Synthesis, electronic properties, and self-assembly on Au{111} of thiolated phenylethynyl phenothiazines

Article abstract of DOI:10.1021/cm901514t

Thiolated phenylethynyl phenothiazines can be regarded as redox-active "alligator clips". They are readily synthesized from ethynyl phenothiazines and show intense blue-green luminescence upon ultraviolet (UV) excitation. Cyclic voltammetry reveals revers

Full text of DOI:10.1021/cm901514t

52 Chem. Mater. 2010, 22, 52–63
DOI:10.1021/cm901514t
Synthesis, Electronic Properties, and Self-Assembly on Au{111} of
Thiolated Phenylethynyl Phenothiazines
Christa S. Barkschat,† Svetlana Stoycheva,‡ Michael Himmelhaus,‡,§ and
€
Thomas J. J. Muller*
,†
†
€
€
Institut fu€r Organische Chemie und Makromolekulare Chemie, Heinrich-Heine-Universitat Dusseldorf,
‡
Universitatsstrasse 1, D-40225 Dusseldorf, Germany and Physikalisch-Chemisches Institut,
€
Ruprecht-Karls-Universita€t Heidelberg, Im Neuenheimer Feld 253, D-69120 Heidelberg, Germany.
Present address: Research & Development Division, Fujirebio, Inc., 51 Komiya-cho, Hachioji-shi,
Tokyo 192-0031, Japan.
€
§
Received June 2, 2009. Revised Manuscript Received November 20, 2009
Thiolated phenylethynyl phenothiazines can be regarded as redox-active “alligator clips”. They are
readily synthesized from ethynyl phenothiazines and show intense blue-green luminescence upon
ultraviolet (UV) excitation. Cyclic voltammetry reveals reversible oxidation, also after the chemi-
sorption of in situ liberated thiols on gold electrodes. The chemisorption on Au{111} was studied by
ellipsometry, contact angle measurements, X-ray photoelectron spectroscopy, and infrared reflection
absorption spectroscopy, and the obtained data support the formation of self-assembled monolayers,
depending on the structure of the substrates. These findings classify them as promising candidates for
molecular wires switchable by redox manipulation.
Introduction
gold to form SAMs via the exposition of well-defined gold
substrates to solutions of the sulfur-functionalized mole-
cules.5 These “alligator clips”6 are able to bind functional
molecules covalently to Au{111} surfaces. Adsorbed on gold,
phenyl derivatives,7,8 conjugated biphenyls7,9 and oligo-
phenyls,7,9,10 oligothiophenes,7 porphyrine derivates,9
phenanthrenes,11 fullerenes,12 and optically active naph-
thalenes13 were studied in break junction experiments,
with respect to single-molecule conductance, one-bit ran-
dom access memory and, especially, as conductive mole-
cular wires. Among the many heteroaromatic systems,
phenothiazines, their derivatives, and oligomers are highly
interesting building blocks for rigid-rod and wire-like
molecular modules for single-molecule electronics, as a
Functional organic π-systems1 are of fundamental
importance in the miniaturization of electronic devices
in particular, because they could serve as molecular
switches, wires, and transistors.2 In combination with
the molecule-based bottom-up approach to nanoscale
structures, self-assembled monolayers (SAMs) on well-
defined metal surfaces have become a ground-breaking
strategy for the development of molecular electronics.3 In
recent years, many investigations of SAMs of organic
molecules on gold surfaces have been conducted.4 Thiols,
thiol esters, and disulfides can be easily chemisorbed on
*Author to whom correspondence should be addressed. E-mail: ThomasJJ.
Mueller@uni-duesseldorf.de.
(1) For a monography, see, e.g., Functional Organic Materials;
Synthesis, Strategies, and Applications; M€uller, T. J. J., Bunz, U. H.
F., Eds.; Wiley-VCH: Weinheim, Germany, 2007.
(2) (a) Timp, G. L. Nanotechnology; AIP Press/Springer: New York,
1999. (b) Nanoelectronics and Information Technology: Advanced
Electronic Materials and Novel Devices Information Technology;
Waser, R., Ed.; Wiley-VCH: Weinheim, Germany, 2003. (c) Joachim,
C.; Gimzewski, J. K.; Aviram, A. Nature 2000, 408, 541–548.
(d) Forrest, S. R. Nature 2004, 428, 911–918.
(6) For selected examples of “alligator clips”, see, e.g.: (a) Berry, J. F.;
Cotton, F. A.; Murillo, C. A. Organometallics 2004, 23, 2503–2506.
(b) Maya, F.; Flatt, A. K.; Stewart, M. P.; Shen, D. E.; Tour, J. M. Chem.
Mater. 2004, 16, 2987–2997. (c) Seminario, J. M.; Zacarias, A. G.;
Tour, J. M. J. Am. Chem. Soc. 1999, 121, 411–416. (d) Pearson, D. L.;
Tour, J. M. J. Org. Chem. 1997, 62, 1376–1387.
(7) Tour, J. M.; Jones, L.II; Pearson, D. L.; Lamba, J. S.; Burgin, T.;
Whitesides, G. W.; Allara, D. L.; Parikh, A. N.; Atre, S. J. Am.
Chem. Soc. 1995, 117, 9529–9534.
(3) Tour, J. M. Molecular Electronics: Commercial Insights, Chemis-
try, Devices, Architecture and Programming; World Scientific: River
Edge, NJ, 2003.
(8) Reed, M. A.; Zhou, C.; Muller, C. J.; Burgin, T. P.; Tour, J. M.
Science 1997, 278, 252–254.
(9) Tour, J. M.; Rawlett, A. M.; Kozaki, M.; Yao, Y.; Jagessar, R. C.;
Dirk, S. M.; Price, D. W.; Reed, M. A.; Zhou, C.-W.; Chen, J.;
Wang, W.; Campbell, I. Chem.;Eur. J. 2001, 7, 5118–5134.
(10) Bumm, L. A.; Arnold, J. J.; Cygan, M. T.; Dunbar, T. D.; Burgin,
T. P.; Jones, L.II; Allara, D. L.; Tour, J. M.; Weiss, P. S. Science
1996, 271, 1705–1706.
(11) (a) Dameron, A. A.; Ciszek, J. W.; Tour, J. M.; Weiss, P. S. J. Phys.
Chem. B 2004, 108, 16761–16767. (b) Ciszek, J. W.; Tour, J. M.
Tetrahedron Lett. 2004, 45, 2801–2803.
(12) Shirai, Y.; Cheng, L.; Chen, B.; Tour, J. M. J. Am. Chem. Soc. 2006,
128, 13479–13489.
