for 2), 1423090 (for 3), 1423091
(for 4) and 1423092 (for 5) contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre.
Synthesis of Cs6[Cu(
hydroxide monohydrate (50.0 mmol, 8.40 g, 8.33 equiv.) and
L
-tartH–2-κ2O2,O3)2]·8H2O (5): Caesium
-(+)-
L
tartaric acid (8.00 mmol, 1.20 g, 1.33 equiv.) were mixed together
in distilled water (4 mL) at 4 °C. The reaction mixture was warmed
to room temperature and copper(II) hydroxide (6.00 mmol, 585 mg,
1.00 equiv.) in water (1 mL) was added dropwise. The deep-blue
reaction mixture was stirred for 2 h at room temperature and centri-
fuged. The concentration of the deep-blue solution with KOH
yielded 5 as blue crystals after several weeks and, as side-products,
black CuO and red Cu2O (2.60 g). The product was very soluble in
water (pH 12.5), and a small amount seemed to decompose to
insoluble CuO; dried crystals of 4 are very hygroscopic and decom-
pose within a few hours. UV/Vis: λmax(H2O) = 671 nm; λmax(crystal) =
591sh, 669 nm.
Synthesis
of
Li7[Cu8(L L-tartH–1-
-tartH–2-κ4O1,O2,O3,O4)4(
κ3O1,O2,O4)2 (H2O)6]NO3·19H2O (1): A copper nitrate solution
(7 mL, 0.772 mol L–1, 5.40 mmol, 4.00 equiv.) was added dropwise
to a solution of L-(+)-tartaric acid (4.05 mmol, 608 mg, 3.00 equiv.)
and lithium hydroxide monohydrate (14.8 mmol, 624 mg,
11.0 equiv.) in distilled water (5 mL). The resulting deep-blue reac-
tion solution was stirred for 60 min at room temperature and fil-
tered. Methanol (1 mL) was added to the filtrate (1 mL). After some
months of slow evaporation at 4 °C, a few blue blocks of 1 among
an amorphous blue powder were collected. The product is very
soluble in water, and loss of crystallinity occurs after a few days at
room temperature.
Solubility Studies in General: Solubility studies were performed
in aqueous solution under ambient conditions. Copper hydroxide
Synthesis of K2Na4[Cu(
sodium -(+)-tartrate tetrahydrate (7.80 mmol, 2.20 g, 1.00 equiv.)
L
-tartH–2-κ2O2,O3)2]·12H2O (2): Potassium
L
was prepared as described above;
L-(+)-tartaric acid (Fluka, ≥99.5 %)
was dissolved in distilled water (5 mL) at 65 °C. Then sodium
hydroxide (12.5 mmol, 0.500 g, 1.60 equiv.), potassium hydroxide
(12.1 mmol, 0.800 g, 1.55 equiv.) and copper hydroxide (10.3 mmol,
1.00 g, 1.32 equiv.) were added stepwise to the clear solution. The
resulting deep-blue reaction mixture was stirred at 65 °C for another
20 min, filtered and centrifuged. Concentration of the viscous deep-
blue solution with KOH yielded 2 as blue rods after several weeks
(1.044 g, yield: 36.0 %). The product was very soluble in water
(pH 12.5) and a small amount seemed to decompose to insoluble
black CuO; drying in vacuo resulted in the loss of crystallinity. UV/
Vis: λmax(H2O) = 673 nm; λmax(crystal) = 656 nm. The powder X-ray
diffraction pattern gave good agreement with that calculated on
the basis of the single-crystal diffraction studies and indicated the
was used without any further purification. The 1.0
M
NaOH solution
(Fischer Chemical, factor limits 0.999–1.001) was standardized
against potassium hydrogen phthalate (Aldrich, ≥99.95 %) as a pri-
mary standard. Solid compounds were separated from the reaction
solution by using a glass filter crucible and a Büchner flask.
