Synthesis and photochemical evaluation of iodinated squarylium cyanine dyes

Article abstract of DOI:10.1002/hlca.200590084

Several (multiply) iodinated squarylium cyanine dyes of type 1 and 8 (see Scheme and Table), derived from 1,3-benzothiazole and 6-iodo-1,3-benzothiazole, were synthesized as potential new photosensitizers, with absorptions in the 700-nm region. Their ability to generate singlet oxygen (¹O ₂) was assessed by luminescence-decay measurement in the near-IR. Some of these new dyes show interesting photophysical properties, and may be potentially used in photodynamic therapy (PDT).

Full text of DOI:10.1002/hlca.200590084

Helvetica Chimica Acta ± Vol. 88 (2005)
1135
Synthesis and Photochemical Evaluation of Iodinated Squarylium Cyanine
Dyes
a
a
a
a
b
by Paulo F. Santos* ), Lucinda V. Reis ), Isabel Duarte ), Jo aÄ o P. Serrano ), Paulo Almeida ),
c
c
Anabela S. Oliveira ), and Luís F. Vieira Ferreira )
a) Departamento de Química and Centro de Química ± Vila Real,
Universidade de Tr a s-os-Montes e Alto Douro, Apartado 1013, PT-5000-911 Vila Real
(
fax: ‡ 351259350480; e-mail: psantos@utad.pt)
b) Departamento de Química and Unidade de I&D de Materiais T eà xteis e Papeleiros,
Universidade da Beira Interior, PT-6201-001 CovilhaÄ
c) Centro de Química-Física Molecular, Complexo Interdisciplinar, Instituto Superior T e cnico,
PT-1049-001 Lisboa
Several (multiply) iodinated squarylium cyanine dyes of type 1 and 8 (see Scheme and Table), derived from
1
,3-benzothiazole and 6-iodo-1,3-benzothiazole, were synthesized as potential new photosensitizers, with
1
absorptions in the 700-nm region. Their ability to generate singlet oxygen ( O
2
) was assessed by luminescence-
decay measurement in the near-IR. Some of these new dyes show interesting photophysical properties, and may
be potentially used in photodynamic therapy (PDT).
Introduction. ± Photodynamic therapy (PDT) is a fast-developing therapeutic
modality that allows the selective destruction of diseased tissue, especially for the
treatment of cancer [1]. PDT involves the administration of a photosensitizing drug
that is preferentially taken up or retained by neoplastic cells. The drug itself is harmless
and has no effect on either healthy or abnormal tissue. However, when the tissue that
contains the photosensitizer is irradiated with light of appropriate wavelength, the
sensitizer is excited to a short-lived singlet state, which, after intersystem crossing
(
ISC), relaxes to the more-stable, longer-lived triplet state. This triplet-activated
sensitizer can then interact with a suitable biological substrate, by proton or electron
transfer, leading to radical formation (type-I reaction), or transfer its energy directly to
3
1
ground-state oxygen ( O ), generating highly reactive singlet oxygen ( O ) in a type-II
2
2
reaction. Both processes may lead to damage or destruction of cellular components,
and, eventually, cause cell death restricted to the illuminated area [2]. It is widely
1
believed that O , resulting from the conversion of dissolved ground-state triplet
2
3
oxygen ( O ), is the primary cytotoxic agent responsible for tumor-cell inactivation [3].
2
So far, hematoporphyrin derivatives have been the most commonly and widely used
drugs for medical applications of PDT, but several disadvantages of these ꢀfirst-
generationꢁ photosensitizers ± namely chemical heterogeneity, relatively poor photo-
efficiency, and prolonged skin photosensitivity ± have led, in recent years, to great
efforts in the search for other classes of sensitizers with improved medical properties
[
4]. Besides selectivity for cancer cells and the lack of collateral toxic and phototoxic
effects, any potential PDT drug candidate must fulfill several other multidisciplinary
ꢂ 2005 Verlag Helvetica Chimica Acta AG, Zürich

1
136
Helvetica Chimica Acta ± Vol. 88 (2005)
5
À1
À1
requirements such as 1) strong absorption (e > 10 m cm ) within the so-called
phototherapeutic windowꢁ (600 ± 1000 nm), in which light is only weakly scattered by
ꢀ
living tissue, and where its penetration depth increases; and 2) the ability to generate
1
cytotoxic O [5].
2
1
We have recently explored the O production efficiency of several squarylium
2
cyanines [6], a particular class of the large family of squarylium dyes. Despite their
special properties, such as sharp and intense absorption in the VIS and near-IR region,
as well as considerable thermo and photostability, which have turned them into
particularly attractive substrates for technological applications in the domain of
photonics [7], they have been scarcely studied with regard to PDT [8]. Our previously
obtained, encouraging results [6] prompted us to improve the capacity of this type of
1
dyes for O generation. The introduction of heavy atoms (such as iodine) in the
2
aromatic moiety of the dye seemed to be a straightforward way of increasing the ISC
from the singlet to the triplet state of the sensitizer through enhancement of spin ±
1
orbital coupling, and, consequently, the efficiency of O production.
2
Herein, we report the synthesis and photophysical properties (quantum yields of
O production) of several iodinated squarylium cyanine dyes that readily absorb within
1
2
the ꢀphototherapeutic windowꢁ. Some non-halogenated analogues were also prepared,
1
and their O2 quantum yields were compared with the quantum yields of the
halogenated analogues. Thereby, preference was given to aminosquarylium dyes,
since the incorporation of amino moieties usually induces a convenient bathochromic
shift in the absorption maximum of the compound relative to the unsubstituted,
neutral parent squarylium dyes [9]. Also, amino groups offer a convenient way of
linking the dye to biological substrates, and the cationic nature of these dyes may
also be an advantage, since cationic cyanines seem to be potential sensitizers for PDT
[
10].