(13) Zhu, Y.; Gergel, N.; Majumdar, N.; Harriot, L. R.; Bean, J. C.; Pu,
L. Org. Lett. 2006, 8, 355–358.
(4) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides,
G. M. Chem. Rev. 2005, 105, 1103-1169 and references therein.
(5) For reviews on self-assembled monolayers of thiolates on metals,
see, e.g.: (a) Tao, F.; Bernasek, S. L. Chem. Rev. 2007, 107, 1408–
1453. (b) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.;
Whitesides, G. M. Chem. Rev. 2005, 105, 1103–1170. (c) Kriegisch, V.;
Lambert, C. Top. Curr. Chem. 2005, 258, 257–313. (d) Otsubo, T.;
Aso, Y.; Takimiya, K. J. Mater. Chem. 2002, 12, 2565–2575.
(e) Ulman, A. Acc. Chem. Res. 2001, 34, 855–863. (f) Ulman, A.
Chem. Rev. 1996, 96, 1533–1554. (g) Kumar, A.; Abbott, N. L.; Kim,
E; Biebuyck, H. A.; Whitesides, G. M. Acc. Chem. Res. 1995, 28,
219–226.
r
pubs.acs.org/cm
Published on Web 12/10/2009
2009 American Chemical Society

Article
Chem. Mater., Vol. 22, No. 1, 2010 53
€
consequence of their electronic properties. In particular,
their reversible formation of stable radical cations,14 their
tunable redox and fluorescence properties,15 and their
tendency to self-assemble on surfaces by π-π interac-
tions16 make them eligible as redox-switchable molecular
entities. However, the design of reversible phenothiazine
redox systems for SAMs on gold have remained unex-
plored so far. In addition, the inherent folded conforma-
tion of phenothiazines,17 with a folding angle of 158.5°,
represents an intriguing new aspect for the formation of
self-assembled monolayers (SAMs) of this class of com-
pounds. Furthermore, the transformation of phenothia-
zines into stable planar radical cations with excellent
delocalization18 qualifies them as excellent models for
switchable conductive or semiconductive molecular
wires. In continuation of our studies directed to synthe-
size and study (oligo)phenothiazine-based functional
π-systems,19 we now have focused on thiolated phenyl-
ethynyl phenothiazines as redox-active “alligator clips”.
Here, we report the synthesis and their electronic proper-
ties, as studied by cyclic voltammetry (CV), as well as
spectroscopic and spectrometric methods. Furthermore,
to gain insight into their applicability in molecular elec-
tronics also from a practical point of view, their chemi-
sorption and SAM formation on Au{111} was studied by
ellipsometry, contact angle measurements, X-ray photo-
electron spectroscopy (XPS), and infrared reflection
absorption spectroscopy (IRRAS).
Melting points (uncorrected values): Buchi melting point B-540,
1
Stuart Scientific SMP 3, heating rate = 5 K/min; H and 13C
NMR spectra: Bruker ARX 300, Varian VXR 400S; CDCl3
(locked to Me4Si),20 [D6]DMSO (locked to Me4Si). The assign-
ments of quaternary C, CH, CH2 and CH3 have been made using
DEPT spectra. Infrared (IR) analysis: Perkin-Elmer FT-IR
spectrometer 1000, Bruker Vector 22. UV/vis: Perkin-Elmer
UV/vis Spectrometer Lambda 16, Hewlett-Packard Model 8452
A. Fluorescence spectra: Perkin-Elmer LS 50 B (irradiation at
∼10 nm less in energy than the longest wavelength absorption
maximum). MS: Finnigan MAT 90, Finnigan MAT 95 Q,
Finnigan TSQ 700 or JEOL JMS-700. Elemental analyses were
performed in the Microanalytical Laboratories of the Depart-
€
ment Chemie, Ludwig-Maximilians-Universitat Munchen,
and of the Organisch-Chemisches Institut, Ruprecht-Karls-
€
€
Universitat Heidelberg.
Electrochemistry. Cyclic voltammetry experiments (EG & G
potentiostatic instrumentation) were performed under argon in
dry and degassed CH2Cl2 at room temperature and at scan rates
of 100, 250, and 500 mV/s. The electrolyte was 0.10 M Bu4NPF6.
The working electrode was a platinum disk (1 mm in diameter),
the counter electrode was a platinum wire, and the reference
electrode was an Ag/AgCl electrode. For documentation, the
potentials were corrected to the internal standard of Fc/Fcþ in
CH2Cl2 (E00/þ1 =450 mV).21
Synthesis of Thioacetic Acid S-[4-(10-hexyl-10H-phenothia-
zin-3-ylethynyl)-phenyl] Ester (3a). 3-Ethynyl-10-hexyl-10H-
phenothiazine (1a)19c (297 mg (0.97 mmol)) was placed in a
100-mL Schlenk flask under argon. Through a short column
(diameter (Ø)=1.5 cm, L=3 cm) filled with dry, basic alumina
(dried at 70 °C for 1 d in a dry oven and cooled under an
atmosphere of argon) that was placed on top of the Schlenk
flask; subsequently, 8 mL of dry N-ethyl diisopropyl amine and
8 mL of dry THF were filtered in the Schlenk flask. After deg-
assing the light yellow solution with argon for 10 min, 257 mg
(0.92 mmol) of 1-(S-acetylthio)-4-iodo benzene (2)22 and 33 mg
(0.03 mmol) of Pd(PPh3)4 were added and, with mixing, was
heated for 3 days to 40 °C, using a oil bath). The solvents were
removed in vacuo and the residue was triturated with diethyl
ether and filtered. The solvents of the filtrate were removed in
vacuo and the residue was chromatographed on silica gel (Ø=
2.5 cm, L=25 cm) with diethyl ether/pentane (1:25) to give 291
mg (69%) of 3a as a light orange oil. Rf (diethyl ether/pentane
1:25)=0.45. MS (FABþ) m/z (%): 459 (19), 458 (46), 457 (Mþ,
100), 330 (10). IR (KBr), ν: 3059 cm-1 (w), 2950 (ss), 2927 (ss),
2855 (s), 2202 (m), 1850 (w, br), 1712 (ss), 1589 (s), 1574 (s), 1501
(ss), 1468 (ss), 1397 (s), 1335 (s), 1296 (m), 1251 (s), 1195 (m),
1130 (m, sh), 1118 (s, br), 1015 (m), 949 (m), 882 (m), 827 (ss),
749 (ss), 621 (s). UV/vis (CH2Cl2), λmax (ε): 242 nm (21400), 280
(33500), 304 (28000), 358 (15500). Anal. Calcd. for C28H27NOS2
(457.6): C, 73.48; H, 5.95; N, 3.06; S, 14.01. Found: C, 73.41; H,
6.17; N, 2.97; S, 13.89.
Experimental Section
General Remarks. All manipulations (syntheses, cyclic vol-
tammetry, and preparations of SAMs) were conducted under an
inert atmosphere of argon, and the solvents were dried by
standard methods. Reagents were purchased and used without
further purification, unless noted otherwise.
Column Chromatography. Silica gel 60 (Merck, Darmstadt,
Germany), mesh 70-230, was used. Thin layer chromatography
(TLC): silica gel plates (60 F254, Merck, Darmstadt, Germany).
(14) (a) Oka, H. J. Mater. Chem. 2008, 18, 1927–1934. (b) Okamato, T.;
Kuratsu, M.; Kozaki, M.; Hirotsu, K.; Ichimura, A.; Matsushita, T.;
Okada, K. Org. Lett. 2004, 6, 3493–3496. (c) Sun, D.; Rosokha, S. V.;
Kochi, J. K. J. Am. Chem. Soc. 2004, 126, 1388–1401. (d) Kochi, J. K.;
ꢀ
Rathore, R.; Le Magueres, P. J. Org. Chem. 2000, 65, 6826–6836.
(e) Nishinaga, T.; Inoue, R.; Matsuura, A.; Komatsu, K. Org. Lett.
2002, 4, 1435–1438. (f) Pan, D.; Philips, D. L. J. Phys. Chem. A 1999,
103, 4737–4743.
€
(15) (a) Sailer, M.; Franz, A. W.; Muller, T. J. J. Chem.;Eur. J. 2008,
14, 2602–2614. (b) Franz, A. W.; Popa, L. N.; M€uller, T. J. J.
Tetrahedron Lett. 2008, 49, 3300–3303. (c) Franz, A. W.; Popa,
L. N.; Rominger, F.; M€uller, T. J. J. Org. Biomol. Chem. 2009, 7,
469–475.
Thioacetic Acid S-[4-(10,100-dihexyl-10H,100H-[3,30]bipheno-
thiazinyl-7-ylethynyl)-phenyl] Ester (3b). 3-Ethynyl-10,100-di-
hexyl-10H,100H-[3,30]-biphenothiazinyl (1b)19f (162 mg (0.27 mmol))
was placed in a 100-mL Schlenk flask under argon. Through a short
column (Ø=1.5 cm, L=3 cm) filled with dry, basic alumina (dried at
€
(16) Barkschat, C. S.; Guckenberger, R.; Muller, T. J. J. Z. Naturforsch.
2009, 64b, 707–718.
(17) McDowell, J. J. H. Acta Crystallogr., Sect. B: Found. Crystallogr.
1976, B32, 5–10.
(18) Uchida, T.; Ito, M.; Kozawa, K. Bull. Chem. Soc. Jpn. 1983, 56,
577–582.
€
(19) (a) Muller, T. J. J. Tetrahedron Lett. 1999, 40, 6563–6566.
€
(b) Kramer, C. S.; Zeitler, K.; M€uller, T. J. J. Org. Lett. 2000, 2,
€
3723–3726. (c) Kramer, C. S.; M€uller, T. J. J. Eur. J. Org. Chem. 2003,
(20) Hesse, M.; , Meier, H.; , Zeeh, B. Spektroskopische Methoden in der
Organischen Chemie; Georg Thieme Verlag: Stuttgart, New York,
1991; p 69.
(21) Zanello, P. In Ferrocenes; Togni, A., Hayashi, T., Eds.; VCH:
Weinheim, New York, Basel, Cambridge, Tokyo, 1995, 317-430.
(22) Gryko, D. T.; Clausen, C.; Roth, K. M.; Dontha, N.; Bocian, D. F.;
Kuhr, W. G.; Lindsey, J. S. J. Org. Chem. 2000, 65, 7345–7355.
3534–3548. (d) Sailer, M.; Nonnenmacher, M.; Oeser, T.; M€uller, T. J. J.
Eur. J. Org. Chem. 2006, 423–435. (e) Bucci, N.; M€uller, T. J. J.
Tetrahedron Lett. 2006, 47, 8323–8327. (f) Bucci, N.; M€uller, T. J. J.
€
Tetrahedron Lett. 2006, 47, 8329–8332. (g) Hauck, M.; Schonhaber,
J.; Zucchero, A. J.; Hardcastle, K. I.; M€uller, T. J. J.; Bunz, U. H. F.
J. Org. Chem. 2007, 72, 6714–6725.

54 Chem. Mater., Vol. 22, No. 1, 2010
Barkschat et al.
70 °C for 1 d in a dry oven and cooled under an atmosphere of argon)
that was placed on top of the Schlenk flask subsequently 10 mL of dry
N-ethyl diisopropyl amine and 10 mL of dry THF were filtered in the
Schlenk flask. After degassing the light yellow solution with argon for
10 min, 73 mg (0.26 mmol) of 1-(S-acetylthio)-4-iodo benzene (2)22
and 16 mg (0.01 mmol) of Pd(PPh3)4 were added and, with mixing,
was heated for 3 days to 40 °C in an oil bath). The solvents were
removed in vacuo and the residue was chromatographed on silica gel
(Ø=2.5cm,L=25 cm) with diethyl ether/pentane (1:25) to give after
trituration with pentane 86 mg (44%) of 3b as a yellow resin. Rf
(diethyl ether/pentane 1:25) =0.43. MS (FABþ) m/z (%): 741 (14),
740 (34), 739 (68), 738 (Mþ, 100), 737 (13), 654 (14), 653 (16). IR
(KBr), ν: 2954 cm-1 (m), 2927 (m), 2855 (w), 2205 (w), 2066 (w),
2010 (w), 1750 (w), 1704 (s), 1688 (s, sh), 1628 (m), 1579 (w), 1562
(w), 1460 (ss), 1417 (w), 1398 (w), 1381 (w), 1335 (w), 1241 (m, br),
1193 (w), 1119 (w, br), 808 (w), 748 (w). UV/vis (CH2Cl2), λmax (ε):
236 nm (38900), 260 (42500, sh), 290 (69100), 308 (49900, sh),
376 (27100). HRMS calcd. for C46H46N2OS3: 738.2772; Found:
738.2764. Anal. Calcd. for C46H46N2OS3 (739.0): C, 74.76; H, 6.27;
N, 3.79; S, 13.01; Found: C, 74.58; H, 6.19; N, 3.85; S, 13.10.
Thioacetic Acid S-{4-[7-(4-acetylsulfanylphenylethynyl)-10-
methyl-10H-phenothiazin-3-ylethynyl]-phenyl} Ester (5a) and Thio-
acetic Acid S-[4-(7-ethynyl-10-methyl-10H-phenothiazin-3-ylethynyl)-
phenyl] Ester (5b). 3,7-Diethynyl-10-methyl-10H-phenothiazine
(4a)19c (147 mg (0.56 mmol)) and 39 mg (0.03 mmol) of Pd-
(PPh3)4 were placed in a 50-mL Schlenk flask under argon.
Through a short column (Ø=1.5 cm, L=3 cm) filled with dry,
basic alumina (dried at 70 °C for 1 d in a dry oven and cooled
under an atmosphere of argon) that was placed on top of the
Schlenk flask subsequently 10 mL of dry N-ethyl diisopropyl
amine and 10 mL of dry THF were filtered in the Schlenk flask.
After degassing the light yellow solution with argon for 10 min,
300 mg (1.08 mmol) of 1-(S-acetylthio)-4-iodo benzene (2)22 was
mixed and heated for 2 days to 40 °C in an oil bath. The solvents
were removed in vacuo and the residue was triturated with THF
and filtered. The solvents of the filtrate were removed in vacuo,
and the residue was chromatographed on silica gel (Ø=2.5 cm,
L = 25 cm) with dichloromethane/hexane (1:20) and gradually
increasing the amount of dichloromethane to give 26 mg (11%)
of 5a as a dark yellow solid and 152 mg (0.27 mmol=50%) of 5b
as a dark yellow powder.
Through a short column (Ø=1.5 cm, L=3 cm) filled with dry, basic
alumina (dried at 70 °C for 1 d in a dry oven and cooled under an
atmosphere of argon) that was placed on top of the Schlenk flask
subsequently 10 mL of dry N-ethyl diisopropyl amine and 10 mL of
dry THF were filtered in the Schlenk flask. After degassing the light
yellow solution with argon for 10 min 174 mg (0.62 mmol) of 1-(S-
acetylthio)-4-iodo benzene (2)22 and 29 mg (0.02 mmol) of Pd-
(PPh3)4 and, with mixing, was heated for 5 days to 40 °C in an oil
bath). The solvents were removed in vacuo, and the residue was
chromatographed on silica gel (Ø=4 cm, L=20 cm) with dichlo-
romethane/pentane (1:3) and gradually increasing the amount of
dichloromethane to give after trituration with pentane 105 mg
(28%) of 5c as a yellow powder. Mp. 118-122 °C. Rf (dichlo-
romethane/pentane 1:4) = 0.34. MS (FABþ) m/z (%): 1197 (16),
1196 (35), 1195 (66), 1194 (95), 1193 (Mþ, 100), 1192 (17), 1152 (10),
1151 (14), 1151 (9), 1122 (9), 441 (13), 419 (16). IR (KBr), ν: 2954
cm-1 (m, br), 2926 (m), 2855 (w), 2199 (w, br), 1710 (s), 1624 (w),
1581 (w), 1460 (ss), 1377 (m), 1355 (m), 1242 (s), 1191 (m), 1153 (m),
1108(m),1015(w),948(w),876(w),820(s),808(s), 732(w).UV/vis
(CH2Cl2), λmax (ε): 238 (59700, sh), 292 (132700), 384 (53900). Anal.
Calcd. for C74H71N3O2S5 (1194.7): C, 74.39; H, 5.99; N, 3.52; S,
13.42; Found: C,74.59; H, 6.05; N, 3.68; S, 13.28.
Preparation of Sample Films of 3a, 5a, and 5b on Gold. A
freshly prepared 0.1 mM solution of either 3a, 5a, or 5b (10 mL)
in THF was degassed with argon for 10 min using a long thin
syringe needle. A piece of a gold-coated silicon wafer (Au
thickness of 100 nm, 5 nm of titanium as a adhesion promoter,
obtained from Georg Albert Beschichtungen, Heidelberg,
Germany) ∼20 mm ꢀ 20 mm in size and 0.4 mL of a 0.1 M
aqueous solution of NH4OH then were added. Degassing with
argon was continued for 10 min to ensure a thorough mixing of
the aqueous solution and the organic solvent. The substrate was
left in solution under argon for different exposure times (10 min,
30 min, 60 min (1 h), 300 min (5 h), 1440 min (24 h), 7200 min
(120 h)). The wafer piece then was taken out of the solution and
rinsed with THF and ethanol in an ultrasonic bath. Finally, the
substrate bearing an ultrathin film of one of the compounds 3a,
5a, or 5b was dried in a constant stream of dry nitrogen and
stored under an inert gas atmosphere until further analysis.
Surface Characterization. Ellipsometry. Spectral ellipsome-
try was performed ex situ, using a J. A. Woollam M-44 spectral
ellipsometer. For calibration (i.e., the determination of the
incident angle), thermally oxidized SiO2/Si wafers were utilized.
The optical constants of the thin gold layer were obtained from
measurements on freshly evaporated uncoated samples. For
data analysis, a three-layer model (air/organic layer/gold) was
applied, whereby the optical properties of the organic layer were
described by a Cauchy model, assuming a refractive index of
1.490 at 500 nm, which had given good agreement with XPS
results in a prior study on aromatic SAMs.23
5a: MP = 150-151 °C. Rf (dichloromethane/hexane ratio =
1:20)=0.22. MS (EIþ, 70 eV) m/z (%): 412 (29), 411 (Mþ, 100),
370 (10), 369 (33), 368 (19), 355 (10), 354 (35), 353 (22). IR
(KBr), ν: 3279 cm-1 (m, sh), 2900 (w), 2200 (w), 2100 (w), 1694
(m), 1628 (w), 1579 (w), 1504 (w), 1469 (ss), 1393 (m), 1336 (s),
1269 (m), 1154 (w), 1128 (m), 1090 (w), 1016 (w), 955 (w), 824 (s).
UV/vis (CH2Cl2), λmax (ε): 248 (20000), 282 (44900), 306
(29300), 362 (7300), 358 (15500). HRMS calcd. for C25H17-
NOS2: 411.0751; Found: 411.0758 (MS).
Contact Angle Measurements. Advancing contact angles were
€
5b: Mp. 224-225 °C. MS (EIþ, 70 eV) m/z (%): 563 (25), 562
(39), 561 (Mþ, 100), 521 (21), 520 (35), 519 (89), 518 (23), 478
(15), 477 (36), 476 (24), 463 (18), 462 (43), 461 (40), 460 (11). IR
(KBr), ν: 2200 cm-1 (w), 1700 (s), 1639 (ss, br), 1543 (s), 1490
(ss), 1393 (w), 1337 (s), 1269 (w), 1155 (m), 1127 (m), 1015 (w),
954 (w), 882 (w), 826 (m). UV/vis (CH2Cl2), λmax (ε): 290 (55400,
sh), 306 (61600), 380 (21400). Anal. Calcd. for C33H23NO2S3
(561.7): C, 70.56; H, 4.13; N, 2.49; S, 17.12. Found: C, 70.45; H,
4.19; N, 2.53; S, 16.85.
Thioacetic Acid S-{4-[700-(4-acetylsulfanylphenylethynyl)-10,
100,1000-trihexyl-10H,100H,1000H-[3,30,70,300]terphenothiazin-7-
ylethyynyl]-phenyl} Ester (5c). 7,700-Diethynyl-10,100,1000-tri-
hexyl-10H,100H,1000H-[3,30,70,300]-terphenothiazine (4b)19c (279 mg
(0.31 mmol)) was placed in a 50-mL Schlenk flask under argon.
measured with a Kruss G1 goniometer, using purified deionized
water (Milli-Q plus system, Millipore, Eschborn, Germany).
Infrared Spectroscopy (IRRAS). Infrared spectra were taken
with a dry-air purged Bio-Rad spectrometer (model FTS 175c)
that was equipped with a liquid-nitrogen-cooled MCT detector,
an aluminum wire-grid polarizer, and a multiangle reflection
unit set to an incidence angle of 75°. All spectra were norma-
lized to the spectrum of a gold substrate coated with a perdeut-
erated SAM (octadecanethiol-d37) acquired under the same con-
ditions.
(23) Stoycheva, S.; Himmelhaus, M.; Fick, J.; Korniakov, A.; Grunze,
M.; Ulman, A. Langmuir 2006, 22, 4170–4178.

Article
Chem. Mater., Vol. 22, No. 1, 2010 55
Scheme 1. Synthesis of Phenylethynyl (Oligo)Phenothiazinyl Thioacetates 3 and 5
Table 1. Absorptiona and Selected Emission Datab of
(Oligo)Phenothiazinyl Thioacetates 3 and 5
absorption
emission
max,em [nm] (Φf)
Stokes shift,
4υ~ [cm-1
compound
λmax,abs [nm]
λ
]
3a
3b
5a
5b
5c
242, 280, 304, 358
236, 260, 290, 308, 376
248, 282, 306, 362, 358
290, 306, 380
491 (0.51)
6200
238, 292, 384
490 (0.41)
5600
a Recorded in CH2Cl2. b Recorded in CHCl3 with perylene as a
standard.
Figure 2. Normalized absorption (solid line) and emission (dashed line)
spectra of thioacetate 5c (recorded in dichloromethane, T = 298 K).
Table 2. Oxidation Potentials of Thioacetates 3 and after Chemisorption
of Their In Situ Generated Thiols on the Gold Electrodesa
Thioacetatesb
Thiols after Chemisorptionc
0/þ1
þ1/þ2
0/þ1
E
1/2
[mV]
E
1/2
[mV]
E
E1pa
[mV]
E2pa
[mV]
E1pc
[mV]
1/2
compound
[mV]
3ad
3be
781
437
617f
599
562g
765
865
739
a Solvent, T = 20 °C, v = 100 mV/s, electrolyte: nBu4NþPF6-, Au
working electrode, Pt counter electrode, Ag/AgCl reference electrode.
b Normalized to the internal standard Fc/Fcþ in CH2Cl2 (E0
=
0/þ1
Figure 1. Normalized absorption (solid line) and emission (dashed line)
spectra of thioacetate 3b (recorded in dichloromethane, T=298 K).
450 mV).21 c Without normalization to Fc/Fcþ. d Recorded in THF.
eRecorded in CH2Cl2. f ΔEpa - Epc =34 mV. gΔEpa - Epc =17 mV.
X-ray Photoelectron Spectroscopy (XPS). X-ray photoelec-
tron spectroscopy was performed with a Leybold Max 200 spec-
trometer applying an Al KR X-ray source operated at 200 W and
a spherical sector-type analyzer. Survey spectra were acquired at
a constant pass energy of 96 eV, detail scans at 48 eV. The bin-
ding energy scale was referenced to the Au 4f7/2 peak at 84.0 eV.
For details of the spectra evaluation, we refer to the literature.23
established in our laboratories,19a,c,f and its application to
terminal phenothiazinyl alkynes 1 and diynes 4 with 1-(S-
acetylthio)-4-iodo benzene (2)22 causes the formation of
phenylethynyl (oligo)phenothiazinyl thioacetates 3 and 5
in moderate to good yields (see Scheme 1). The structures
of the thioacetates 3 and 5 are unambiguously supported
by 1H and 13C NMR, and IR spectroscopy, as well as mass
spectrometry and correct combustion analysis. The elec-
tronic properties of the phenylethynyl phenothiazinyl thio-
acetates 3 and 5 have been investigated by optical spec-
troscopy (see Table 1). UV/vis and selected fluorescence
Results and Discussion
Synthesis and Electronic Properties of the Phenylethynyl
Phenothiazinyl Thioacetates. The synthesis of alkynylated
phenothiazines by Sonogashira coupling was previously

56 Chem. Mater., Vol. 22, No. 1, 2010
Barkschat et al.
Figure 3. (a) Cyclic voltammogramof thioacetate3a in THF atv=100 mV/s (platinum electrode). (b) Cyclic voltammogramafter chemisorption ofthe in-
situ-formed thiols (prepared from c0(3a)=10-3 mol/L in THF by adding 3 mL of 0.1 M aqueous ammonium hydroxide solution in the presence of a Au
electrode and leaving the electrode in the solution for 30 min; the modified electrode then was rinsed with CH2Cl2, water, and THF) on a gold electrode at
v=100 mV/s. (c) Modified electrode at v=100 mV/s (curve I) and v=1000 mV/s (curve II) (gold electrode). (d) Cyclic voltammogram of the electrolyte
solution depicted in panels (a)-(c) after exchange of the 3a modified electrode against a platinum working electrode.
spectra reveal that absorbance occurs at the edge to the
UV edge to the visible and oligophenothiazines 3b and 5c
display considerable greenish-blue luminescence with
51% and 41% fluorescence quantum yield and large Stokes
shifts (Δυ~ = 5600-6200 cm-1) (see Figures 1 and 2).
As previously demonstrated for amine-terminated con-
jugated molecular wires,24 the chemisorption of the phe-
nothiazines 3 was first probed by semiquantitative cyclo-
voltammetric studies in the anodic region (scan area up to
1.0 V). Chemisorption experiments on a gold electrode
were directly carried out from argon degassed THF
solutions of the thioacetates of 3 by immersing the gold
electrode. Then, by in situ deprotection of the thioace-
tates with aqueous ammonia, the liberated thiol function-
alities were reacted with the electrode surface for 30 min.
After rinsing the electrode with dichloromethane, water,
and THF, the prepared electrodes were plunged into the
electrolyte solution of the cyclic voltammetry cell and the
cyclic voltammograms were recorded (see Table 2, as well
as Figures 3 and 4).
of unsymmetrical substitution of the dyad. After depro-
tection and chemisorption of the thiols on gold electro-
des, the reversibility of the first oxidations is maintained
showing the expected narrow differences of anodic and
cathodic peak potentials in the slightly distorted bell-
shaped cyclic voltammograms characteristic for elec-
trode-adsorbed redox-active molecules with Nernstian
behavior (see Figures 3b and 4b). This is also supported
by the linear correlation between peak currents and scan
rate (see Figure 3c).25 Upon increasing the scan rate from
100 mV/s to 1000 mV/s, an approximate 10-fold increase
of anodic and cathodic peak currents can be detected. The
gradual depression of the oxidative current peak poten-
tials (Figures 3b and 4b) converges to constant values that
are then maintained over multiple sweeps and also at two
different scan rates (see Figure 3c). Hence, electropoly-
merization of the chemisorbed species to conjugated
polymers can be excluded. Since perfect orientation of
the thiol substrates on the gold electrode surface cannot
be expected upon chemisorption, it appears that multiple
redox cycles, leading to constant peak currents, indicate a
post-chemisorption self-organization on the electrode
surface.
