Complexes of a porphyrin-like N4-donor Schiff-base macrocycle

Article abstract of DOI:10.1039/c3dt33057e

A metal-free 16-membered N₄-donor [1 + 1] Schiff-base macrocycle was isolated as HL^Pr·2acid, by 1:1 condensation of diphenylamine-2,2′-dicarboxaldehyde (1), dipropylene triamine and an excess of either acetic or formic acid, or as HL^Pr·(tosylic acid) when just one equivalent of tosylic acid was employed. Interestingly, the acid-free synthesis employed for the 14-membered analogue HL^Et failed to generate pure HL^Pr macrocycle, but nevertheless the crude product obtained was able to be used in subsequent complexation reactions to form five mononuclear complexes: Zn^IIL^Pr(BF₄) ·H₂O·0.5IPA (where IPA is isopropylalcohol), [Cu ^IIL^Pr](BF₄), [Ni^IIL ^Pr](BF₄), [Co^IIL^Pr](BF ₄)·0.5H₂O, [Fe^IIIL^Pr(NCS) ₂]·1.5H₂O. Crystal structure determinations show that, like the HL^Et analogues, [Ni^IIL^Pr] (BF₄) features a square planar N₄ coordinated Ni ^II centre and [Fe^IIIL^Pr(NCS) ₂]·0.15MeOH·0.2H₂O features an octahedral N₆ coordinated Fe^III centre (two NCS anions bound axially). In both cases the N₄-donor macrocycle is bound equatorially to the metal ion. Cyclic voltammograms of the BF₄ complexes were carried out in MeCN vs. 0.01 mol L^-1 AgNO₃/Ag and revealed multiple redox processes. The Zn^II complex exhibits multiple ligand-centered redox processes. Interestingly, [Ni^IIL ^Pr](BF₄) has two reversible redox processes, at E _m = +0.38 (ΔE = 0.06 V) and -1.7 V (ΔE = 0.06 V), whereas the previously reported analogue [Ni^IIL^Et](BF ₄) had a process at E_pc = +0.59 V with only a weak return wave. Likewise, [Cu^IIL^Pr](BF₄) has a reversible process, at E_m = -1.17 V (ΔE = 0.06 V) plus a process at E_pc = +0.45 V, whereas previously reported [Cu^IIL ^Et](BF₄) only featured irreversible processes, with the oxidation occurring at E_pc = +0.50 V.

Full text of DOI:10.1039/c3dt33057e

Dalton
Transactions
View Article Online
View Journal
PAPER
Complexes of a porphyrin-like N4-donor Schiff-base
macrocycle†
Cite this: DOI: 10.1039/c3dt33057e
Rajni K. Wilson (née Sanyal) and Sally Brooker*
A metal-free 16-membered N4-donor [1 + 1] Schiff-base macrocycle was isolated as HLPr·2acid, by 1 : 1
condensation of diphenylamine-2,2’-dicarboxaldehyde (1), dipropylene triamine and an excess of either
acetic or formic acid, or as HLPr·(tosylic acid) when just one equivalent of tosylic acid was employed. Inter-
estingly, the acid-free synthesis employed for the 14-membered analogue HLEt failed to generate pure
HLPr macrocycle, but nevertheless the crude product obtained was able to be used in subsequent com-
plexation reactions to form five mononuclear complexes: ZnIILPr(BF4)·H2O·0.5IPA (where IPA is isopropyl-
alcohol), [CuIILPr](BF4), [NiIILPr](BF4), [CoIILPr](BF4)·0.5H2O, [FeIIILPr(NCS)2]·1.5H2O. Crystal structure
determinations show that, like the HLEt analogues, [NiIILPr](BF4) features a square planar N4 coordinated
NiII centre and [FeIIILPr(NCS)2]·0.15MeOH·0.2H2O features an octahedral N6 coordinated FeIII centre (two
NCS anions bound axially). In both cases the N4-donor macrocycle is bound equatorially to the metal ion.
Cyclic voltammograms of the BF4 complexes were carried out in MeCN vs. 0.01 mol L−1 AgNO3/Ag and
revealed multiple redox processes. The ZnII complex exhibits multiple ligand-centered redox processes.
Interestingly, [NiIILPr](BF4) has two reversible redox processes, at Em = +0.38 (ΔE = 0.06 V) and −1.7 V
(ΔE = 0.06 V), whereas the previously reported analogue [NiIILEt](BF4) had a process at Epc = +0.59 V with
only a weak return wave. Likewise, [CuIILPr](BF4) has a reversible process, at Em = −1.17 V (ΔE = 0.06 V)
plus a process at Epc = +0.45 V, whereas previously reported [CuIILEt](BF4) only featured irreversible
processes, with the oxidation occurring at Epc = +0.50 V.
Received 20th December 2012,
Accepted 28th January 2013
DOI: 10.1039/c3dt33057e
www.rsc.org/dalton
Advantages of the macrocyclic approach include: control of
nuclearity and solubility, enhanced thermodynamic stability
Introduction
The reports by Curtis,1 Busch2 and Jäger3 in the early 1960s of the (macrocyclic effect), and the ability to fine-tune the metal ion
metal ion template syntheses of tetradentate Schiff-base macro- environment and stabilise redox products. It is well estab-
cycles led to wide adoption of template cyclisation methods and lished that the degree of unsaturation and the size of the
an explosion of interest in such systems.4 This interest was macrocycle have a pronounced effect on the relative stability of
further fuelled, especially ‘downunder’,5 by the first metal ion the various oxidation states available to the encapsulated tran-
templated synthesis of dimetallic Schiff-base macrocycles sition metal ion.20,29,30 For example, an electrochemical study
reported by Robson in 1970.6 Such macrocyclic complexes7–11 of a family of Ni(II) complexes of Me6[16]aneN4, [15]aneN4 and
continue to be actively sought and developed due a myriad of [14]aneN4 (Fig. 1, macrocycle size indicated by [xx]) revealed
potential uses, for example in medicine (e.g., PDT, MRI),12–14 as that these N4-donor macrocycles could stabilise the unusual
catalysts,15–17 bioinorganic mimics15,18–21 or single molecule Ni(III) and Ni(I) oxidation states.31 Further, the Ni2+/Ni3+ couple
magnets,22–26 as well as their relationship to naturally occurring shifted in favour of Ni(III) as the ring size decreased (E1/2
=
macrocycles such as porphyrins, chlorins and corrins.27,28
+1.3, +0.90 and +0.69 V for Me6[16]aneN4, [15]aneN4 and
[14]aneN4 respectively) whereas the Ni+/Ni2+ couple shifted in
favour of Ni(I) as the ring size was increased (E1/2 = −1.70,
−1.57 and −1.40 V for [14]aneN4, [15]aneN4 and [16]aneN4
respectively).31 Changing the macrocycle size can also shift the
position of the Ni(II) d–d band. For example, a pronounced red
shift was observed in Ni([16]aneN4)Br2 (416 nm) vs. Ni([14]-
aneN4)Br2 (370 nm, smaller ring and stronger ligand field).32
However, the effect is not always so pronounced: for example,
in [Cu(Bzo2[14]aneN4)]-(ClO4)2 vs. [Cu(Bzo2[16]aneN4)](ClO4)2
Department of Chemistry and MacDiarmid Institute for Advanced Materials and
Nanotechnology, University of Otago, PO Box 56, Dunedin 9054, New Zealand.
E-mail: sbrooker@chemistry.otago.ac.nz; Fax: +64 3 479 7906; Tel: +64 3 4797919
†Electronic supplementary information (ESI) available: NMR spectra; IR spectra;
UV-Vis spectra; full electrochemical information; detailed results of the CSD
L
survey; perspective view of the cation of [NiIILPr](BF4) and [FeIII Pr(NCS)2·0.15-
MeOH·0.2H2O. CCDC 913820 and 913821. For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/c3dt33057e
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 1 Selected literature examples of 14- (dotted box), 15- and 16- (solid box)
membered N4-donor macrocycles, as complexes with the chelate rings sizes
noted, along with the 16-membered macrocycle of interest in the present study,
[MLPr x+
] .
Fig. 2 The Schiff-base macrocycle HLPr made without acid was impure but was
suitable for metallation as shown. The nickel(II) complex can also be prepared by
metal template cyclisation. Z is not always present.
the Cu(II) d–d band position was reported to be unshifted
within experimental error, 490 nm vs. 491 nm.33
[NiLEt](ClO4), [NiLPr](ClO4), Cu(HLEt)(ClO4), [CuLPr](ClO4),
[Co(HLEt)]-(ClO4) and [CoLPr](ClO4)·H2O. None of these complexes
were structurally, electrochemically or magnetically character-
ised. We subsequently reported that a family of mid-to-late
transition metal complexes of HLEt exhibits interesting struc-
tural, redox and magnetic behaviour.38 In particular, [CoIILEt]-
(BF4)·H2O and FeIIILEt(BF4)2·2H2O·MeCN have low-spin and
intermediate-spin states, respectively, consistent with the
14-membered (LEt)− macrocycle providing a strong ligand field,
similar to that imposed by the porphyrins (16-membered).
Hence the impact of employing the larger, 16-membered,
macrocycle HLPr on the properties of the resulting mid-to-late
transition metal complexes has been investigated and we
report the results of this study here.
A search of the CSD (version 5.33)34 for 16-membered N4-
donor macrocycles (i.e. porphyrin-sized, Fig. 1) with “any
bond” type containing four-coordinate Cu(II), Ni(II) or Co(II), or
four/five/six-coordinate Fe(III), but excluding “porphyrin” and
“corrin” structures, resulted in a total of only 38 hits featuring
6666-chelate rings (see ESI, Table S1 and Fig. S1†). Of these 38
structures there are no four, five or six coordinate Fe(III) com-
plexes. Of the four-coordinate complexes, 2 are of Co(II), 13 are
of Cu(II) and the remaining 23 are of Ni(II). Both of the 4-coor-
dinate Co(II) complexes have tetrahedral Co(II) geometries; it is
worth noting that further searches revealed that there are
3 five-coordinate low-spin Co(II) complexes (of tetra-imine
macrocycles) which are more relevant to the present study (see
ESI, Table S1 and Fig. S2†).35–37 All 23 of the four coordinate
Ni(II) complexes are square planar (hence diamagnetic):
remarkably 2 of them are of all-amine macrocycles and 1 is of
a two-imine macrocycle. The remaining 20 are of all-imine
macrocycles. Another point of interest is that further searches
Results and discussion
Synthesis of metal-free macrocycle
showed that there are only 3 Cu(II) and 6 Ni(II) six-coordinate Unlike for HLEt,38 the synthesis of the metal-free Schiff-base
complexes, and no five-coordinate examples.
macrocycle HLPr proved to be non-trivial. Utilizing the simple
We are currently exploring the properties of complexes of experimental protocol that worked well for HLEt, i.e. refluxing
porphyrin-like N4-donor Schiff-base macrocycles38 (and larger a 1 : 1 mixture of 1 : diamine in MeOH for 3 hours before
N6-donor macrocycles39) derived from the head unit diphenyl- rotary evaporating off the solvent and drying in vacuo, resulted
amine-2,2′-dicarboxaldehyde40 (1, Fig. 2). Prior to our work, mostly in the desired Schiff-base macrocycle (NH, 5 aromatic
some complexes of HLEt (14-membered ring) and HLPr (16- and 3 alkyl signals observed as expected in the H NMR spec-
1
membered ring, Fig. 1 and 2) were prepared from 1 and diethy- trum in CDCl3), however some dialdehyde always remained
lene triamine or dipropylene triamine, respectively, by the (Fig. S3†). Whilst the resulting oil proved suitable for sub-
metal template method by the groups of Lewis41 and sequent metallations (Fig. 2), further efforts were made to
Black.42,43 Black42 reported that attempts to prepare the metal- obtain HLPr cleanly.
free Schiff-base macrocycles HLEt and HLPr failed, resulting
It is well established that Schiff-base formation is pH
instead in polymeric materials. Hence they used the metal dependent, with the optimal pH usually being slightly acidic.44
template cyclization method, and reported the formation of Hence, the condensation was repeated in the presence of acid,
Dalton Trans.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
using a similar protocol to one employed in the literature (in CvO stretch of unreacted aldehyde. In all four cases the for-
MeOH at RT, then taken to dryness).45 The acids employed mation of the imine is clear, with a strong sharp CvN stretch
were acetic, formic and tosylic, which have pKa values of 4.79, at 1627–1628 cm−1. The spectra of HLPr·2HOAc, HLPr·2HCO2H
3.77, and −2.80 respectively in H2O.46 With an excess of either and HLPr·C7H8SO3 also have signals pertaining to the respect-
acetic or formic acid (one equivalent was insufficient, see ive organic acids, for example the peaks present at 605 and
ESI†), the resulting bright red oil gave a clean but complicated 626 cm−1 (in neat acetic acid) are now present at 615 and
1H NMR spectrum in CDCl3 (Fig. S4–S5†), consistent with the 648 cm−1 in HLPr·2HOAc. Similarly the peaks present at 561,
formation of the desired Schiff base macrocycle (no signals for 672 and 813 cm−1 (in neat tosylic acid) are now present at 567,
any starting materials). The ratio of the acid to macrocycle 681 and 822 cm−1 in HLPr·C7H8SO3. The sharp CvO stretch
signals is 2 : 1 in both cases, implying that the macrocycles present at 1703 cm−1 in neat acetic acid shifts to lower energy,
have been isolated as HLPr·2HOAc and HLPr·2HCO2H, respect- appearing as a weak shoulder at 1689 cm−1. The broad
ively. This was subsequently confirmed by elemental analysis 1686 cm−1 CvO stretch in neat formic acid presumably also
results (Tables S2 and S3†).
shifts to lower frequencies, so may be obscured by the imine
However, the 1H and 13C NMR spectra obtained in CDCl3 stretch of the macrocycle.
both feature two sets of signals, in 1 : 1 ratio, which appear to
Conductivity measurements on HLPr·C7H8SO3 were per-
correspond to separate molecules as no cross-coupling peaks formed in MeOH, DMF and DMSO, giving molar conductivities
are observed between the two sets. A possible reason for this is of 60, 45 and 23 Ω−1 cm2 mol−1 respectively (literature range
that one set of signals (NH, 5 overlapping aromatic and 3 alkyl for a 1 : 1 electrolyte 70–90 Ω−1 cm2 mol−1 in MeOH, 65–90
signals) belongs to HLPr and the other set of signals (NH, 5 Ω−1 cm2 mol−1 in DMF and 23–42 Ω−1 cm2 mol−1 in DMSO).47
overlapping aromatic and 3 alkyl signals) arises from either a All of these values are at the low end or slightly lower than the
hydrogen bonded species HLPr(HOAc)x or a bigger ring-size of literature ranges for 1 : 1 electrolytes (possibly due to the large
1
macrocycle, for example (2 + 2). When the H NMR spectra of size of the cation), but indicate that this compound exists
HLPr·2HOAc and HLPr·2HCO2H are instead obtained in the mostly as (H2LPr)+ and C7H8SO3− ions in these solvents.
protic solvent CD3OD (Fig. S6–S7†) a single set of 5 overlapping
peaks in the aromatic region and 3 sharp peaks in the propy-
Synthesis of macrocyclic complexes
lene region is observed (with the expected cross-correlation
peaks), consistent with the hydrogen-bonding hypothesis. The mononuclear macrocyclic complexes of (LPr)− were syn-
Furthermore,
a
variable-temperature 1H NMR study on thesised as shown in Fig. 2. For reactions involving metallation
HLPr·2HCO2H in CDCl3 showed that on increasing the temp- of the free macrocycle, the crude macrocycle prepared without
erature to 323 K, one set of peaks sharpens whilst the other set acid template was utilized. This gave, after vapour diffusion of
broadens out, indicating that the two compounds undergo diethyl ether into the concentrated reaction solutions,
fluxional changes on different timescales (Fig. S12†). Also, the ZnIILPr(BF4)·H2O·0.5IPA (where IPA is isopropyl alcohol) as a
ratio of the sharper peaks to the broadened peaks is 2 : 1 at dark orange crystalline solid (57%), [CuIILPr](BF4) as orange
323 K whereas at 298 K the ratio is 1 : 1, and at 252 K, the ratio crystals (55%) and [CoIILPr](BF4)·0.5H2O as a black crystalline
is 1 : 2, so they appear to be in equilibrium with one another. solid (71%). [FeIIILPr(NCS)2]·1.5H2O was obtained as a dark
On cooling, the broad triplets at lower chemical shift at 323 K green solid (45% yield) which precipitated out of the reaction
separated into two sets of broad triplets at 252 K.
solution. An attempted synthesis of the zinc(II) complex in pyri-
In contrast, when one equivalent of the stronger acid tosylic dine, as was successfully carried out for the HLEt analogue,38
acid was employed, HLPr·C7H8SO3 precipitated as a pale yellow resulted in an oil with low percentages of C, H and N so this
solid, in 35% yield. The 1H NMR spectrum of a freshly pre- complex was instead prepared in IPA. The attempted synthesis
pared suspension of this pale yellow solid in (CD3)2SO showed of the HLPr analogue of the dark green intermediate spin [FeIII-
the same complexity as that of the acetic and formic acid LEt](BF4)2·2H2O·MeCN,38 using the same protocol as for the
experiments. However, over a period of 2–3 days the suspen- HLEt complex (in 1 : 1 DCM : MeCN), resulted instead in a dark
sion slowly dissolved to give a pale yellow solution which gave red solid with low microanalytical data (Table S4 and
a simple 1H NMR spectrum, with only trace amounts of the Fig. S19†). Repeating the reaction under argon and then
signals belonging to the other species (Fig. S8–S10†). The ratio opening it up to air to allow oxidation to FeIII resulted in a
of the acid to macrocycle signals is 1 : 1 and this was sub- similar solid with low microanalytical data (Table S5 and
sequently confirmed by elemental analysis results. Single crys- Fig. S20†), so this was not pursued further. The nickel
tals of this compound have not been forthcoming.
complex, [NiIILPr](BF4), could be prepared either by metallation
The infrared spectra of HLPr·2HOAc, HLPr·2HCO2H and of the crude macrocycle, or by nickel(II) templated [1 + 1]
HLPr·C7H8SO3 (Fig. S17 and S18, ESI†) showed that no precur- Schiff-base condensation of 1 with dipropylene triamine in
sors are present (no NH-stretch at 3360 and 3287 cm−1; no refluxing MeOH.