Results and Discussion. ± 1. Synthesis. Our strategy for the synthesis of iodinated
squarylium dyes with amino functions started with (the unsuccessful) attempt of direct
iodination of the cyanine 1c with ICl, a method recently described for the iodination of
squarylium hemicyanines [8c]. Since 1c is a relatively valuable substrate, further efforts
on either direct or multistep iodination seemed not to be worthwhile, and we, thus, used
as starting compound 6-iodo-2-methyl-1,3-benzothiazole (2), which was easily pre-
pared through classical aromatic iodo-dediazotation (Scheme). Thus, nitration of 2-
methyl-1,3-benzothiazole (3) with HNO /H SO , followed by reduction of the resulting
3
2
4
nitro compound 4 with 10% Pd/C in the presence of HCO NH yielded the aromatic
2
4
amine 5. The latter was converted to 2 in good overall yield through diazotation with
NaNO , followed by in situ treatment with aqueous KI.
2
Quaternization of the free base 2 with an excess of hexyl or dodecyl iodide provided
the salts 6a and 6b, respectively, which were then condensed with squaric acid (ˆ 3,4-
dihydroxycyclobut-3-ene-1,2-dione) to the corresponding squarylium cyanine dyes 1 in
refluxing BuOH/pyridine. Methylation of 1 with ꢀmethyl triflateꢁ (ˆCF SO Me)
3
3
afforded the intermediary methyl ethers 7, whose MeO groups could be substituted
with different aromatic amines to afford the desired iodinated dyes 8a ± 8c. The non-
halogenated analogues 8d and 8e (for structures, see Table below) were similarly
prepared from the corresponding non-iodinated analogues of 7 according to [9a].

Helvetica Chimica Acta ± Vol. 88 (2005)
1137
Scheme
a) HNO
e) C
3
, H
13I or C12
2
SO
H
4
. b) 10% Pd/C, HCO
25I, MeCN, reflux. f) Squaric acid, BuOH, pyridine, reflux. g) CF
NH ) or 3-iodobenzene-amine (3-I-C -NH ), CH
2
NH
4
, MeOH/THF 2 :1. c) NaNO
2
, aq. HCl, 08. d) aq. KI, 08 ! 30 to 408.
SO Me, CH Cl . h) Aniline
Cl
6
H
3
3
2
2
(C
6
H
5
2
6
H
4
2
2
2
.
2
. Photophysical Properties. The iodinated squarylium cyanine dyes 1a,b and 8a ± c
exhibit absorptions within the ꢀphototherapeutic windowꢁ, with high molar absorption
coefficients e. As can be seen from the Table, these compounds, upon irradiation at
1
3
37 nm, produce singlet oxygen ( O ) in variable quantum yields f. In contrast to
2
alkylamino- and (dialkylamino)-substituted squarylium cyanines [9], the presence of
aromatic residues at the central, four-membered ring had a negligible effect (or no
effect at all) on the VIS spectrum, when compared to that of the corresponding
unsubstituted compound. However, the introduction of I-atoms at the termini of the
chromophoric, heteroaromatic system caused a bathochromic shift of ca. 12 ± 15 nm
(
compare lmax value of 1a with that of 1c [6a] in the Table; see also Fig. 1).
1
As expected, all the iodinated dyes produced O2 more efficiently than the
corresponding non-halogenated analogues. Among the dyes with the same carbon
skeleton, the compounds with the highest numbers of I-atoms invariably gave rise to
the highest efficiency.
1
The relative enhancement in O quantum yields f of the aminosquarylium dyes 8
2
due to the ꢀinternal heavy-atom effectꢁ depends on the previous presence of I-atoms in
the molecule. In fact, while iodination of the non-halogenated substrate 8d (to formally

1
138
Helvetica Chimica Acta ± Vol. 88 (2005)
1
Table. UV/VIS Spectral Data,Yield of Singlet Oxygen ( O
2
) Production,and Quantum Yield for Compounds 1
2 2
and 8. In CH Cl at ambient temperature; excitation wavelength 337.1 nm (or 355 nm; see Exper. Part).
Compound
l
max [nm] (log e)
Yield [%]
Quantum yield
1
1
1
8
8
8
8
8
a
b
c )
a
b
c
d
e
684 (5.47)
684 (5.46)
672 (5.39)
678 (5.35)
675 (5.20)
675 (5.28)
663 (5.29)
660 (5.19)
62
47
76
90
83
34
60
65
0.32 Æ 0.02
0.22 Æ 0.02
0.26 Æ 0.02
0.46 Æ 0.03
0.66 Æ 0.05
0.33 Æ 0.02
0.32 Æ 0.02
0.39 Æ 0.03
a
a) Data from [6a].
Fig. 1. UV/VIS Absorption spectra of selected squarylium cyanine dyes. In CH
2
Cl
2
at ambient temperature.

Helvetica Chimica Acta ± Vol. 88 (2005)
1139
produce 8a and 8e) results in a consistent contribution to f of 0.07 per I-atom,
additional iodination of 8a,e to 8b induces an increment in f that is more pronounced
with the introduction of iodine in aniline (0.20) than in the heteroaromatic ring (0.135
per I-atom).
1
In general, amino-substituted squarylium cyanine dyes show quantum yields for O
2
generation superior to those of the parent, unsubstituted compounds. In spite of the
possibility that an eventual enhancement is originated by the ꢀexternal heavy-atom
effectꢁ, which is absent in the zwitterionic squarylium dyes 1, a possibly higher degree of
rigidification in compounds 8 (due to restricted rotation about the CÀN bond) may
also contribute to the observed difference. Restriction to free rotation eventually leads
to a decrease of the efficiency of the non-radiative decay by photo-isomerization. As a
1
consequence, there is an increase of the efficiency of ISC and O production.
2
Hindered rotation of the amino group in 8 was also suggested by the two very broad,
1
separated H-NMR signals for the methine H-atoms at ambient temperature, indicating
non-equivalence of chemical environments and local field inhomogeneity. Upon
recording the same spectra at 60 ± 708, the two signals collapsed to a slightly broadened
singlet of intermediate chemical shift.
We have reported before that related squarylium cyanine dyes derived from
1
benzothiazole and containing pendant N-hexyl groups give rise to enhanced O2
quantum yields compared to those of the N-ethyl analogs [6b]. Interestingly, in all of
1
the new dyes, increasing the N-alkyl chain from C H to C H diminished the O
6
13
12 25
2
quantum yields (Table). For shorter N-alkyl chains (less than six C-atoms), the increase
in chain length decreases the relative importance of the non-radiative pathways of
deactivation (photo-isomerization) [6]. However, for longer alkyl chains (more than
six C-atoms), the major effect is probably the increase of internal conversion
(associated with vibrational relaxation), which becomes more important in this case
relative to photo-isomerization and ISC processes.
In conclusion, taking into consideration the absorption and quantum yields for O2
generation, some of the newly synthesized dyes could be tested as PDT sensitizers.
1
The authors are grateful to FCT, Lisbon, POCTI, and FEDER, for their financial support (Project POCTI/
3
2915/QUI/00).
Experimental Part
General. All reagents were purchased from Sigma-Aldrich and used without further purification. Solvents
were of anal. grade. All reactions were monitored by thin-layer chromatography (TLC) using 0.25-mm Al-
backed silica gel plates (Merck 60 F254). Melting points (m.p.) were measured in open capillaries on a Büchi 530
melting-point apparatus; uncorrected. IR Spectra were recorded on a Mattson 5000 FT-IR spectrophotometer;
À1
1
in cm . UV/VIS Spectra were recorded on a Perkin-Elmer Lambda-6 instrument; lmax in nm (log e). H- and
1
3
C-NMR spectra were recorded on Bruker ACP-250, ARX-300, and ARX-400 spectrometers; chemical shifts d
in ppm rel. to residual solvent signals or Me Si, coupling constants J in Hz. Fast-atom-bombardment electron-
4
impact mass spectra (FAB-EI-MS) were recorded on Micromass AutoSpec-M and AutoSpec-Q spectrometers
operating at 70 eV, using a 4-nitrobenzyl alcohol as matrix; in m/z.
2
-Methyl-6-nitro-1,3-benzothiazole (4). To an ice-cooled soln. of 2-methyl-1,3-benzothiazole (3; 5.87 g,
9 mmol) in H SO (23 ml) was slowly added a mixture of HNO (4.7 ml) and H SO (3.5 ml). The mixture was
stirred for 3 h at r.t., poured on ice, and left overnight. The resulting precipitate was collected by filtration,
washed several times with cold H O, and recrystallized from MeOH to afford 4 (4.86 g, 64%). Yellowish crystals.
M.p. 166.5 ± 167.58 (lit. 168 ± 1708 [11]). IR (KBr): 3097w, 3066w, 1600w, 1569m, 1506s, 1436m, 1332s, 1267s,
3
2
4
3
2
4
2

1
140
Helvetica Chimica Acta ± Vol. 88 (2005)
1
1
238m, 1174m, 1116m, 1037w, 993w, 916w, 902m, 821m, 752m, 717m, 671w, 647w, 547w. H-NMR (400 MHz,
CDCl
3
): 2.91 (s, 3 H); 8.02 (d, J ˆ 9.0, 1 H); 8.32 (dd, J ˆ 2.3, 9.0, 1 H); 8.76 (d, J ˆ 2.3, 1 H). FAB-MS: 195
‡
(
[M ‡ H] ).
6
-Amino-2-methyl-1,3-benzothiazole (5). To a soln. of 4 (2.0 g, 10 mmol) in THF (40 ml) and MeOH
(
20 ml), 10% Pd/C (750 mg) and HCOONH (9.74 g, 150 mmol) were added, and the mixture was stirred at r.t.
4
for 3 d. The mixture was filtered through a bed of Celite, washing with hot MeOH. The filtrate was evaporated to
dryness, the resulting residue was dissolved in CHCl , and extracted with brine. The org. layer was dried
), filtered, and evaporated in vacuo to afford 5 (1.55 g, 92%).White crystals. M.p. 119 ± 1208 (lit. 125 ±
3
(
2 4
Na SO
1
1
D
268 [12]). IR (KBr): 3376m, 3315m, 3207m, 1637m, 1604s, 1560m, 1517m, 1465s, 1432m, 1288m, 1228s, 1166s,
1
049w, 829m, 817m, 728w, 649m. H-NMR (400 MHz, CDCl
3
): 2.74 (s, 3 H); 3.70 (br. s, 2 H, exchanging with
‡
2
O); 6.78 (dd,J ˆ 2.3, 8.6, 1 H); 7.05 (d, J ˆ 2.3, 1 H); 7.69 (d, J ˆ 8.6, 1 H). FAB-MS: 165 ([M ‡ H] ).
