Isolation and structure of a hydrogen-bonded 2,2':6',2'-terpyridin-4'-one acetic acid adduct

Article abstract of DOI:10.1071/CH13571

Herein, we report the crystal structure of a key intermediate in the synthesis of 4'-substituted-terpyridines. Our findings confirm that the terpyridin-4'-one intermediate as generated from the condensation reaction of the corresponding triketone precursor with ammonium acetate is isolated as a hydrogen-bonded adduct with acetic acid, and not, as previously reported, as the acetate salt of a protonated pyridine nitrogen. This finding provides a rationale for the behaviour and structure of substituted terpyridin-4'-ones and pyridones in both the solid state and in solution. CSIRO 2014.

Full text of DOI:10.1071/CH13571

CSIRO PUBLISHING
Aust. J. Chem. 2014, 67, 651–656
Full Paper
http://dx.doi.org/10.1071/CH13571
Isolation and Structure of a Hydrogen-bonded 2,29:69,200-
Terpyridin-49-one Acetic Acid Adduct
A
A
A
A
Pas Florio, Campbell J. Coghlan, Chih-Pei Lin, Kei Saito, Eva M.
A
A
A B
,
Campi, W. Roy Jackson, and Milton T. W. Hearn
A
School of Chemistry, Monash University, Clayton, Vic. 3800, Australia.
Corresponding author. Email: milton.hearn@monash.edu
B
Herein, we report the crystal structure of a key intermediate in the synthesis of 40-substituted-terpyridines. Our findings
confirm that the terpyridin-40-one intermediate as generated from the condensation reaction of the corresponding triketone
precursor with ammonium acetate is isolated as a hydrogen-bonded adduct with acetic acid, and not, as previously
reported, as the acetate salt of a protonated pyridine nitrogen. This finding provides a rationale for the behaviour and
structure of substituted terpyridin-40-ones and pyridones in both the solid state and in solution.
Manuscript received: 21 October 2013.
Manuscript accepted: 27 November 2013.
Published online: 6 January 2014.
Introduction
larger scales and led to yields routinely in excess of ,53% of the
acetic acid adduct over two steps from ethyl picolinate. The
proton NMR spectrum of this compound showed a 1 : 1 ratio of
the terpyridin-40-one to acetic acid. The spectrum showed a dis-
tinct acetate singlet peak at 2.11ppm, integrating to exactly three
protons (1H NMR) as well as acetate peaks at 21.33 and
174.6 ppm in the carbon NMR spectrum. Microanalysis was also
consistentwitha 1 : 1 ratiooftheterpyridin-40-one 2toaceticacid.
Lunkwitz et al.[9] have reported that under the conditions
described by Constable and Ward,[8] the product isolated from
the reaction was the acetate salt 3 (Fig. 1) in which the nitrogen
of the central ring was protonated (although in this case, the
authors did not report recrystallization of their product). The
melting point of their ‘salt’ (128–1318C) was very similar to that
of the product obtained by us (132–1368C), again much lower
than that reported by Constable and Ward (166–1688C).[8] The
authors also reported that the only way to obtain the terpyridin-
40-one 2 in its free base form was to use ammonia as the nitrogen
source in the condensation reaction (Scheme 1).[9]
Terpyridines have attracted considerable interest,[1] particularly
as a class of ligand whose metal complexes have found broad
application, including but not limited to catalysis,[2] medicinal
chemistry,[3] medical imaging,[4] interactive polymers,[5] and
sensors.[6] We recently have pursued the synthesis of a variety of
substituted terpyridines as part of our investigations into their
use as ligands in protein fractionation via immobilized metal-ion
affinity chromatography (IMAC).[7] For this purpose, we ini-
tially adopted, in part, a synthetic strategy originally described
by Constable and Ward.[8] A key transformation in this synthesis
was the conversion of the triketone precursor 1 into the corre-
sponding terpyridin-40-one 2, in a condensation reaction that
utilizes ammonium acetate as the nitrogen source (Scheme 1).
Constable and Ward reported the isolation of the terpyridin-
40-one 2 as the free base, with a melting point of 166–1688C, in
80 % yield.[8] In our hands, however, this reaction has consis-
tently yielded a stable, highly crystalline solid with a sharp
melting point (132–1368C), which is significantly lower than
that reported by Constable and Ward. In order to verify the
identity of the product of this important reaction, the procedure
previously described by Constable and Ward[8] using ammoni-
um acetate was replicated (without recrystallization from etha-
nol) and crystals of the product subjected to detailed structural
analysis, including X-ray diffraction analysis. In addition, this
condensation reaction was also carried out using other ammoni-
um salts.
The synthesis of terpyridin-40-one 2 using ammonium ace-
tate, as shown in Scheme 1, has also been reported by Holbrey
et al.,[10] Potts and Konwar,[11] Murguly et al.,[12] and Wieprecht
et al.;[13] however, the melting point of the product was only
reported by Holbrey et al.[10] (165–1668C), and with no 1H NMR
data included.
X-Ray Crystallography
Single crystals of the terpyridin-40-one–acetic acid product
suitable for X-ray crystallography were obtained by recrystal-
lization of the product from dimethylformamide. The crystal
structure of this product shows a hydrogen-bonded adduct 4
(Fig. 2). The crystal structure shows the terpyridone was present
as its keto tautomer with the carbonyl oxygen hydrogen bonded
to the hydrogen of the hydroxy group of the acetic acid, with a
Results and Discussion
The reaction of the triketone 1 with ammonium acetate using the
conditions described in the Experimental section (i.e. without
recrystallization from ethanol) generated in these small-scale
reactions a white crystalline solid in 32% yield (based on the
isolation of only the first crop of crystals for the determination of
crystal structure). This reaction has been repeated many times at
[12]
˚
hydrogen bond length of 1.740 A. Further, Murguly et al.
Journal compilation Ó CSIRO 2014
www.publish.csiro.au/journals/ajc

652
P. Florio et al.
O
O
H-(N2) in the central ring to N(1) and N(3) in the peripheral
˚
pyridine rings are 2.302 and 2.271 A respectively, longer than in
the Murguly et al. structure and hence weaker. The bond angles
for N(1) ꢀ ꢀ ꢀ H–N(2) and N(3) ꢀ ꢀ ꢀ H–N(2) are 105.628 and
106.358 respectively. These angles are significantly less than
the optimal 1808 suggested in the IUPAC document defining the
hydrogen bond[14] and slightly less than the 1108 angle the
document gives as a preferred lower limit.
NH4OAc
O
O
N
H
EtOH, reflux
N
N
N
N
1
2
Scheme 1. The condensation reaction of triketone 1 with ammonium
acetate as reported by Constable and Ward.[8]
The bond lengths are all within normal ranges and similar to
those for the keto tautomer reported by Murguly et al.[12] There
is no p stacking and no other interactions were evident within
the packing arrangements of the crystal structure (Fig. 3).
Thus, the crystal structure of the product from the condensa-
tion of triketone 1 with ammonium acetate generates the acetic
acid adduct 4, which clearly shows the terpyridone acting as a
hydrogen-bond acceptor to acetic acid, and not as the acetate salt
of a terpyridone protonated on the nitrogen atom in the central
ring as previously described.[9] The carbonyl peak at 174.6 ppm
in the carbon NMR spectrum was also consistent with the
presence of acetic acid and not the acetate anion and in accord
with literature precedents of 176.9 (acetic acid) and 182.6
(acetate) ppm respectively for the carbon NMR spectrum of
acetic acid and acetate salts.[15]
O
Ϫ
OAc
ϩ
N
H
H
N
N
3
Fig. 1. Reported structure of the terpyridone acetate salt 3.
O(1)
O(2)
Moreover, we found that the ‘counterion’ could not be
washed out by organic solvents, i.e. the product could be washed
with diethyl ether as was also the case in the procedure of
Constable and Ward and crystals suitable for X-ray structure
determination obtained by cooling a warm DMF solution.
However, treatment of a chloroform solution of the acetic acid
adduct 4 with a saturated aqueous sodium bicarbonate solution
yielded the free base 2, in quantitative yield, with a melting point
(165–1698C) nearly identical in value to that reported by
H(2A)
N(3)
N(2)
O(3)
N(1)
1
Constable and Ward.[8] The H NMR spectrum for this com-
pound was consistent with pure terpyridin-40-one 2. The acetate
peak previously reported to exist for 4 was not present. In
addition, we subsequently found that recrystallization from
boiling absolute ethanol results in loss of the acetic acid from
the adduct and isolation of crystals of the free base, i.e. the same
product as reported by Constable and Ward[8] (as well as
Holbrey et al.[10] and also Potts and Konwar[11]).
Fig. 2. Crystal structure of the hydrogen-bonded complex 4 formed
between the terpyridin-40-one and acetic acid.
Despite the simplicity of the terpyridone-acetic acid adduct
4 as revealed from the crystal structure, this structure is unique
for an acid adduct. Hitherto, the only structure reported in the
literature related to the 4-pyridone or 4-hydroxypyridine com-
pound class involves an adduct of the 4-hydroxypyridine tauto-
mer with tartronic acid.[16] In this structure, an oxygen of the
carboxylate anion is hydrogen-bonded to the protonated pyr-
idinium cation. There is also a hydrogen bond from the other
oxygen of the carboxylate anion to the hydrogen of the phenolic
OH of the 4-hydroxypyridine.
have reported the crystal structure of the free base 2, which
exists as a dimer, composed of a 1 : 1 mixture of the keto and
enol tautomers. They reported a hydrogen bond length of
˚
1.702 A between the keto carbonyl and the enol hydroxyl
groups. In comparison, as found for the acetic acid adduct in the
current study, the bond lengths between atoms involved in the
hydrogen bonding between the acetic acid and the keto oxygen
˚
˚
(O(2)–H(2A) is 0.82 A, H(2A) ꢀ ꢀ ꢀ O(1) is 1.74 A, and O(2) to O
˚
(1) is 2.509 A) are shorter than expected. In addition, the bond
angle for O(2)–H(2A) ꢀ ꢀ ꢀ O(1) is 154.918, significantly smaller
than the optimum value of 1808 suggested in the International
Union of Pure and Applied Chemistry (IUPAC) document
defining the hydrogen bond.[14] These factors suggest that the
crystal packing forces are driving this pair far closer than the
ideal and hence this is a higher-energy H-bond than the analo-
gous bond in the 1 : 1 keto-enol crystal.
It should be noted that whereas the condensation reaction of
the triketone 1 with ammonium acetate furnished an acetic acid
adduct 4, this adduct could be directly employed in subsequent
reactions without causing any side reactions or a decrease
in product yields. These reactions included O-triflation, PCl5/
POCl3 halogenation, and O-alkylation with an alkyl bromide.
Lunkwitz et al.[9] have reported that different ammonium
salts can be used as a nitrogen source in the condensation
reaction (Scheme 1), resulting in the corresponding product
assigned as the terpyridone salt, including the formate, carbon-
ate, and bicarbonate. We thus carried out analogous reactions of
terpyridone 1 with ammonium formate and benzoate to ascertain
whether the product was isolated as the free base, a salt or as the
hydrogen-bonded adduct.
The cisoid orientation of the three nitrogen atoms in the
acetate complex was similar to their orientation in the keto
tautomer in the dimer reported by Murguly et al.[12] For exam-
ple, these workers concluded that strong intramolecular hydro-
gen bonds existed between the N–H ꢀ ꢀ ꢀ N of the keto tautomer,
˚
with N(2B) and N(3B) ꢀ ꢀ ꢀ H(1NB) having distances of 2.196A.
In the acetic acid adduct 4, the analogous bond lengths from

Crystal Structures of Terpyridone Adducts
653
Fig. 3. Packing arrangement of the hydrogen-bonded complex 4.
N(1)
N(3)
N(2)
O(1)
O(2)
N(4)
N(6)
Fig. 5. Packing arrangement of (a) the hydrogen bonded keto-enol dimer
of 2; and (b) 2, showing non-planarity of peripheral rings in the enol
tautomers.
N(5)
(Fig. 4) in contrast to the acetic acid adduct 4, i.e. there was no
evidence for the presence of either a formic acid adduct or a
formate salt of a protonated terpyridone. The crystal structure
shows a hydrogen-bonded dimer composed of an equimolar
mixture of the keto and enol tautomers of the free base 2, in
accord with the conclusion of Murguly et al.[12] but with sig-
nificant differences in the packing arrangements. The keto
tautomer again shows a cisoid orientation of the three nitrogen
atoms, strong NH ꢀ ꢀ ꢀ N intramolecular hydrogen bonds (2.205
Fig. 4. Crystal structure of the hydrogen-bonded keto-enol dimer of 2.
Ammonium formate gave crystals of a product 5 whose 1H
NMR spectrum showed a distinct formate proton peak at
8.37 ppm, integrating to one proton, as well as a formate
carbonyl peak at 164.65 ppm in the carbon NMR spectrum,
consistent with the presence of formic acid and not the formate
anion (lit.[15] 166.3 (formic acid) and 171.3 (formate) ppm
respectively). Microanalysis was also consistent with a 1 : 1
ratio of the terpyridin-40-one 2 to formic acid. The melting point
of 5 (115–1178C) was consistent with that reported by Lunkwitz
et al. (104–1148C crude, 120–1258C recrystallized) previously
described as the corresponding formate ‘salt’.[9]
˚
and 2.211 A respectively) from H(N2) in the central ring to N(1)
˚
and N(3) and a hydrogen bond (1.619 A) from the carbonyl
oxygen to the H of the enol tautomer. The enol tautomer shows a
transoid orientation of the three nitrogen atoms, consistent with
other findings made with the related 40-ethoxy-5,500-dimethyl-
terpyridine[17] and 40-amino-terpyridine.[18]
Fig. 5a and b shows the packing arrangement of 2 from
different axes, clearly different to the parallel sheets, herring-
bone pattern reported by Murguly et al. for the dimer.[12]
A
X-Ray Crystallography
notable difference is the non-planarity of the peripheral rings in
the enol tautomer (Fig. 5b). These packing arrangements may be
a direct consequence of the differences in the solvents used for
the crystallization of the free base 2, i.e. evaporative
The formate adduct 5 was recrystallized from acetonitrile to give
single crystals suitable for X-ray crystallography. Interestingly,
formic acid was not present in the crystal structure so obtained

654
P. Florio et al.
crystallization from an acetonitrile solution (present study)
compared with evaporative crystallization from chloroform
solution as previously employed. There is actually a CHCl3 of
crystallization present in the structure of Murguly et al.,[12] the
‘proton’ of which acts as a hydrogen-bond donor to the lone pair
electrons of the OH group of the enol tautomers. This contrasts
with the acetic acid adduct 4 in which the hydrogen bond is from
the acetic acid OH group to the oxygen of the keto tautomers.
There is no solvent present in our structure 2; hence a completely
different packing arrangement is not unexpected.
In general, pyridones have been shown to exist predominant-
ly as the keto tautomer, both in solution and in the solid
state.[12,19] In the absence of a suitable hydrogen-bond donor,
they have been reported to hydrogen bond to themselves, with
the pyridone NH serving as a hydrogen-bond donor to the keto
oxygen.[20,21] However, when presented with organic
Brønsted[22] or Lewis acids,[23] pyridones preferentially hydro-
gen-bond to the donor acid via the keto group, rather than
become protonated to form salts. This property is consistent
with the tendency of the keto group to hydrogen bond, possibly
due to the significant delocalization of the nitrogen lone pair
onto the keto oxygen increasing its Lewis basicity.
Conclusion
This study has examined the propensity of the condensation
reaction conditions stipulated by Constable and Ward[8] to fur-
nish a stable hydrogen-bonded adduct of the desired terpyridin-
40-one. This adduct has been demonstrated to consist of acetic
acid hydrogen-bonded to the keto tautomer of the terpyridin-4-
one, rather than the acetate salt of the protonated terpyridin-40-
one as has been previously reported. The structure also does not
involve a hydrogen bond from the acid carbonyl to a 4-hydro-
xypyridine moiety. These findings corroborate other reports that
pyridones will complex or hydrogen-bond to a suitable donor,
further stabilizing the thermodynamically stable keto tautomer.
This donor can be another terpyridin-40-one in its enol tauto-
meric state, or alternatively, a hydrogen-bonded complex to an
organic acid, such as acetic acid. Whereas crystals of the formic
acid and benzoic acid adducts suitable for X-ray analysis could
not be obtained, it is likely that these adducts are of a similar
structure to that found for the acetic acid adduct.
Experimental
General
From these results, it can be concluded that in the absence of
a suitable hydrogen-bond donor, a terpyridone will also hydro-
gen bond to itself. In contrast to simple monocyclic pyridones,
however, and perhaps for steric reasons, rather than the keto
group hydrogen bonding to the central pyridone NH, the
phenolic OH of the enol tautomer serves as hydrogen-bond
donor to the keto oxygen of the keto tautomer of the terpyridone
(as evident in Fig. 4). The current study has demonstrated, for
the first time, that in the presence of an organic acid, terpyridone
2, like pyridone itself, will also form a stable hydrogen-bonded
adduct 4 with acetic acid acting as a hydrogen-bond donor to the
terpyridone keto oxygen (Fig. 2), rather than form an acid–base
conjugate with the nitrogen in the central pyridine ring or with
either nitrogen in the peripheral rings.
The use of ammonium benzoate in the condensation reaction
(Scheme 1) has not been previously reported. Reaction of the
triketone 1 with ammonium benzoate gave a crystalline solid 6
whose 1H NMR spectrum showed aromatic signals integrating
for a 1 : 1 ratio to the 40-terpyridone and whose 13C NMR
spectrum showed a benzoate carbonyl peak at 170.18 ppm
consistent with the presence of benzoic acid and not the benzo-
ate anion (lit.[15] 172.6 (benzoic acid) and 177.6 (benzoate) ppm
respectively). Microanalysis was also consistent with a 1 : 1 ratio
of the terpyridin-40-one 2 to acid. However, crystals suitable for
X-ray analysis have not yet been obtained.
Nuclear magnetic resonance spectra were recorded at 300 MHz
(1H) or 75 MHz (13C) with a Bruker DPX-300 spectrometer. The
NMR spectra refer to solutions in deuterated base-washed
(Na2CO3) CDCl3 where tetramethylsilane (TMS) was used as
the internal standard (d 0.00 ppm) for 1H spectra and the residual
solvent peak used as an internal reference in 13C NMR spectra.
Chemical shift values (d) are given in parts per million (ppm)
relative to the residual solvent peak (or TMS) and coupling
constants are given in Hertz (J, Hz). Infrared spectra were
recorded on a Perkin Elmer 1600 series Fourier-transform
infrared spectrometer. Low-resolution electrospray ionization
mass spectra (in positive mode (ESIþ)) were recorded on a
Micromass Platform II API QMS electrospray mass spectrom-
eter. Melting points were determined using a Buchi B-545 or a
Gallenkamp melting point apparatus with a digital thermometer
and are uncorrected. Reagents and compounds were purchased
from Sigma–Aldrich, Castle Hill, NSW. Microanalyses were
performed by the Campbell Microanalytical Laboratory,
Department of Chemistry, University of Otago, Dunedin. The
acetic acid adduct was recrystallized from DMF to furnish
crystals suitable for X-ray analysis. The formic and benzoic acid
adducts were recrystallized from acetonitrile.
2,20:60,200-Terpyridin-40-(10H)-one acetic acid adduct 4
A solution of 1,5-bis(20-pyridyl)pentane-1,3,5-trione 1 (2.0 g,
7.46 mmol) and excess ammonium acetate (4.0 g, 51.89 mmol,
6.95 equiv.) in absolute ethanol (50 mL) was heated at reflux for
6 h. The resulting dark-brown solution was concentrated under
vacuum to half the original volume and cooled to room tem-
perature. The solid that precipitated was collected by filtration
and washed with diethyl ether to furnish 4 as an off-white
crystalline solid (0.74 g, 32 %), mp 132–1368C (lit.[9] 128–
1318C). dH (300 MHz, CDCl3) 2.11 (s, 3H, OAc), 7.38 (s, 2H,
H30, H50), 7.43 (ddd, J 7.5, 4.8, 1.2, 2H, H5, H500), 7.88 (apparent
td, J ,7.8, 1.5, 2H, H4, H400), 7.96 (d, J 8.1, 2H, H3, H300), 8.78
(d, J 4.8, 2H, H6, H600). dC (75 MHz, CDCl3) 21.33 (CH3),
113.31 (C30, C50), 120.52 (C3, C300), 125.19 (C5, C500), 137.64
(C4, C400), 144.59 (C20, C60), 148.34 (C2, C20þ0), 149.42 (C6,
C600), 174.57 (COOH), 181.58 (C40). m/z (ESI ) 521.1 [(free
base)2 þ Na]þ (100 %), 499 [(free base)2 þ H]þ (73 %), 272.1
The electrospray ionization mass spectra (ESI-MS), run in
positive mode, for the terpyridone adducts 4, 5, and 6, and the
free base 2 showed peaks that corresponded to either the free
base 2 or to its dimer. Whereas hydrogen-bonded complexes
have been observed by mass spectroscopy, their observation has
been found to be sensitive to the conditions of spectra acquisi-
tion, reflecting the strength of the non-covalent interac-
tion.[24,25] In the current study, we have not been able to
observe similar hydrogen-bonded complexes in the ESI-MS
with our terpyridin-40-one adducts. What is interesting is that the
Mþ ion peak corresponding to the self-associating dimer is
prominent in the mass spectra of the complexes as well as the
free base. This finding is consistent with previous observa-
tions[12] of the preferred self-association via intramolecular
hydrogen bonding of the terpyridine monomer, both in solution
and the solid state.

Crystal Structures of Terpyridone Adducts
655
[free base þ Na]þ (7 %), 250.1 [free base þ H]þ (16 %). Anal.
Calc. for C17H15N3O3: C 66.01, H 4.89, N 13.58. Found:
C 66.14, H 4.72, N 13.80 %.
(5 mL). The solution was dried over MgSO4, filtered, and con-
centrated to furnish 2 as an off-white solid after drying under
high vacuum at ambient temperature (201 mg, ,100 %), mp
165–1698C.(lit.[8] 166–1688C; lit.[9] 153–1588C). dH (300 MHz,
CDCl3) 7.12 (s, 2H, H30, H50), 7.43 (ddd, J 7.2, 4.8, 1.2, 2H, H5,
H500), 7.88 (apparent td, J ,7.8, 1.8, 2H, H4, H400), 7.95 (d, J 7.8,
2H, H3, H300), 8.79 (apparent dt, J 4.5, ,1.5, 2H, H6, H600). dC
(75 MHz, CDCl3) 113.86 (C30, C50), 120.59 (C3, C300), 125.21
(C5, C500), 137.77 (C4, C400), 144.37 (C20, C60),þ 148.90
(C2, C200), 149.68 (C6, C600), 181.53 (C40). m/z (ESI ) 521.1
[(M)2 þ Na]þ (95 %), 499 [(M)2 þ H]þ (100 %), 272.1 [M þ
Na]þ (5 %), 250.1 [M þ H]þ (23 %).
2,20:60,200-Terpyridin-40-(10H)-one formic acid adduct 5
A solution of the triketone 1 (2.0 g, 7.46 mmol) and excess
ammonium formate (3.27 g, 51.86 mmol, 6.95 equiv.) in abso-
lute ethanol (50 mL) was heated at reflux for 6 h. The resulting
dark-brown solution was concentrated under vacuum to
approximately half the original volume and cooled to room
temperature. The resulting crystalline solid was collected by
filtration and washed with diethyl ether to furnish an off-white
solid (1.93 g). The crude solid was suspended in chloroform and
filtered to furnish 5, after concentration of the filtrate, as a beige
solid (1.05 g, 48 %), mp 115–1178C (lit.[9] 104–1148C crude,
120–1258C, recryst.). dH (300 MHz, CDCl3) 7.35 (s, 2H, H30,
H50), 7.46 (ddd, J 7.5, 4.8, 1.2, 2H, H5, H500), 7.91 (apparent td, J
,7.8, 1.5, 2H, H4, H400), 7.98 (apparent dt, J 7.8, ,1.1, 2H, H3,
H300), 8.37 (s, 1H, HCOOH), 8.80 (apparent dt, J 4.8, ,1.2, 2H,
H6, H600). dC (75 MHz, CDCl3) 113.32 (C30, C50), 120.91 (C3,
C300), 125.68 (C5, C500), 138.02 (C4, C400), 145.23 (C20, C60),
148.35 (C2, C200), 149.79 (C6, C600), 164.65 (COOH), 181.59
(C40). m/z (ESIþ) 521.1 [(free base)2 þ Na]þ (95 %), 499 [(free
base)2 þ H]þ (100 %), 272.1 [free base þ Na]þ (6 %), 250.1
[free base þ H]þ (24 %). Anal. Calc. for C16H13N3O3: C 65.08,
H 4.44, N 14.23. Found: C 65.21, H 4.15, N 14.48 %.
Recrystallization of the acetic acid adduct 4 from boiling
absolute ethanol also resulted in the isolation of the free base
2,20:60,200-terpyridin-40-(10H)-one (2). The melting point and
spectral data (1H and 13C NMR) were consistent with those
reported above.
X-ray Crystal Structure Determinations
Single crystals of 2 and 4, suitable for X-ray diffraction, were
coated in viscous oil and mounted on a glass fibre. Full spheres
(2ymax 50–558) of CCD area-detector diffractometer data were
measured (Bruker X8 Apex II CCD, f- and v-scans, mono-
˚
chromatic Mo Ka radiation, l ¼ 0.71073 A) using the Bruker
Apex2 software suite,[26] yielding Nt reflections, these merging
to N unique data (Rint quoted) after empirical and multi-scan
absorption correction (SADABS).[27] All structures were solved
by standard methods and refined by full-matrix least-squares on
F2 (SHELX97)[27] with anisotropic displacement parameter
forms for non-hydrogen atoms. Plausible positions of hydrogen
atoms attached to N or O were located in the difference Fourier
map and were refined with N–H and O–H distances restrained to
2,20:60,200-Terpyridin-40-(10H)-one benzoic acid adduct 6
A solution of the triketone 1 (2.0 g, 7.46 mmol) and excess
ammonium benzoate (7.22 g, 51.89 mmol, 6.95 equiv.) in
absolute ethanol (50 mL) was heated at reflux for 6 h. Initially an
attempt was made to crystallize the product directly from the
crude reaction mixture; however, the propensity of ammonium
benzoate to readily crystallize in ethanol made the standard
procedure of fractional crystallization difficult to perform. The
resulting dark-brown solution was concentrated under vacuum
and the mixture resuspended in ethyl acetate. The mixture was
filtered, and the excess ammonium benzoate collected and
washed with additional ethyl acetate (3 ꢁ ,10 mL). The com-
bined filtrate was concentrated, and the resulting brown solid
was recrystallized from ethanol to yield a solid that was col-
lected by filtration and washed with diethyl ether to furnish 6 as
an off-white solid (1.14 g, 41 %), mp 110–1138C. dH (300 MHz,
CDCl3) 7.39 (s, 2H, H30, H50), 7.43–7.49 (m, 4H, H5, H500,
benzoic acid, meta,), 7.58 (d, J 7.5, 1H, benzoic acid, para), 7.90
(apparent td, J ,7.8, 1.5, 2H, H4, H400), 8.00 (d, J 8.1, 2H, H3,
H300), 8.16 (d, J 6.9, 2H, benzoic acid, ortho), 8.85 (d, J 4.5, 2H,
H6, H600). dC (75 MHz, CDCl3) 113.67 (C30, C50), 120.86 (C3,
C300), 125.56 (C5, C500), 128.45 (benzoic acid, meta), 130.23
(benzoic acid, ortho, ipso), 133.05 (benzoic acid, para), 137.91
(C4, C400), 144.93 (C20, C60), 148.66 (C2, C20þ0), 149.70 (C6,
C600), 170.18 (COOH), 181.83 (C40). m/z (ESI ) 521.1 [(free
base)2 þ Na]þ (74 %), 499 [(free base)2 þ H]þ (100 %), 272.1
[free base þ Na]þ (6 %), 250.1 [free base þ H]þ (24 %). Anal.
Calc. for C22H17N3O3: C 71.15, H 4.61, N 11.31. Found: C
71.20, H 4.53, N 11.39 %.
˚
reasonable values (0.88–0.98 A) apart from that of the O–H in 4,
which was left to ride unrestrained; all other hydrogen atoms
were placed in calculated positions and constrained to ride on
the parent atoms with Uiso(H) ¼ 1.2.Ueq(C).
2,20:60,200-Terpyridin-40-(10H)-one Acetic Acid Adduct 4
C17H15N3O3, M ¼ 309.32, Cambridge Crystallographic Data
Centre (CCDC) 870316. Colourless rectangular, 0.30 ꢁ 0.20
ꢁ 0.10 mm3, monoclinic, space group P21/c (no. 14), a
˚
¼ 10.3271(4), b ¼ 5.2838(2), c ¼ 26.6702(11) A, b ¼ 91.255
3
˚
(2)8,
V ¼ 1454.95(10) A ,
Z ¼ 4,
Dc ¼ 1.412 g cmꢂ3
,
F000 ¼ 648, T ¼ 123(1)K, 2ymax ¼ 55.08, 11802 reflections col-
lected, 3332 unique (Rint ¼ 0.0556). Final GooF ¼ 1.056,
R1 ¼ 0.0502, wR2 ¼ 0.1129, R indices based on 2311 reflections
with I . 2sigma(I) (refinement on F2), 213 parameters, 0
restraints. Lp and absorption corrections applied,
m ¼ 0.099 mmꢂ1
.
2,20:60,200-Terpyridin-40-(10H)-one 2
C15H11N3O, M ¼ 249.27, CCDC 870317. Colourless rectan-
gular, 0.40 ꢁ 0.20 ꢁ 0.10 mm3, monoclinic, space group C2/c
˚
(no. 15), a ¼ 28.9398(8), b ¼ 8.0158(2), c ¼ 22.3854(7)ꢂA3,
3
˚
b ¼ 111.237(3)8, V ¼ 4840.2(2) A , Z ¼ 16, Dc ¼ 1.368 g cm
,
F000 ¼ 2080, T ¼ 123(1)K, 2ymax ¼ 55.08, 17129 reflections
collected, 5496 unique (Rint ¼ 0.0463). Final GooF ¼ 1.053,
R1 ¼ 0.0489, wR2 ¼ 0.1123, R indices based on 4214 reflections
with I . 2sigma(I) (refinement on F2), 353 parameters, 0
restraints. Lp and absorption corrections applied,
2,20:60,200-Terpyridin-40-(10H)-one 2
The terpyridone–acetic acid adduct 4 (250 mg, 0.81 mmol) was
dissolved in chloroform (2 mL) and washed with saturated
sodium bicarbonate (5 mL) and saturated sodium chloride
m ¼ 0.089 mmꢂ1
.

656
P. Florio et al.
Supplementary Material
[9] R. Lunkwitz, G. Pabst, G. Scherr, US Patent 6 784 296 2004.
[10] J. D. Holbrey, G. J. T. Tiddy, D. W. Bruce, J. Chem. Soc., Dalton
Selected interatomic distances for 2 and 4 and torsion angles for
the enol tautomer of 2 are available from the Journal’s website.
Crystallographic information files for 2 and 4 have been
deposited with the Cambridge Crystallographic Data Centre and
assigned the deposit codes 870317 and 870316 respectively.
These can be obtained free of charge via http://www.ccdc.cam.
ac.uk/data_request/cif (verified 5 December 2013).
Trans. 1995, 1769. doi:10.1039/DT9950001769
[11] K. T. Potts, D. Konwar, J. Org. Chem. 1991, 56, 4815. doi:10.1021/
JO00015A050
[12] E. Murguly, T. B. Norsten, N. Branda, J. Chem. Soc., Perkin Trans. 2
1999, 2789. doi:10.1039/A904323C
[13] T. Wieprecht, J. Xia, U. Heinz, J. Dannacher, G. Schlingloff, J. Mol.
Catal. A 2003, 203, 113. doi:10.1016/S1381-1169(03)00406-0
[14] E. Arunan, G. R. Desiraju, R. A. Klein, J. Sadlej, S. Scheiner,
I. Alkorta, D. C. Clary, R. H. Crabtree, J. J. Dannenberg, P. Hobza,
H. G. Kjaergaard, A. C. Legon, B. Mennucci, D. J. Nesbitt, Pure Appl.
Chem. 2011, 83, 1637. doi:10.1351/PAC-REC-10-01-02
[15] E. Pretsch, J. Seibl, T. Clerc, W. Simon, Tables of Spectral Data for
Structure Determination of Organic Compounds 2nd Edn 1989
(Springer-Verlag: Berlin).
Acknowledgements
Financial support as an ARC Linkage Grant from the Australian Research
Council (ARC) and Novo Nordisk A/S, which included a scholarship (to
CJC), is gratefully acknowledged. We wish to thank Priscilla Johnston for
assistance with the crystal structures and Lachlan Schwarz for assistance
with the graphics.
[16] T. Fukunaga, S. Kashino, H. Ishida, Acta Crystallogr. 2004, C60,
o718.
References
[17] R. A. Fallahpour, M. Neuburger, M. Zehnder, Polyhedron 1999, 18,
2445. doi:10.1016/S0277-5387(99)00147-3
[18] R. A. Fallahpour, M. Neuburger, M. Zehnder, New J. Chem. 1999, 23,
[1] U. S. Schubert, H. Hofmeier, G. R. Newkome, Modern Terpyridine
Chemistry 2006 (Wiley-VCH: Weinheim).
53. doi:10.1039/A806690F
[2] A. Winter, G. R. Newkome, U. S. Schubert, ChemCatChem. 2011, 3,
1384. doi:10.1002/CCTC.201100118
[3] J. B. Stimmel, F. C. Kull, Nucl. Med. Biol. 1998, 25, 117. doi:10.1016/
S0969-8051(97)00151-0
[4] J. Costa, R. Ruloff, L. Burai, L. Helm, A. E. Merbach, J. Am. Chem.
Soc. 2005, 127, 5147. doi:10.1021/JA0424169
[5] I. Eryazici, O. k. Farha, O. C. Compton, C. Stern, J. T. Hupp,
S. T. Nguyen, J. Chem. Soc., Dalton Trans. 2011, 9189.
[6] A. D’Ale´o, E. Cecchetto, L. De Cola, R. M. Williams, Sensors 2009, 9,
3604. doi:10.3390/S90503604
[19] P. Beak, Acc. Chem. Res. 1977, 10, 186. doi:10.1021/AR50113A006
[20] P. G. Jones, Acta Crystallogr. 2001, C57, 880.
[21] A. Tyl, M. Nowak, J. Kusz, Acta Crystallogr. 2008, C64, 661.
[22] C. Li, P. D. Robinson, J. D. Dyer, Acta Crystallogr. 2006, C62, o336.
[23] D. Cook, Can. J. Chem. 1963, 41, 515. doi:10.1139/V63-071
[24] K. Yamaguchi, J. Mass Spectrom. 2003, 38, 473. doi:10.1002/
JMS.488
[25] J. H. Clark, M. Green, R. Madden, C. D. Reynolds, Z. Dauter,
J. M. Miller, T. Jones, J. Am. Chem. Soc. 1984, 106, 4056. doi:10.1021/
JA00326A042
[7] F. H. Arnold, Nat. Biotechnol. 1991, 9, 151. doi:10.1038/NBT0291-151
[8] E. C. Constable, M. D. Ward, J. Chem. Soc., Dalton Trans. 1990, 1405.
doi:10.1039/DT9900001405
[26] Bruker Apex2 v2.0 2005 (Bruker AXS: Madison, WI).
[27] G. M. Sheldrick, Acta Crystallogr. 2008, A64, 112.