A convenient synthesis of isocyclam and [16]aneN4 and the photophysics of their dicyanochromium(III) complexes

Article abstract of DOI:10.1016/j.ica.2009.09.020

The syntheses of the tetraazamacrocyclic ligands 1,4,7,11-tetraazacyclotetradecane (isocyclam) and 1,5,9,13-tetraazacyclohexadecane ([16]aneN₄) in two steps starting from the corresponding tetraamine and diethylmalonate is reported. The trans-dicyanochromium(III) complexes, trans-[Cr(isocyclam)(CN)₂]PF₆ and trans-[Cr([16]aneN₄)(CN)₂]PF₆ have also been prepared. Both are ²E_g emitters with 0-0 band emission wavelengths at 721.2 and 704.8 nm, respectively. The isocyclam complex has a room temperature excited state lifetime of 147 μs in aqueous solution which increases to 215 μs upon macrocyclic N-H deuteration, whereas the corresponding lifetime of the [16]aneN₄ complex is 25 μs and is unaffected by macrocyclic N-H deuteration. The implications of the temperature dependence of the excited state lifetimes are also presented.

Full text of DOI:10.1016/j.ica.2009.09.020

Inorganica Chimica Acta 363 (2010) 157–162
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
A convenient synthesis of isocyclam and [16]aneN4 and the photophysics
of their dicyanochromium(III) complexes
*
David L. Grisenti, Mary beth Smith, Luxi Fang, Nicholas Bishop, Paul S. Wagenknecht
Department of Chemistry, Furman University, Greenville, SC 29613, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 31 July 2009
Received in revised form 2 September 2009
Accepted 4 September 2009
Available online 13 September 2009
The syntheses of the tetraazamacrocyclic ligands 1,4,7,11-tetraazacyclotetradecane (isocyclam) and
1,5,9,13-tetraazacyclohexadecane ([16]aneN4) in two steps starting from the corresponding tetraamine
and diethylmalonate is reported. The trans-dicyanochromium(III) complexes, trans-[Cr(isocy-
clam)(CN)2]PF6 and trans-[Cr([16]aneN4)(CN)2]PF6 have also been prepared. Both are 2Eg emitters with
0–0 band emission wavelengths at 721.2 and 704.8 nm, respectively. The isocyclam complex has a room
temperature excited state lifetime of 147
cyclic N–H deuteration, whereas the corresponding lifetime of the [16]aneN4 complex is 25
unaffected by macrocyclic N–H deuteration. The implications of the temperature dependence of the
excited state lifetimes are also presented.
l
s in aqueous solution which increases to 215
l
s upon macro-
s and is
Keywords:
l
Chromium(III) complexes
Photochemistry
Photophysics
Tetraazamacrocycle
Isocyclam
Ó 2009 Elsevier B.V. All rights reserved.
1. Introduction
Thus, we initially set out to design streamlined syntheses of these
ligands using more benign reaction conditions. Herein we present
Tetraazamacrocyclic ligands have been the subject of a signifi-
cant number of studies exploiting their metal ion size selectivity,
strong binding constants, and ease of functionalization [1,2].
Among potential applications of the metal complexes of these li-
gands are the selective removal of toxic metals from waste streams
[3], radiotherapy [4], and contrast agents for magnetic resonance
imaging [5]. Of the simple ligands in this category (Fig. 1), all but
isocyclam and [16]aneN4 are commercially available. Our own
interest in these ligands revolves around the synthesis and photo-
physical characterization of their dicyanochromium(III) com-
plexes. We have chiefly been interested in the effect of steric
constraint on the mechanism of excited state deactivation [6],
but more recently have been interested in preparing a set of closely
related complexes with which to study energy transfer as a func-
tion of subtle changes in thermodynamic driving force [7]. With re-
spect to this goal, we are interested in complexes of the general
formula trans-[Cr(N4)(CN)2]+ (where N4 is a tetraazamacrocyclic li-
gand). Thus, we set out to prepare the two remaining congeners,
trans-[Cr([16]aneN4)(CN)2]PF6 and trans-[Cr(isocyclam)(CN)2]PF6
[8].
a convenient synthesis of these ligands along with the new ligand
dioxoisocyclam. We also present the synthesis and photophysical
characterization of trans-[Cr([16]aneN4)(CN)2]PF6, trans-[Cr(isocy-
clam)(CN)2]PF6, and for comparison purposes, cis-[Cr([13]aneN4)-
(CN)2]PF6.
2. Experimental
2.1. Materials and methods
3,6-Diaza-1,9-nonanediamine (2,2,3-tet) was prepared accord-
ing to the literature procedure [11]. cis-[Cr([13]aneN4)Cl2]PF6 was
prepared by metathesis of cis-[Cr([13]aneN4)Cl2]Cl [12]. Chroma-
tography was performed using activated alumina (Neutral, Brock-
man I, 150 mesh). All measurements were performed under an
air atmosphere unless otherwise noted. UV–Vis absorption spectra
were recorded using a Varian Cary-50 spectrophotometer. Emis-
sion spectra were recorded on an SLM Aminco instrument running
DataMax software. Emission lifetimes were measured using as the
excitation source a Photon Technology International (PTI) GL-3300
pulsed nitrogen laser fed into a PTI GL-302 dye laser. Data was col-
lected on an OLIS SM-45 EM fluorescence lifetime measurement
system and analyzed using OLIS SpectralWorks. The dye laser
was operated at 440 nm, corresponding to 4A2g ? 4T2g (Oh) excita-
tion of the trans-[Cr(N4)(CN)2]+ complexes.