Study Considering CuH–4: Cu(OH)2 (292.7 mg, 3.000 mmol) was
suspended in 1
M NaOH (6 mL, 6.000 mmol) and stirred for 1 h. The
blue reaction mixture was filtered and the light-blue solid washed
with acetone and dried at 50 °C overnight. The filtrate was clear
and colourless, yield 291.7 mg.
Study Considering CuLH–4: Cu(OH)2 (292.7 mg, 3.000 mmol) and
L-(+)-tartaric acid (450.2 mg, 3.000 mmol) were suspended in 1 M
Eur. J. Inorg. Chem. 2016, 1798–1807
www.eurjic.org
1805
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[19] S. C. Khurana, I. J. Nigam, J. Inorg. Nucl. Chem. 1978, 40, 159–160.
[20] L. Meites, J. Am. Chem. Soc. 1949, 71, 3269–3275.
[21] V. T. A. Kereichuk, I. Churikova, Zh. Neorg. Khim. 1978, 23, 2436–2441.
[22] V. Simeon, K. Voloder, O. A. Weber, Anal. Chim. Acta 1969, 44, 309–313.
[23] K. J. Sircar, K. L. Yadava, J. Chem. Eng. Data 1982, 27, 231–233.
[24] J. H. Dunlop, R. D. Gillard, G. Wilkinson, J. Chem. Soc. A 1966, 1260–1264.
[25] L. Johansson, R. Larsson, Chem. Scr. 1975, 7, 67–75.
[26] L. Johansson, Chem. Scr. 1975, 7, 102–110.
[27] R. Treptow, Inorg. Chem. 1970, 9, 2583–2585.
[28] A. Cotton, Ann. Chim. Phys. 1896, 8, 347–432.
[29] G. S. Manku, A. N. Bhat, B. D. Jain, J. Inorg. Nucl. Chem. 1969, 31, 2533–
2543.
[30] Y.-H. Liu, S.-H. Lee, J.-C. Chiang, P.-C. Chen, P.-H. Chien, C.-I. Yang, Dalton
Trans. 2013, 42, 16857–16867.
[31] F. Jian, P. Zhao, Q. Wang, J. Coord. Chem. 2005, 58, 1133–1138.
[32] C. K. Prout, J. R. Carruthers, F. J. C. Rossotti, J. Chem. Soc. A 1971, 3336–
3342.
[33] H. Soylu, Hacettepe Bull. Nat. Sci. Eng. 1982, 61–73.
[34] J. H. L. Randy, D. Down, Environmental Instrumentation and Analysis
Handbook, Wiley, Hoboken, NJ, 2005, p. 471–472.
NaOH (6 mL, 6.000 mmol) and stirred for 1 h. The blue reaction
mixture was filtered and the light-blue solid washed with acetone
and dried at 50 °C overnight. A clear deep-blue solution and a light-
blue solid compound were isolated, yield 79.9 mg.
Study Considering CuL2H–4: Cu(OH)2 (292.7 mg, 3.000 mmol) and
L-(+)-tartaric acid (900.4 mg, 6.000 mmol) were dissolved in 1 M
NaOH (6 mL, 6.000 mmol) and stirred for 1 h. The deep-blue
solution was filtered and no remaining solid compounds were ob-
served.
Quantum Chemical Calculations and Continuous Shape Meas-
ures: All calculations were performed by using the Orca (version
3.0.3)[53] software package. Structures were optimized with Ahlrich's
latest version of the tzvp base set (in Orca: Def2/tzvp)[54] and with
five different functionals (Becke ′88 exchange and Perdew ′86 corre-
lation BP, the TPSS meta-GGA functional, the hybrid version of TPSS
named TPSSh, the one-parameter hybrid version of the Perdew–
Burke–Ernzerhof GGA functional named PBE0 and the popular
B3LYP functional)[55]. COSMO[56] was used to simulate aqueous con-
ditions. Atom-pairwise dispersion correction of the DFT energy was
performed with Becke–Johnson damping.[57] Found stationary
points were confirmed by subsequent frequency analyses at the
corresponding level of theory. Coordinates for Cu2rac-L2H–4 were
[35] A. E. Martell, R. J. Motekaitis, Determinations and Use of Stability Con-
stants, 2nd ed., Wiley, New York, 1992.