Thus, in multisweep experiments with a platinum
working electrode, the oxidations of acetates 3 to the
monocations and dications are fully reversible, in accor-
dance with Nernstian behavior (see Figures 3a and 4a).
All multisweep experiments (three cycles) were conducted
with an initial potential of 0 V and reversal potentials of
0.9-1.0 V. For acetate 3b, expectedly, two distinctly
separated, reversible oxidations can be found, because
After the multisweep experiments, the electrolyte solu-
tions were checked for eventually desorbed molecules.
The modified Au electrodes were replaced by a platinum
(25) Bard, A. J.; Faulkner, L. R. In Electrochemical Methods;Funda-
mentals and Applications, 2nd Edition; John Wiley and Sons: New York,
Chichester, Weinheim, Brisbane, Singapore, Toronto, 2001; Chapter 14.3,
p 589.
(24) Lambert, C.; Kriegisch, V. Langmuir 2006, 22, 8807–8812.

Article
Chem. Mater., Vol. 22, No. 1, 2010 57
Figure 4. (a) Cyclic voltammogram of thioacetate 3b in CH2Cl2 at v=100 mV/s (platinum electrode). (b) Cyclic voltammogram after chemisorption of the
in-situ-formed thiols (preparedfromc0(3b)=10-3 mol/L in THF byadding 3 mLof0.1 M aqueous ammonium hydroxidesolutionin the presence ofan Au
electrode and leaving the electrode in the solution for 30 min; the modified electrode then was rinsed with CH2Cl2, water, and THF) on an Au electrode at
v=100 mV/s. (c) Cyclic voltammogram of the electrolyte solution depicted in panels (a)-(c) after exchange of the 3b modified electrode against a platinum
working electrode.
˚
Table 3. Layer Thickness by Ellipsometry (A) and Contact Angle (°) for Ultrathin Layers of Compounds 3a, 5a, and 5b on Gold for Various Incubation Times
3a
5a
5b
˚
thickness [A]
˚
thickness [A]
˚
thickness [A]
incubation time [min]
contact angle [°]
contact angle [°]
contact angle [°]
10
30
60 (1 h)
12.7
12.2
15.2
15.4
13.6
15.6
79
87
84
90
88
89
11.9
11.9
14.6
17.8
18.3
19.2
74
74
76
75
73
73
18.5
18.2
20.8
23.5
24.3
27.1
69
77
76
76
74
75
300 (5 h)
1440 (24 h)
7200 (120 h)
working electrode and cyclic voltammetry of the electro-
lyte solutions was performed (see Figures 3d and 4c),
indicating that, within the applied voltage range and after
multiple oxidation and reduction cycles (see Figures 3b,
3c, and 4b), neither leaching from the electrode nor
degradation of the chemisorbed species can be found. In
contrast to pyrroles and thiophenes, phenothiazines are
quite stable, under the conditions of one-electron oxida-
tions,15 even if they are chemisorbed, as in the presented
cases. Therefore, (oligo)phenothiazines chemisorbed via
thiol ligation should be stable enough and should behave
favorably for the formation of ultrathin films of self-
assembled monolayers.
Preparation and Investigation of Sample Films of 3a, 5a,
and 5b on Gold. For preparing ultrathin films by self-
assembly, the thioacetates 3a, 5a, and 5b were subjected to
chemisorption experiments with suitable gold-coated si-
licon wafers to gain insight into the extent to which these
complex molecules are able to form ordered and dense SAMs,
which is a prerequisite for their practical application in
molecular electronics. The experiments were performed
Figure 5. Layer thicknesses of the films of compounds 3a, 5a, and 5b on
gold determined by ellipsometry, compared to calculated monolayer
thicknesses of 3a and 5a (dashed line) and 5b (solid line).
in the presence of aqueous ammonium hydroxide solu-
tion to promote the base-mediated saponification of the
thioacetates, thus leading to the in situ liberation of thiols

58 Chem. Mater., Vol. 22, No. 1, 2010
Barkschat et al.
Figure 6. Detail spectra of N 1s, C 1s, S 2p, and Au 4f of samples 3a, 5a, and 5b on gold.
Table 4. Peak Positions of the Detail Spectra of Samples 3a, 5a, and
5b on Gold
Table 5. Calculated Layer Thicknesses (by Attenuation of the Gold Signal)
and Stoichiometric Ratios C 1s/N 1s und C 1s/S 2p of Samples 3a, 5a, and
5b on Gold Determined by the Statistic and Layer Models
compound
N 1s
C 1s
S 2p3/2
Au 4f7/2
C 1s/S 2p
layer thickness by statistic layer
modela model modela model
C 1s/N 1s
3a
5a
5b
399.3
399.6
399.6
284.2
284.2
284.3
162.5
163.1
163.4
84.0
84.0
84.0
statistic layer
a
˚
XPS [A]
compound
3a
5a
5b
13.5 (20)
16.9 (20)
20.8 (26)
13.3 (13)
19 (11.5)
12.6 (9.3)
9.3 18.5 (26)
12.6 17.9 (23)
20
16
26.6
that can undergo the self-assembly on a gold surface. This
strategy allows the triggering of the chemisorptions pro-
cess and, thus, the SAM formation, which is another
interesting aspect of the systems of the present study, in
view of the formation of complex molecular electronics
structures.
9
32 (29)
a Calculated theoretical values are given in parentheses.
˚
thickness of 26 A, also assuming an orthogonal orienta-
tion of the molecules on the surface.
Ellipsometry (shown in Figure 5) indicates that all
representatives relatively quickly (300 min) form com-
plete films, which only slightly have a tendency to reor-
ganize with time (7200 min). The films of compound 5a
and 5b correspond to the calculated theoretical mono-
layer thicknesses with orthogonal orientation of the mole-
cules in good accordance. However, films of the N-hexyl-
Kinetic Measurements. Ultrathin layers of compounds
3a, 5a, or 5b on gold were prepared with various incuba-
tion times (10 min, 30 min, 60 min, 300 min, 1440 min,
7200 min) in a 0.1 mM THF solution. An estimation of
the film formation kinetics was then obtained by imme-
diately measuring the thickness of layers and their contact
angle (see Table 3) after removal of the samples from the
incubation solution and subsequent rinsing, as detailed in
the Experimental Section.
˚
substituted derivative 3a reveal a deviation by almost 5 A.
Presumably, the hexyl side chains hamper a close packing
within the layer. According to these results, all three mole-
cules give rise to monolayer formation without competing
multilayer generation.
The water contact angle of the samples almost remains
constant over time. With angles of 71° (5b) and 73° (5a),
good accordance with the literature data for aromatic
SAMs on gold is obtained.27 Sample 3a furnishes an angle
According to force field calculations, the intramolecu-
lar distances between the thiol S atoms and the distal-ring
˚
C atoms in compounds 3a and 5a are ∼16 A. Upon
˚
adding the Au-S bond length of 2 A, the bond length
˚
of the ethynyl moiety (1.2 A), and the extension of the
26
hexyl side chain (1.5 A), a monolayer of orthogonally
˚
˚
arranged molecules should be ∼20 A thick. For com-
pound 5b, the intramolecular distance between both
˚
terminal thiol S atoms calculates to 22 A, and addition
(27) (a) de Boer, B.; Meng, H.; Perepichka, D. F.; Zheng, J.; Frank,
M. M.; Chabal, Y. J.; (b) Bao, Zh. Langmuir 2003, 19, 4272–4284.
(c) Tai, Y.; Shaporenko, A.; Rong, H.-T.; Buck, M.; Eck, W.; Grunze,
M.; Zharnikov, M. J. Phys. Chem. B 2004, 108, 16806–16810.
(d) Kang, J. F.; Ulman, A.; Liao, S.; Jordan, R. Langmuir 2001, 17,
95–106.
of the Au-S and S-H bonds yields a calculated layer
(26) Ulman, A. In An Introduction to Ultrathin Organic Films; Academic
Press: Boston, 1991.

Article
Chem. Mater., Vol. 22, No. 1, 2010 59
Figure 7. Detail spectra of N 1s, C 1s, S 2p, and Au 4f of sample 5b on gold with the Voigt fit functions.
of 90° that can be rationalized by the hydrophobic char-
acter and influence of the N-hexyl side chain.
Voigt profiles. Since the spectra of all three molecules
turned out to be very similar, only the evaluated detail
spectra of sample 5b are depicted in Figure 7, including
their respective Voigt fits, as an example. As previously
reported for mercapto biphenyl spectra, satisfactory fit-
ting of the C 1s peak could be achieved only when five
individual Voigt profiles with different positions were assu-
med, because of its asymmetry at the higher-binding-
energy side.
The causes for this asymmetry are related to the struc-
ture of the main emission peak, which is broadened by
various chemical shifts, the varying attenuation of the
photoemission of the different C atoms, according to
their position within the layer, and the appearance of
so-called “shakeup signals” within the aromatic matrix.29
The layer thicknesses determined by this method yield
slightly smaller values, in comparison to the theoretically
calculated thicknesses and those determined by ellipso-
metry. This deviation could have its origin in the uncer-
tainty in determining the ellipsometry refractive indices
and/or the error in the mean free path length (λ) used in
XPS analysis.23 Despite these minor differences, the
stoichiometric C 1s/S 2p ratios of samples 5a and 5b
determined by the layer model fit very nicely with the
theoretically calculated values (see Table 5). The devia-
tion is larger if the statistic model is used. This better
agreement with the layer model indicates a well-ordered
and oriented alignment of the molecules. In contrast, for
sample 3a, the statistic model has obviously better applic-
ability, thus indicating an irregular distribution of the
molecules within the film.
Further information on coverage and chemical com-
position of the films was obtained by XPS of the samples
with incubation times of 7200 min. Expectedly, carbon,
sulfur, and nitrogen were detected, as shown by the detail
spectra (see Figure 6) and the corresponding peak posi-
tions (shown in Table 4). The bond energies were normal-
ized to the Au 4f7/2 peak at 84.0 eV. The absence of oxygen
indicates the successful deprotection of the thioester. The
peak positions of all three molecules are almost identical,
because of their similar chemical structure and, thus,
similar chemical shift of the respective moieties. The S 2p
peaks at 163 eV indicate thiolate formation.28
The Au 4f and C 1s intensities are indicative for
decreasing layer thicknesses from 5b over 5a to 3a. The
exact layer thicknesses were determined from the attenua-
tion of the Au signal (λ=35), which is due to the organic
overlayer, using both a statistic and a layer model for
calculation of the stoichiometric ratios of C 1s/S 2p and
C 1s/N 1s (see Table 5). In the statistic model, an irregular
distribution of the molecules within the layer is assumed,
which is synonymous with an arbitrary height distribu-
tion of the individual atoms comprising the molecules
within the film. Therefore, on average, all different types
of atoms within the molecule experience the same at-
tenuation of their photoemission. In contrast, the layer
model assumes a defined alignment and orientation of the
molecules, resulting in a well-defined height profile of the
film and, thus, different attenuation for atoms located in
different height levels. Therefore, based on the agreement
of the predictions made by these two models, e.g., in terms
of calculated stoichiometric ratios, evidence on the mo-
lecular structure of the film may be obtained. For such
quantification, the different emission peaks were fitted to
For the C 1s/N 1s ratios, the best-fitting results were
obtained for sample 5b. However, note that, because of
the low intensities of the S 2p and N 1s signals leading to
inaccuracies upon fitting, a significant error is made for
€
(28) Himmelhaus, M.; Gauss, I.; Buck, M.; Eisert, F.; Woll, C.; Grunze,
(29) Shaporenko, A.; Heister, K.; Ulman, A; Grunze, M.; Zharnikov,
M. J. Phys. Chem. B 2005, 109, 4096–4103.
M. J. Electron Spectrosc. Relat. Phenom. 1998, 92, 139–149.

60 Chem. Mater., Vol. 22, No. 1, 2010
Barkschat et al.
the calculation of these stoichiometric ratios. Therefore,
these data should rather be considered as a qualitative
affirmation of the film formation than a precise way to
determine the exact packing density of the molecules.
According to ellipsometry, contact angle, and XPS data,
all three molecules form monolayers on gold relatively
quickly (within 300 min (or 5 h)). Still, after 7200 min
(120 h), sample 3a has a lower layer thickness, indicating a
lower degree of coverage. The undefined nature of the
layer is also supported by the C 1s/S 2p ratio and could be
caused by the disturbing effect of the hexyl side chain.
The aforementioned results indicate the formation of
well-ordered monolayers, in particular, for the moieties
5a and 5b, the structure of which thus may be further
analyzed by IRRAS to obtain insight into the average
molecular orientation. This can be achieved by taking
advantage of the dipole selection rule for IR spectroscopy
on metal surfaces, which states that an IR-allowed reso-
nance is only observable in IRRAS if it is not oriented
parallel to the metal surface. Accordingly, depending on
the appearance or disappearance of characteristic group
resonances, the average orientation of the adsorbed
molecules can be assessed.