The literature syntheses for isocyclam [9] and [16]aneN4 [10]
involve cyclization reactions using tosyl-protected polyamines, fol-
lowed by detosylation reactions in concentrated acid solutions.
* Corresponding author. Tel.: +1 864 294 2905; fax: +1 864 294 3559.
E-mail address: paul.wagenknecht@furman.edu (P.S. Wagenknecht).
0020-1693/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2009.09.020

158
D.L. Grisenti et al. / Inorganica Chimica Acta 363 (2010) 157–162
2.2.3. Synthesis of 1,4,8,12-tetraazacyclotetradecane-5,7-dione
(dioxoisocyclam)
This macrocycle was prepared in an analogous fashion to that
for 1,5,9,13-tetraazacyclohexadecane-6,8-dione but starting with
2,2,3-tet (10 g, 62.4 mmol), diethyl malonate (9.99 g, 62.4 mmol)
in anhydrous ethanol (1 L), yielding 2.68 g (20%) of a white crystal-
line solid. No further purification was required for this material. 1H
NMR (CDCl3): d 1.60 (bs, 2H), 1.70 (quintet, 2H), 2.66 (m, 6H), 2.76
(t, 2H), 3.18 (s, 2H), 3.37 (quartet, 2H), 3.43 (quartet, 2H), 7.50 (bs,
1H), 7.70 (bs, 1H). Anal. Calc. for C10H20N4O2ꢃ½H2O: C, 50.61; H,
8.92; N, 23.62. Found: C, 50.69; H, 9.12; N, 23.29%.
2.2.4. Synthesis of 1,4,7,11-tetraazacyclotetradecane (isocyclam)
This was prepared in an analogous fashion to that for [16]aneN4
but starting with 1,4,8,12-tetraazacyclotetradecane-5,7-dione
(1.00 g, 4.38 mmol) and BH3ꢃTHF (1.0 M, 57.4 mL) resulting in
white waxy solid (0.72 g, 82%). 1H NMR (CDCl3): d (ppm) 1.66 (q,
4H), 2.10 (bs, 4H), 2.65 (m, 16H). This material is sufficiently pure
for metal complexation. However, the material can be further puri-
fied by sublimation (75 °C, 300 mTorr). Anal. Calc. for C10H24N4: C,
59.96; H, 12.08; N, 27.97. Found: C, 59.98; H, 11.88; N, 27.89%.
Fig. 1. Simple tetraazamacrocyclic ligands.
2.2. Syntheses
2.2.1. Synthesis of 1,5,9,13-tetraazacyclohexadecane-6,8-dione
A 2 L round bottom flask (RBF) was charged with anhydrous
ethanol (1.2 L), 1,5,9,13-tetraazatridecane, (12.15 g, 64.5 mmol)
and diethyl malonate (10.33 g, 64.5 mmol). The contents were
heated at reflux for seven days under a nitrogen atmosphere, after
which the solvent volume was reduced to ꢀ40 mL. The resulting
solution was refrigerated at ꢁ15 °C for several days to induce crys-
tal growth. White crystals were collected by suction filtration,
washed with ice-cold acetonitrile (2 ꢂ 5 mL) and cold diethyl ether
(2 ꢂ 10 mL) and dried in vacuo (3.00 g, 18.1%). 1H NMR (CDCl3): d
1.66(m, 8H), 2.60(t, 4H), 2.65(t, 4H), 3.15(s, 2H), 3.41(q, 4H),
7.62(bs, 2H). This material is sufficiently pure for the next step.
However, the product can be further purified by recrystallization
from absolute ethanol. Anal. Calc. for C12H24N4O2: C, 56.23; H,
9.44; N, 21.86. Found: C, 56.37; H, 9.52; N, 21.93%.
2.2.5. Synthesis of trans-[Cr([16]aneN4)Cl2]PF6
The preparation of the crude light-green trans-[Cr([16]a-
neN4)Cl2]Cl using CrCl3 (0.328 g, 2.07 mmol), Zn dust (ꢀ0.007 g),
and [16]aneN4 (0.472 g, 2.07 mmol) followed the procedure by
Eriksen and Mønsted [13] and resulted in 0.442 g of crude trans-
[Cr([16]aneN4)Cl2]Cl. This solid was dissolved in aqueous NH4Cl
(3 M, ꢀ3 mL) using sonication. The green solution was filtered to
remove an insoluble green material that was judiciously rinsed
with 3 M aqueous NH4Cl until the filtrate was colorless. To the
green filtrate was added saturated aqueous NH4PF6 (ꢀ4 mL). The
resulting light-green precipitate was collected by filtration and
washed with water (2 ꢂ 8 mL) and diethyl ether (3 ꢂ 5 mL) and
dried (0.487 g, 49% overall yield). Anal. Calc. for C12H28N4CrCl2PF6:
C, 29.04; H, 5.69; N, 11.29. Found: C, 28.74; H, 5.55; N, 10.98%.
2.2.6. Synthesis of trans-[Cr([16]aneN4)(CN)2]PF6
A 10 mL oven dried RBF was charged with trans-[Cr([16]a-
neN4)Cl2]PF6 (0.100 g, 0.21 mmol) and 1.2 mL anhydrous DMSO.
The mixture was heated to 62 °C and upon complete dissolution
of the complex, finely ground NaCN (0.084 g, 1.7 mmol) was added.
After heating and stirring for 75 min, the contents were filtered
(while still warm) to remove insoluble impurities, which were
rinsed with an additional 8–10 drops of room temperature DMSO
until the filtrate was no longer colored. The product was oiled
out by the addition of diethyl ether (30 mL). The supernatant
was discarded and this step was repeated to ensure the complete
removal of DMSO. To the orange oil was added water (4 mL) fol-
lowed by saturated aqueous NH4PF6 (4 mL), resulting in an orange
precipitate. The product was collected by filtration, and washed
with water (2 ꢂ 1 mL), ethanol (2 ꢂ 1 mL), and ether (2 ꢂ 5 mL)
(0.048 g, 49.5%). Anal. Calc. for C14H28N6CrPF6: C, 35.22; H, 5.91;
N, 17.60. Found: C, 35.56; H, 5.84; N, 17.62%. Further purification
has been performed by column chromatography using Al2O3 and
eluting with CH3CN.
2.2.2. Synthesis of 1,5,9,13-tetraazacyclohexadecane ([16]aneN4)
An oven dried 2-neck 100 mL RBF equipped with a condenser
and kept under a positive pressure of nitrogen was charged with
1,5,9,13-tetraazacyclohexadecane-6,8-dione (1.00 g, 3.90 mmol),
then BH3ꢃTHF (1.0 M solution in THF, 51.5 mL) was added through
a septum. When the evolution of hydrogen had moderated, the
rubber septum was replaced with a glass stopper. After 72 h at re-
flux the contents were cooled to RT and methanol (6 mL) was
added dropwise to quench the excess borane. After bubbling
ceased, HCl (3.0 M, 10 mL) was added dropwise. The solvent vol-
ume was reduced to ꢀ10 mL using rotary evaporation. Methanol
(20 mL) was added to the flask and again the solvent volume was
reduced to ꢀ10 mL. This process was repeated three more times
to aid in the removal of trimethylborate; on the last methanol
addition, the solution was evaporated to dryness. The white solid
was taken up in HCl (3 M, 40 mL) and the solution was extracted
with CHCl3 (2 ꢂ 20 mL) to remove organic impurities. The organic
layers were discarded and aqueous KOH (15 M, 20 mL) was added
dropwise to the aqueous layer to increase the pH > 13. The product
was extracted from this basic solution into CHCl3 (5 ꢂ 20 mL) and
the combined extracts were collected and dried over K2CO3. Evap-
oration of chloroform yielded a white waxy solid (0.444 g, 50%). 1H
NMR (CDCl3): d 1.70(q, 8H), 1.82(bs, 4H), and 2.70(t, 16H). This
material is sufficiently pure for use in metal complexation. How-
ever, the product can be further purified by sublimation (60 °C,
250 mTorr). Anal. Calc. for C12H28N4: C, 63.11; H, 12.36; N, 24.53.
Found: C, 62.9; H, 12.10; N, 24.38%.
2.2.7. Synthesis of cis/trans-[Cr(isocyclam)Cl2]PF6
A crude mixture of cis/trans-[Cr(isocyclam)Cl2]Cl was prepared
using isocyclam and CrCl3(THF)3 according to the method of
Swisher et al [14]. For metathesis, cis/trans-[Cr(isocyclam)Cl2]Cl
(0.526 g, 1.47 mmol) was dissolved in aqueous NH4Cl (3.0 M,
8 mL). To the filtrate was added saturated aqueous NH4PF6
(4 mL) and the olive brown precipitate was collected by filtration
and washed with cold distilled water (2 ꢂ 5 mL) and diethyl ether

D.L. Grisenti et al. / Inorganica Chimica Acta 363 (2010) 157–162
159
(2 ꢂ 4 mL) (0.517 g, 75%). Anal. Calc. for C10H24N4CrCl2PF6: C,
3. Results and discussion
25.65; H, 5.17; N, 11.97. Found: C, 25.35; H, 5.01; N, 11.69%.
3.1. Ligand syntheses
2.2.8. Synthesis of trans-[Cr(isocyclam)(CN)2]PF6
The general scheme for synthesis of the ligands follows a gen-
eral method first used by Tabushi et al. [15] for the synthesis of
C-alkylated cyclams and consists of a cyclization reaction between
the appropriate tetraamine and diethylmalonate to obtain the
macrocyclic diamide ligands, followed by a borane reduction to
give the tetraazamacrocycles (Fig. 2). The ligand dioxoisocyclam
has not been previously reported whereas the ligand dioxo-
[16]aneN4 was previously reported but without synthetic details
or analytical data [16]. The syntheses of the macrocyclic diamides
are very reliable with the products simply crystallizing out of the
concentrated reaction mixture after several days to a week. Precur-
sor tetraamine, 1b, used for the synthesis of [16]aneN4 is commer-
cially available, whereas precursor tetraamine, 1a, for the synthesis
of isocyclam can be prepared in large quantities from the conden-
sation of diethylenetriamine with acrylonitrile followed by cata-
lytic hydrogenation [11]. The macrocyclic diamides thus obtained
are of sufficient purity for use in the borane reduction. The borane
reduction of 2a to give 3a (isocyclam) proceeds with good yield,
typically between 80% and 90%. The analogous reduction of 2b to
give 3b ([16]aneN4), on the other hand, consistently gives lower
yields, typically near 50%. These macrocyclic tetraamines are of
sufficient purity to use in metal complexation reactions. In our
hands, we find this synthetic scheme more convenient than the
known literature syntheses of isocyclam [9] and [16]aneN4 [10].
It also provides a synthetic strategy for the synthesis of C-function-
alized isocyclam and [16]aneN4 ligands by starting with C-alkyl-
ated diethylmalonate [15].
To a 25 mL oven dried RBF was added cis/trans-[Cr(isocy-
clam)Cl2]PF6 (0.200 g, 0.427 mmol) and 3.0 mL anhydrous DMSO.
The contents in the flask were brought to 65 °C and when the com-
plex was dissolved, NaCN (0.113 g, 2.30 mmol) was added. After
75 min the flask was cooled to room temperature. The DMSO
was extracted using diethyl ether (5 ꢂ 20 mL). To the orange
brown oil that remained was added aqueous NH4PF6 (saturated,
4 mL) and the mixture was stirred in an ice bath for 10 min. The or-
ange brown precipitate was collected by filtration and washed
with H2O (2 ꢂ 1 mL), ethanol (2 ꢂ 1 mL), and diethyl ether
(2 ꢂ 5 mL). After drying, the crude material (0.118 g) was dissolved
in 30 mL CH3CN and dry loaded onto 3 mL of Al2O3 and then eluted
down an Al2O3 column (7 cm ꢂ 2 cm) using CH3CN. The yellow
band was collected and the solvent volume was reduced to
ꢀ5 mL. Diethyl ether was added to the concentrated yellow solu-
tion and the yellow precipitate was collected by filtration, washed
with diethyl ether (3 ꢂ 5 mL) and dried (0.0467 g, 24%). Anal. Calc.
for C12H24N6CrPF6: C, 32.08; H, 5.38; N, 18.70. Found: C, 32.37; H,
5.28; N, 18.72%.
2.2.9. Synthesis of cis-[Cr([13]aneN4)(CN)2]PF6ꢃH2O
The following procedures were performed in the absence of
light. To
a 25 mL oven dried RBF was added cis-[Cr([13]a-
neN4)Cl2]PF6 (0.200 g, 0.44 mmol) and 3 mL anh. DMSO. The flask
was heated to 62 °C and when the complex was completely dis-
solved finely ground NaCN (0.116 g, 2.38 mmol) was added. The
contents were stirred for 75 min. at 62 °C. The DMSO was extracted
(5 ꢂ 20 mL) with diethyl ether. To the dark brown colored oil was
added 8 mL of a 1.0 M solution of NH4PF6. The crude orange precip-
itate was collected by filtration then dissolved using a minimum
volume of warm water. NH4PF6 (1.4 g) was added and after refrig-
eration overnight, the orange precipitate was collected by filtration
and washed with diethyl ether (0.067 g, 35%). Anal. Calc. for
C11H22N6CrPF6ꢃH2O: C, 29.15; H, 5.34; N, 18.54. Found: C, 29.37;
3.2. Metal complex syntheses
Initial metallation of the macrocyclic ligands to give the
[Cr(N4)Cl2]Cl follows known literature procedures. For the case of
isocyclam [12,14], CrCl3(THF)3 is used for the metallation accord-
ing to the method used successfully for cyclam [17], whereas for
the case of [16]aneN4 [13], CrCl3 in the presence of catalytic Zn is
used for the metallation. We have also attempted the metallation
of [16]aneN4 using CrCl3(THF)3 [12] but have had poor results. To
render the complexes soluble in organic solvent for the replace-
ment of Clꢁ ligand by CNꢁ, metathesis of the Clꢁ counterion for
H, 5.21; N, 18.56%. UV–Vis(CH3CN)(
and 338 nm (112). Raman shift
(C„N) (cmꢁ1): 2144 and 2123.
e
Mꢁ1 cmꢁ1): 458 nm(125)
m
2.3. N–H deuteration of trans-[Cr(N4)(CN)2]PF6
ꢁ
PF6 was performed. An excess of NH4Cl was added during the
A 25 mL round bottom flask was charged with the complex
(20 mg), D2O (20 mL), and 1 drop of triethylamine. The solution
was stirred for 4 h in the case of the [16]aneN4 complex and 16 h
for the isocyclam complex. The solvent was removed on a vacuum
line. Deuteration was determined to be >90% by IR spectroscopy.
metathesis in order to suppress any aquation reactions [6a,18].
We were particularly interested in preparing the dicyano com-
plexes because of the unique photochemical behavior of the known
[CrN4(CN)2]+ complexes and our interest in the effect of macrocycle
geometry on their photobehavior [6]. Ligand exchange of Clꢁ by
CNꢁ using a modification of the literature scheme for trans-[Cr(cy-
clam)(CN)2]+ [8a] gave the desired complexes in moderate yield.
Whereas [13]aneN4 has only been observed to form complexes
with the cis-configuration with Cr(III) (presumably due to the rel-
atively small hole-size) [12], and [16]aneN4 preferentially forms
complexes of the trans-configuration with Cr(III) [12,13], isocyclam
complexes can be either cis or trans [12,14].
The [Cr(isocyclam)(CN)2]PF6 initially obtained requires purifica-
tion using chromatography (neutral Al2O3, CH3CN). Dry loading an
acetonitrile solution of the complex onto Al2O3 was performed be-
cause the crude material does not have good enough solubility in
CH3CN for initial loading onto the column. We are confident that
the product from this synthesis and purification is the trans-isomer
for three reasons. First, the related cyclam complex, trans-[Cr(cy-
clam)(CN)2]ClO4, is prepared from a cis/trans mixture of [Cr(cy-
clam)Cl2]ClO4 and results exclusively in the trans isomer [8a].
Fig. 2. Synthesis of macrocyclic diamide ligands dioxoisocyclam (2a), dioxo-
[16]aneN4 (2b), isocyclam (3a), and [16]aneN4 (3b).