[36] P. Gans, A. Sabatini, A. Vacca, Ann. Chim. 1999, 89, 45–49.
[37] P. Gans, A. Sabatini, A. Vacca, Talanta 1996, 43, 1739–1753.
[38] O. Masson, B. Steele, J. Chem. Soc. Trans. 1899, 75, 725–734.
[39] J. Packer, I. Wark, J. Chem. Soc. Trans. 1921, 1348–1355.
[40] A. V. Ablov, G. A. Popovich, Zh. Obshch. Khim. 1954, 24, 974–978.
[41] R. Scholder, R. Felsenstein, A. Apel, Z. Anorg. Allg. Chem. 1933, 45, 138–
144.
obtained from the molecular structure of Li4[Cu2(rac-tartH–2
-
κ2O1,O2:κ2O3,O4)2]·11.75H2O previously described,[2] whereas the in-
itial structure for Cu2L2H–4 was prepared by using GaussView
v. 5.0.8.[58] Continuous shape measures were calculated by using
the SHAPE program established by Alvarez et al.[59]
[42] H. G. v. Schnering, Angew. Chem. Int. Ed. 1965, 4, 1051–1060; Angew.
Chem. 1965, 77, 1090–1090.
[43] S. Alvarez, P. Alemany, D. Casanova, J. Cirera, M. Llunell, D. Avnir, Coord.
Chem. Rev. 2005, 249, 1693–1708.
[44] J. Zhao, D.-S. Li, Y.-P. Wu, W.-W. Dong, Y.-Y. Wang, Q.-Z. Shi, Inorg. Chim.
Acta 2010, 363, 957–960.
[45] S.-M. Fang, Q. Zhang, M. Hu, E. C. Sañudo, M. Du, C.-S. Liu, Inorg. Chem.
2010, 49, 9617–9626.
Supporting Information (see footnote on the first page of this
article): Details of the combined potentiometric UV/Vis titration, OR-
TEP plots, structural data including bonds and angles, conforma-
tional analysis, hydrogen bonds and continuous shape measures,
UV/Vis spectra and powder diffraction data.
[46] J. B. Goodenough, Magnetism and the Chemical Bond, Wiley, New York–
London, 1963, vol. 1, p. 157–180.
[47] J. Cirera, P. Alemany, S. Alvarez, Chem. Eur. J. 2004, 10, 190–207.
[48] P. Kubelka, J. Opt. Soc. Am. 1948, 38, 1067.
Keywords: Copper · Fehling's solution · Stability constants ·
UV/Vis spectroscopy
[49]
G. M. Sheldrick, Acta Crystallogr., Sect. A 2008, 64, 112–122.
[50] L. J. Farrugia, J. Appl. Crystallogr. 2012, 45, 849–854.
[51] C. B. Hübschle, G. M. Sheldrick, B. Dittrich, J. Appl. Crystallogr. 2011, 44,
1281–1284.
[1] H. Fehling, Ber. Dtsch. Chem. Ges. 1848, 106–113.
[52] WinXPOW, v.2.21, STOE & Cie GmbH, Darmstadt, 2007.
[2] S. Albrecht, P. Klüfers, Z. Anorg. Allg. Chem. 2013, 639, 280–284.
[3] F. A. Cotton, G. Wilkinson, C. A. Murillo, M. Bochmann, Advanced Inor-
ganic Chemistry, 6th ed., Wiley, Chichester, UK, 1999, p. 869.
[4] The European Pharmacopeia Commission, in: Europäisches Arzneibuch,
Deutscher Apothekerverlag, 8th ed., Stuttgart, and Govi Verlag, Esch-
born, Germany, 2014, p. 700.