6
-Iodo-2-methyl-1,3-benzothiazole (2). To an ice-cooled soln. of 5 (1.48 g, 9.02 mmol) in conc. HCl (4 ml)
and H O (4 ml), a soln. of NaNO (1.14 g, 16 mmol) in H O (5.2 ml) and, 30 min later, a soln. of KI (2.59 g,
6 mmol) in H O (3 ml), were added dropwise. After stirring for 10 min., the mixture was allowed to gradually
reach r.t., and then heated at 30 ± 408 for 15 min. The resulting dark cake was dissolved in CH Cl and washed
with 5% aq. NaHSO soln., until a pale yellow color was obtained. The org. layer was dried (Na SO ) and
2
2
2
1
2
2
2
3
2
4
evaporated in vacuo, and the resulting residue was recrystallized from EtOH to afford 2 (1.70 g, 79%). White
crystals. M.p. 140 ± 1418 (lit. 140 ± 1418 [13]). IR (KBr): 1511m, 1430s, 1390w, 1371w, 1297w, 1268m, 1238m,
1
1
8
178m, 1166m, 858w, 817s, 740w, 665w, 651w. H-NMR (400 MHz, CDCl
3
): 2.81 (s, 3 H); 7.65 ± 7.73 (m, 2 H);
‡
.14 (d, J ˆ 1.5, 1 H). FAB-MS: 276 ([M ‡ H] ).
General Procedure for the Synthesis of the Quaternary Ammonium Salts 6. Asoln. of 2 (1.0 mmol) and 1-
iodohexane or 1-iodododecane (3.0 ± 5.0 mmol) in MeCN (25 ml) was heated at reflux for 5 d. After cooling,
Et
Et
2
O was added, the precipitated ammonium salt was collected by vacuum filtration, washed several times with
O, and dried in vacuo to afford spectroscopically pure 6. The combined etheral filtrates were evaporated in
2
vacuo, and the remaining residue, consisting mainly of unreacted starting materials, was heated at reflux for
another 5 d, followed by the same workup as described above. This process was repeated three times to achieve
suitable yields of 6 (see below).
3
-Hexyl-6-iodo-2-methyl-1,3-benzothiazol-3-ium Iodide (6a). Yield: 64%. White crystals. M.p. 202 ± 2038.
1
IR (KBr): 3050w, 2952w, 2923w, 2852w, 1565w, 1513w, 1452w, 1382s, 1319w, 823w. H-NMR (400 MHz, CDCl
3
/
CD
3
OD): 0.79 (t, J ˆ 6.9, 3 H); 1.24 ± 1.28 (m, 4 H); 1.36 ± 1.40 (m, 2 H); 1.81 ± 1.85 (m, 2 H); 3.68 (s, 3 H); 4.64
‡
(
t, J ˆ 7.9, 2 H); 7.70 (d, J ˆ 8.9, 1 H); 8.03 (dd, J ˆ 1.4, 8.9, 1 H); 8.48 (d, J ˆ 1.3, 1 H). FAB-MS: 360 (M ).
3
-Dodecyl-6-iodo-2-methyl-1,3-benzothiazol-3-ium Iodide (6b). Yield: 20%. White crystals. M.p. 191 ± 1938.
1
IR (KBr): 2915s, 2848s, 1569w, 1511w, 1465w, 1450w, 1382m, 1326w, 802w, 719w, 557w, 530w. H-NMR
(
(
4
400 MHz, CDCl
3
): 0.86 (t, J ˆ 6.8 Hz, 3 H); 1.23 ± 1.34 (m, 16 H); 1.43 ± 1.47 (m, 2 H); 1.90 ± 1.93 (m, 2 H); 3.39
s, 3 H); 4.78 (t, J ˆ 7.8, 2 H); 7.76 (d, J ˆ 8.9, 1 H); 8.04 (dd, J ˆ 1.3, 8.8, 1 H); 8.66 (d, J ˆ 1.4, 1 H). FAB-MS:
‡
44 (M ).
General Procedure for the Synthesis of Squarylium Dyes 1. The novel squarylium dyes 1 were prepared, in
analogy to a literature procedure [14], by condensing the iodides 6 with squaric acid (ˆ 3,4-dihydroxycyclobut-
-ene-1,2-dione) in BuOH/pyridine at reflux.
3
(
4Z)-4-[(3-Hexyl-6-iodo-1,3-benzothiazol-3-ium-2-yl)methylidene]-2-[(Z)-(3-hexyl-6-iodo-1,3-benzothia-
1
zol-2(3H)-ylidene)methyl]-3-oxocyclobut-1-en-1-olate (1a) ). Yield: 62%. Metallic, green crystals. M.p. 2618
(
1
1
dec.). UV/VIS (CH
187w, 1093m, 1074w, 954w, 755w, 514w. H-NMR (400 MHz, CDCl
2 H); 1.67 ± 1.77 (m, 4 H); 4.02 (br. t, J ˆ 7.4, 4 H); 5.96 (br. s, 2 H); 6.86 (d, J ˆ 8.5, 2 H); 7.63 (d, J ˆ 8.6, 2 H);
2
Cl
2
): 684 (5.47). IR (KBr): 2921w, 2848w, 1641w, 1590w, 1573w, 1459s, 1431s, 1355w, 1243s,
1
3
): 0.89 (br. t, J ˆ 6.8, 6 H); 1.31 ± 1.42 (m,
‡
‡
7.78 (s, 2 H). HR-FAB-MS: 797.0220 ([M ‡ H] , C32
H
35
I
2
N
2
O
2
S
2
; calc. 797.0230).
(
4Z)-4-[(3-Dodecyl-6-iodo-1,3-benzothiazol-3-ium-2-yl)methylidene]-2-[(Z)-(3-dodecyl-6-iodo-1,3-ben-
1
zothiazol-2(3H)-ylidene)methyl]-3-oxocyclobut-1-en-1-olate (1b) ). Yield: 47%. Metallic, red crystals. M.p.
2
1
508 (dec.). UV/VIS (CH
2
Cl
2
): 684 (5.46). IR (KBr): 2921m, 2850w, 1666w, 1587m, 1571m, 1459s, 1432s, 1388m,
1
369s, 1353m, 1330m, 1241s, 1143w, 1099s, 1076m, 1031w, 960w, 784w, 761w, 738w, 520w. H-NMR (400 MHz,
CDCl
3
): 0.87 (br. t, J ˆ 6.6, 6 H); 1.25 ± 1.39 (m, 36 H); 1.74 ± 1.76 (m, 4 H); 3.99 (br. t, J ˆ 7.5, 4 H); 5.84 (br. s,
‡
2
H); 6.84 (d, J ˆ 8.7, 2 H ) ; 7.61 (d, J ˆ 8.6, H); 7.76 (s, 2 H). HR-FAB-MS: 965.2073 ([M ‡ H] ,
‡
C
44
H
59
I
2
N
2
O
2
S
2
; calc. 965.2108).
1)
Since the chromophore is fully delocalized, the systematic name of only one resonance structure is given
the one shown in the Scheme or Table).
(

Helvetica Chimica Acta ± Vol. 88 (2005)
1141
General Procedure for the Synthesis of the Methylated Squarylium Dyes 7. To a soln. of 1 (ca. 0.5 mmol) in
anh. CH Cl (50 ml), vigorously stirred under N atmosphere, CF SO CH (4 equiv.) was added at r.t. After 3 ±
h, the mixture was quenched with cold 5% aq. NaHCO soln. The org. layer was dried (Na SO ) and
evaporated, and the resulting residue was recrystallized from CH Cl /MeOH.
2
2
2
3
3
3
5
3
2
4
2
2
3
-Hexyl-2-[(Z)-{3-[(Z)-(3-hexyl-6-iodo-1,3-benzothiazol-2(3H)-ylidene)methyl]-2-methoxy-4-oxocyclo-
1
but-2-en-1-ylidene}methyl]-6-iodo-1,3-benzothiazol-3-ium Trifluoromethanesulfonate (7a) ). Yield: 99%. Met-
allic, green crystals. M.p. 2508 (dec.). UV/VIS (CH
2
Cl
2
): 655 (5.48). IR (KBr): 2927w, 2865w, 1650w, 1575w,
1
1
500w, 1421s, 1351m, 1322w, 1243s, 1209s, 1145s, 1114s, 1025m, 806w, 744w, 636w, 518w. H-NMR (400 MHz,
CDCl
3
): 0.87 (br. t, J ˆ 6.6, 6 H); 1.30 ± 1.41 (m, 12 H); 1.76 ± 1.82 (m, 4 H); 4.33 (br. t, J ˆ 7.0, 4 H); 4.61 (s, 3 H);
‡
5
.98 (br. s, 2 H); 7.11 (d, J ˆ 8.7, 2 H); 7.75 (d, J ˆ 8.6, 2 H); 7.91 (s, 2 H). HR-FAB-MS: 811.0361 (M ,
‡
C
33
H
37
I
2
N
2
O
2
S
2
; calc. 811.0380).
3
-Dodecyl-2-[(Z)-{3-[(Z)-(3-dodecyl-6-iodo-1,3-benzothiazol-2(3H)-ylidene)methyl]-2-methoxy-4-oxocy-
1
clobut-2-en-1-ylidene}methyl]-6-iodo-1,3-benzothiazol-3-ium Trifluoromethanesulfonate (7b) ). Yield: 89%.
Metallic, red crystals. M.p. 222 ± 2248. UV/VIS (CH
2
Cl
2
): 655 (5.51). IR (KBr): 2921w, 2850w, 1654w, 1577w,
1
1
6
2
502w, 1425s, 1353m, 1243s, 1209m, 1143s, 1025w, 746w, 636w, 516w. H-NMR (400 MHz, CDCl
3
): 0.86 (br. t, J ˆ
.7, 6 H); 1.22 ± 1.39 (m, 36 H); 1.79 (br. s, 4 H); 4.33 (br. s, 4 H); 4.61 (s, 3 H); 5.97 (br. s, 2 H); 7.11 (d, J ˆ 8.8,
‡
‡
H); 7.75 (d, J ˆ 8.6, 2 H); 7.90 (s, 2 H). HR-FAB-MS: 979.2237 (M , C45
H
61
I
2
N
2
O
2
S
2
; calc. 979.2264).
General Procedure for the Synthesis of the Aminosquarylium Dyes 8. The appropriate aromatic amine
(
sixfold excess) was added to a soln. of 7 (ca. 0.4 mmol) in anh. CH
mixture was stirred at r.t. for 6 d in the presence of catalytic amounts of Et
dimethylamino)pyridine). The mixture was extracted with cold H O, the org. layer was separated by
decantation, dried (Na SO ), and evaporated in vacuo. The resulting residue was dissolved in MeOH, and to this
soln. was added 14% aq. KI soln. (approximately the same volume). After 2 h, the precipitated dye, bearing the
iodide counter-ion, was collected by filtration, washed with H O, and recrystallized from CH Cl /MeOH/Et O to
2
Cl
2
(40 ml) under N
2
atmosphere, and the
3
N and DMA P ˆ( 4-
(
2
2
4
2
2
2
2
afford 8a ± c. Compounds 8d and 8e were prepared analogously, but starting from the corresponding non-
iodinated dyes (Table).
3
-Hexyl-2-[(Z)-{3-[(Z)-(3-hexyl-6-iodo-1,3-benzothiazol-2(3H)-ylidene)methyl]-2-(phenylamino)-4-oxo-
1
cyclobut-2-en-1-ylidene}methyl]-6-iodo-1,3-benzothiazol-3-ium Iodide (8a) ). Yield: 90%. Violet crystals. M.p.
2
1
6
2
708 (dec.). UV/VIS (CH
349m, 1247s, 1172m, 1147m, 1076w, 1024w, 983w, 796w, 759w, 518w. H-NMR (400 MHz, CD
H); 1.30 ± 1.52 (m, 14 H); 1.83 (br. s, 2 H); 3.71 (br. s, 2 H); 4.53 (br. s, 2 H); 5.02 (br. s, 1 H); 6.95 ± 7.11 (m,
2
H); 7.31 ± 7.44 (m, 6 H); 7.46 ± 7.95 (m, 4 H); 10.92 (s, 1 H, exchanging with D O). C-NMR (62.5 MHz,
2
Cl
2
): 678 (5.35). IR (KBr): 3457w, 2925w, 2852w, 1631w, 1594w, 1542w, 1492m, 1434s,
1
2
Cl
2
): 0.89 (br. s,
1
3
(
1
D
6
)DMSO, Tˆ 608): 13.4; 21.5; 25.3; 26.6; 30.4; 46.3; 87.0; 88.4; 114.9; 124.2; 126.4; 129.0; 129.5; 130.6; 136.1;
‡
‡
40.2; 160.3; 177.6. HR-FAB-MS: 872.0690 (M , C38
H
40
I
2
N
3
OS
2
; calc. 872.0702).
3
-Hexyl-2-[(Z)-{3-[(Z)-(3-hexyl-6-iodo-1,3-benzothiazol-2(3H)-ylidene)methyl]-2-[(3-iodophenyl)ami-
1
no]-4-oxocyclobut-2-en-1-ylidene}methyl]-6-iodo-1,3-benzothiazol-3-ium Iodide (8b) ). Yield: 83%. Purple
crystals. M.p. 2428 (dec.). UV/VIS (CH
2
Cl
2
): 675 (5.20). IR (KBr): 3434w, 2925w, 2854w, 1637w, 1579w, 1529w,
1
1
492m, 1463m, 1444s, 1347m, 1267m, 1247m, 1170m, 1149w, 1066w, 987w, 802w, 771w, 520w. H-NMR (250 MHz,
(
2
D
6
)DMSO, T ˆ 608): 0.87 (br. t, J ˆ 6.5, 6 H); 1.26 (br. s, 12 H); 1.62 (br. s, 4 H); 4.13 (br. s, 4 H); 5.70 (br. s,
13
2
H); 7.33 ± 7.52 (m, 4 H); 7.72 ± 7.86 (m, 4 H); 8.40 (s, 2 H); 10.71 (br. s, 1 H, exchanging with D O). C-NMR
(
1
62.5 MHz, (D
6
)DMSO, Tˆ 608): 13.6; 21.8; 25.5; 26.8; 30.6; 46.5; 87.1; 88.7; 94.6; 115.1; 123.3; 129.6; 130.7;
‡
‡
30.9; 132.2; 135.0; 136.2; 137.6; 140.3; 159.8; 174.9. HR-FAB-MS: 997.9641 (M , C38
H
39
I
3
N
3
OS
2
; calc.
9
97.9669).
Dodecyl-2-[(Z)-{3-[(Z)-(3-dodecyl-6-iodo-1,3-benzothiazol-2(3H)-ylidene)methyl]-2-[(3-iodophenyl)-
1
amino]-4-oxocyclobut-2-en-1-ylidene}methyl]-6-iodo-1,3-benzothiazol-3-ium Iodide (8c) ). Yield: 34%. Blue
crystals. M.p. 2408 (dec.). UV/VIS (CH
2
Cl
2
): 675 (5.28). IR (KBr): 3434w, 2921w, 2850w, 1635w, 1579w, 1533w,
1
1
442s, 1382w, 1347w, 1247m, 1174w, 1149w, 1064w, 981w, 761w, 518w. H-NMR (400 MHz, CDCl
3
): 0.86 (br. s,
6
H); 1.23 (br. s, 34 H); 1.53 (br. s, 4 H); 1.88 (br. s, 2 H); 3.80 (br. s, 2 H); 4.61 (br. s, 2 H); 5.03 (br. s, 1 H); 6.89
‡
‡
(
br. s, 1 H); 7.13 (t, J ˆ 7.7, 2 H); 7.53 ± 7.90 (m, 8 H). HR-FAB-MS: 1166.1571 (M , C50
H
63
I
3
N
3
OS
2
; calc.
1
166.1547).
2
-[(Z)-{3-[(Z)-(3-Hexyl-1,3-benzothiazol-2(3H)-ylidene)methyl]-4-oxo-2-(phenylamino)cyclobut-2-en-1-
1
ylidene}methyl]-3-hexyl-1,3-benzothiazol-3-ium Iodide (8d) ). Yield: 60%. Blue crystals. M.p. 260 ± 2628. UV/
VIS (CH
2
Cl
2
): 663 (5.29). IR (KBr): 3444w, 2927w, 2856w, 1634w, 1545w, 1495w, 1442s, 1355w, 1258s, 1175w,
1
1
1
1
156w, 1137w, 983w, 750w, 505w. H-NMR (300 MHz, (D
6 H); 3.82 (br. s, 2 H); 4.39 (br. s, 2 H); 5.00 (br. s, 1 H); 6,61 (br. s, 1 H); 7.36 ± 7.53 (m, 11 H); 7.70 (br. s, 2 H);
0.63 (br. s, 1 H, exchanging with D
6
)DMSO/CDCl
3
): 0.89 (br. s, 6 H); 1.31 ± 1.86 (m,
O). 1 C-NMR (75 MHz, (D
3
)DMSO/CDCl
): 13.5; 21.9; 25.7; 27.1; 30.8;
2
6
3

1
142
Helvetica Chimica Acta ± Vol. 88 (2005)
‡
4
C
6.7; 87.3; 112.1; 121.9; 124.0; 124.8; 126.2; 127.4; 128.6; 135.9; 140.1; 160.2; 175.0. HR-FAB-MS: 620.2774 (M ,
‡
38
H
42
N
3
OS
2
; calc. 620.2769).
2
-[(Z)-{3-[(Z)-(3-Hexyl-1,3-benzothiazol-2(3H)-ylidene)methyl]-2-[(3-iodophenyl)amino]-4-oxocyclo-
1
but-2-en-1-ylidene}methyl]-3-hexyl-1,3-benzothiazol-3-ium Iodide (8e) ). Yield: 65%. Blue crystals. M.p. 254 ±
2
1
568. UV/VIS (CH
176w, 1156m, 1138w, 986w, 747w, 506w. H-NMR (300 MHz, (D
2
Cl
2
): 660 (5.19). IR (KBr): 3435w, 2858w, 2921w, 1635w, 1581w, 1536w, 1444s, 1353w, 1254s,
1
6
)DMSO/CDCl
3
): 0.89 (t, J ˆ 6.7, 6 H); 1.29 (br.
s, 12 H); 1.67 (br. s, 4 H); 4.14 (br. s, 4 H); 5.05 (br. s, 1 H); 6.40 (br. s, 1 H); 7.26 ± 7.38 (m, 4 H); 7.48 ± 7.52 (m,
4
1
3
H); 7.67 ± 7.69 (m, 2 H); 7.84 (d, J ˆ 7.8, 2 H); 10.65 (br. s, 1 H, exchanging with D
2
O). C-NMR (75 MHz,
(
D
6
)DMSO/CDCl
3
): 13.6; 21.9; 25.7; 27.0; 30.7; 46.4; 87.1; 94.1; 112.7; 122.3; 122.7; 124.8; 127.4; 127.5; 130.3;
‡
‡
1
31.9; 134.5; 137.4; 140.1; 159.2; 175.0. HR-FAB-MS: 746.1737 (M , C38
H
41IN
3
OS
2
; calc. 746.1736).
1
Quantum Yields f of Singlet Oxygen ( O
2
) Generation. Experiments were performed at r.t. with a PTI-
1
PL2300 N
2
laser (337.1 nm, 0.60 ns pulses, 1.1 mJ/pulse) with air-equilibrated samples. The emission from O
2
was detected at 908 to the incident laser beam ± after passing through a 1270-nm interference filter (Corion) or a
combination of the interference filter with a long-wave pass filter (1200 nm; CVI Laser Corp.) to reduce
scattered radiation from the laser (shortly after the laser pulse) as much as possible ± by means of a 5-MHz Ge
photodiode (Judson J16-8SP-R05 M-HS). The signal was amplified with a pre-amplifier (Oriel 70732, 350 MHz;
and Thorn EMI Electron Tubes A1 and/or A2).
Decay curves were obtained with the help of an 8-bit AD-converter/recorder system (Fast TR50; 50 MHz),
and each curve was the average of at least 50 decays. Decays were fitted with a single-exponential function to
obtain the initial luminescence intensity at time t ˆ 0 (or the intensity at any other time t) after the laser pulse
(
Fig. 2) [15]. Phenazine in CH
2
Cl
2
(f ˆ 0.89) [15a], with an absorbance A of 0.30 [15c] at the wavelength of
excitation (337.1 nm), was used as a reference. Deoxygenated solns. were obtained by bubbling N
2
2
for ca.
0 min. Accurate calibration curves, with linear dependence of 1
O
2
emission intensity vs. laser energy were
obtained by placing neutral density filters in the excitation laser path length. Acomparison of the slopes for the
test samples and the reference yielded the quantum yield f in a straightforward manner (Fig. 2).
1
Fig. 2. Plots of initial singlet oxygen ( O
2
) luminescence intensity vs. laser dose for optically matched samples of
Cl at ambient temperature; excitation wavelength: 337.1 nm. The
compound 1a (~) and phenazine (
&
). In CH
2
2
quantum yield f was derived from the ratio of slopes (f ˆ 0.32 Æ 0.02 for 1a).
The third harmonic (355 nm, ca. 6-ns FWHM, 2 ± 15 mJ/pulse) of a Nd-YAG laser (Thomson-CSF Sagsa
1
2
2-10) could also be used as an excitation source. In this case, NaNO solns. were used to attenuate the excitation
energy and to ensure that the emission intensity was a linear function of the laser energy. The results obtained
independently with the two lasers were the same within experimental error, and most work was, thus, performed
with the N laser.
2

Helvetica Chimica Acta ± Vol. 88 (2005)
1143
There was no dye aggregation in CH
2
Cl
2
at A ˆ 0.30 ± 0.50 for all dyes under study, as carefully verified by
À6
À5
ground-state absorption studies at 10 ± 10 m. There was also no dependence on concentration of the quantum
1
1
yield of
2 2
O production in this range for both excitation wavelengths (337 and 355 nm; see above). The O
phosphorescence lifetime for air-equilibrated solns. of phenazine in CH
2
Cl
2
(at A337 ˆ 0.30) was 88 Æ 2 ms. For all
1
other compounds under study, a similar lifetime was observed for O
2
emission.
REFERENCES
[1] T. J. Dougherty, C. J. Gomer, B. W. Henderson, G. Jori, D. Kessel, M. Korbelik, J. Moan, Q. Peng, J. Natl.
Cancer Inst. 1998, 90, 889.
[
[
2] R. Bonnett, ꢀChemical Aspects of Photodynamic Therapyꢁ, Gordon & Breach, Amsterdam, 2000.
3] M. J. Niedre, A. J. Secord, M. S. Patterson, B. C. Wilson, Cancer Res. 2003, 63, 7986; Z. Diwu, J. W. Lown,
Pharmacol. Ther. 1994, 63, 1.
[
[
[
4] M. Wainwright, Chem. Soc. Rev. 1996, 25, 351; ꢀPhotodynamic Tumour Therapy ± 2nd and 3rd Generation
Photosensitizersꢁ, Ed. J. C. Moser, Hardwood Academic Publishers, Amsterdam, 1998.
5] G. Jori, J. Photochem. Photobiol.,B 1996, 36, 87; D. Wöhrle, A. Hirth, T. Bogdahn-Rai, G. Schnurpfeil, M.
Shopova, Russ. Chem. Bull. 1998, 47, 807.
6] a) P. F. Santos, L. V. Reis, P. Almeida, A. S. Oliveira, L. F. V. Ferreira, J. Photochem. Photobiol.,A 2003,
160, 159; b) P. F. Santos, L. V. Reis, P. Almeida, J. P. Serrano, A. S. Oliveira, L. F. V. Ferreira, J. Photochem.
Photobiol.,A 2004, 163, 267.
[
[
[
7] J. Fabian, H. Nakazumi, M. Matsuoka, Chem. Rev. 1992, 92, 1197; K.-Y. Law, Chem. Rev. 1993, 93, 449;
A. P. Piechowski, G. R. Bird, D. L. Morel, E. L. Stogryn, J. Phys. Chem. 1984, 88, 934; ꢀInfrared Absorbing
Dyesꢁ, Ed. M. Matsuoka, Plenum Press, New York, 1990.
8] a) D. Ramaiah, A. Joy, N. Chandrasekhar, N. V. Eldho, S. Das, M. V. George Photochem. Photobiol. 1997,
65, 783; b) D. Ramaiah, K. T. Arun, D. Suresh, E. Bernd, U.S. Patent 2002123532, 2002; c) A. R. Morgan,
M. E. Menes, R. Wang, WO Patent 44742, 2000.
9] a) L. V. Reis, J. P. C. Serrano, P. Almeida , P. F. Santos, Synlett 2002, 10, 1617; b) S. H. Kim, S. H. Hwang,
J. J. Kim, C. M. Yoon, S. R. Keum, Dyes Pigm. 1998, 37, 145; c) S. H. Kim, S. K. Han, Color. Technol. 2001,
117, 61.
[10] G. L. Indig, Recent Res. Dev. Pure Appl. Chem. 1999, 3, 9; A. R. Oseroff, in ꢀPhotodynamic Therapy ± Basic
Principles and Clinical Applicationsꢁ, Eds. B. W. Henderson, T. J. Dougherty, Marcel Dekker, New York,
1992, p. 79 ± 96.
[
[
[
[
[
11] I. Tabakovic, M. Trkovnik, M. Batusik, K. Tabakovic, Synthesis 1979, 590.
12] A. I. Kiprianov, I. K. Ushenko, E. D. Sych, J. Gen. Chem. U.S.S.R. 1945, 15, 200.
13] M. Q. Doja, J. C. Banerji, J. Indian Chem. Soc. 1958, 35, 779.
14] L. Tong, P. Bixian, Dyes Pigm. 1998, 39, 201.
15] a) J. C. Scaiano, R. W. Redmond, B. Mehta, J. T. Arnason, Photochem. Photobiol. 1990, 52, 655; b) R.
Schmidt, C. Tanielian, R. Dunsbach, C. Wolff, J. Photochem. Photobiol.,A 1994, 79, 11; c) M. Mir, L. M. G.
Jansen, F. Wilkinson, J. L. Bourdelande, J. Photochem. Photobiol.,A 1998, 113, 113.
Received November 2,2004