CvO stretch at 1674 cm−1) whereas whilst that of the crude
The four tetrafluoroborate complexes are readily soluble in
HLPr macrocycle (synthesized without addition of acid) pre- CH3CN, DMF and CH3NO2 and partially soluble in CH3OH.
dominantly contains signals pertaining to the desired product, The neutral molecule [FeIIILPr(NCS)2]·1.5H2O is readily soluble
a small shoulder at ca. 1683 cm−1 likely corresponds to the in CH2Cl2, CH3CN and CHCl3 and partially soluble in CH3OH.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.

View Article Online
Paper
Dalton Transactions
1013–1052 cm−1 and 519–521 cm−1, due to the tetrafluoro-
borate anion.
The UV-Vis spectra of all these very intensely coloured com-
plexes, in MeCN (Table 1, Fig. S23†), were characterised by
intense charge transfer and π–π* transitions between
300–550 nm. At longer wavelengths, no further bands are
observed for ZnIILPr(BF4)·H2O·0.5IPA [346 (4168), 460 (2980)
and 527 nm (1429 dm3 mol−1 cm−1)]. In contrast, [CuIILPr]-
(BF4), [NiIILPr](BF4), [CoIILPr](BF4)·0.5H2O and [FeIIILPr(NC-
S)2]·1.5H2O all have an additional transition, at 808 (496), 615
(1701), 673 (219) and 772 nm (2626 dm3 mol−1 cm−1), respect-
ively. Whilst some of these bands could be d–d in origin, for
example the 808 nm band for [CuIILPr](BF4) and 673 nm for
Fig. 3 1H NMR spectrum of [NiIILPr](BF4) in CD3CN.
The H and 13C NMR spectra obtained for [NiIILPr](BF4) in [CoIILPr](BF4)·0.5H2O, some/all of them are likely to be LMCT
CD3CN were assigned (Fig. 3 and S21†) on the basis of chemi- bands, which effectively obscure any d–d bands in the spectra.
cal shift, peak area integration and multiplicity, and two The effect of increasing the macrocycle size from HLEt to HLPr
dimensional gCOSY, HSQC and gHMBC techniques. Upon is to red shift the bands observed in the visible spectra of the
coordination of nickel(II) any symmetry the metal-free macro- NiII and CuII complexes (Fig. 4 and S24†): [NiIILEt](BF4)·H2O at
cycle (HLPr) might have had is lost, resulting in separate 545 nm vs. 615 nm for [NiIILPr](BF4); [CuIILEt](BF4)·H2O at
signals for every proton in the macrocycle. In contrast to the 690 nm vs. 808 nm for [CuIILPr](BF4).
1
situation with the NMR spectra of the organic species (above)
Room temperature effective magnetic moments (μeff) were
all of these signals correspond to one species, as there are obtained for all of the complexes. A value close to the spin
many cross-correlations in the 2D spectra. The two distinct
imine signals occur at 7.70 and 7.26 ppm. It was also found
that there is no apparent tendency for this complex to become
paramagnetic by binding additional donors as the addition of
a drop of pyridine to the NMR sample did not alter the signals
observed.
Table 1 Comparison of λmax/nm (ε/dm3 mol−1 cm−1), in MeCN, of the com-
plexes of HLEt and HLPr
Complex
LEt
LPr
ZnIIL(BF4)a
[CuIIL](BF4)b
325 (7350), 390 (9340),
460 (13 860)
325 (6610), 355 (6620),
388 (7019), 465 (11 970),
690 (450)
320 (10 300), 410 (4380),
455 (9085), 545 (2290)
420 (2231), 490 (3495),
600 (915)
346 (4168), 460 (2980),
527 (1429)
326 (5854), 386 (4643),
470 (10 209), 808 (496)
The positive ion electrospray mass spectra of ZnIILPr(BF4)·
H2O·0.5IPA, [CuIILPr](BF4) and [NiIILPr](BF4) complexes show a
common peak, the fragment [MLPr]+ which is associated with
the loss of the tetrafluoroborate anion whereas the mass
spectra of [CoIILPr](BF4)·0.5H2O and [FeIIILPr(NCS)2]·1.5H2O
have a peak due to [MLPr-H]+, associated with loss of an
additional proton and the tetrafluoroborate or thiocyanate co-
ligand.
[NiIIL](BF4)c
[CoIIL](BF4)d
[FeIIIL(NCS)2]e
326 (6815), 409 (3739),
480 (8286), 615 (1701)
408 (1397), 494 (2546),
673 (219)
323 (5869), 379 (7020),
423 (6301), 772 (2626)
410 (7210), 870 (1750)
The molar conductivity values in MeCN for [CoIILPr]-
(BF4)·0.5H2O [NiIILPr](BF4), [CuIILPr](BF4) and ZnIILPr(BF4)·
H2O·0.5IPA, of 137, 153, 185 and 156 Ω−1 cm2 mol−1, respect-
ively, are in or slightly above the literature range for a 1 : 1
electrolyte in MeCN (120–160 Ω−1 cm2 mol−1, but values as
high as 199 Ω−1 cm2 mol−1 have been used to characterise 1 : 1
electrolytes in MeCN).47 The molar conductivity for
[FeIIILPr(NCS)2]·1.5H2O in MeCN was 12 Ω−1 cm2 mol−1 con-
sistent with it being neutral and non-ionic in MeCN.
a [ZnIILEt(py)](BF4) and ZnIILPr(BF4)·H2O·0.5IPA. b [CuIILEt](BF4)·H2O
and [CuIILPr](BF4). c [NiIILEt](BF4)·H2O and [NiIILPr](BF4). d [CoIILEt]-
(BF4)·H2O
and
[CoIILPr](BF4)·0.5H2O.
e [FeIII
L
Et(NCS)2]
and
[FeIII Pr(NCS)2]·1.5H2O.
L
The infrared spectra (Fig. S22†) of all five complexes had
imine stretches in the range 1611–1629 cm−1, mostly at
somewhat lower frequencies than in crude HLPr (1628 cm−1).
In the analogous complexes of the smaller HLEt macrocycle
the imine stretch occurred at higher frequencies, in the range
1633–1644 cm−1 (cf. 1629 cm−1 for the metal-free macrocycle
HLEt). The CuN stretch of the thiocyanate ions in [FeIIILPr(NC-
S)2]·1.5H2O is observed at 2101 cm−1 and is consistent with
terminal N-bound thiocyanate (literature range ≥2050 cm−1),48
which was subsequently confirmed by an X-ray crystal struc-
Fig. 4 UV-Vis spectra of complexes in CH3CN at 0.1 mmol L−1: [CuIILPr](BF4)
(solid line) and [CuIILEt](BF4)·H2O (dashed line). Insert: expansion of the
ture determination. The other complexes showed two bands, 600–1100 nm region.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
only value for one unpaired electron (1.73 BM) was observed
for [CuIILPr](BF4) (1.76 BM). The magnetic moment for [NiIILPr]-
(BF4) was observed to be 0 BM, consistent with the diamag-
netic NMR spectrum and a square planar geometry, as seen
for all 23 related examples (Table S1†). The [CoIILPr]-
(BF4)·0.5H2O complex μeff value (2.31 BM) is within the range
(2.1–2.9 BM) expected for low-spin CoII at RT, as seen for the
analogous complex [CoIILEt](BF4)·H2O (2.06 BM)38 and
reported for [Co(phthalocyanine)] and [Co(salen)]:49 this is a
first for N4-donor 16-membered macrocycles featuring less
than 4 imine donors (Table S1†). The [FeIIILPr(NCS)2]·1.5H2O
complex has μeff = 2.02 BM, consistent with it being a low-spin
octahedral FeIII complex, as seen for the analogous complex
[FeIIILEt(NCS)2] (2.55 BM).38 These values lie above the spin
only moment expected for one unpaired electron (μso = 1.73
BM) due to orbital contributions. These results clearly show
that despite being 16-membered HLPr provides a strong ligand
field similar to that imposed by 14-membered HLEt, as it gives
the same set of spin states including that of low-spin CoII and
low-spin FeIII-thiocyanate complexes.
Table 2 Comparison of selected bond lengths [Å], angles [°] and other para-
meters (oop means out of plane) for [NiIILPr](BF4) and [NiIILEt](BF4)
LEt
LPr
Bond length [Å]
Ni(1)–N(1)
Ni(1)–N(2)
Ni(1)–N(3)
Ni(1)–N(4)
1.869(3)
1.861(7)
1.924(4)
1.830(7)
1.867(2)
1.894(2)
1.942(2)
1.899(2)
Bond angle [°]
N(1)–Ni(1)–N(2)
N(1)–Ni(1)–N(4)
N(3)–Ni(1)–N(2)
N(3)–Ni(1)–N(4)
N(1)–Ni(1)–N(3)
N(2)–Ni(1)–N(4)
Twist of phenyl rings [°]
cis N–Ni–N angles [°]
trans N–Ni–N angles [°]
Ave. Ni–N distance [Å]
Ni oop from N4 plane [Å]
Nimine oop of attached phenyl ring [Å]
94.9(3)
95.3(3)
85.2(3)
84.6(3)
179.3(4)
169.8(19)
47.3
84.6–95.3
169.8–179.3
1.871
89.48(10)
90.51(10)
85.98(10)
94.26(10)
174.78(10)
174.32(9)
63.4
86.0–94.3
174.3–174.8
1.901
0.002
0.136, 0.182
0.028
0.050, 0.066
donors), imposing a square planar environment. The NiII ion
is located 0.028 Å out of the mean plane defined by the N4
donor atoms.
X-ray crystal structures
Crystal structure determinations were carried out on [NiIILPr]-
(BF4) and [FeIIILPr(NCS)2]·0.15MeOH·0.2H2O. Both complexes
are monometallic, but the NiII ion is in a square planar N4-
donor environment whereas the FeIII ion is in an octahedral
N6-donor geometry.
The shortest Ni–N bond distance involves the deprotonated
nitrogen atom, Ni(1)–N(1) [1.867(2) Å] which is located in the
negatively charged portion of the macrocycle. The two crystal-
lographically independent Ni–Nimine bond lengths are equiva-
lent within experimental error and are of intermediate lengths,
Ni(1)–N(2) [1.894(2) Å] and Ni(1)–N(4) [1.899(2) Å]. The
Ni–Namine bond is the longest, Ni(1)–N(3) [1.942(2) Å]. Overall
the structure is similar to that of [NiIILEt](BF4) (Table 2) although,
except for Ni–N1, the Ni–N bond lengths are longer and the
twist angle between the phenyl rings is greater in [NiIILPr](BF4).
The increase from a 6655 chelate ring size pattern in [NiIILEt]-
(BF4) to a 6666 pattern in [NiIILPr](BF4) (Fig. 1) has also led to
the pair of head unit moiety cis N–Ni–N angles being much
closer to 90° (89.48, 90.51°) than they could be in the (LEt)−
analogue (94.9, 95.3°), despite these being 6-membered in
both complexes. A more detailed analysis of the structures of
[NiIILPr](BF4) and [NiIILEt](BF4) is presented in the ESI
(Fig. S62–S63 and Tables S13–S14†), and is consistent with the
low symmetries revealed by the NMR spectra (Fig. 3).
Dark red-brown single crystals of [NiIILPr](BF4) were
obtained by diethyl ether vapour diffusion into a MeOH solu-
tion of the sample and the crystal structure determined
(Fig. 5, Table 2). The complex crystallizes in the monoclinic
space group P21/n with the entire complex in the asymmetric
unit. The monoanionic tetradentate macrocyclic ligand is
bound to the metal through all four nitrogen atoms (one
amine, one deprotonated amine and two imine nitrogen
Black block crystals of [FeIIILPr(NCS)2]·0.15MeOH·0.2H2O
were obtained by slow evaporation of a MeOH solution of
[FeIIILPr(NCS)2]·1.5H2O and the crystal structure determined
(Fig. 6, Table 3). In addition to the macrocycle, two axially
coordinated (trans) thiocyanate anions are also bound to the
FeIII centre, forming an octahedral N6 coordination environ-
ment. There are two entire [FeIIILPr(NCS)2] complexes in the
asymmetric unit. The Fe(1) complex has no disorder whereas
in the Fe(2) complex both NCS ions and a central part of the
dipropylene triamine alkyl chain are disordered over two sites
(occupancy of 0.8 : 0.2).
Fig. 5 Perspective view of the cation of [NiIILPr](BF4) (bold line) overlaid with
that previously reported for [NiIILEt](BF4) (dashed line). Non acidic hydrogen
−
Again the Fe–N1 (deprotonated amine) bond is the shortest
and the Fe–Namine bond is the longest (Table 3). The four
atoms and BF4 anions omitted for clarity. Atoms used to fit the two structures
are numbered.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.

View Article Online
Paper
Dalton Transactions
pair of cis N–Fe–N angles made by the head unit moiety drop
from an average of 95.6° to 88.73°, closer to 90°.
Electrochemistry
Electrochemical studies were carried out on 1 mmol L−1 MeCN
solutions of the four tetrafluoroborate complexes vs.
a
0.01 mol L−1 AgNO3/Ag reference electrode. In this cell the
Fc/Fc+ couple occurs at Em 0.079 V with ΔE 78 mV. Multiple
redox events occurred in all cases (Table 4): both the CuII and
NiII complex showed some reversible processes (Fig. 7–9 and
Table 4).
There are multiple redox processes present in the cyclic vol-
tammogram (CV) of ZnIILPr(BF4)·H2O·0.5IPA, which are attribu-
ted to ligand based processes as ZnII is redox inactive in this
range of potentials.
The CVs (Fig. 7 and 8) of [CuIILPr](BF4) show a clear oxi-
dation process at an Epc of +0.45 V which has weaker, but dis-
tinct, return waves at Epa = +0.39 V, +0.30 V and +0.20 V
(Fig. S29, Table S6†). On scanning to negative potentials a
reversible reduction process at Em = −1.17 V with ΔE = 0.06 V
is observed (Fig. 8, Tables 4 and S8,† Epa = −1.20 V, Epc = −1.14
V). Controlled potential coulometry experiments, starting from
fresh bright orange solutions, at +0.57 V (removal of 0.78 e−
equivalents) and −1.30 V (addition of 2.3 e− equivalents)
resulted in colour changes to forest green and dark purple
(Fig. S28†), respectively. In both cases subsequent CVs were
considerably altered (see ESI, Fig. S33, S38 and S39†), likely
due to these corresponding to chemically irreversible ligand
redox events. Details of these experiments are provided in the
ESI (Fig. S26–S40, Tables S6–S8†).
Fig. 6 Perspective view of one of the two independent molecules of
[FeIIILPr(NCS)2]·0.15MeOH·0.2H2O (bold line) overlaid with that previously
reported for [FeIIILEt(NCS)2]·NO2Me (dashed line). Non acidic hydrogen atoms
and solvent of crystallization omitted for clarity. Atoms used to fit the two struc-
tures are NI, N2, N3, N4 and Fe1.
Table 3 Comparison of selected bond lengths [Å], angles [°] and other para-
meters (oop means out of plane) for [FeIIILPr(NCS)2]·0.15MeOH·0.2H2O and
[FeIIILEt(NCS)2]·NO2Me
LEt
LPr
Bond length [Å]
Fe(1)–N(1)/Fe(2)–N(8)
Fe(1)–N(2)/Fe(2)–N(9)
Fe(1)–N(3)/Fe(2)–N(10)
Fe(1)–N(4)/Fe(2)–N(11)
Fe(1)–N(50)/Fe(2)–N(70)
Fe(1)–N(60)/Fe(2)–N(80)
1.858(5)
1.912(5)
1.969(5)
1.912(5)
1.939(6)
1.949(6)
1.869(3), 1.883(3)
1.962(3), 1.966(3)
2.067(4), 2.068(3)
1.961(3), 1.964(3)
1.932(3), 1.933(3)
1.931(3), 1.931(3)
The CVs of [CuIILPr](BF4) and [CuIILEt](BF4)·H2O in MeCN
(Fig. S41†) show considerable differences, especially with
regard to the presence of a reversible redox process at −1.20 V
in the CV of the present complex. The oxidative process at
+0.50 V in [CuIILEt](BF4)·H2O shifts to a slightly lower value of
+0.45 V for [CuIILPr](BF4). The CV’s obtained after coulometry
experiments are also significantly different. For example, the
quasi-reversible process observed at +0.31 V after coulometry
for [CuIILEt](BF4)·H2O is not observed for [CuIILPr](BF4).
The CVs (Fig. 7 and 9) of [NiIILPr](BF4) reveal a reversible
oxidation process, at Em = +0.35 V with ΔE = 0.06 V is observed
(Table 4 and ESI, Table S9,† Epc = +0.38 V, Epc = +0.32 V). This
occurs at a lower potential than in the analogue [NiIILEt]-
(BF4)·H2O (Epc = +0.59 Epa = +0.54 V), as observed for the above
pair of copper(II) complexes. On scanning to negative poten-
tials, a reversible reduction process at Em = −1.67 V with ΔE =
0.06 V is observed (Tables 4 and S10,† Epa −1.70 V, Epc −1.64 V).
Coulometry experiments, starting with fresh bright green
solutions, at +0.51 V (removal of 0.84 e−) and at −1.80 V
(addition of 1.1 e−) resulted in colour changes to dark blue
Bond angle [°]
N(1)–Fe(1)–N(2)/
N(8)–Fe(2)–N(9)
95.6(2)
95.6(2)
84.7(2)
84.1(2)
177.9(3)
168.7(2)
88.98(13), 89.07(13)
88.57(14), 88.31(14)
91.51(14), 92.44(13)
90.96(14), 90.17(14)
178.94(14), 178.39(14)
177.44(14), 176.78(14)
N(1)–Fe(1)–N(4)/
N(8)–Fe(2)–N(11)
N(3)–Fe(1)–N(2)/
N(10)–Fe(2)–N(9)
N(3)–Fe(1)–N(4)/
N(10)–Fe(2)–N(11)
N(1)–Fe(1)–N(3)/
N(8)–Fe(2)–N(10)
N(2)–Fe(1)–N(4)/
N(9)–Fe(2)–N(11)
Twist of phenyl rings [°]
cis N–Fe–N angles [°]
trans N–Fe–N angles [°]
50.0
84.1–95.6
168.8–177.9
58.2, 61.9
88.6–91.5, 88.3–92.4
177.4–178.9,
176.8–178.4
1.965, 1.970
0.002, 0.021
0.056, 0.187; 0.146,
0.193
Ave. Fe–N distance [Å]
Fe oop from N4 plane [Å]
Nimine oop of attached phenyl
ring [Å]
1.913
0.004
0.145, 0.153
crystallographically independent Fe–Nimine bond lengths are and dark red (Fig. S52†), respectively. In both cases, the sub-
equivalent within experimental error but are longer than in the sequent CVs were considerably altered (Fig. S47, S48 and
(LEt)− analogue (Table 3). The change in chelate ring size from S53†). Details of these experiments are provided in the ESI
6655, in [Fe(LEt)(NCS)2], to 6666, in [Fe(LPr)(NCS)2] (Fig. 1), once (Fig. S42–S55, Tables S9–S12†).
again causes the phenyl rings of the head unit moiety to be
more highly twisted to facilitate optimal coordination, and the [NiIILEt](BF4)·H2O (Fig. S55†), in particular in exhibiting two
The CVs of [NiIILPr](BF4) are remarkably different from
Dalton Trans.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Table 4 Comparison of the electrochemical data obtained by cyclic voltammetry in MeCN at 100 mV s−1 for 1 mmol L−1 solutions of the ZnII, CuII, NiII and CoII com-
plexes of the macrocyclic ligands (LPr − and (LEt −
)
)
Complex
Em or Epc and/or Epa, and, where applicable, {ΔE}, in V vs. 0.01 mol L−1 AgNO3/Ag
ZnIILPr
ZnIILEt
CuIILPr
CuIILEt
NiIILPr
NiIILEt
CoIILPr
CoIILEt
+0.32c
−0.97d
−0.48d
−0.58d
+0.37 [+0.32] {0.05}b
+0.45 [+0.39, +0.30, +0.20]b
+0.50 [+0.45] {0.05}b
+0.35 {0.06}a
+0.82c
+0.75c
+0.94c
+0.65c
−1.17 {0.06}a
−0.81d
−1.67 {0.06}a
−1.56d
+0.59 [+0.54] {0.05}a
+0.66, +0.84 [+0.61]b
+0.55c
−1.60d
−0.88d
−0.96d
+1.45c
a Reversible, hence Em and {ΔE} are given. b Processes with distinct but lower intensity return waves, hence Epa [Epc] and {ΔE} are given.
c Irreversible cathodic process hence only Epc given. d Irreversible anodic process hence only Epa given.
Fig. 9 CVs of [NiIILPr](BF4) in MeCN, from −0.2 → +1.0 → −2.0 → −0.2 V, at
different scan rates 25, 50, 100, 200, 300 and 400 mV s−1 before conducting a
controlled potential coulometry experiment.
reversible redox processes (at −1.70 and +0.38 V), perhaps due
to the larger, more flexible (LPr)− macrocycle being more
accommodating. As observed for the [CuIILPr](BF4) complex,
the quasi-reversible process at +0.45 V after coulometry at
Fig. 7 CVs, run from Vi → positive limit → negative limit → Vi, before conduct-
+0.71 V for [NiIILEt](BF4)·H2O is not observed for [NiIILPr](BF4).
ing controlled potential coulometry experiments: (bottom) [CuIILPr](BF4) and
No reversible processes were seen for [CoIILPr](BF4)·0.5H2O,
only two closely spaced oxidation waves at Epc +0.66 and ca.
+0.84, with Epa = +0.61 (Fig. S57 and S58†), which were not
seen for [CoIILEt](BF4)·H2O in MeCN (Fig. S59†).
(top) [NiIILPr](BF4), as 1 mmol L−1 solutions in MeCN (100 mV s−1, 0.1 mol L−1
NBu4PF6, platinum electrode, versus 0.01 mol L−1 AgNO3/Ag).
Conclusion
The preparation of a range of metal-free forms of the larger,
16-membered, [1 + 1] Schiff-base tetradentate macrocycle HLPr
has been described: unlike the smaller 14-membered analogue
HLEt, HLPr is not obtained cleanly without adding acid. Five
complexes of (LPr)− have also been synthesised: ZnIILPr(BF4)·
H2O·0.5IPA, [CuIILPr](BF4), [NiIILPr](BF4), [CoIILPr](BF4)·0.5H2O
and [FeIIILPr(NCS)2]·1.5H2O. Two complexes, [NiIILPr](BF4) and
[FeIIILPr(NCS)2]·0.15MeOH·0.2H2O, have been structurally
characterised, adding valuable new examples of the relatively
uncommon 6666 chelate set for non-porphyrin macrocyclic
Fig. 8 CVs of [CuIILPr](BF4) in MeCN, from −0.7 → −1.7 → −0.7 V at different
scan rates 50, 100, 200, 300 and 400 mV s−1 before conducting a controlled
coulometry experiment.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.

View Article Online
Paper
Dalton Transactions
complexes. Indeed the Fe(III) complex is the first structurally
HLPr·2HOAc. To a bright yellow solution of 1 (50.2 mg,
characterised example of such a, non-porphyrin, complex. Ana- 0.223 mmol) in MeOH (20 mL) was added a solution of dipropy-
logous to [CoIILEt](BF4)·H2O, [CoIILPr](BF4)·0.5H2O is also low- lene triamine (29.2 mg, 0.223 mmol) in MeOH (1 mL) result-
spin, a first for a cobalt complex of a 16-membered N4-donor ing in no colour change. Acetic acid (2 drops) was added
macrocycle featuring less than four imine donors. This is also resulting in no colour change. The resulting bright yellow solu-
consistent with this larger, 16-membered, (LPr)− macrocycle tion was stirred at RT for 48 hours. Solvent was removed
still providing a strong field ligand. Interestingly, the electro- under reduced pressure resulting in a bright orange crystalline
chemical behaviour of the tetrafluoroborate complexes of solid (92 mg, 94%). (Found: C, 65.04; H, 7.58; N, 12.79%. Calc.
(LPr)− is quite different to the analogous (LEt)− complexes. In for [C20H24N4·2HOAc] (440.53 g mol−1): C, 65.43; H, 7.32;
particular, in contrast to the (LEt)− complexes, both [CuIILPr]- N 12.72%). IR (ATR): ν/cm−1 = 3013, 2929, 2835, 1627, 1577,
(BF4) and [NiIILPr](BF4) exhibit some reversible redox 1552, 1510, 1495, 1450, 1396, 1312, 1278, 1200, 1157, 1118,
processes.
1040, 1009, 989, 855, 747, 648, 615, 469, 452. Signals for species
A 1H NMR (500 MHz, CDCl3, 298 K): δ (ppm) = 2.05 (td, J = 7.5,
6.0, 4.4 Hz, 4H, H10), 2.98 (t, J = 6.5 Hz, 4H, H9), 3.77 (m, 2H,
H8), 6.97 (m, 2H, H4), 7.29 (ddd, J = 8.5, 7.2, 1.7 Hz, 2H, H3),
7.40 (m, 2H, H6), 7.55 (m, 2H, H3), 8.50 (s, 1H, H7), 11.10 (s,
1H, NHa). 13C NMR (126 MHz, CDCl3, 298 K): δ (ppm) = 28.17
(C10), 46.44 (C9), 59.25 (C8), 118.60 (C6), 120.74 (C4), 123.58
Experimental
General
All the instruments, and procedures carried out, were as pre- (C2), 131.19 (C5), 132.44 (C3), 143.07 (C1), 163.25 (C7). Signals
1
viously reported,38 except for the following. Magnetic data for for species B H NMR (500 MHz, CDCl3, 298 K): δ (ppm) = 2.14
[FeIIILPr(NCS)2]·1.5H2O
and
recorded at 298 K using a Quantum Design Physical Property 4H, H28), 6.97 (m, 2H, H24), 7.27 (m, 2H, H25), 7.40 (m, 2H,
Measurement System (PPMS) equipped with vibrating H26), 7.52 (m, 2H, H23), 8.44 (s, 1H, H27), 10.80 (s, 1H, NHe).
[CoIILPr](BF4)·0.5H2O
were (m, 4H, H30), 2.98 (t, J = 6.5 Hz, 4H, H29), 3.62 (t, J = 8.0 Hz,
a
sample mount with an applied field of 1 T. Magnetic measure- 13C NMR (126 MHz, CDCl3, 298 K): δ (ppm) = 28.43 (C30),
ments for [CuIILPr](BF4) and [NiIILPr](BF4) were carried out 46.95 (C29), 59.69 (C28), 118.42 (C26), 120.52 (C4), 123.81
using a Johnson-Matthey MSB-MKI magnetic susceptibility (C22), 130.99 (C25), 132.12 (C23), 143.33 (C21), 162.47 (C27).
balance. All magnetic data were corrected for diamagnetic con- ESI(+) MS (m/z) (DCM): [C20H24N4 + H] expected 321.2102,
tributions using Pascal’s constants. Elemental analyses were found 321.2074.
carried out by the Campbell Microanalytical Laboratory at the
University of Otago.
HLPr·2HCO2H. To a bright yellow solution of 1 (72.5 mg,
0.322 mmol) in MeOH (25 mL) was added a solution of dipropy-
In all the reactions MeOH and MeCN were HPLC grade lene triamine (42.2 mg, 0.322 mmol) in MeOH (1 mL) result-
whereas IPA, DCM and pyridine were reagent grade and used ing in no colour change. Formic acid (2 drops) was added
without further purification. All reactions were conducted in resulting in no colour change. The resulting bright yellow solu-
air unless otherwise stated. Compound 1 was prepared accord- tion was stirred at RT for 48 hours. Solvent was removed
ing to the literature method42 as was Fe(py)4(NCS)2.50
under reduced pressure resulting in a bright orange crystalline
Crude HLPr. To a bright yellow refluxing solution of 1 solid (128 mg, 96%). (Found: C, 64.60; H, 7.12; N, 13.80%.
(100 mg, 0.444 mmol) in MeOH (25 ml) was added a solution Calc. for [C20H24N4·2HCOOH] (412.49 g mol−1): C, 64.06;
of dipropylene triamine (58.3 mg, 0.444 mmol) in MeOH H, 6.84; N 13.58%). IR (ATR): ν/cm−1 = 3116, 3014, 2934, 2836,
(1 mL) resulting in no colour change. The resulting solution 2775, 2679, 1627, 1578, 1510, 1452, 1375, 1313, 1265, 1199,
was refluxed for 3 hours producing a yellow/orange solution. 1157, 1119, 1039, 970, 907, 850, 747, 643, 470. Signals for
Solvent was removed under reduced pressure and the yellow species A 1H NMR (500 MHz, CDCl3, 298 K): δ (ppm) = 1.99
oil dried in vacuo to yield a bright yellow crystalline solid (p, J = 6.0 Hz, 4H, H10), 2.89 (t, J = 6.0 Hz, 4H, H9), 3.77 (m,
(132 mg, 71%). (Found: C, 69.67; H, 6.95; N, 14.99%. Calc. for 4H, H8), 6.97 (m, 1H, H4), 7.29 (ddd, J = 8.5, 7.2, 1.7 Hz, 2H,
[C20H24N4·0.2dialdehyde·1.6MeOH] (416.74 g mol−1): C, 70.32; H5), 7.43 (m, 2H, H6), 7.58 (dd, J = 7.8, 1.6 Hz, 2H, H3), 8.51
H, 7.88; N, 14.12%). IR (ATR): ν/cm−1 = 3119, 2929, 2824, 1683, (s, 1H, H7), 11.05 (s, 1H, NHa). 13C NMR (126 MHz, CDCl3,
1628, 1606, 1577, 1606, 1450, 1378, 1308, 1196, 1155, 1117, 298 K): δ (ppm) = 29.03 (C10), 46.67 (C9), 59.33 (C8), 118.34
1039, 963, 891, 847, 744, 640, 602, 584, 552, 468, 445, 431, 419. (C6), 120.65 (C4), 123.51 (C2), 131.11 (C5), 132.18 (C3), 143.09
1H NMR (500 MHz, CDCl3, 298 K): δ (ppm) = 1.89 (m, 4H, (C1), 162.77 (C7). Signals for species B 1H NMR (500 MHz,
H10), 2.69 (t, J = 6.0 Hz, 4H, H9), 3.76 (m, 4H, H8), 6.96 (t, J = CDCl3, 298 K): δ (ppm) = 2.18 (m, 4H, H30), 3.00 (t, J = 7.6 Hz,
7.5 Hz, 2H H4), 7.30 (t, J = 7.7 Hz, 2H, H5), 7.50 (d, J = 8.3 Hz, 4H, H29) 3.63 (t, J = 8.0 Hz, 4H, H28), 6.94 (d, J = 7.4 Hz, 2H,
2H, H6), 7.68 (dd, J = 7.7, 1.5 Hz, 2H, H3), 11.35 (s, 1H, NHa). H24), 7.27 (m, 2H, H25), 7.39 (d, J = 7.9 Hz, 2H, H26), 7.51 (m,
13C NMR (126 MHz, CDCl3, 298 K): δ (ppm) = 30.56 (C10), 2H, H23), 8.43 (s, 1H, H27), 10.96 (s, 1H, NHe). 13C NMR
46.52 (C9), 58.68 (C8), 117.60 (C6), 120.43 (C4), 123.37 (C2), (126 MHz, CDCl3, 298 K): δ (ppm) = 28.39 (C30), 47.10 (C29),
130.98 (C5), 131.20 (C3), 143.20 (C1), 161.44 (C7). ESI(+) MS 59.81 (C28), 118.47 (C26), 120.53 (C24), 123.77 (C22), 130.99
(m/z) (DCM): [C20H24N4
+
H] expected 321.2043, found (C25), 132.24 (C23), 143.31 (C21), 162.58 (C27). ESI(+) MS (m/z)
321.2074.
(DCM): [C20H24N4 + H] expected 321.2085, found 321.2074.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
HLPr·C7H8SO3. To a bright yellow solution of dialdehyde 1 dried in vacuo (114 mg, 55%). (Found: C, 51.11; H, 5.13; N,
(87.6 mg, 0.389 mmol) in MeOH (13 mL) was added dipropy- 11.82%. Calc. for [C20H23N4BF4Cu] (469.78 g mol−1): C, 51.13;
lene triamine (51.03 mg, 0.389 mmol) in MeOH (1 mL) result- H, 4.93; N, 11.93%). IR (ATR): ν/cm−1 = 3255, 2924, 2880, 1629,
ing in colour change. After 15 min of stirring at RT, tosylic acid 1607, 1553, 1536, 1459, 1432, 1366, 1317, 1298, 1279, 1199,
(74.1 mg, 0.389 mmol) in MeOH (2 mL) was added to the 1157, 1128, 1013, 896, 873, 831, 796, 750, 638, 599, 568, 519,
bright yellow solution resulting in a bright orange colour. Pre- 470, 458. ESI(+) MS (m/z) (MeCN): [CuIILPr]+ expected 382.1249,
cipitation of a pale yellow solid was observed after 15–20 min found 382.1213. UV-vis (MeCN): λmax/nm (ε/dm3 mol−1 cm−1) =
of stirring. This pale yellow solid was collected by filtration 326 (5854), 386 (4643), 470 (10 209), 808 (496). μ (Gouy) = 1.76
and dried in vacuo (67 mg, 35%). (Found: C, 65.92; H, 6.61; N, BM. Λm (MeCN) = 185 Ω−1 cm2 mol−1
.
11.28%. Calc. for [C20H24N4·C7H8SO3] (492.64 g mol−1): C,
[NiIILPr](BF4). To a bright yellow refluxing solution of 1
65.83; H, 6.55; N, 11.37%). IR (ATR): ν/cm−1 = 3057, 3016, (52 mg, 0.231 mmol) in methanol (25 mL) was added nickel(II)
2943, 2865, 1629, 1608, 1580, 1517, 1456, 1388, 1343, 1319, tetrafluoroborate hexahydrate (78.5 mg, 0.231 mmol) in metha-
1224, 1174, 1149, 1119, 1033, 1009, 967, 909, 822, 798, 754, nol (2 mL). No colour change was observed. A solution of
1
741, 681, 660, 567, 465. H NMR (500 MHz, CDCl3, 298 K): δ dipropylene triamine (30 mg, 0.231 mmol) in methanol
(ppm) = 2.00 (m, 4H, H10), 2.28 (s, 3H, H21), 3.10 (t, J = 6.6 (10 mL) was added dropwise, causing an immediate dark red
Hz, 4H, H9), 3.70 (m, 4H, H8), 7.02 (ddd, J = 8.0, 5.9, 2.3 Hz, colour change. The resulting solution was refluxed for 1.5 h
2H, H4), 7.10 (m, 2H, H24), 7.35 (m, 4H, H5/6), 7.47 (m, 2H, and the colour had changed to dark green. This dark green solu-
H23), 7.62 (dd, J = 7.4, 1.2 Hz, 2H, H3), 8.55 (s, 1H, H7), 10.75 tion was concentrate under reduced pressure and finally sub-
(s, 1H, NHa). 13C NMR (126 MHz, CDCl3, 298K): δ (ppm) = jected to diethyl ether vapour diffusion. The resulting dark
20.76 (C21), 26.92 (C10), 45.75 (C9), 57.61 (C8), 118.15 (C6), green crystals were collected by filtration and dried in vacuo
120.66 (C4), 123.39 (C2), 125.47 (C24), 128.02 (C23), 131.19 (86 mg, 79%). (Found: C, 51.66; H, 5.11; N, 12.01%. Calc. for
(C5), 132.40 (C3), 137.55 (C22), 142.26 (C1), 145.74 (C25), [C20H23N4BF4Ni] (464.92 g mol−1): C, 51.67; H, 4.99; N,
163.48 (C7). ESI(+) MS (m/z) (MeOH): [C40H49N8]+ expected 12.05%). IR (ATR): ν/cm−1 = 3264, 2953, 1611, 1593, 1555,
641.4027, found 641.4075.
1537, 1461, 1433, 1397, 1326, 1313, 1286, 1253, 1233, 1202,
ZnIILPr(BF4)·H2O·0.5IPA. To a bright yellow solution of di- 1162, 1149, 1108, 1050, 951, 925, 867, 750, 636, 611, 589, 521,
aldehyde 1 (83.9 mg, 0.37 mmol) in IPA (25 mL) was added 466, 428. 1H NMR (500 MHz, CD3CN, 298 K): δ/ppm = 1.80 (m,
dipropylene triamine (48.9 mg, 0.37 mmol) in IPA (1 mL) 3H, H12/12′/9′), 2.19 (m, 1H, H11), 2.36 (ddd, J = 13.2, 9.3, 4.8
resulting in no colour change. This bright yellow solution was Hz, 1H, H10), 2.52 (s, 1H, NH3), 2.67 (m, 2H, H9/11′), 3.04
stirred at RT for 24 h and taken to dryness giving yellow oil. (ddd, J = 17.1, 11.4, 5.3 Hz, 1H, H10′), 3.21 (dt, J = 11.5, 5.5 Hz,
The yellow oil was redissolved in IPA and a solution of zinc(II) 1H, H8′), 3.31 (ddd, J = 8.4, 6.5, 3.1 Hz, 1H, H8), 3.40 (ddd, J =
tetrafluoroborate hydrate (89 mg, 0.37 mmol) in IPA (3 mL) 18.6, 13.7, 7.1 Hz, 2H, H13/13′), 6.84 (m, 1H, H19), 6.92 (m,
was added resulting in an immediate orange precipitate. A solu- 1H, H4), 6.96 (d, J = 8.5 Hz, 1H, H17), 7.05 (d, J = 8.4 Hz, 1H,
tion of triethylamine (37.7 mg, 0.37 mmol) was added in IPA H6), 7.14 (ddd, J = 8.5, 7.1, 1.6 Hz, 1H, H18), 7.20 (ddd, J = 8.5,
(1 mL) resulting in dissolution of the orange precipitate and 7.2, 1.6 Hz, 1H, H5), 7.50 (td, J = 7.7, 1.4 Hz, 2H, H3/20), 7.62
producing a dark orange solution. This dark orange solution (s, 1H, H14), 7.70 (s. 1H, H7). 13C NMR (126 MHz, CD3CN,
was concentrated under reduced pressure and finally subjected 298 K): δ/ppm = 24.79 (C12), 28.18 (C9), 49.15 (C11), 50.03
to diethyl ether vapour diffusion. The resulting dark orange (C10), 55.36 (C13), 57.99 (C8), 119.98 (C19), 120.87 (C4), 124.06
crystalline solids were collected by filtration and dried in vacuo (C17), 125.28 (C6), 127.82 (C15), 128.58 (C2), 133.55 (C20),
(111 mg, 57%). (Found: C, 49.43; H, 5.77; N, 10.47%. Calc. for 134.02 (C18), 134.09 (C5), 134.34 (C3), 148.30 (C16), 149.00
[C20H23N4BF4Zn·H2O·0.5IPA] (519.68 g mol−1): C, 49.69; H, (C1), 162.77 (C14), 163.88 (C7). ESI(+) MS (m/z) (MeCN):
5.62; N, 10.78%). IR (ATR) ν/cm−1 = 3569, 3278, 2930, 2870, [NiIILPr]+ expected 377.1267, found 377.1271. UV-vis (MeCN):
1629, 1594, 1554, 1535, 1515, 1488, 1461, 1433, 1403, 1309, λmax/nm (ε/dm3 mol−1 cm−1) = 326 (6815), 409 (3739), 480 (8286),
1192, 1157, 1030, 872, 751, 637, 597, 559, 520, 474, 412. ESI(+) 615 (1701). μ (Gouy) = 0 BM. Λm (MeCN) = 153 Ω−1 cm2 mol−1
MS (m/z) (MeCN): [ZnIILPr]+ expected 383.1183, found [CoIILPr](BF4)·0.5H2O. To a bright yellow solution of the
383.1209. UV-vis (MeCN): λmax/nm (ε/dm3 mol−1 cm−1) = 346 crude macrocycle HLPr (41.6 mg, 0.130 mmol) in pyridine
(4168), 460 (2980), 527 (1429). Λm (MeCN) = 156 Ω−1 cm2 mol−1
(15 mL) was added a solution of cobalt(II) tetrafluoroborate
.
.
[CuIILPr](BF4). To a bright yellow solution of the crude hexahydrate (44.3 mg, 0.130 mmol) in pyridine (3 mL). The
macrocycle HLPr (143 mg, 0.446 mmol) in a 1 : 1 mixture of bright yellow solution immediately turned dark brown in
methanol and dichloromethane (30 mL) was added copper(II) colour. The resulting solution was stirred at RT overnight. This
tetrafluoroborate hydrate (106 mg, 0.445 mmol) in methanol was then concentrated to 20 mL under reduced pressure and
(2 mL). The bright yellow solution immediately turned dark subjected to diethyl ether vapour diffusion. The resulting dark
brown/orange in colour. Triethylamine (45 mg, 0.446 mmol) in brown/black crystalline solid was collected by filtration and
methanol (1 mL) resulting in obvious colour change. The dried in vacuo (44 mg, 71%). (Found: C, 50.27; H, 4.72; N,
volume of the dark brown/orange solution was concentrated to 11.65%. Calc. for [C20H23N4BF4Co·0.5H2O] (474.17 g mol−1): C,
20 mL and subjected to diethyl ether vapour diffusion. The 50.66; H, 5.10; N, 11.82%). IR (ATR): ν/cm−1 = 3060, 1621,
resulting bright orange crystals were collected by filtration and 1603, 1557, 1540, 1461, 1438, 1402, 1309, 1239, 1201, 1157,
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.

View Article Online
Paper
Dalton Transactions
1052, 750, 699, 646, 627, 563, 543, 518, 472, 455, 431, 404. ESI
Acknowledgements
(+) MS (m/z) (MeCN): [CoIIILPr-H]+ expected 377.1143, found
377.1171. UV-vis (MeCN): λmax/nm (ε/dm3 mol−1 cm−1) = 408 We are grateful to the University of Otago for the award of a
(1397), 494 (2546), 673 (219). μ (PPMS at RT) = 2.31 BM. Λm Post-graduate Scholarship (to RKW) and for funding this
(MeCN) = 137 Ω−1 cm2 mol−1
.
research. We thank the MacDiarmid Institute for funding the
[FeIIILPr(NCS)2]·1.5H2O. Under an Ar atmosphere, to purchase of the SQUID and PPMS magnetometers, Dr Jeff
a
bright yellow suspension of the crude macrocycle Tallon (IRL) for facilitating access to them, and Dr Humphrey
HLPr (80.3 mg, 0.250 mmol) in MeOH (30 mL) was added FeII L.C. Feltham (Brooker group) for measuring the room temp-
(py)4(SCN)2 (122.4 mg, 0.250 mmol) as a solid. The bright erature magnetic moments. We are very grateful to Professor
yellow solution immediately turned dark orange/brown in Alan Bond (Monash) for his helpful comments and advice on
colour. After an hour of stirring under Ar, the dark orange solu- the electrochemical results, Dr Mervyn Thomas (Otago) for
tion was exposed to air and the solution slowly turned dark obtaining the NMR spectra and the Campbell Microanalytical
green in colour. This solution was left exposed to air for Laboratory for carrying out the microanalyses on the com-
5 hours. At the end of 3 hours, a dark green precipitate pounds reported in this paper.
was observed. This dark green precipitate was obtained by fil-
tration and dried in vacuo (58 mg, 45%). (Found: C, 51.14; H,
4.66; N, 16.06; S, 12.02%. Calc. for [C22H23N6S2Fe·1.5H2O]
Notes and references
(518.45 g mol−1): C, 50.97; H, 5.05; N, 16.21; S, 12.37%). IR
(ATR): ν/cm−1 = 3192, 2927, 2101, 1615, 1593, 1556, 1541,
1459, 1434, 1409, 1291, 1233, 1207, 1159, 1126, 1093, 1077,
1061, 1038, 958, 946, 888, 875, 851, 815, 765, 752, 696, 651,
642, 610, 583, 562, 526, 482, 469, 440, 428, 412. ESI(+) MS
(m/z) (MeCN): [FeIIILPr-H]+ expected 374.1181, found
1 N. F. Curtis and D. A. House, Chem. Ind., 1961, 1708–1709.
2 G. A. Melson and D. H. Busch, Proc. Chem. Soc., London,
1963, 223, 223–224.