[53] F. Neese, Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2012, 2, 73–78.
[54] A. Schäfer, C. Huber, R. Ahlrichs, J. Chem. Phys. 1994, 100, 5829–5835.
[55] BP: A. D. Becke, Phys. Rev. A 1988, 38, 3098–3100; J. P. Perdew, Phys. Rev.
B 1986, 33, 8822–8824; TPSS and TPSSH: J. Tao, J. P. Perdew, V. N. Starov-
erov, G. E. Scuseria, Phys. Rev. Lett. 2003, 91, 146401; B3LYP: A. D. Becke,
Phys. Rev. A 1988, 38, 3098–3100; C. Lee, W. Yang, R. G. Parr, Phys. Rev. B
1988, 37, 785–789; PBE0: C. Adamo, V. Barone, J. Chem. Phys. 1999, 110,
6158–6170.
[56] S. Sinnecker, A. Rajendran, A. Klamt, M. Diedenhofen, F. Neese, J. Phys.
Chem. A 2006, 110, 2235–2245.
[57] A. D. Becke, E. R. Johnson, J. Chem. Phys. 2005, 122, 154101; E. R. John-
son, A. D. Becke. J. Chem. Phys. 2005, 123, 024101; E. R. Johnson, A. D.
Becke. J. Chem. Phys. 2006, 124, 174104; S. Grimme, J. Antony, S. Ehrlich,
H. Krieg, J. Chem. Phys. 2010, 132, 154104.
[58] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R.
Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nak-
atsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Mont-
gomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers,
K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari,
A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M.
Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli,
[5] US Pharmacopeial Convention, United States Pharmacopeia, 38 National
Formulary 33, Rockville, 2015, vol. 1, p. 1888.
[6] F. Bullnheimer, E. Seitz, Ber. Dtsch. Chem. Ges. 1899, 2347–2352.
[7] F. Bullnheimer, E. Seitz, Ber. Dtsch. Chem. Ges. 1900, 817–823.
[8] W. Traube, Ber. Dtsch. Chem. Ges. 1921, 54, 3220–3232.
[9] P. Pfeiffer, H. Simons, E. Schmitz, Z. Anorg. Allg. Chem. 1948, 256, 318–
342.
[10] R. Belford, R. Missavage, J. Chem. Soc., Chem. Commun. 1971, 1970–1971.
[11] R. J. Missavage, R. L. Belford, I. C. Paul, J. Coord. Chem. 1972, 2, 145–157.
[12] B. Y. N. D. Chasteen, R. L. Belford, N. Chasteen, Inorg. Chem. 1970, 9, 169–
175.
[13] J. Lefebvre, J. Chim. Phys. Phys. -Chim. Biol. 1957, 54, 601–619.
[14] E. Bottari, M. Vicedomini, J. Inorg. Nucl. Chem. 1971, 33, 1463–1473.
[15] K. Blomqvist, E. R. Still, Inorg. Chem. 1984, 23, 3730–3734.
[16] L. Johansson, Acta Chem. Scand. A 1980, 34, 495–506.
[17] J. Piispanen, L. H. J. Lajunen, Acta Chem. Scand. Sect. A 1995, 49, 241–
247.
[18] M. Rashidipour, Z. Derikvand, A. Shokrollahi, Z. Mohammadpour, A. Az-
adbakht, Arab. J. Chem. 2013, DOI: 10.1016/j.arabjc.2013.12.010.
Eur. J. Inorg. Chem. 2016, 1798–1807
www.eurjic.org
1806
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth,
P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B.
Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian 09, revision E.01,
Gaussian, Inc., Wallingford, CT, 2009.
[59] M. Pinsky, D. Avnir, Inorg. Chem. 1998, 37, 5575–5582.
Received: February 19, 2016
Published Online: March 31, 2016
Eur. J. Inorg. Chem. 2016, 1798–1807
www.eurjic.org
1807
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim