Phenothiazine electrophores immobilized on periodic mesoporous organosilicas by ion exchange

Article abstract of DOI:10.1039/c9nj04661e

Two different phenothiazines carrying quaternary ammonium groups in the side chain have been synthesized and fully characterized. These compounds were immobilized by ion exchange on the surface of a periodic mesoporous organosilica (PMO) and a neat silica SBA-15 material bearing sulphonate groups covalently bound to the pore surface. The resulting porous and electrochromophoric materials were studied by means of solid state CP-MAS NMR, IR, UV/Vis and fluorescence spectroscopy, nitrogen adsorption/desorption measurements and cyclic voltammetry. Independent of the nature of the support, the colour of these materials changes to pink by irradiation with light, indicating the formation of phenothiazine radical cations. These species, which turned out to be highly stable even in the presence of atmospheric oxygen, were characterized by EPR spectroscopy.

Full text of DOI:10.1039/c9nj04661e

NJC
View Article Online
View Journal | View Issue
PAPER
Phenothiazine electrophores immobilized on
periodic mesoporous organosilicas by ion
exchange†
Cite this: New J. Chem., 2019,
43, 16396
Bjorn Schafgen,a Hilla Khelwati,b Dominique F. Bechtel,a Annelies DeCuyper,a
¨
¨
¨
c
a
Axel Schussler,a Adam Neuba,
Antonio J. Pierik,
Stefan Ernst,‡a
Thomas J. J. Muller *b and Werner R. Thiel
*
a
¨
Two different phenothiazines carrying quaternary ammonium groups in the side chain have been
synthesized and fully characterized. These compounds were immobilized by ion exchange on the surface of
a periodic mesoporous organosilica (PMO) and a neat silica SBA-15 material bearing sulphonate groups
covalently bound to the pore surface. The resulting porous and electrochromophoric materials were studied
by means of solid state CP-MAS NMR, IR, UV/Vis and fluorescence spectroscopy, nitrogen adsorption/
desorption measurements and cyclic voltammetry. Independent of the nature of the support, the colour of
these materials changes to pink by irradiation with light, indicating the formation of phenothiazine radical
cations. These species, which turned out to be highly stable even in the presence of atmospheric oxygen,
were characterized by EPR spectroscopy.
Received 10th September 2019,
Accepted 17th September 2019
DOI: 10.1039/c9nj04661e
rsc.li/njc
and their oligomers particularly interesting as redox switchable
1. Introduction
7
8
¨
luminophores, dye sensitizers in Gratzel-type solar cells, and
Fully reversible oxidation or reduction is rather associated with as donors in photoinduced intramolecular electron transfer
inorganic than with organic compounds. However, there are systems.9
also some small, purely organic compounds showing this
Some years ago these features motivated us to covalently
behaviour. During the past two decades, the phenothiazine immobilize phenothiazines in the pores of periodically struc-
motif became increasingly dominating for applications in tured mesoporous silica such as MCM-41.10 Thereafter, a few
organic electronics. Phenothiazines undergo a reversible one- reports of others dealing with phenothiazines on silica surfaces
electron oxidation at low potentials.1 Fine-tuning of the redox appeared.11 We have particularly focused on materials with
potential is possible by choosing a suitable substitution covalently grafted phenothiazines since simply adsorbed systems
pattern. The oxidation process results in stable and intensely lack stability. Accordingly, different protocols for covalently grafting
coloured radical cations, which has widely been used for a phenothiazines on silica-based materials with high specific surface
quantitative determination of this class of compounds.2 Based areas were established. One critical point is the right choice of the
on their electronic properties, phenothiazines have gained linker between the phenothiazine core and the silica. Following
multiple applications, not only in organic electronics but also the carbamate approach, commercially available 3-(triethoxysilyl)-
in medicine,3 biochemistry,4 material sciences5 or polymer propylisocyanate was reacted with a phenothiazine functionalized
chemistry.6 In addition, the unusual combination of both alcohol. Subsequently, the resulting product was covalently
luminescence and redox activity has made phenothiazines immobilized in the pores of MCM-41, by applying post-synthetic
grafting. Oxidation with (NO)BF4 resulted in long-lived, red-coloured,
surface bound radical cations. However, according to this protocol,
a
¨
¨
Fachbereich Chemie, Technische Universitat Kaiserslautern, Erwin-Schrodinger-Str. 54,
only about 15 wt% of the organic material could be ligated to the
support. For higher loadings of phenothiazines inside the pores
with a more homogeneous distribution of the redox-active sites
along the pores, we had to switch to an in situ method. This
means, that triethoxysilyl-functionalized phenothiazine precursors
had to be applied in the reaction mixture for the formation of the
support. This was not feasible with the carbamate linker since it is
readily hydrolysed by the amine catalyst required for the support
D-67663 Kaiserslautern, Germany. E-mail: thiel@chemie.uni-kl.de
b
c
¨
Institut fur Organische Chemie und Makromolekulare Chemie,
¨
¨
¨
¨
Heinrich-Heine-Universitat Dusseldorf, Universitatsstraße 1, D-40225 Dusseldorf,
Germany. E-mail: ThomasJJ.Mueller@uni-duesseldorf.de
¨
Department Chemie, Universitat Paderborn, Warburger Straße 100,
D-33098 Paderborn, Germany
† Electronic supplementary information (ESI) available: NMR, EPR, IR, UV/Vis
and mass spectra, TGA and EDX data, SEM images. See DOI: 10.1039/c9nj04661e
‡ In memoriam to Prof. Dr Ing. Stefan Ernst who deceased on Jan. 28, 2019.
16396 | New J. Chem., 2019, 43, 16396--16410 This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
Paper
NJC
synthesis from tetraethoxysilane. To overcome this problem, urea compounds were recorded using a Shimadzu IRAffinity-1
based linkers were implemented. The corresponding triethoxysilyl- device. An ATR-FT-IR spectrometer PerkinElmer Spectrum 100
functionalized phenothiazine urea derivatives were therefore containing a diamond coated ZnSe-window was used for mea-
obtained by reacting amino-functionalized phenothiazines with suring the infrared spectra of the PMOs with a spectral range of
3-(triethoxysilyl)propylisocyanate. This strategy of in situ processing 4000–650 cmꢀ1 and a resolution of ꢁ4 cmꢀ1. Intensities of the
allowed increasing the loading to 30 wt% of organic material IR absorptions are indicated by w (weak), m (medium) and
inside the pores while maintaining an acceptable quality of pore- s (strong). The solid-state UV/Vis data were recorded on a
ordering. Again one-electron oxidation gave red-coloured radical PerkinElmer Lambda 18 UV/Vis spectrometer in a spectral
cations for monophenothiazines, while the oxidation of pheno- range from 900–200 nm and with a step size of 1 nm. TGA
thiazine dyads resulted in green diradical dications. Finally solid analyses were performed on a Setaram SETSYS 16/18 TGA/DTG
state cyclic voltammetry proved that the reversible electrochemical system. Elemental analyses were performed on a PerkinElmer
oxidation of the surface-bound phenothiazine is possible.
Series II Analyzer 2400 and an Elementar Vario Micro Cube at the
We herein present a further development of the urea Microanalytical Laboratories of the Institut fu¨r Pharamazeutische
¨
approach by using ionic end-groups that allow an electrostatic Chemie, Heinrich-Heine-Universitat, Du¨sseldorf, Germany and
immobilization of the phenothiazine motif. We see a clear with an Elementar Vario Micro Cube in the Analytical Labora-
advantage in this approach providing a moderate mobility of tories at the TU Kaiserslautern, Germany. For column chromato-
phenothiazines along the pores, while keeping binding to the graphy and TLC, analytic silica gel 60, mesh 70–230 or 60 F254
support. A phenylene-bridged periodic mesoporous organo silica gel plates were used. Electron capture ionization mass
silica of SBA 15 structure was used as the support here. Such spectra were measured with a triple quadrupole mass spectro-
materials possess more hydrophobic pore walls and a higher meter TSQ 7000 from Finnigan MAT. The measurements of the
mechanical and chemical stability compared to neat silica melting points were carried out with a B-540 Bu¨chi melting point
supports. These properties are advantageous for using the apparatus. EPR spectra were recorded with a Bruker Elexsys E580
materials in ion exchanging reactions.
X-band EPR spectrometer working in the perpendicular mode
at room temperature coupled to a dielectric resonator ER
4118X-5-MD (Bruker). The samples were measured in capillary
tubes with a microwave frequency of 9.72 ꢁ 0.01 GHz and a
microwave power of 0.2–20 mW. The standard modulation
amplitude for these radical compounds was 0.1–0.25 mT based
on the spectral resolution of the sample.
2. Experimental
2.1. General considerations
All reactions leading to low-molecular-weight compounds were
carried out under an inert atmosphere of nitrogen using septum
and cannula techniques in heated one- or multi-necked flasks or
Schlenk tubes. All reagents, catalysts and solvents were purchased
reagent grade and used without further purification. THF, diethyl
ether, dichloromethane, methanol and 1,4-dioxane were dried and
distilled according to standard procedures. The phenothiazine
derivatives 2, 3, 5, BTEB (11) (1,4-(bistriethoxysilyl)benzene)
and SBA-15 (18) were prepared according to procedures published
in the literature.12–14
2.3. Textural properties
X-Ray powder diffraction (PXRD) patterns of the silica samples
were recorded on a Siemens/Bruker AXS Type D 5005 instrument
using Cu-Ka radiation (l = 0.15405 nm) in a 2Y range of 0.51 o
2Y o 51 with a step range of 0.041. N2-adsorption–desorption
isotherms, pore size distributions as well as additional textural
properties of the materials were determined at ꢀ196 1C by a
Quantachrome Autosorb-1-Instrument sorption analyser. Before
analysis, the samples were degassed at 150 1C for at least 12 h
under vacuum and then the adsorption–desorption procedure
2.2. Spectroscopic characterization
The 1H, 13C and 135-DEPT NMR spectra were recorded on was conducted by passing nitrogen into the sample, which was
Bruker Avance III 300 and Avance III 600 instruments. The kept under liquid nitrogen. The average pore sizes of the samples
1
chemical H NMR shifts d are reported in ppm relative to the were estimated using the BJH approach based on the Kelvin
signals of the residual protons in the deuterated solvents equation while assuming a cylindrically shaped porous structure.
(MeOH-d4, acetone-d6, or DMSO-d6). In describing the multi- The specific surface areas were calculated by means of the
plicity of the individual signals, the following common abbrevia- Brunauer–Emmett–Teller (BET) equation and the pore size
tions were used: s (singlet), d (doublet), t (triplet), q (quartet), distribution curves were analysed with the adsorption branch by
m (multiplet). In the description of the high resolution 13C NMR the BJH method. The scanning electron microscope (SEM) images
spectra of the low-molecular-weight compounds primary carbon were taken on a Jeol JSM 6490 LA field emission device with
nuclei were designated with CH3, secondary with CH2, tertiary acceleration voltage of 15 kV. The silica particles were deposited
with CH, and quaternary with Cquat.. The assignments were made from suspension onto silicon wafers and thin Au films (approx.
by using DEPT spectra. Solid-state 13C CP-MAS NMR and 29Si CP- 10 nm) were deposited on the samples for imaging.
MAS NMR spectra were recorded on a 500 MHz Bruker Avance III
2.4. Electrochemistry
spectrometer with a spinning frequency of 11 000 Hz and
resonance frequencies of 125 MHz and 99 MHz for 13C or 29Si Cyclic voltammetry measurements of the low-molecular-weight
nuclei, respectively. IR spectra for the low-molecular-weight compounds were carried out in a small volume glass cell (4 mL)
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 16396--16410 | 16397

View Article Online
NJC
Paper
with a three electrode arrangement. The experiments were per- (37.4 g, 104 mmol) in acrylonitrile (214 mL, 3.23 mol) a solution of
formed under argon atmosphere in dry and degassed dichloro- Triton B (40 wt%, 3.83 mL, 22.0 mmol) was added dropwise under
methane at room temperature at scan rates of 100, 250, 500 and vigorous stirring. After complete addition, the solution was stirred
1000 mV sꢀ1. The electrolyte was Bu4NPF6 (0.025 M). The working at 0 1C for 5 min and a colour change from ochre to brown was
electrode was a 1 mm platinum disk, the counter electrode observed. TLC indicated the completion of the reaction. The
was a platinum wire and the reference electrode was an excessive acrylonitrile was removed under reduced pressure and
Ag/AgCl electrode. All potentials were corrected to the internal the crude product was adsorbed onto Celites and purified by
standard of decamethylferrocene/ferrocenium in dichloro- column chromatography on silica gel (hexane/acetone 8 : 2).
methane (E00/+1 = ꢀ0.095 V). The cyclic and square-wave voltam- The product was obtained as a colourless, fluffy powder (31.2 g,
metry measurements of the materials containing immobilized 75.9 mmol, 74%). Mp 146–148 1C. Rf (hexane/acetone, 8 : 2)
phenothiazines were performed with the Metrohm-Autolab 0.26. 1H NMR (600 MHz, DMSO) d 7.42 (d, 4J = 2.3 Hz, 2 H), 7.39
potentiostat PGSTAT 101 in a CH2Cl2 or MeCN/0.1 M tBu4NPF6 (dd, 3J = 8.7 Hz, 4J = 2.3 Hz, 2 H), 7.04 (d, 3J = 8.7 Hz, 2 H), 4.18
solution with the following three electrode arrangement: GC (t, 3J = 6.6 Hz, 2 H), 2.90 (t, 3J = 6.6 Hz, 2 H). 13C NMR (125 MHz,
working electrode (d = 2 mm), Ag/0.01 M AgNO3/MeCN-reference DMSO) d 142.9 (Cquat.), 130.3 (CH), 129.2 (CH), 126.1 (Cquat),
electrode and platinum wire counter electrode. For these 118.4 (Cquat.), 117.7 (CH), 114.6 (Cquat.), 42.5 (CH2), 15.7 (CH2).
measurements, the materials were deposited on the GC work- IR n~ [cmꢀ1] 3086 (vw), 2990 (vw), 2961 (vw), 2905 (vw), 2648 (vw),
ing electrode as follows: 2–4 mg of the material were suspended 2249 (vw), 1867 (vw), 1846 (vw), 1585 (vw), 1477 (w), 1452 (s),
in water or ethanol and sonicated for several minutes. One 1408 (w), 1391 (w), 1335 (m), 1317 (m), 1288 (w), 1248 (m), 1227
small drop of the fine dispersion was placed on the GC (w), 1194 (m), 1165 (w), 1155 (w), 1113 (w), 1096 (w), 1080 (w), 1070
electrode tip (d = 2 mm) and heated for 30 min at 80 1C to (w), 1045 (w), 1028 (vw), 989 (vw), 901 (vw), 864 (m), 797 (s), 789
evaporate the solvent. All measurements were carried out under (m), 752 (m), 741 (w), 700 (vw), 650 (w), 606 (w). MS (EI) m/z (%)
argon atmosphere, with absolute and vented solvents.
412 (37, [C15H1081Br2N2S]+), 410 (81, [C15H1079Br81BrN2S]+), 408
(37, [C15H10Br279N2S]+), 372 (52, [C13H881Br2NS]+), 370 (100,
[C13H879Br81BrNS]+), 368 (48, [C13H879Br2NS]+), 358 (44, [C12H6
-
2.5. Synthetic procedures
81
N,N,N-Trimethyl-3-(10H-phenothiazin-10-yl)propan-1-ammonium Br2NS]+), 356 (89, [C12H679Br81BrNS]+), 354 (42, [C12H679Br2NS]+),
trifluoromethanesulphonate (4). 3-(10H-Phenothiazin-10-yl)propan- 291 (59, [C13H881BrNS]+), 289 (54, [C13H879BrNS]+), 196 (65,
1-amine (3) (317 mg, 1.23 mmol) and potassium carbonate [C12H6NS]+).
(1.71 g, 12.3 mmol) were suspended in methanol (12 mL) and
3-(3,7-Dibromo-10H-phenothiazin-10-yl)propan-1-amine (7).
methyl trifluoromethylsulphonate (1.35 mL, 12.3 mmol) was Anhydrous indium trichloride (3.31 g, 15.0 mmol) and sodium
added dropwise to the solution while a colour change from borohydride (1.70 g, 45.0 mmol) were stirred in THF (50 mL) at
orange to yellow could be observed. The reaction was stirred room temp for 1 h, whereas a colour change from greenish to
at room temp for 30 min and afterwards at 60 1C for 16 h. greyish could be observed. Afterwards, compound 6 (6.15 g,
All volatile components were removed under reduced pressure, 15.0 mmol) was added to the reaction mixture which was
while the residue was dissolved in dichloromethane and stirred at room temp for another 4 h. The solution was
filtrated from the residual solid. The solvent was removed quenched by the dropwise addition of 3 M hydrochloric acid
under reduced pressure and the product was suspended in (50 mL) and stirred under reflux for 1 h. Afterwards methanol
diethyl ether (2 ꢂ 12 mL). The product was dried under high (25 mL) was added and the solution was stirred again under
vacuum and was obtained as a colourless powder (332 mg, reflux for 1 h. The organic solvents were removed under reduced
1.11 mmol, 90%). Mp 168–170 1C. Rf (methanol) 0.31. 1H NMR pressure and the remaining aqueous solution was extracted with
3
4
(300 MHz, MeOH) d 7.25 (mc, 2 H), 7.19 (dd, J = 7.6 Hz, J = diethyl ether (50 mL). The acidic, aqueous solution was then
3
4
1.6 Hz, 2 H), 7.06 (dd, J = 8.1 Hz, J = 0.9 Hz, 2 H), 6.99 (mc, adjusted with sodium hydroxide to pH 10 and the product was
2 H), 4.11 (t, 3J = 6.3 Hz, 2 H), 3.45 (mc, 2 H), 3.01 (s, 9 H), 2.25 extracted with diethyl ether (3 ꢂ 50 mL). The combined organic
(mc, 2 H). 13C NMR (125 MHz, MeOH) d 146.2 (Cquat.), 128.8 phases were dried with anhydrous magnesium sulphate and the
(CH), 128.6 (CH), 127.4 (Cquat.), 124.3 (CH), 117.3 (CH), 65.7 solvents were removed under reduced pressure while the crude
(CH2), 53.5 (CH3), 44.7 (CH2), 21.8 (CH2). IR n~ [cmꢀ1] 3065 (vw), product was adsorbed onto Celites. After purification by column
3042 (vw), 2965 (vw), 2868 (vw), 1593 (w), 1570 (w), 1485 (m), chromatography on silica gel (gradient of hexane/acetone 9 : 1
1479 (m), 1460 (s), 1445 (m), 1422 (w), 1400 (w), 1339 (w), 1254 with 1% triethylamine to 7 : 3 with 1% triethylamine), the product
(vs), 1221 (s), 1200 (w), 1157 (s), 1140 (s), 1128 (m), 1109 (w), was obtained as a greenish powder (6.09 g, 13.5 mmol, 91%). Mp
1028 (s), 961 (w), 939 (w), 918 (w), 908 (w), 764 (s), 754 (w), 729 141 1C. Rf (hexane/acetone 1 : 1 with 1% triethylamine) 0.42.
(m), 637 (s). MS (EI) m/z (%) 448 (1, [C19H23F3O3N2S2]+), 299 (10, 1H NMR (300 MHz, DMSO-d6) d 7.93 (s, 3 H), 7.47–7.35
3
[C18H23N2S]+) 284 (41, [C17H20N2S]+), 239 (100, [C15H14NS]+), (m, 4 H), 7.02 (d, J = 8.9 Hz, 2 H), 3.95 (t, 3J = 6.9 Hz, 2 H),
3
212 (18, [C13H10NS]+), 199 (50, [C12H8NS]+). Anal. calcd for 2.92–2.78 (m, 2 H), 1.94 (tt, J = 7.2 Hz, 6.9 Hz, 2 H). 13C NMR
C18H23N2S [299.5] C 50.88, H 5.17, N 6.25, S 14.30; found C 50.83, (150 MHz, DMSO-d6) d 143.6 (Cquat.), 130.4 (CH), 129.2 (CH), 125.9
H 5.36, N 6.09, S 14.05.
(Cquat.), 117.9 (CH), 114.4 (Cquat.), 43.9 (CH2), 36.5 (CH2), 24.3
3-(3,7-Dibromo-10H-phenothiazin-10-yl)propanenitrile (6). To (CH2). IR n~ [cmꢀ1] 3092 (vw), 3057 (vw), 2934 (vw), 2872 (vw), 1585
an ice cooled suspension of 3,7-dibromo-10H-phenothiazine (5) (w), 1479 (m), 1456 (s), 1387 (w), 1327 (w), 1294 (w), 1252 (m), 1229
16398 | New J. Chem., 2019, 43, 16396--16410 This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
Paper
NJC
(m), 1150 (w), 1109 (w), 1082 (w), 1057 (w), 868 (w), 806 (m), 750 (Cquat.), 44.5 (CH2), 37.7 (CH2), 28.3 (CH3), 26.6 (CH2). IR n~
(m), 650 (w). MS (EI) m/z (%) 416 (36, [C15H1481Br2N2S]+), [cmꢀ1] 2974 (w), 2926 (w), 2864 (w), 1701 (m), 1680 (m), 1472
414 (70, [C15H1479Br81BrN2S]+), 412 (34, [C15H1479Br2N2S]+), (vs), 1429 (m), 1400 (m), 1391 (m), 1364 (m), 1348 (w), 1333 (w),
398 (20, [C15H1281Br2NS]+), 396 (39, [C15H1279Br81BrNS]+), 394 1292 (w), 1261 (s), 1238 (s), 1207 (m), 1163 (s), 1109 (m), 1080 (w),
(19, [C15H1279Br2NS]+), 372 (12, [C13H881Br2NS]+), 370 (24, [C13H8- 1042 (w), 1020 (w), 1007 (vw), 988 (w), 951 (vw), 951 (vw), 874 (m),
79Br81BrNS]+), 368 (12, [C13H879Br2NS]+), 358 (48, [C12H681Br2NS]+), 851 (m), 810 (s), 793 (m), 783 (w), 750 (m), 691 (vs), 650 (vw). MS
356 (85, [C12H679Br81BrNS]+), 354 (42, [C12H679Br2NS]+), 291 (EI) m/z (%) 520 (32, [C28H28N2O2S3]+), 464 (24, [C24H19N2O2S3]+), 446
(21, [C13H881BrNS]+), 289 (20, [C13H879BrNS]+), 277 (29, [C12H6- (6, [C24H19N2OS3]+), 420 (13, [C23H19N2S3]+), 376 (8, [C21H14NS3]+),
81BrNS]+), 196 (73, [C12H6NS]+).
tert-Butyl-(3-(3,7-dibromo-10H-phenothiazin-10-yl)propyl)- C 64.58, H 5.42, N 5.38, S 18.47; found C 64.83, H 5.69, N 5.12, S 18.17.
carbamate (8). Compound 7 (4.28 g, 9.60 mmol), triethylamine 3-(3,7-Di(thiophen-2-yl)-10H-phenothiazin-10-yl)-N,N,N-tri-
362 (100, [C20H12NS3]+). Anal. calcd for [C28H28N2O2S3] [519.7]
(2.63 mL, 19.2 mmol) and di-tert-butyl dicarbonate (2.51 g, methylpropan-1-ammonium trifluoromethanesulphonate (10).
11.5 mmol) were dissolved in dichloromethane (9.6 mL) and Compound 9 (200 mg, 385 mmol) was dissolved in dichloro-
4-dimethylaminopyridine (116 mg, 0.96 mmol, 10 mol%) was methane (3.8 mL) and trifluoroacetic acid (237 mL, 3.08 mmol)
added in small portions to the reaction mixture. After a reaction was added to the solution. The reaction mixture was stirred at
time of 1 h at 40 1C the solvent was evaporated, the crude 40 1C for 16 h. Volatile components were removed under
product was adsorbed onto Celites and purified by column reduced pressure and the residue was dissolved in methanol
chromatography on silica gel (hexane/acetone 8 : 2). The product was (3.8 mL). Potassium carbonate (530 mg, 3.85 mmol) and methyl
isolated as yellow crystals (4.36 g, 8.45 mmol, 88%). Mp 70–72 1C. trifluoromethylsulphonate (630 mg, 3.85 mmol) were added
Rf (hexane/acetone 7 : 3 with 1% triethylamine) 0.39. 1H NMR and the reaction mixture was stirred at room temp for 30 min
(300 MHz, DMSO-d6) d 7.38–7.33 (m, 4 H), 6.96 (d, 3J = and afterwards at 60 1C for 16 h. The evaporable components
7.7 Hz, 2 H), 6.86 (t, 3J = 4.9 Hz, 1 H), 3.83 (t, 3J = 6.5 Hz, were removed under reduced pressure and the crude product
2 H), 3.01 (m, 2 H), 1.76 (m, 2 H), 1.35 (s, 9 H). 13C NMR was adsorbed onto aluminium oxide and purified by column
(75 MHz, DMSO-d6) d 155.6 (Cquat.), 143.8 (Cquat.), 130.3 (CH), chromatography on aluminium oxide (gradient of dichloro-
129.1 (CH), 125.6 (Cquat.), 117.6 (CH), 114.1 (Cquart.), 77.5 (Cquat.), methane to dichloromethane/methanol 3%). The product
44.5 (CH2), 37.6 (CH2), 28.2 (CH3), 26.4 (CH2). IR n~ [cmꢀ1] 3395 could be obtained as yellow powder (181 mg, 296 mmol,
1
(vw), 3379 (vw), 2976 (w), 2926 (w), 2868 (vw), 2851 (vw), 1694 (m), 77%). Mp 128–130 1C. Rf (methanol) 0.10. H NMR (600 MHz,
3
3
1686 (m), 1516 (w), 1506 (m), 1481 (w), 1452 (vs), 1389 (m), DMSO) d 7.52–7.49 (m, 4 H), 7.43 (dd, J = 5.1 Hz, J = 0.9 Hz,
3
1364 (m), 1327 (w), 1267 (m), 1248 (s), 1202 (w), 1163 (s), 1109 2 H), 7.16 (d, J = 8.3 Hz, 2 H), 7.11 (mc, 2 H), 7.04 (dd, 3J =
(w), 1082 (w), 1040 (vw), 1003 (w), 959 (vw), 932 (vw), 868 (m), 3.6 Hz, 3J = 1.1 Hz, 2 H), 4.21 (t, 3J = 6.9 Hz, 2 H), 3.80 (mc, 2 H),
802 (s), 779 (w), 750 (m), 652 (w). MS (EI) m/z (%) 516 3.05 (s, 9 H), 2.50 (mc, 2 H). 13C NMR (125 MHz, MeOH-d4)
(14, [C20H2281Br2N2O2S]+), 514 (32, [C20H2279Br81BrN2O2S]+), d 145.1 (Cquat.), 144.1 (Cquat.), 131.5 (Cquat.), 129.2 (CH), 127.5
512 (14, [C20H2279Br2N2O2S]+), 460 (13, [C16H1481Br2N2O2S]+), (Cquat.), 126.4 (CH), 125.6 (CH), 125.6 (CH), 123.9 (CH), 117.6
458 (29, [C16H1479Br81BrN2O2S]+), 456 (13, [C16H1479Br2N2O2S]+), (CH), 65.8 (CH2), 53.7 (CH3), 45.0 (CH2), 22.0 (CH2). IR: n~ [cmꢀ1
]
357 (62, [C12H779Br81BrNS]+), 277 (30, [C12H779BrNS]+), 196 (39, 3034 (vw), 2963 (w), 1603 (vw), 1472 (s), 1427 (m), 1402 (m), 1346
[C12H6NS]+), 102 (100, [C5H9O2]+). Anal. calcd for C20H22Br2- (w), 1339 (w), 1258 (vs), 1223 (m), 1200 (w), 1153 (m), 1084 (s),
N2O2S [514.3] C 46.71, H 4.31, N 5.45, S 6.23; found C 46.98, 1045 (s), 1028 (vs), 1015 (s), 966 (w), 943 (vw), 910 (w), 872 (m),
H 4.17, N 5.32, S 6.23.
tert-Butyl-(3-(3,7-di(thiophen-2-yl)-10H-phenothiazin-10-yl)- ESI-HRMS calcd for [C26H27N2S3 ] 463.1336; found 463.1336.
propyl)carbamate (9). Compound 8 (514 mg, 1.00 mmol), BTEB-PMO (12). P123 (39.4 g, 6.79 mmol) was dissolved in a
851 (m), 799 (vs), 752 (m), 737 (w), 692 (s), 662 (vw), 637 (s), 610 (w).
+
4,4,5,5-tetramethyl-2-(thiophen-2-yl)-1,3,2-dioxaborolane (462 mg, mixture of water (1.40 L) and concentrated hydrochloric acid
2.20 mmol), caesium fluoride (973 mg, 6.40 mmol) and tetrakis- (7.92 mL). The solution was stirred at room temp for 2 h. After
(triphenylphosphane)palladium(0) (69 mg, 0.06 mmol, 6 mol%) this time it was cooled to 5 1C and BTEB (11) (40.0 g, 99.3 mmol)
were suspended in 1,4-dioxane (10 mL) and the reaction was added and the temperature of the reaction mixture was
mixture was stirred under reflux for 14 h. After cooling to room kept under 5 1C for 1 h. Then the temperature was raised to
temp the solvent was removed under reduced pressure, the 40 1C and the stirring was continued for another 4 h whereby
residue was adsorbed onto Celites and was purified by column a colourless solid precipitated. After keeping the colourless
chromatography on silica gel (hexane/acetone 8 : 2). The product suspension at 100 1C for additional 24 h without stirring,
could be isolated as yellow crystals (401 mg, 770 mmol, 77%). Mp the white solid was filtered off and washed to neutral pH
1
138–141 1C. Rf (hexane/acetone, 7 : 3) 0.34. H NMR (600 MHz, with water. To improve the pore structure the material was
DMSO) d 7.48 (dd, 3J = 5.1 Hz, 4J = 1.1 Hz, 2 H), 7.46–7.42 (m, 6 H), suspended in a closed vessel in water (800 mL) and kept at
3
3
3
7.04 (dd, J = 9.1 Hz, 2 H), 7.04 (d, J = 9.1 Hz, 2 H), 6.92 (t, J = 100 1C without stirring for 24 h. To remove the template, the solid
5.5 Hz, 1 H), 3.91 (t, 3J = 7.0 Hz, 2 H), 3.07 (mc, 2 H), 1.84 (mc, 2 H), was first filtered off and then suspended in a mixture of ethanol
1.36 (s, 9 H). 13C NMR (75 MHz, DMSO-d6) d 155.6 (Cquat.), 143.4 (1.2 L) and concentrated hydrochloric acid (12 mL) and stirred
(Cquat.), 142.2 (Cquat.), 128.5 (Cquat.), 128.5 (CH), 125.0 (CH), 124.9 under reflux for 18 h. The colourless solid was filtered off, washed
(CH), 123.7 (CH), 123.6 (Cquat.), 123.1 (CH), 116.1 (CH), 77.5 with ethanol (2 ꢂ 400 mL) and diethyl ether (2 ꢂ 400 mL) and
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 16396--16410 | 16399

View Article Online
NJC
Paper
then dried under vacuum. The latter procedure of template (99 MHz) d ꢀ62.7, ꢀ72.4, ꢀ80.6, ꢀ92.5, ꢀ101.3. PXRD 0.73 (s),
elimination was repeated two more times. The colourless solid 1.36 (w), 1.59 (w). BET Aspec. (m2 ꢀ1) 742, total pore volume
g
was finally dried to constant weight yielding 21.9 g. 13C CP-MAS (BJH, cm3 gꢀ1) 0.97, pore diameter (BJH, Å) 52.2. IR (ATR) n~ (cmꢀ1
)
NMR (125 MHz) d 131.9, 67.8, 55.8, 14.0. 29Si CP-MAS NMR 3346 (w), 3104 (w), 1632 (w), 1207 (w), 1152 (m), 1041 (s), 1022 (s),
(99 MHz) d ꢀ65.9, ꢀ72.8, ꢀ80.7. PXRD 0.66 (s), 1.29 (w), 1.52 (w). 920 (m), 808 (w), 767 (w). UV/Vis (nm) 227, 263, 270, 276. Anal.
BET Aspec. (m2 gꢀ1) 1066; total pore volume (BJH, cm3 gꢀ1) 1.37; found C 33.65, H 3.81, N 0.94, S 1.89.
pore diameter (BJH, Å) 56.3. IR (ATR) n~ [cmꢀ1] 3377 (w),
PT1/PT2-BTEB-PMO (16 and 17). In a typical exchange reac-
3059 (w), 2978 (w), 2893 (w), 1387 (w), 1151 (s), 1050 (s), tion PMO 15 (0.40 g, 211 mmol according to N content) was
1021 (s), 926 (m), 780 (m). UV/Vis (nm) 230, 270, 276. Anal. mixed with phenothiazines 4 or 10 (113 mg, 274 mmol of 4,
found C 41.36, H 4.58.
168 mg, 274 mmol of 10) and suspended in acetonitrile (5 mL).
SH-BTEB-PMO (13). In a typical grafting reaction 12 (5.00 g) The yellow suspension was stirred at 40 1C for 48 h. The solid
and 3-MPTS (25.0 g, 121 mmol) were suspended in toluene was filtered off and washed with acetonitrile (3 ꢂ 20 mL) and
(33 mL) and stirred under heating to reflux for 72 h. After water (2 ꢂ 20 mL). In the last step, the obtained product was
cooling to room temp the product was filtered off, washed with dried under vacuum to give 0.37 g respectively 0.39 g of a
toluene (3 ꢂ 50 mL) and dried at 70 1C for 18 h to give 6.09 g of yellowish solid. 16: 13C CP-MAS NMR (125 MHz) d 133.0, 126.6,
a colourless solid. 13C CP-MAS NMR (125 MHz) d 132.5, 68.0, 115.8, 57.6, 52.1, 22.0, 17.9, 11.8. 29Si CP-MAS NMR (99 MHz)
56.5, 46.0, 25.4, 14.3, 9.8, 7.2. 29Si CP-MAS NMR (99 MHz) d ꢀ61.5, ꢀ71.2, ꢀ80.4, ꢀ102.4. PXRD 0.64 (s), 1.30 (w), 1.54 (w).
d ꢀ50.4, ꢀ60.5, ꢀ72.7, ꢀ81.0. PXRD 0.68 (s), 1.29 (w), 1.51 (w). BET Aspec. (m2 gꢀ1) 688, total pore volume (BJH, cm3 gꢀ1) 0.96,
BET Aspez. (m2 gꢀ1) 827; total pore volume (BJH, cm3 gꢀ1) 1.10; pore diameter (BJH, Å) 53.5. IR (ATR) n~ [cmꢀ1] 3374 (m), 3056
pore diameter (BJH, Å) 56.1. IR (ATR) n~ (cmꢀ1) 3375 (w), (w), 2967 (w), 1635 (w), 1462 (w), 1382 (w), 1151 (m), 1039 (s),
2976 (w), 1384 (w), 1148 (m), 1034 (s), 1019 (s), 915 (m), 807 1020 (s), 912 (m). UV/Vis (nm) 228, 256, 270, 276, 299, 520. Anal.
(w), 780 (m). UV/Vis (nm) 229, 271, 277. Anal. found C 38.89, found C 36.38, H 3.97, N 1.20, S 2.67. 17 : 13C CP-MAS NMR
H 4.86, S 4.09.
(125 MHz) d 140.4, 131.5, 125.6, 114.2, 56.1, 51.0, 16.9, 10.0. 29Si
SH-BTEB-PMO (13). In a typical grafting reaction 12 (5.00 g) CP-MAS NMR (99 MHz) d ꢀ62.4, ꢀ72.0, ꢀ80.7, ꢀ92.6, ꢀ101.4.
and 3-MPTS (25.0 g), 121 mmol, (3-mercaptopropyl)trimethoxysilane PXRD 0.69 (s), 1.37 (w), 1.59 (w). BET Aspec. (m2 gꢀ1) 671, total
were suspended in toluene (33 mL) and stirred under heating to pore volume (BJH, cm3 gꢀ1) 0.91, pore diameter (BJH, Å) 52.1.
reflux for 72 h. After cooling to room temp the product was filtered IR (ATR) n~ (mꢀ1) 3340 (m), 3059 (w), 1632 (w), 1474 (w), 1382 (w),
off, washed with toluene (3 ꢂ 50 mL) and dried at 70 1C for 18 h to 1152 (m), 1038 (s), 1022 (s), 918 (m), 809 (w). UV/Vis (nm) 229, 279,
give 6.09 g of a colourless solid. 13C CP-MAS NMR (125 MHz) 285, 342, 548. Anal. found C 35.74, H 4.03, N 0.82, S 3.50.
d 132.5, 68.0, 56.5, 45.75, 25.4, 13.9, 9.8. 29Si CP-MAS NMR
(99 MHz) d ꢀ50.1, ꢀ60.4, ꢀ72.7, ꢀ81.0. PXRD 0.68 (s), 1.29 (w), 84.6 mmol according to the S content) was degassed at 50 1C
1.51 (w). BET Aspec. (m2 gꢀ1) 827; total pore volume (BJH, cm3 gꢀ1
for 3 h in vacuum and, after cooling to room temp, suspended
Chemical oxidation of PT1-BTEB-PMO (16). 16 (348 mg,
)
1.10; pore diameter (BJH, Å) 56.1. IR (ATR) n~ (cmꢀ1) 3375 (w), 2976 (w), in dry dichloromethane (5 mL). After addition of (NO)BF4
1384 (w), 1148 (m), 1034 (s), 1019 (s), 915 (m), 807 (w), 780 (m). (19.7 mg, 169 mmol) the colour changed from beige to raspberry.
UV/Vis (nm) 229, 271, 277. Anal. found C 38.89, H 4.86, S 4.09. The solid was filtered off, washed with dichloromethane (3 ꢂ 25 mL)
SO3H-BTEB-PMO (14). The thiopropyl functionalized PMO and then dried in vacuum. 299 mg of a pink solid was obtained.
13 (1.00 g) was suspended in H2O2 (35%, 100 mL) and the 13C CP-MAS NMR (125 MHz) d 132.3, 126.6, 117.9, 68.8, 51.8, 20.7,
mixture was stirred at 60 1C for 24 h. The product was filtered 17.3, 13.8, 8.8. 29Si CP-MAS NMR (99 MHz) d ꢀ71.6, ꢀ80.5,
off and washed with water (100 mL) and ethanol (100 mL). The ꢀ103.3. PXRD 0.79 (s), 1.74 (w). BET Aspec. (m2 gꢀ1) 467, total
resulting colourless solid was dried at 60 1C for 18 h to give pore volume (BJH, cm3 gꢀ1) 0.60, pore diameter (BJH, Å) 38.3.
0.86 g of a colourless solid. 13C CP-MAS NMR (125 MHz) IR (ATR) n~ (cmꢀ1) 3287 (w), 1381 (w), 1152 (m), 1040 (s), 917 (m),
d 131.5, 125.8, 68.0, 55.8, 51.8, 16.4, 14.0, 9.9. 29Si CP-MAS 811 (w), 760 (w). UV/Vis (nm) 234, 272, 276, 298, 341. Anal.
NMR (99 MHz) d ꢀ62.7, ꢀ72.5, ꢀ81.7, ꢀ95.1, ꢀ101.9. PXRD found C 34.97, H 3.88, N 0.98, S 2.13.
0.72 (s), 1.36 (w), 1.61 (w). BET Aspec. (m2
g
ꢀ1) 687, total pore
SH-SBA-15 (19). SBA-15 (18) (2.00 g) and 3-MPTS (10.0 g,
volume (BJH, cm3 gꢀ1) 0.95, pore diameter (BJH, Å) 49.5. IR (ATR) 48.4 mmol) were suspended in toluene (20 mL) and stirred for
n~ (cmꢀ1) 3364 (w), 1151 (m), 1035 (s), 1022 (s), 912 (m), 813 (w). 72 h under reflux. After cooling to room temp the product was
UV/Vis (nm) 230, 271, 277. Anal. found C 32.15, H 4.93, S 1.61. filtered off, washed with toluene (3 ꢂ 50 mL) and dried at 70 1C
PySO3-BTEB-PMO (15). The functionalized PMO 14 (0.85 g) for 18 h to give 2.31 g of a colourless solid. 13C CP-MAS NMR
was suspended in pyridine (7 mL) and stirred at room temp for (125 MHz) d 45.9, 23.5, 7.4. 29Si CP-MAS NMR (99 MHz) d ꢀ53.9,
16 h. The light beige solid was filtered off and washed with ꢀ63.4, ꢀ72.8, ꢀ107.9, ꢀ116.9. PXRD 0.71 (s) 1.65 (w). BET Aspec.
toluene (3 ꢂ 50 mL). To remove some adsorbed residues of (m2
g g
ꢀ1) 289; total pore volume (BJH, cm3 ꢀ1) 0.45; pore
pyridine, the solid was stirred for additional 3 h in toluene diameter (BJH, Å) 38.3. IR (ATR) n~ (cmꢀ1) 3357 (w), 2929 (w),
(18 mL). After the product was filtered off and washed again 1051 (w), 932 (m), 800 (m), 696 (w). Anal. found C 9.23, H 2.59,
with toluene (3 ꢂ 50 mL), it was dried for 3 h under vacuum to S 7.87.
yield 0.75 g of a greyish solid. 13C CP-MAS NMR (125 MHz)
SO3H-SBA-15 (20). 19 (1.00 g) was suspended in H2O2 (35%,
d 131.5, 126.8, 56.4, 52.4, 16.3, 13.7, 10.2. 29Si CP-MAS NMR 100 mL) and stirred at 60 1C for 24 h. After cooling to room
16400 | New J. Chem., 2019, 43, 16396--16410 This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
Paper
NJC
temp the product was filtered off, washed with water (100 mL)
and ethanol (100 mL) and dried at 60 1C for 18 h to give 0.87 g
of a colourless solid. 13C CP-MAS NMR (125 MHz) d 52.8,
17.1, 10.4. 29Si CP-MAS NMR (99 MHz) d ꢀ58.2, ꢀ68.1, ꢀ102.4,
ꢀ110.4. PXRD 0.68 (s), 1.61 (w). BET Aspec. (m2 gꢀ1) 581; total pore
volume (BJH, cm3 gꢀ1) 0.75; pore diameter (BJH, Å) 50.5. IR (ATR)
n~ (cmꢀ1) 3350 (w), 1637 (w), 1046 (s), 955 (m), 799 (m). Anal. found
C 3.26, H 2.32, S 1.97.
PySO3-SBA-15 (21). 20 (0.50 g) was suspended in pyridine
(4 mL) and stirred at room temp for 16 h. The light beige
product was filtered off and washed with toluene (3 ꢂ 50 mL).
To remove some adsorbed residues of pyridine the solid was
stirred for additional 3 h in toluene (10 mL). After the solid was
filtered off and was washed again with toluene (3 ꢂ 50 mL), the
resulting material was dried for 3 h under vacuum to yield
0.43 g of a greyish solid. 13C CP-MAS NMR (125 MHz) d 140.9,
126.8, 52.8, 17.1, 10.3. 29Si CP-MAS NMR (99 MHz) d ꢀ58.1,
ꢀ67.9, ꢀ92.5, ꢀ101.6, ꢀ111.1. PXRD 0.69 (s), 1.41 (w), 1.64 (w).
BET Aspec. (m2 gꢀ1) 557; total pore volume (BJH, cm3 gꢀ1) 0.74;
pore diameter (BJH, Å) 50.5. IR (ATR) n~ (cmꢀ1) 3351 (w), 1636 (w),
1041 (s), 957 (m), 797 (m). UV/Vis (nm): 255. Anal. found C 5.58,
H 2.08, N 0.76, S 1.96.
PhenoSO3-SBA-15 (22). 21 (0.25 g, 136 mmol according to N
content) was mixed with 4 (79.4 mg, 177 mmol) and suspended
in acetonitrile (2.4 mL). The yellow suspension was stirred at
40 1C for 48 h. The product was filtered off and washed with
acetonitrile (3 ꢂ 20 mL) and water (2 ꢂ 20 mL). In the last
step the resulting material was dried under vacuum to give
0.18 g of a yellow-greyish solid. 13C CP-MAS NMR (125 MHz)
d 143.8, 139.5, 126.5, 115.2, 63.2, 51.6, 42.0, 16.1, 8.7. 29Si
CP-MAS NMR (99 MHz) d ꢀ60.2, ꢀ66.2, ꢀ91.8, ꢀ102.9, ꢀ111.8.
PXRD 0.70 (s), 1.40 (w), 1.64 (w). BET Aspec. (m2 gꢀ1) 448; total
pore volume (BJH, cm3 gꢀ1) 0.65; pore diameter (BJH, Å) 49.1.
IR (ATR) n~ (cmꢀ1) 3354 (w), 1049 (s), 960 (m), 801 (m). UV/Vis
(nm): 207, 253, 306, 514. Anal. found C 6.96, H 2.01, N 0.73,
S 2.52.
Scheme 1 Retrosynthetic analysis of substituted phenothiazine salts.
centres will be more homogeneous than by applying a classical
post-synthetic grafting procedure, where the pores get blocked
by the covalent binding of large dye molecules preferentially at
the pore entrances. The covalent surface-bound anionic groups
required for an electrostatic immobilization of the cationic dye
molecules are introduced by small precursors that can diffuse
easily into the pores.
To realize this strategy, N-alkylated phenothiazines were
synthesized bearing quaternary ammonium groups as a counter-
part to the anionic species on the scaffold. A retrosynthetic
analysis of trimethylammonium substituted phenothiazines
suggests the formation of the ammonium salt by the reaction
of a primary amine with an alkylating agent (Scheme 1). In case
of p-expanded phenothiazines we found that a Boc-protected
amine serves as the optimal precursor for the quaternation
reaction. For both target compounds, the amine functionality
can be introduced by reduction of the corresponding nitrile.
Suzuki-cross-coupling reaction seemed to be most suitable
for the introduction of (hetero)aryl substituents in the
3,7-positions and the use of a Boc-protected amine as the
(hetero)arylhalide source is rational.
A literature known access to cyano substituted phenothia-
zines is provided by the Michael addition of acrylonitrile to
10H-phenothiazine (1).12 Under standard reduction conditions
with lithium aluminium hydride in boiling diethyl ether nitrile
3. Results and discussion
3.1. Synthesis and properties of phenothiazinyl alkyl
ammonium salts
Immobilization of functional sites on the surface of an inert
support material can be realized by three different ways of
interaction: (1) adsorption, which in general is a reversible
process, (2) covalent binding or (3) electrostatic interaction.
In contrast to covalent binding, electrostatic immobilization
gives the functional sites more mobility, which can be of
benefit for some applications. Here, we report on the electro-
static immobilization of rather large, cationic phenothiazines
on a high specific area, periodic mesoporous, BTEB based
SBA-15 organosilica material that carries covalently grafted
anionic groups. The immobilization process relies on an ion
exchange process.
By using this route to a material containing redox-active dye
molecules it can be expected that the distribution of the redox
Scheme 2 Syntheses of phenothiazinyl amines 3 and 7.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 16396--16410 | 16401

View Article Online
NJC
Paper
Scheme 3 Synthesis of the dithienylated, Boc-protected, amino alkyl
substituted phenothiazine 9.
2 can be transformed to the corresponding primary amine 3 in
excellent yields (Scheme 2).12
Scheme 4 Syntheses of the alkyl ammonium functionalized phenothiazines
4 and 10.
Based on this procedure the synthesis of p-expanded, dithie-
nylated phenothiazine derivatives with substituents at the
3- and 7-positions was carried out. Starting from 5 the cyanoalkyl
chain was introduced. Due to the instability of the cyanoalkyl
chain of compound 6 towards strong reducing agents like lithium
aluminium hydride an alternative reduction procedure first
published by B. Singaram et al. was probed.15 Under rather mild
reduction conditions, using indium trichloride and sodium
borohydride, the nitrile was transformed into the corresponding
amine 7. After workup of the crude product, the free base amine 7
could be obtained with excellent yield (Scheme 2).
Table 1 Selected absorption and emission spectra and CV data of
compounds 4 and 10
Absorption
max,abs/nm
Compound (e 103/Mꢀ1 cmꢀ1
Stokes
shift
l
Emission
E00/+1
,
)
l
max,em/nm Dvꢀ1/cmꢀ1 E0+1/+2/V
4
10
255 (40), 305 (5)
289 (53), 342 (18) 489
—
—
8800
0.78
0.72, 1.26
The direct use of the free amine base 7 in the following and d 3.05 for 10. In the 13C NMR spectra, the corresponding
cross-coupling reaction only led to low yields due to the carbon nuclei resonate at d 53.3 for compound 4 and at 53.7 for
formation of side products, catalyst deactivation and inefficient compound 10. Spectroscopic data of all compounds in this
column chromatography. However, Boc-protection of the manuscript can be found in the ESI.†
amine group giving compound 8 in almost 90% yield prevented
The electronic properties of the phenothiazinyl ammonium
the formation of by-products in the subsequent two-fold thie- salts 4 and 10 were investigated by absorption and emission
nylation of the phenothiazine core. A Suzuki-coupling protocol spectroscopy and by cyclic voltammetry in the anodic region
turned out to be the best choice for the cross-coupling of the (up to 1.7 V). The corresponding data are summarized in Table 1.
two electron-rich substrates giving compound 9. The carbon–carbon In the absorption spectrum of compound 4 the two typical intense
bond formation was carried out using established reaction condi- absorption bands of phenothiazine derivatives appear at about
tions from our group with tetrakis(triphenylphosphine)palladium(0) 255 nm and 305 nm, representing p–p*-transitions, which was
as a catalyst and, due to its good solubility in 1,4-dioxane, caesium shown by comparison with the unsubstituted phenothiazine.18,19
fluoride as the base (Scheme 3).
No fluorescence was detected for compound 4.
With both derivatives in their amine resp. Boc-protected
Compound 10 shows as well two absorption maxima at 289
amine form the generation of the ammonium salts was the next and 342 nm and furthermore a blue-green fluorescence at 489 nm
step. Methyl iodide proved to be not suitable for this purpose upon excitation. The emission hardly overlaps with the absorp-
since the electrochemical oxidation potential of iodide to tion resulting in a large Stokes shift of 8800 cmꢀ1, which
iodine lies in the same range as for phenothiazine and its indicates a significant change in the geometry occurring
derivatives.16,17 Methyl triflate (MeOTf) has similar reactivity as during the transition from the ground state to the excited
methyl iodide and it is easy to manage. The amine base 3 reacted state.20 For the evaluation of the oxidation potentials and the
smoothly with methyl triflate to give the desired ammonium salt 4 reversibility of the first oxidation process cyclic voltammetry
in excellent yield. In terms of the Boc-protected amine 9, the target measurements were performed in solution. Compound 4
compound 10 could be obtained by quantitative deprotection shows the typical behaviour of a quasi-reversible one electron
with trifluoroacetic acid (TFA) and subsequent quaternation with process in the anodic area, with an oxidation potential
0/+1
ox
methyl triflate (Scheme 4).
observed at E
= 0.78 V. The cyclic voltammogram of
The molecular structures of both ammonium salts 4 and 10 the dithienylated compound 10 shows two well-separated
0/+1
were unambiguously supported by spectroscopic characterisa- single-electron processes, with oxidation potentials at E
=
ox
1
+1/+2
tion, elemental analysis and HRMS. In their H NMR spectra, 0.72 V and E
= 1.26 V (see ESI†), the more positive and
ox
the ammonium salts 4 and 10 can clearly be identified by the irreversible process is assigned to the oxidation of one of the
resonances of the trimethylammonium group at d 3.01 for 4 thiophene rings.
16402 | New J. Chem., 2019, 43, 16396--16410 This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
Paper
NJC
well as of the phenothiazine-exchanged products 16 and 17
the typical methods of material chemistry were applied.
The complete set of data can again be found in the ESI.†
In its powder XRD, 12 shows the expected d(100), d(110) and
d(200) reflexes of a 2D hexagonal p6mm structure, typical for
SBA-15-like materials.22 This structure remains pristine through-
out the complete modification processes (Fig. 1) which is not the
case if ammonia was used instead of pyridine to generate a
sulphonate salt containing PMO (see comparison in the ESI†).
In addition, SEM images also show a structurally ordered, SBA-15
like material for 16, even after the whole series of post synthetic
modifications (Fig. 1). According to nitrogen physisorption
measurements, the specific surface area of 1066 m2 gꢀ1 for 12
decreases continuously to 688 and 671 m2 gꢀ1 for 16 or 17,
respectively. All materials show type IV adsorption and desorption
isotherms with a large hysteresis which characterizes mesoporous
compounds (Fig. 1). The pore size distributions, following the BJH
approach, stay stable at around 50 Å over all reaction steps, which
already indicates moderate loadings on the material since
the pore size should decrease significantly if a high amount of
functionalities would have been immobilized inside the pores.
The pore volume decreases by 0.4 cm3 gꢀ1 throughout the
functionalisations while the wall thickness decreases due to a
little collapsing in the structure of the material. When 4 and 10
are immobilized, the wall thickness increases by around 2 nm
in case of 16 and by about 1 nm for 17. This difference can be
Scheme 5 Synthesis and functionalization of the BTEB based meso-
porous material followed by the immobilization of the phenothiazines 4
(R = H) resp. 10 (R = thiophenyl). Conditions: (a) 3-MPTS, toluene, 72 h,
reflux; (b) H2O2(aq.), 24 h, 60 1C; (c) pyridine, 16 h, rt.
3.2. Phenothiazines immobilized on BTEB-PMO as a scaffold.
a. Material synthesis and characterization. As a scaffold for
immobilization of the phenothiazines 4 and 10 a BTEB based
periodic mesoporous organosilica (12) was synthesized from
1,4-dibromobenzene, following modified procedures from the
literature.13,21 The material was subsequently functionalized by
post synthetic modification to obtain the pyridinium sulpho-
nate loaded PMO 15 which was then used to immobilize
phenothiazines 4 and 10 onto its surface through ion exchange
reactions (Scheme 5). For the chemical, structural and textural
characterization of 12, 15, their intermediates 13 and 14 as
Fig. 1 Powder XRD patterns (top left), nitrogen physisorption isotherms (top right), BJH and DFT pore size distributions (bottom left) and SEM image for
16 as representative (bottom right).
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 16396--16410 | 16403

View Article Online
NJC
Paper
Table 2 Overview of characteristic data for the PMOs (d100 is the d(100) spacing, a0 is the cell parameter (a0 = 2d100/O3), Vp is the pore volume acc. to
BJH, Dp(BJH) is the average pore diameter according to BJH, wt is the wall thickness (a0 ꢀ Dp(BJH)), Aspec(BET) is the specific surface area according to BET)
Material
d100/nm
a0/nm
Vp(BJH)/cm3 gꢀ1
Dp(BJH)/nm
wt/nm
Aspec(BET)/m2 gꢀ1
Loadinga/mmol gꢀ1
12
13
14
15
16
13.4
13.0
12.3
12.1
13.8
—
15.4
15.0
14.2
14.0
15.9
—
1.37
1.10
0.95
0.97
0.96
0.60
0.91
5.63
5.61
4.95
5.22
5.35
3.83
5.21
9.77
9.39
9.25
8.78
10.6
—
1066
827
687
742
688
467
671
—
1276
502
589
o461
o332
o480
+
16ꢃ
17
12.8
14.8
9.59
a
Calcd according to the sulphur content of the materials. For the materials containing phenothiazines, the value is a maximum of loading, since it
is calculated for a complete ion exchange.
explained by the size of the dyes: 4 is smaller than 10 and at d 115.8 and 22.0 can be assigned to the molecular structure
therefore it is fitting better into the pores, which leads to a of phenothiazine 4. Signals that also could be part of the
presumably higher loading and finally to an overall thicker wall spectrum of 4, such as the peaks at d 126.6 and 52.1, are
when 16 and 17 are compared. The data obtained from powder overlapping with peaks that are already present in the 13C
XRD and physisorption measurements for the PMOs are CP-MAS NMR of 15. In the 29Si CP-MAS NMR spectrum of 16,
summarized in Table 2.
the three T branches are observed at d ꢀ61.5 (T1, R-Si(OSi)1(OR0)2),
To prove the successful immobilization, a comparison of the ꢀ71.2 (T2, R-Si(OSi)2(OR0)1) and ꢀ80.4 (T3, R-Si(OSi)3) being
13C CP-MAS NMR spectra of PMOs 16 and 17 and the high typical for silica centres functionalized with aromatic substituents
resolution 13C{1H} NMR spectra of the corresponding precursor R.23,24 In addition, a peak at d ꢀ102.4 is present and can be
compounds 4 and 10 is given in Fig. 2. For PMO 16 the signals assigned to a Q3 branch (Si(OSi)3(OR)1) which shows a small
destructive effect that occurs during the oxidation process leading
to the sulphonic acid intermediate 14 since these silica centres are
not anymore connected to a phenylene bridge. In case of 17, the
peaks at d 140.4 and 114.2 in the 13C CP-MAS NMR have counter-
parts in the high resolution 13C{1H} NMR spectrum of 10 and
therefore prove the immobilisation onto the PMO’s surface.
Beside the three T branches there are two additional peaks for
the scaffold in the 29Si CP MAS NMR spectrum at d ꢀ92.6 and
ꢀ101.4, which can be assigned to the Q2 and the Q3 branch,
again originating from the oxidation of the thiol 13.10,25
According to elemental analysis, using the sulphur content
for calculation, the loading of 4 on 16 is 243 mmol gꢀ1 and the
loading of 10 on 17 is 167 mmol gꢀ1. This supports the
argument given above for the difference of the wall thicknesses
since the smaller molecule 4 is fitting better into the pores
than 10. In comparison to our previous work, the loading is
lower, caused by the synthetic approach. While in the work of
Hemgesberg et al. a sol–gel process was applied to synthesize
a phenothiazine bearing mesoporous material, we here use
post-synthetic grafting followed by ion exchange. The pre-
cursors 4 and 10 have to diffuse into already existing pores
instead of their concomitant formation with the precursor.26
Since ion exchange is a typical equilibrium reaction, a complete
exchange cannot be expected. This also lowers the loading.
Additionally, the aromatic wall of the PMOs lowers the overall
loading of the dyes calculated in mmol gꢀ1 since the relative
mass of the anchoring points is clearly lower than in a neat
silica material synthesized from TEOS. One of the strong
advantages of the current approach is that phenothiazines are
immobilized via ionic interactions so that they can hop inside
the pores from sulphonate to sulphonate anion, which should
lead to a more homogenous distribution of the dyes. Another
advantage is the more robust scaffold, coming from the
Fig. 2 13C and 29Si CP-MAS NMR spectra of 16 (top) and 17 (bottom) with
high resolution 13C{1H} NMR spectra of 4 and 10 (both grey) for comparison.
Rotational sidebands in the 13C spectra are marked with *.
16404 | New J. Chem., 2019, 43, 16396--16410 This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
Paper
NJC
aromatic walls, which leads to a material that is more tolerable approximately five times higher upon light exposition (Fig. 3). One
to harsher reaction conditions, overall leading to a material reason for this observation is that chemical oxidation causes a
with better ordering.27,28
loss of immobilized phenothiazine 16 since the loading after the
b. Spectroscopy. After successfully immobilizing pheno- treatment is 332 mmol gꢀ1 and therefore approx. 130 mmol gꢀ1
thiazines via ionic interaction onto mesoporous organosilicas, lower than prior to oxidation. Furthermore, it might be possible
the goal was to generate stable radicals on these materials. that the collapsed structure resulting from (NO)BF4 treatment has
With 16 two routes were tested: on the one hand 16 was a negative influence on the radical formation. However, over-
oxidized with (NO)BF4, accordinꢀgly to our previous work, to oxidation of the phenothiazine moiety to the corresponding
obtain a radical cation with BF4 as a counter ion.26 On the S-oxide should be the most important factor in this context. The
other hand this PMO was stored on a window bench for 24 h g-value is in both cases 2.0055 and therefore close to the data of
since we noticed a colour change when the sample was exposed previously reported stable phenothiazine radical cations with
to daylight. Photo activation of phenothiazines leading to g = 2.0051.30 Also the spectral shape due to the anisotropic
stable radical cations is well known in the literature so that unresolved 14N and 1H hyperfine interactions is typical for
we decided to compare these two strategies.29 Oxidation with phenothiazine in a sterically restricted environment.
(NO)BF4 leads to a collapse in the structural ordering as can be
In the solid state UV/Vis spectra the radical cation is hardly
seen in comparison to the powder XRD and the physisorption detectable after oxidation with (NO)BF4, while the band at
data. While the d(100) reflex in the powder XRD shifts to a around 540 nm, which can be assigned to the radical, is
higher angle, the d(110) and d(200) reflexes completely disappear. clearly visible after light irradiation (Fig. 3).31 A further
Both underline the decrease in structural ordering of the material. difference between the two materials is a band at 341 nm
In addition, the specific surface area (467 m2 gꢀ1) is by 221 m2 gꢀ1 appearing after chemical oxidation. A possible explanation
lower than before the chemical oxidation and the pore diameter for this observation is that the phenothiazine gets over-
collapses from 53.5 to 38.3 Å (Fig. 3). Still the reduced structural oxidized, presumably at the sulphur atom, creating a new
ordering is not as low as in our previous work demonstrating the chromophore. In addition, the visual impression of the sample
robust character of the BTEB based PMO materials.26 Finally, as is different: while irradiation with light leads to an intensely
the EPR spectra show, the amount of radicals on the material is pink material, the one oxidized with (NO)BF4 was only slightly
Fig. 3 Powder XRD patterns (top left), nitrogen physisorption isotherms and BJH pore size distributions (top right), EPR spectra (bottom left) and UV/Vis
spectra (bottom right) for 16 after treatment with light (black) and (NO)BF4 (red).
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 16396--16410 | 16405

View Article Online
NJC
Paper
Fig. 4 EPR (top) and UV/Vis spectra (bottom) of 17 after treatment with
light.
Fig. 5 Cyclic and square-wave voltammograms of 16 (top) and 17
(bottom), measured with a scan rate of 100 mV sꢀ1 in a 1 M solution of
tBu4NPF6 in dichloromethane as electrolyte and with a platinum electrode
as working and as counter electrode and a Ag/0.01 M AgNO3/MeCN
system as reference electrode.
pink. Pink is the prototypical colour of phenothiazine radical
cations.32
In case of material 17, being stored under ambient light the
same way as discussed above, there are also radicals formed
that can be detected by EPR (g value: 2.0047) and by solid state
UV/Vis spectroscopy showing a band at 548 nm. The EPR signal
is slightly sharper due to the lack of hyperfine couplings of the are E0o/x+1 = 0.55 and E0o/x+1 = 0.53 V for the PMOs, respectively. The
protons at the 3- and 7-position in compound 17. This signal second oxidation process of the dithienylated phenothiazine
becomes rather intense when the PMO was stored under cannot be observed in the solid 17, an observation that was
ambient light near by the window bench for 24 h (Fig. 4). In also reported by Hemgesberg et al. with a covalently grafted
this case, a dusky pink solid results. Both oxidized materials, dithienylated phenothiazine on a SBA scaffold.26 As the electro-
+
16ꢃ and 17ꢃ+, maintain their colour even when they are stored chemical data of compounds 4 and 10 were recorded in CH2Cl2
in the dark for ten months, still showing intense signals in the suspension, the PMO scaffold provides a more polar environ-
EPR (see the ESI† for comparison), which proves the persistence ment, rationalizing the slightly lower redox potentials of the
of the radicals formed. The high stability can be explained by phenothiazine sites in the materials 16 and 17 compared to
the environment inside the pores. Since not every available their precursors.
pyridinium sulphonate was used for ion exchange (equilibrium
After the generation and analysis of these persistent radicals
reaction), there is still a considerable amount of pyridinium we took a closer look to the behaviour of material 16. A sample
sulphonate inside the pores and, thus, close to the radical cations. of 16 was stored in the dark directly after preparation and was
These pyridinium sulphonates stabilize the radical cations to a then treated under constant UV light, while aliquots were taken
great extent providing them their long-lived character.31
after some intervals. These samples were again stored under
In solid state cyclic and square-wave voltammetry measure- exclusion of light. Fig. 6 shows the process of radical formation:
ments of materials 16 and 17 both show a one electron process the intensity of the EPR signal increases over the irradiation
with a quasi-reversible behaviour (Fig. 5). The oxidation potentials time until 24 h were reached. It is also notable that no radicals
16406 | New J. Chem., 2019, 43, 16396--16410 This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
Paper
NJC
Scheme 6 Synthesis and functionalization of SBA-15 followed by the
immobilization of the phenothiazines 4. Conditions: (a) 3-MPTS, toluene,
72 h, reflux (b) H2O2(aq.), 24 h, 60 1C (c) pyridine, 16 h, rt.
obtained the typical 2D hexagonally ordered structure. How-
ever, a closer look at the physisorption data shows that there
are some differences compared to the organosilica materials.
The specific surface area is by 448 m2 gꢀ1 over 200 m2 gꢀ1 lower
than for the PMO derivative 16. Also the average pore size (49.1 Å)
is slightly smaller (Fig. 7).
The pore volume decreases significantly when the thiopropyl
chains are grafted onto the material indicating the high load-
ing of 2.45 mmol gꢀ1. After oxidation to the sulphonic acid,
the pore volume increases since the reaction process leads to a
loss of grafted species on the scaffold now possessing a
sulphonate loading of 614 mmol gꢀ1. Finally, the pore volume
decreases again when compound 4 is immobilized. The wall
thickness increases while the grafting procedure, again show-
ing a high content of thiopropyl chains. Thereafter, the wall
thickness slightly decreases until the final compound 22
is obtained. The data obtained from powder XRD and physi-
Fig. 6 EPR spectra after different radiation times with UV light (top) and
after reduction with ascorbic acid (bottom).
were found when the freshly prepared sample was directly sorption measurements for the SBA-15s are compared in
stored in the dark showing that no radicals are formed during Table 3.
immobilization and that radiation with light is essential to
To our surprise the loading of 4 on 22 was only 175 mmol gꢀ1
form these radicals. In addition, we tested the reduction of the and therefore 68 mmol gꢀ1 lower than for PMO 16. A possible
+
radical 16ꢃ with ascorbic acid. Therefore, the material was explanation is that on the sulphonate step PMO 15 only
suspended in a saturated solution of ascorbic acid in degassed contains a slightly lower loading of sulphonate groups with
water and subsequently washed for several times with degassed 589 mmol gꢀ1 than the SBA-15 21 with 611 mmol gꢀ1, so the
water to remove excessive ascorbic acid. The EPR spectrum difference before ion exchange is not that as significant as
shows that the radicals nearly disappear due to the reduction expected. In addition, the PMO derivative 15 has an average
indicating the chemical reversibility of the radical formation pore diameter of 52.2 Å, while neat silica SBA-15 21 has with
(Fig. 6). The small residual signal in the EPR spectrum after 50.5 Å a slightly smaller pore size. The surface area is with
reduction can be explained with the hydrophobic character of 742 m2 gꢀ1 for 15 higher than for 21 with 557 m2 gꢀ1. Finally,
the BTEB-PMOs.33 It is likely that the aqueous solution of the structural ordering is higher for PMOs than for SBA-15s
ascorbic acid does not reach every phenothiazine radical since the d(110) and d(200) reflexes are hardly detectable in
in the rather hydrophobic pores so that some of the radical case of the SBA-15s (Fig. 7). In combination, these effects
species remain after the reduction process of more accessible surpass the small difference in sulphonate loading leading to
radicals.
PMO 16 that is with respect to previous aspects superior to the
SBA-15 derivative 22.
3.3. Phenothiazines immobilized on SBA-15 as a scaffold
b. Spectroscopy. In the EPR and solid state UV/Vis spectra
a. Material synthesis and characterization. For exploring the radical species of 22 is also clearly visible after storage on a
whether the aromatic framework of the BTEB-PMOs has more window bench (Fig. 8). The g-value (2.0056) is nearly the same
influence on the material properties than a lower loading as for 16, showing that the difference in the scaffold has
of phenothiazine, we synthesized the neat silica SBA-1514 virtually no influence on the radical cation.
material 18. This was functionalized the same way as the PMOs
Therefore the presence of phenylene moieties in the PMO
(Scheme 6) leading to material 22 loaded with 4. Again, we framework is not responsible for the light-driven oxidation of
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 16396--16410 | 16407

View Article Online
NJC
Paper
Fig. 7 Powder XRD patterns (top left), nitrogen physisorption isotherms (top right), BJH and DFT pore size distributions (bottom left) and SEM image for
22 as representative (bottom right).
Table 3 Overview over characteristic data for the SBA-15s (d100 is the d(100) spacing, a0 is the cell parameter (a0 = 2d100/O3), Vp is the pore volume,
Dp(BJH) is the average pore diameter according to BJH, wt is the wall thickness (a0 ꢀ Dp(BJH)), Aspec(BET) is the specific surface area according to BET)
Material
d100/nm
a0/nm
Vp/cm3 gꢀ1
Dp(BJH)/nm
wt/nm
Aspec(BET)/m2 gꢀ1
Loadinga/mmol gꢀ1
18
19
20
21
22
12.6
12.4
13.0
12.8
12.6
14.6
14.4
15.0
14.8
14.6
0.96
0.45
0.75
0.74
0.65
4.94
3.83
5.05
5.05
4.91
9.66
10.6
9.95
9.75
9.69
840
289
581
557
448
—
2454
614
611
Z393
a
Calcd according to the sulphur content of the materials. For the material containing phenothiazines, the value is a maximum of loading, since it
is calculated for a complete ion exchange.
the phenothiazines. Since we had not observed a comparable phenothiazine centres by applying commonly used (NO)BF4,
behaviour for the previously published, covalently immobilized resulted in several products of different chemical structure.
systems, we assume that the ionic nature of the phenothiazines In contrast, simple irradiation of the phenothiazine containing
clearly favours a rather rapid oxidation under light.
samples by visible light produced the desired phenothiazine
radical cations without any side products. This is in contrast to
materials in which the phenothiazine moieties are covalently
bound to the surface of mesoporous silica materials. These
radical cations are highly stable at room temp and under
4. Conclusions
Periodic mesoporous organosilica and neat silica materials ambient atmosphere and their formation is not dependent on
carrying electrochromophores were obtained by ion exchange the support but only on the mode of immobilization. By apply-
reactions with phenothiazine functionalized quaternary ammo- ing appropriate reducing agents such as ascorbic acid an almost
nium salts. Full characterization of these materials and their complete regeneration of the neutral phenothiazine sites can be
organic precursor compounds was carried out. According to achieved. In combination with solid-state cyclic voltammetry,
UV/Vis spectroscopic data, oxidation of the immobilized these experiments prove that the highly reversible redox
16408 | New J. Chem., 2019, 43, 16396--16410 This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
Paper
NJC
I. E. Kochevar and R. W. Redmond, J. Am. Chem. Soc., 1995,
117, 10871.
3 J. M. Ford, E. P. Bruggemann, I. Pastan, M. M. Gottesman
and W. N. Hait, Cancer Res., 1990, 50, 1748; B. Weiss,
W. C. Prozialeck and T. L. Wallace, Biochem. Pharmacol.,
1982, 31, 2217; I. Forrest and F. Forrest, Biochim. Biophys.
Acta, 1958, 29, 441; Y. Iida, Bull. Chem. Soc. Jpn., 1971,
44, 663; J.-C. Moutet and G. Reverdy, Nouv. J. Chim., 1983,
7, 105; B. Varga, A. Csonka, A. Csonka, J. Molnar, L. Amaral
and G. Spengler, Anticancer Res., 2017, 37, 5983; K. Pluta,
B. Morak-Mlodawska and M. Jelen, Eur. J. Med. Chem., 2011,
46, 3179.
4 S. Taniguchi, N. Suzuki, M. Masuda, S. Hisanaga,
T. Iwatsubo, M. Goedert and M. Hasegawa, J. Biol. Chem.,
2005, 280, 7614; E. A. Weinstein, T. Yano, L.-S. Li,
D. Avarbock, A. Avarbock, D. Helm, A. A. McColm,
K. Duncan, J. T. Lonsdale and H. Rubin, Proc. Natl. Acad.
Sci. U. S. A., 2005, 102, 4548; E. Akoury, M. Pickhardt,
M. Gajda, J. Biernat, E. Mandelkow and M. Zweckstetter,
Angew. Chem., Int. Ed., 2013, 52, 3511–3515; E. Nishiwaki,
H. Nakagawa, M. Takasaki, T. Matsumoto, H. Sakurai and
M. Shibuya, Heterocycles, 1990, 31, 1763; J. Decuyper,
J. Piette, M. Lopez, M. P. Merville and A. Vorst, Biochem.
Pharmacol., 1984, 33, 4025; A. G. Motten, G. R. Buettner and
C. F. Chignell, Photochem. Photobiol., 1985, 42, 9; H. Fujita
and I. Matsuo, Chem.-Biol. Interact., 1988, 66, 27.
Fig. 8 EPR (top) and UV/Vis spectra (bottom) of 22 after treatment with
light.
5 B. Wang, Y. Wang, J. Hua, Y. Jiang, J. Huang, S. Qian and
H. Tian, Chem. – Eur. J., 2011, 17, 2647; R. Y. Lai, X. Kong,
S. A. Jenekhe and A. J. Bard, J. Am. Chem. Soc., 2003,
125, 12631; S. Xu, T. Liu, Y. Mu, Y.-F. Wang, Z. Chi, C.-C.
Lo, S. Liu, Y. Zhang, A. Lien and J. Xu, Angew. Chem., Int. Ed.,
2015, 54, 874; E. A. Weiss, M. J. Tauber, R. F. Kelley,
M. J. Ahrens, M. A. Ratner and M. R. Wasielewski, J. Am.
Chem. Soc., 2005, 127, 11842; S. Ardo, Y. Sun, A. Staniszewski,
F. N. Castellano and G. J. Meyer, J. Am. Chem. Soc., 2010,
132, 6696; A. Higuchi, H. Inada, T. Kobata and Y. Shirota, Adv.
Mater., 1991, 3, 549; Z.-S. Huang, H. Meier and D. Cao,
J. Mater. Chem. C, 2016, 4, 2404; R. S. Nobuyasu, Z. Ren,
G. C. Griffiths, A. S. Batsanov, P. Data, S. Yan, A. P. Monkman,
M. R. Bryce and F. B. Dias, Adv. Opt. Mater., 2016, 4, 597;
M. Cheng, C. Chen, X. Yang, J. Huang, F. Zhang, B. Xu and
L. Sun, Chem. Mater., 2015, 27, 1808.
character of the phenothiazines does not change due to the
electrostatic immobilization on inert supports. The results
described above now enable a simple access to a broad variety
of electrochemically stable, redox-active hybrid materials by ion
exchange.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
6 S. A. Jenekhe, L. Lu and M. M. Alam, Macromolecules, 2001,
34, 7315; X. Kong, A. P. Kulkarni and S. A. Jenekhe, Macro-
molecules, 2003, 36, 8992; D.-H. Hwang, S.-K. Kim,
M. J. Park, J. H. Lee, B. W. Koo, I. N. Kang, S. H. Kim and
T. Zyung, Chem. Mater., 2004, 16, 1298; Y. Zou, W. Wu,
G. Sang, Y. Yang, Y. Liu and Y. Li, Macromolecules, 2007,
40, 7231; M. J. Park, J. Lee, J.-H. Park, S. K. Lee, J.-I. Lee,
H.-Y. Chu, D.-H. Hwang and H.-K. Shim, Macromolecules,
2008, 41, 3063.
The authors wish to thank the Deutsche Forschungsge-
meinschaft for financial support of this work (DFG projects
MU 1088/13-1 and TH 550/20-1).
Notes and references
1 For a comprehensive review on the chemistry of phenothia-
zines, see e.g. R. A. Aitken and K. M. Aitken, 1,4-Thiazines
and their Benzo Derivatives, Comprehensive Heterocyclic
Chemistry III, 2008, ch. 8.09, vol. 8, p. 607.
¨
7 M. Hauck, M. Stolte, J. Schonhaber, H. G. Kuball and
T. J. J. Mu¨ller, Chem. – Eur. J., 2011, 17, 9984; C. S.
¨
2 R. McIntyre and H. Gerischer, Ber. Bunsen-Ges. Phys. Chem.,
1984, 88, 963; G. Diehland and U. Karst, J. Chromatogr. A,
2000, 890, 281; C. Garcia, G. A. Smith, W. G. McGimpsey,
Kramer, K. Zeitler and T. J. J. Mu¨ller, Tetrahedron Lett.,
2001, 42, 8619; D. Urselmann, K. Deilhof, B. Mayer and
T. J. J. Mu¨ller, Beilstein J. Org. Chem., 2016, 12, 2055;
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 16396--16410 | 16409

View Article Online
NJC
Paper
M. Sailer, A. W. Franz and T. J. J. Mu¨ller, Chem. – Eur. J., 20 L. Yang, J.-K. Feng and A.-M. Ren, J. Org. Chem., 2005,
2008, 14, 2602; M. Sailer, M. Nonnenmacher, T. Oeser and
T. J. J. Mu¨ller, Eur. J. Org. Chem., 2006, 423.
70, 5987.
¨
21 B. Schafgen, O. D. Malter, E. Kaigarula, A. Schu¨ßler, S. Ernst
8 A. P. Kulkarni, T. W. Kwon and S. A. Jenekhe, J. Phys. Chem.
B, 2005, 109, 19584; T. Meyer., D. Ogermann, A. Pankrath,
K. Kleinermanns and T. J. J. Mu¨ller, J. Org. Chem., 2012,
77, 3704.
9 N. Bucci and T. J. J. Mu¨ller, Tetrahedron Lett., 2006, 47, 8323;
N. Bucci and T. J. J. Mu¨ller, Tetrahedron Lett., 2006, 47, 8329;
S. Bay, T. Villnow, G. Ryseck, V. Rai-Constapel, P. Gilch and
T. J. J. Mu¨ller, ChemPlusChem, 2013, 78, 137.
and W. R. Thiel, Microporous Mesoporous Mater., 2017,
251, 122; S. Shylesh, M. Jia, A. Seifert, S. Adappa, S. Ernst
and W. R. Thiel, New J. Chem., 2009, 33, 717; S. Shylesh,
A. Wagener, A. Seifert, S. Ernst and W. R. Thiel, Angew.
Chem., Int. Ed., 2010, 49, 184; F. Rajabi, D. Schaffner,
S. Follmann, C. Wilhelm, S. Ernst and W. R. Thiel, Chem-
CatChem, 2015, 7, 3513.
22 Y. Goto and S. Inagaki, Chem. Commun., 2002, 2410.
¨
10 Z. Zhou, A. W. Franz, M. Hartmann, A. Seifert, T. J. J. Mu¨ller 23 L. Wang, S. Shylesh, D. Dehe, T. Philippi, G. Dorr, A. Seifert,
and W. R. Thiel, Chem. Mater., 2008, 20, 4986; A. W. Franz,
Z. Zhou, M. Hartmann, R. N. Klupp Taylor, M. Jia, S. Ernst
Z. Zhou, R. Turdean, A. Wagener, B. Sarkar, M. Hartmann,
and W. R. Thiel, ChemCatChem, 2012, 4, 395.
S. Ernst, W. R. Thiel and T. J. J. Mu¨ller, Eur. J. Org. Chem., 24 C. Yoshina-Ishii, T. Asefa, N. Coombs, M. J. MacLachlan and
2009, 3895; Z. Zhou, A. W. Franz, S. Bay, B. Sarkar, A. Seifert, G. A. Ozin, Chem. Commun., 1999, 2539.
P. Yang, A. Wagener, S. Ernst, M. Pagels, T. J. J. Mu¨ller and 25 Z. Zhou, Q. Meng, A. Seifert, A. Wagener, Y. Sun, S. Ernst
W. R. Thiel, Chem. – Asian J., 2010, 5, 2001; M. Fingerle,
M. Hemgesberg, S. Lach, W. R. Thiel and C. Ziegler, Thin
Solid Films, 2015, 591, 8.
and W. R. Thiel, Microporous Mesoporous Mater., 2009, 121,
145–151.
26 M. Hemgesberg, B. Bayarmagnai, N. Jacobs, S. Bay, S. Follmann,
C. Wilhelm, Z. Zhou, M. Hartmann, T. J. J. Mu¨ller, S. Ernst,
G. Wittstock and W. R. Thiel, RSC Adv., 2013, 3, 8242.
11 F. Gai, X. Li, T. Zhou, X. Zhao, D. Lu, Y. Liu and Q. Huo,
J. Mater. Chem. B, 2014, 2, 6306; E. Arzoumanian, F. Ronzani,
A. Trivella, E. Oliveros, M. Sarakha, C. Richard, S. Blanc, 27 N. Mizoshita, T. Tani and S. Inagaki, Chem. Soc. Rev., 2011,
T. Pigot and S. Lacombe, ACS Appl. Mater. Interfaces, 2014,
6, 275.
12 E. F. Godefroi and E. L. Wittle, J. Org. Chem., 1956, 21, 1163.
40, 789.
28 S. Inagaki, S. Guan, Y. Fukushima, T. Ohsuna and
O. Terasaki, J. Am. Chem. Soc., 1999, 121, 9611.
13 K. J. Shea, D. A. Loy and O. Webster, J. Am. Chem. Soc., 1992, 29 N. J. Turro, I. V. Khudyakov and H. van Willigen, J. Am. Chem.
114, 6700.
Soc., 1995, 117, 12273; J. Y. Bae, K. T. Ranjit, Z. Luan,
R. M. Krishna and L. Kevan, J. Phys. Chem. B, 2000, 104, 9661;
J. Y. Bae and L. Kevan, Microporous Mesoporous Mater., 2001, 50, 1.
30 Y. S. Kang, J. A. Jung and L. Kevan, J. Chem. Soc., Faraday
14 S.-Y. Chen, C.-Y. Tang, W.-T. Chuang, J.-J. Lee, Y.-L. Tsai,
J. C. C. Chan, C.-Y. Lin, Y.-C. Liu and S. Cheng, Chem.
Mater., 2008, 20, 3906.
´
´
15 J. Z. Saavedra, A. Resendez, A. Rovira, S. Eagon,
D. Haddenham and B. Singaram, J. Org. Chem., 2012, 77, 221.
16 H. Hayashi and T. Koizumi, Heterocycles, 2016, 92, 1441.
17 I. S. El-Hallag, J. Chil. Chem. Soc., 2010, 55, 67.
18 H. H. Mantsch and J. Dehler, Can. J. Chem., 1969, 47, 3173.
Trans., 1998, 94, 3247; F. Lopez Ruperez, J. C. Conesa and
J. Soria, Spectrochim. Acta, 1984, 40, 1021.
31 M. C. Hovey, J. Am. Chem. Soc., 1982, 104, 4196.
32 I. Jelınek, I. Nemcova and P. Rychlovsk´y, Talenta, 1991,
´
ˇ
´
38, 1309.
´
´
19 J. E. Bloor, B. R. Gilson, R. J. Haas and C. L. Zirkle, J. Med. 33 D. Esquivel, C. Jimenez-Sanchidrian and F. J. Romero-
Chem., 1970, 13, 922. Salguero, Mater. Lett., 2011, 65, 1460.
16410 | New J. Chem., 2019, 43, 16396--16410 This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019