3 E. G. Jäger, Z. Chem., 1964, 4, 437.
4 N. F. Curtis, Supramol. Chem., 2012, 24, 439–447.
5 L. F. Lindoy, Supramol. Chem., 2012, 24, 448–461.
6 N. H. Pilkington and R. Robson, Aust. J. Chem., 1970, 23,
2225–2236.
7 S. M. Nelson, Pure Appl. Chem., 1980, 52, 2461–3476.
8 D. E. Fenton, Pure Appl. Chem., 1986, 58, 1437–1444.
9 L. F. Lindoy, The Chemistry of Macrocyclic Ligand Complexes,
Cambridge University Press, Cambridge, 1990.
10 N. V. Gerbeleu, V. B. Arion and J. Burgess, Template Syn-
thesis of Macrocyclic Compounds, Wiley-VCH, Weinheim,
1999.
374.1189. μ (PPMS at RT) = 2.02 BM. UV-vis (MeCN): λmax
/
nm (ε/dm3 mol−1 cm−1) = 323 (5869), 379 (7020), 423 (6301),
772 (2626). Λm (MeCN) = 12 Ω−1 cm2 mol−1. X-ray quality
crystals were grown by slow evaporation of the reaction
solution, giving black blocks.
X-ray crystallography
Data were collected on a Bruker Kappa Apex II area detector
diffractometer at 89–90 K using graphite monochromated Mo-
Kα radiation (λ = 0.71073 Å). Both data sets were absorption 11 J. E. Beves, B. A. Blight, C. J. Campbell, D. A. Leigh and
corrected using SCALE. The structures were solved using R. T.McBurney, Angew. Chem., Int. Ed., 2011, 50, 9260–9327.
SHELXS-97 and refined against F2 using all data by full-matrix 12 J. L. Sessler and E. Tomat, Acc. Chem. Res., 2007, 40,
least-squares techniques with SHELXL-97.51 All non hydrogen
371–379.
atoms were modelled anisotropically except where noted. 13 C. Platas-Iglesias, Eur. J. Inorg. Chem., 2012, 2012,
Hydrogen atoms were inserted at calculated positions and 2023–2033.
rode on the atoms to which they are attached, except where 14 R. E. Mewis and S. J. Archibald, Coord. Chem. Rev., 2010,
noted. Crystallographic data has been deposited with the Cam- 254, 1686–1712.
bridge Crystallographic Data Centre (reference numbers CCDC 15 J. B. Love, Chem. Commun., 2009, 3154–3165.
913820 and 913821).
16 M. R. Kember, F. Jutz, A. Buchard, A. J. P. White and
[NiIILPr](BF4). The asymmetric unit comprised the entire
C. K. Williams, Chem. Sci., 2012, 3, 1245–1255.
macrocyclic complex. No solvent or disorder was present. The 17 P. L. Arnold, G. M. Jones, S. O. Odoh, G. Schreckenbach,
amine hydrogen atom, (H3x), was found and the x, y, z coordi-
nates freely refined, whilst U(H3x) = 1.2 × U(N3).
N. Magnani and J. B. Love, Nat. Chem., 2012, 4, 221–227.
18 V. McKee, Adv. Inorg. Chem., 1993, 40, 323–399.
2[FeIIILPr(NCS)2]·0.3MeOH·0.4H2O. The asymmetric unit 19 P. Mountford, Chem. Soc. Rev., 1998, 27, 105–116.
contains 2 [FeIIILPr(NCS)2] complexes. The Fe(1) complex has 20 S. Brooker, Coord. Chem. Rev., 2001, 222, 33–56.
no disorder whereas in the Fe(2) complex, both thiocyanate 21 A. M. J. Devoille and J. B. Love, Dalton Trans., 2012, 41,
anions are disordered over two sites with 0.8 : 0.2 occupancy. A
central part of the dipropylene triamine chain is also disor- 22 N. Ishikawa, M. Sugita, T. Ishikawa, S.-y. Koshihara and
dered over two sites with 0.8 : 0.2 occupancy. There is also a Y. Kaizu, J. Am. Chem. Soc., 2003, 125, 8694–8695.
region of partial occupancy solvents (MeOH and H2O). Atoms 23 H. L. C. Feltham, Y. Lan, F. Klöwer, L. Ungur,
65–72.
C71, C81, C90, S71, S81, O90, O100 and O101 were refined
isotropically.
L. F. Chibotaru, A. K. Powell and S. Brooker, Chem.–Eur. J.,
2011, 17, 4362–4365.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
24 A. Yamashita, A. Watanabe, S. Akine, T. Nabeshima, 37 A. G. Kolchinski, B. Korybut-Daszkiewicz, E. V. Rybak-
M. Nakano, T. Yamamura and T. Kajiwara, Angew. Chem.,
Int. Ed., 2011, 50, 4016–4019.
Akimova, D. H. Busch, N. W. Alcock and H. J. Clase, J. Am.
Chem. Soc., 1997, 119, 4160–4171.
25 H. L. C. Feltham, R. Clérac, L. Ungur, V. Vieru, 38 R. Sanyal, S. A. Cameron and S. Brooker, Dalton Trans.,
L. F. Chibotaru, A. K. Powell and S. Brooker, Inorg. Chem.,
2012, 51, 10603–10612.
2011, 40, 12277–12287.
39 S. A. Cameron and S. Brooker, Inorg. Chem., 2011, 50, 3697–3706.
26 S.-Y. Lin, Y.-N. Guo, Y. Guo, L. Zhao, P. Zhang, H. Ke and 40 E. D. Bergmann, M. Rabinovitz and I. Agranat, Bull. Res.
J. Tang, Chem. Commun., 2012, 48, 6924–6926. Counc. Isr. Sect. A: Chem., 1962, 11, 149.
27 K. M. Smith, Porphyrins and Metalloporphyrins, Elsevier/ 41 C. W. G. Ansell, J. Lewis and P. R. Raithby, J. Chem. Soc.,
North-Holland Biomedical Press, New York, 1976. Dalton Trans., 1982, 2557–2559.
28 A. Messerschmidt, R. Huber, T. Poulos and K. Wieghardt, 42 D. S. C. Black and N. E. Rothnie, Aust. J. Chem., 1983, 36,
Handbook of Metalloproteins, John Wiley and Sons, Ltd, Chi-
chester, 2001.
29 A. G. Lappin and A. McAuley, Adv. Inorg. Chem., 1988, 32,
241–295.
30 J. Costamagna, G. Ferraudi, J. Canales and J. Vargas, Coord.
Chem. Rev., 1996, 148, 221–248.
31 F. V. Lovecchio, E. S. Gore and D. H. Busch, J. Am. Chem.
Soc., 1974, 96, 3109–3118.
2395–2406.
43 D. S. C. Black and N. E. Rothnie, Tetrahedron Lett., 1978,
31, 2835–2836.
44 S. Patai, The Chemistry of the Carbon–Nitrogen Double Bond,
John Wiley and Sons, Inc., 1st edn, 1970.
45 C. Amort, M. Malaun, A. Krajete, H. Kopacka, K. Wurst,
M. Christ, D. Lilge, M. O. Kristen and B. Bildstein, Appl.
Organomet. Chem., 2002, 16, 506–516.
32 L. Y. Martin, C. R. Sperati and D. H. Busch, J. Am. Chem. 46 J. P. Guthrie, Can. J. Chem., 1978, 56, 2342–2354.
Soc., 1977, 99, 2968–2981. 47 W. J. Geary, Coord. Chem. Rev., 1971, 7, 81–122.
33 M. Salavati-Niasari and F. Davar, J. Coord. Chem., 2010, 63, 48 K. Nakamoto, Infrared and Raman Spectra of Inorganic
3240–3255.
Coordination Compounds. Part B, John Wiley & Sons, Inc.,
New York, 5th edn, 1997.
49 N. N. Greenwood and A. Earnshaw, Chemistry of the
Elements, 1st edn, Pergamon Press, Exeter, 1984.
34 F. H. Allen, Acta Crystallogr., Sect. B: Struct. Sci., 2002, 58,
380–388.
35 M. L. Caste, C. J. Cairns, J. Church, W. K. Lin,
J. C. Gallucci and D. H. Busch, Inorg. Chem., 1987, 26, 50 Inorg. Synth., ed. G. B. Kauffman, R. A. Albers and
78–83. F. L. Harlan, McGraw-Hill, 1970.
36 A. G. Kolchinski, N. W. Alcock and D. H. Busch, Inorg. 51 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystal-
Chem., 1997, 36, 2754–2759.
logr., 2008, 64, 112–122.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.