160
D.L. Grisenti et al. / Inorganica Chimica Acta 363 (2010) 157–162
Table 1
Second, we have demonstrated that cis-[Cr([13]aneN4)(CN)2]PF6
moves more slowly under the column conditions used to purify
[Cr(isocyclam)(CN)2]PF6. We expect that this is due to the more po-
lar nature of the cis-isomer and expect the same to be true of the
cis-isomer of [Cr(isocyclam)(CN)2]PF6, were it to form. Third,
whereas both cis-[Cr([13]aneN4)(CN)2]PF6 and cis-[Cr(rac-(5,12)-
Me6[14]aneN4(CN)2]Cl [8c] are orange solids with notable asym-
metry evident in the UV–Vis absorptions, the [Cr(isocy-
clam)(CN)2]PF6 complex reported herein is bright yellow with
two unsplit bands in the UV–Vis (vide infra).
The elemental analysis of the [Cr([16]aneN4)(CN)2]PF6 initially
obtained indicates good purity without the need for column chro-
matography. However, the emission data (vide infra) suggests that
this product may be a mixture of isomers or that there may be a
trace impurity. The isomeric possibilities are cis- and trans-iso-
mers, and/or a mixture of isomers relating to the chirality at each
nitrogen [13,19]. After column chromatography, the emission data
is more consistent with a single isomer.
UV–Vis spectral data for Cr(III) tetraazamacrocycle complexes.
Complex
kmax
(
e
)
Ref.
trans-[Cr(cyclam)(CN)2]ClO4
trans-[Cr(isocyclam)(CN) 2]PF6
trans-[Cr([15]aneN4)(CN) 2]ClO4
trans-[Cr([16]aneN4)(CN) 2]PF6
414 (63), 328 (62)
440 (143), 341 (110)
452 (80), 348 (82)
464 (84), 357 (74)
[8b]
this work
[7b,8c]
this work
Table 2
Phosphorescence lifetimes (in
l
s) for [Cr(N4)(CN)2]+ complexes.a
Compound
s
ꢀC)
, H2O (20 , DMSO , 1:1 DMSO/H2O
s
s
(20ꢀC)
(77 K)
trans-
147b
215b
249
[Cr(isocyclam)(CN)2]+
Deutd trans-
404b
24
711
[Cr(isocyclam)(CN)2]+
trans-
38c
217
[Cr([16]aneN4)(CN)2]+
Deutd trans-
27d
2800e
3.3. Photobehavior
[Cr([16]aneN4)(CN)2]+
trans-[Cr(cyclam)(CN)2]+
Deutd trans-
335f
330f
355f
1500f
3060f
3.3.1. UV–Vis absorption spectra
[Cr(cyclam)(CN)2]+
UV–Vis data for the four trans-[Cr(N4)(CN)2]+ complexes
(N4 = cyclam, isocyclam, [15]aneN4, [16]aneN4) demonstrates the
effect of ring geometry on both ligand field strength and molar
absorptivity (Fig. 3, Table 1). For example, the increase in kmax as
one proceeds through this series from cyclam to [16]aneN4 is con-
sistent with previous reports of the ligand field strengths of these
macrocycles toward chromium(III), namely, cyclam > isocy-
clam > [15]aneN4 > [16]aneN4, and is consistent with hole-size
arguments [12,20]. In addition, it is noteworthy that isocyclam
has the largest molar absorptivity, an indication of the reduction
of centrosymmetry (and consequent relaxation of the LaPorte
selection rule) imposed by the ligand on this complex. Unusually
high extinction coefficients were also reported for trans-[Cr(1,4-
C2-cyclam)(CN)2]PF6 and were related to the lack of centrosymme-
try demonstrated by X-ray crystallography [6a].
a
Solutions were 0.01 M in HCl to avoid base quenching. This also suppresses H/D
exchange of the N-deuterated complexes [6b].
b
N2 purged.
c
Biexponential decay trace with a shorter lifetime of 8
Biexponential decay trace with a shorter lifetime of 3
Biexponential decay trace with a shorter lifetime of 860
Value reported in Ref. [8b].
l
l
s.
s.
ls.
d
e
f
impurity. The 77 K spectrum of [Cr(isocyclam)(CN)2]+ is dominated
by the 0 ? 0 emission band (721.2 nm) as is characteristic of com-
plexes with a lower degree of centrosymmetry as would be im-
parted by this macrocycle. In contrast, the 77 K emission
spectrum of [Cr([16]aneN4)(CN)2]+, shows vibronic structure and,
like the centrosymmetric cyclam complex, a relatively weak
0 ? 0 band (704.8 nm).
In addition, these complexes show a significant variation in
their room temperature excited state photobehavior in water, par-
ticularly when compared with the cyclam complex. For example,
the room temperature excited state lifetimes for the complexes
3.3.2. General excited state behavior
The photophysical data for the trans-[Cr(N4)(CN)2]+ complexes
are summarized in Table 2 and Fig. 4. The room temperature emis-
sion spectra are broad and moderately structured with peak posi-
tions similar to that observed for other trans-[Cr(N4)(CN)2]+
complexes that have been characterized as 2Eg (Oh) emitters [21].
Room temperature excitation spectra closely match the UV–Vis
spectra, confirming that the observed emission is not due to an
of cyclam, isocyclam, and [16]aneN4 are 335, 147, and 25
respectively. Noteworthy also is that the isocyclam complex is
slightly quenched by oxygen, i.e., the lifetime drops from 147
to 134 s in air saturated solution. Excited state quenching by oxy-
ls,
ls
l
gen is not observed for the complexes of cyclam or [16]aneN4 and
appears to be unique to the isocyclam complex. The lifetimes are
slightly longer in acidified DMSO (0.01 M HCl) than in aqueous
solution (Table 2). The emission of both the isocyclam and
[16]aneN4 complexes is quenched by base, a feature also reported
for the analogous cyclam [8b] and [15]aneN4 [8c] complexes. That
is, in a pH 9 buffer (phosphate, 0.01 M), the lifetime of the isocy-
140
120
100
80
clam complex drops to 22
complex drops to 12 s.
ls and the lifetime of the [16]aneN4
l
60
Aqueous solutions of neither trans-[Cr(isocyclam)(CN)2]PF6 nor
trans-[Cr([16]aneN4)(CN)2]PF6 show perceptible changes in the
UV–Vis spectrum after 1 h photolysis in a Rayonet Photochemical
Reactor operating with eight 350 nm lamps (ꢀ5 ꢂ 10ꢁ8 ein/s).
Thus, these complexes are reasonably photochemically inert, con-
sistent with other trans complexes of this type [6,8]. An aqueous
solution of the cis-[Cr([13]aneN4)(CN)2]PF6 complex, on the other
40
20
0
300
350
400
450
500
550
600
Wavelength (nm)
hand, has a very short room temperature lifetime (1 ls) and shows
significant decomposition after irradiation for 90 min at 350 nm.
Specifically, the absorbance of the peak at 338 nm drops by 30%;
the peak at 458 shows a similar drop in absorbance and a concom-
Fig. 3. UV–Vis spectra of trans-[Cr(N4)(CN)2]+ complexes: — trans-[Cr([15]aneN4)-
(CN)2]ClO4; - - - trans-[Cr(cyclam)(CN)2]ClO4; - - trans-[Cr([16]aneN4)(CN)2]PF6;
— trans-[Cr(isocyclam)(CN)2]PF6.

D.L. Grisenti et al. / Inorganica Chimica Acta 363 (2010) 157–162
161
Fig. 5. Possible stereoisomers for [16]aneN4 in trans-[Cr([16]aneN4)(CN)2]+.
RT
77K
biexponential excited state decay traces of the deuterated trans-
[Cr([16]aneN4)(CN)2]+ (Table 2) since triethylamine was used to
catalyze the H/D exchange.
In order to more clearly understand the excited state lifetimes
of these complexes, we adopt a formalism used extensively for
chromium(III) complexes which separates the excited state decay
rate constant kobs into temperature dependent and temperature
independent components [23].
670
690
710
730
750
770
Wavelength (nm)
sꢁ1 ¼ kobs ¼ kꢄ þ kðTÞ
ð1Þ
Here,
s is the observed lifetime, k° is the nearly temperature
independent limiting rate constant for relaxation, and k(T) is the
temperature dependent term. The k° term can also be expressed
as the sum of a radiative term, kꢄr , and a non-radiative term, kꢄnr,
and k(T) takes into account all temperature dependent terms.
Let us first consider the 77 K lifetimes of the complexes (Table
2), the inverse of which approximates k°. For the complexes of
RT
77K
4
the type discussed herein, where the 2Eg (Oh) and the A2g (Oh)
electronic states have very similar geometries, the non-radiative
relaxation involves tunneling from the excited state to the ground
state and is significantly affected by high frequency N–H vibra-
tions, and deuteration of the N–H bonds typically increases the
lifetime [24,25]. Note that at 77 K, like the cyclam complex [8b],
both the isocyclam and [16]aneN4 complexes demonstrate a signif-
icant deuterium isotope effect, consistent with the abovemen-
tioned tunneling mechanism. What is curious though, is that the
lifetime of the deuterated isocyclam complex is significantly less
than either the cyclam or [16]aneN4 complexes. A possible expla-
nation requires consideration that the k° term can be expressed
as the sum of a radiative term, kꢄr , and a non-radiative term, kꢄnr,
with the non-radiative term being largely due to the aforemen-
tioned relaxation involving the N–H oscillators. If kꢄr is large en-
ough to compete with kꢄnr for the isocyclam complex, then the
shorter lifetime of the deuterated complex is to be expected. The
reason this might be a factor for the isocyclam complex and not
for either the cyclam or [16]aneN4 complexes is that the value of
kr should increase with a decrease in centrosymmetry of the com-
plex [26] because of the direct dependence of kr on the oscillator
strength [27]. Given the effect of the Laporte selection rule on
the molar absorptivities of the trans-[Cr(N4)(CN)2]+ discussed pre-
viously, it is likely that the oscillator strength of the lower-symme-
try isocyclam complex will be the largest, and thus kr will be the
largest. Such an effect has also been observed for the trans-
[Cr(1,4-C2-cyclam)(CN)2]PF6 complex, where the 1,4-C2-cyclam li-
gand imposes a reduction of centrosymmetry [6a].
More significant differences in photobehavior are evident in the
k(T) term as indicated by the more than tenfold difference in room
temperature excited state lifetime between the cyclam and
[16]aneN4 complexes. Also, whereas the excited state lifetime of
the cyclam complex is nearly identical at 77 K and room tempera-
ture, the room temperature excited state lifetime of the isocyclam
complex is only a little over half of that at 77 K, and the room tem-
perature excited state lifetime of the [16]aneN4 complex is barely
over 10% of that at 77 K. Thus, these new complexes both access
a thermally activated decay mechanism at room temperature. Pos-
sibilities for thermally activated relaxation include photoreaction,
670
690
710
730
750
770
Wavelength (nm)
Fig. 4. Emission spectra of trans-[Cr(isocyclam)(CN)2]PF6 (top) and trans-[Cr([16]a-
neN4)(CN)2]PF6 (bottom).
itant 10 nm redshift; and a new shoulder grows in at 560 nm.
These findings are consistent with previous studies on
[Cr(N4)(CN)2]+ complexes which demonstrate that for trans com-
plexes, the dominant ligand field reaction mode is equatorial am(-
m)ine loss, whereas for the cis complexes, there is also appreciable
CNꢁ loss [22]. For the cases studied herein, productive N-photola-
bilization is precluded by the macrocyclic ligand effect, leaving
only photolabilization of the CNꢁ ligands as
photoreaction.
As discussed with the complex syntheses, the [Cr([16]a-
neN4)(CN)2]+ initially obtained in the synthesis is often pure as
demonstrated by elemental analysis. However, prior to chromatog-
raphy, the excited state decay trace shows clear biexponential
a productive
behavior, with lifetimes of approximately 3 and 25
matography, the decay trace is single exponential with a lifetime
near 25 s. If an aqueous solution (20 mg in 20 mL) of this column
ls. After chro-
l
purified material is stirred at room temperature for 1 h with a drop
of triethylamine added, the resulting species has UV–Vis spectra,
and room temperature and 77 K emission spectra that are indistin-
guishable from the column purified material, yet the biexponential
behavior has been restored. A likely explanation for this phenom-
enon is that the triethylamine catalyzes isomerization through
amine deprotonation and supports the idea that this material is a
mixture of isomers involving the N–H stereochemistry (Fig. 5).
+
+
Experiments on Cr([16]aneN4)Cl2 [13] and Co([16]aneN4)Cl2
[19] have demonstrated the existence of two stereoisomers for
each complex. We have been unable to grow X-ray quality crystals
for the unambiguous assignment of the stereochemistry. This tri-
ethylamine catalyzed isomerization is likely responsible for the

162
D.L. Grisenti et al. / Inorganica Chimica Acta 363 (2010) 157–162
4
and back-intersystem crossing to the T2g (Oh) excited state. Pro-
ductive photoreaction has been ruled out by the aforementioned
extended photolysis studies and back-intersystem crossing has
typically been ruled out for cyano-am(m)ine complexes [8c,22].
Non-productive photoreaction involving Cr–N dissociation fol-
lowed by reassociation (imposed by the macrocycle) is plausible
here and has been suggested for other macrocyclic trans-
[Cr(N4)(CN)2]+ complexes [6b].
Acknowledgements
This material is based upon work supported by the National Sci-
ence Foundation under Grant CHE-709562 and by the Henry Drey-
fus Teacher Scholar Award Program.
References
[1] S.R. Cooper (Ed.), Crown Compounds: Toward Future Applications, VCH, New
York, 1992.
[2] (a) R.D. Hancock, A.E. Martell, Chem. Rev. 89 (1989) 1875;
(b) T.A. Kaden, Chimia 54 (2000) 574.
[3] F. Cuenot, M. Meyer, E. Espinosa, A. Bucaille, R. Burgat, R. Guilard, C. Marichal-
Westric, Eur. J. Inorg. Chem. (2008) 267. and references therein.
[4] M. Shokeen, C. Anderson, Acc. Chem. Res. 42 (2009) 832. and references
therein.
Lifetime data for both the [16]aneN4 and isocyclam complexes
was collected over the range of 5–75 °C in 0.01 M HCl(aq). For
the [16]aneN4 complex, plots of ln(k ꢁ k°) versus Tꢁ1 (Eq. (2),
where k° was set to its 77 K value) were linear and gave a value
of 39 kJ/mol for Ea and A = 3 ꢂ 1011
.
[5] (a) R. Delgado, V. Felix, L.M.P. Lima, D.W. Price, Dalton Trans. (2007) 2734;
(b) S. Wang, T.D. Westmoreland, Inorg. Chem. 48 (2009) 719;
(c) T. Vitha, V. Kubicek, P. Hermann, L. Vander Elst, R.N. Muller, Z.I. Kolar, H.T.
Wolterbeek, W.A.P. Breeman, I. Lukes, J.A. Peters, J. Med. Chem. 51 (2008) 677.
[6] (a) P.S. Wagenknecht, C. Hu, D. Ferguson, L.C. Nathan, R.D. Hancock, J.R.
Whitehead, K. Wright-Garcia, M.T. Vagnini, Inorg. Chem. 44 (2005) 9518;
(b) M.T. Vagnini, N.A.P. Kane-Maguire, P.S. Wagenknecht, Inorg. Chem. 45
(2006) 3789.
k ¼ kꢄ þ AeꢁEa=RT
ð2Þ
Even without consideration of k°, the Arrhenius plot was nearly
linear, indicating the lifetime of the complex over this temperature
range is almost entirely determined by the thermally activated
pathway and explains why the [16]aneN4 complex shows little to
no room temperature deuterium isotope effect on its excited state
lifetime. A simple Arrhenius plot for the corresponding isocyclam
complex data, on the other hand, shows substantial curvature over
the temperature range, indicating that the temperature indepen-
dent term plays a significant role in excited state relaxation at or
near room temperature, explaining why this complex shows a sig-
nificant room temperature deuterium isotope effect on its excited
state lifetime. A three parameter fit according to Eq. (2) returned
values of Ea = 39 kJ/mol, A = 1 ꢂ 1010, and k° = 5900 sꢁ1. Thus, the
reason that the lifetime of the [16]aneN4 complex is almost com-
pletely determined by the thermally activated mechanism,
whereas the k° term is significant for the isocyclam complex lies
in the difference in the pre-exponential factor, not the activation
energies. It is noteworthy that the activation barriers for both com-
plexes are very similar to the value of 38 kJ/mol determined for
trans-[Cr(cyclam)(CN)2]+ [28]. It is also curious that the value for
k° for the [16]aneN4 complex (5900 sꢁ1, which corresponds to
[7] (a) P.S. Wagenknecht, N.A.P. Kane-Maguire, D.G. Speece, N. Helwic, Inorg.
Chem. 41 (2002) 1229;
(b) M.T. Vagnini, W.C. Rutledge, P.S. Wagenknecht, Inorg. Chem., submitted for
publication.
[8] [12]ane-N4 and [13]ane-N4 would presumably give only cis- isomers due to
their small hole sizes. The trans-[Cr(N4)(CN)2] complexes of cyclam and 15-
ane-N4 are known: (a) N.A.P. Kane-Maguire, J.A. Bennett, P.K. Miller, Inorg.
Chem. Acta 76 (1983) L123;
(b) N.A.P. Kane-Maguire, W.S. Crippen, P.K. Miller, Inorg. Chem. 22 (1983) 696;
(c) R.B. Lessard, M.J. Heeg, T. Buranda, M.W. Perkovic, C.L. Schwarz, Y. Rudong,
J.F. Endicott, Inorg. Chem. 31 (1992) 3091.
[9] (a) L. Sabatini, L. Fabbrizzi, Inorg. Chem. 18 (1979) 438;
(b) R.W. Hay, B. Kinsman, C.I. Smith, Polyhedron 14 (1995) 1245.
[10] W.L. Smith, J.D. Ekstrand, K.N. Raymond, J. Am. Chem. Soc. 100 (1978) 3539.
and references therein.
[11] G.R. Brubaker, F.H. Jarke, I.M. Brubaker, Inorg. Chem. 18 (1979) 2032.
[12] E. Itoh, K. Yamamoto, M. Funato, M. Suzuki, A. Uehara, Thermochim. Acta 195
(1992) 51.
[13] J. Eriksen, O. Mønsted, Acta Chem. Scand. A 38 (1984) 775.
[14] R.G. Swisher, G.A. Brown, R.C. Smierciak, E.L. Blinn, Inorg. Chem. 20 (1981)
3947.
[15] I. Tabushi, Y. Taniguchi, H. Kato, Tetrahedron Lett. 18 (1977) 1049.
[16] M. Kodama, H. Anan, T. Koike, E. Kimura, Bull. Chem. Soc. Jpn. 62 (1989) 4044.
[17] J. Ferguson, M.L. Tobe, Inorg. Chim. Acta 4 (1969) 109.
[18] R.G. Swisher, E.L. Blinn, Inorg. Chem. 25 (1986) 1320.
[19] Y. Hung, L.Y. Martin, S.C. Jackels, M. Tait, D.H. Busch, J. Am. Chem. Soc. 99
(1977) 4029.
s
= 170
ls), is not in good agreement with the 77 K excited state
lifetime of 217
l
s. This may simply reflect subtle differences in
the medium (the 77 K experiments were run in a DMSO/H2O glass)
or may indicate a discontinuity in the lifetime versus temperature
behavior that can occur with large changes in solvent mobility as
has been observed for other [Cr(N4)X2]+ complexes [29].
[20] V.J. Thöm, J.C.A. Boeyens, G.J. McDougall, R.D. Hancock, J. Am. Chem. Soc. 106
(1984) 3198.
[21] A.F. Fucaloro, L.S. Forster, S.G. Glover, A.D. Kirk, Inorg. Chem. 24 (1985) 4242.
[22] E. Zinato, Coord. Chem. Rev. 129 (1994) 195.
[23] L.S. Forster, Coord. Chem. Rev. 227 (2002) 59.
[24] J.F. Endicott, R.B. Lessard, Y. Lei, C.K. Ryu, R. Tamilarasan, in: A.B.P. Lever (Ed.),
Excited States and Reactive Intermediates, ACS Symposium Series, vol. 307,
American Chemical Society, Washington, DC, 1986, p. 85.
[25] K. Kühn, F. Wagestian, H. Kupka, J. Phys. Chem. 85 (1981) 665.
[26] J.F. Endicott, R.B. Lessard, D. Lynch, M.W. Perkovic, C.K. Ryu, Coord. Chem. Rev.
97 (1990) 65.
[27] A.W. Adamson, P.D. Fleischauer, in: Concepts of Inorganic Photochemistry,
Wiley and Sons, New York, 1975, p. 16.
[28] J.F. Endicott, R. Tamilarasan, R.B. Lessard, Chem. Phys. Lett. 112 (1984) 381.
[29] (a) A.M. Ghaith, L.S. Forster, J.V. Rund, J. Phys. Chem. 92 (1988) 6197;
(b) M.T. Vagnini, W.C. Rutledge, C. Hu, D.G. VanDerveer, P.S. Wagenknecht,
Inorg. Chim. Acta 360 (2007) 1482.
4. Conclusions
Tetraazamacrocyclic ligands have continued to find applications
involving their metal ion binding constants and selectivity. We
have provided preparative details for new syntheses of two of
the less-utilized ligands in this group that should make these li-
gands more accessible. Chiefly, we have expanded the set of
well-characterized trans-[Cr(cyclam)(CN)2]+ complexes available
for our energy transfer studies.