Figure 8 displays IRRA spectra of SAMs of 3a, 5a, and
5b, prepared on gold wafer pieces and immersed into the
respective solution for 120 h. In Figures 8a and 8b, the
characteristic resonances of the terminal alkynyl group of
5a are shown, which indicate;because of their mere
presence;that the molecules in the film formed by 5a
are oriented toward the surface normal. This observation,
which is in agreement with the results obtained by ellip-
sometry and XPS, is further corroborated by the finger-
print spectra of the SAMs, as shown in Figure 6c. Except
for the highest and lowest modes, indicated at 1577 and
821 cm-1, respectively, all other observable peaks can be
attributed to molecular resonances with an orientation of
their transition dipole moments parallel to the molecular
main axis (for mode assignments, see Table S1 in the
Supporting Information). This actually validates that the
molecular main axis of the molecules must have an orien-
tation very close to the surface normal, which further
suggests a high packing density within the monolayer. In
fact, the sharpness of the peaks and the band intensity of
the features observable in the spectrum of 5a indicates
that this film is comprised of the highest degree of order of
all three molecules studied, and, therefore, also the high-
est molecular density. Otherwise, i.e., in terms of resonant
features, the spectra obtained from 5a and 5b look
basically the same. The spectrum of 3a shows fewer
features, because of the higher symmetry of its phenothia-
zine system, which increases the number of IR-forbidden
vibrational modes. The lower intensity observed here is
consistent with a lower degree of order, which increases
the average tilt angle and, thus, increases the not-obser-
vable contribution of the transition dipole moments in a
direction parallel to the surface.
Figure 8. Reflection absorption infrared spectra of SAMs of 3a, 5a, and
5b, respectively, on gold wafer pieces after 120 h of immersion into the
respective solution: (a) terminal alkynyl CH valence vibration of a SAM
formed by 5a; (b) terminal alkynyl CC valence vibration of the SAM
shown in panel (a); (c) region of the CH stretching vibrations. The spectra
in panel (c) are vertically displaced for clarity; the indicated mode
positions were averaged over those obtained from all three molecules.
For the individual mode positions, as well as further details, we refer to
the Supporting Information. Abbreviations: rg=ring, ν=stretching, β=
bending, δ=deformation, s=symmetric, as=asymmetric.
of the molecular main axis close to the surface normal. For a
more-detailed analysis of the IRRAS data, including the
study of the film formation kinetics, we refer to the Support-
ing Information.
Investigation of the Density of Sample Films. The afore-
mentioned surface analysis clearly shows that 5a, 5b, and
3a form self-assembled monolayers on gold, which are;
particularly for 5a and 5b;well-ordered, given the struc-
tural complexity of the molecules. In particular, it can be
excluded that the molecules form multilayers. However,
the density of the packing that is achieved has not yet
become clear, i.e. whether it corresponds to full coverage,
where each surface site is occupied by a molecule, or
whether film formation levels off at lower coverage. In
particular, for the proposed application as molecular
switches, this is an important question, because any
unoccupied surface site might lead to electrical bypasses
in a later application.
Altogether, the IR spectra are in good agreement with
the results of ellipsometry and XPS, in that, particularly,
the SAMs of 5a and 5b are well-ordered, with an orientation

Article
Chem. Mater., Vol. 22, No. 1, 2010 61
Figure 9. (a) Change of the layer thicknesses of samples 3a, 5a, and 5b with increasing incubation time prior to and after the post-adsorption of
trifluoroethane thiol. (b) F 1s signal in the XPS spectra of samples 3a, 5a, and 5b after the adsorption of trifluoroethane thiol. (c) Relative peak area of the
F 1s peaks of samples of 3a, 5a, and 5b normalized to that of a TFET SAM on gold, in dependence of the incubation time in the respective phenothiazine
derivative solution. The TFET post-adsorption time was kept fixed at 24 h.
Therefore, the density of the SAMs of 3a, 5a, and 5b
was determined by performing post-adsorption experi-
ments with trifluoroethane thiol (TFET). TFET was chosen
because the molecule is small and, thus, may diffuse easily
into the phenothiazine film and, furthermore, can be easily
detected and quantified via XPS, because of its fluorine
moieties. Thus, after preparation of SAMs of 3a, 5a, and
5b on gold as previously described (1-120 h of incu-
bation), the samples were placed into a 1 mM solution of
TFET in ethanol for 24 h. After rinsing with ethanol, the
samples then were investigated by ellipsometry, contact
angle, and XPS measurements.
The layer thicknesses of samples 5a and 5b at shorter
incubation times are slightly increased after post-adsorp-
tion of TFET (Figure 9a), indicating that the native films
still had defects. This causes an increase of the average
layer thickness, as revealed by ellipsometry. At incuba-
tion times of 7200 min (120 h), ellipsometry shows hardly
any increase in layer thicknesses caused by additional
TFET adsorption, thereby once more indicating the
formation of well-ordered SAMs.
In contrast, for sample 3a, significantly different beha-
vior is found. As already observed previously (cf. Figure 5),
the native films become even slightly thinner for long incu-
bation times (>5 h). This decrease of the layer thickness
could be rationalized by the disturbing effect of the hexyl
side chains during the formation of the native film, cau-
sing a higher degree of structural defects. Therefore, dur-
ing the post-adsorption of TFET, the molecules have more
space to adjust and incline, which then may lead to the
deliberation of adsorption sites for TFET. Therefore, in
particular, for long incubation times, the post-adsorption
experiment gives a significant increase in layer thickness
due to TFET adsorption. This finding is in accordance
with the poor SAM formation behavior of 3a, as already
revealed by the surface analysis.
Interestingly, the post-adsorption does not significan-
tly change the contact angles (data not shown) for any of
the three molecules, most likely because of the small size
of the molecule, which is probably buried underneath the
phenothiazine layer.
The exact amount of adsorbed TFET was determined
from the intensity of the XPS F 1s peak, relative to that
obtained from a complete monolayer of TFET on gold
(see Figure 9b). Expectedly, the F 1s peak is observed at
686.8 eV. For sample 5b the F 1s signal is fairly weak
(6.9%). The film for sample 5a has adsorbed more TFET
(18.5%), and, finally, the film for sample 3a contains a
significant amount of fluorine (48.2%).
In Figure 9c, the relative peak area of the XPS F 1s
signal of the three samples is plotted as a function of the
incubation time of the substrates into the respective
phenothiazine derivative solutions. For the films of sam-
ple 5b, the amount of fluorine is very low, in comparison
to that of the other two molecules, and even decreases
with increasing incubation time, thus indicating almost-
complete shielding of the surface, because of increased
coverage and better ordering of this phenothiazine deri-
vative, in corroboration of the surface analysis previously
described. The films for sample 5a show higher TFET
adsorption, with only little dependence on the incubation
time (the 300-min data point is apparently an outlier).

62 Chem. Mater., Vol. 22, No. 1, 2010
Chart 1. SAM of Thiolated Phenylethynyl Phenothiazines on Au{111}
Barkschat et al.
While the overall surface analysis of sample 5b gave
evidence for a well-ordered monolayer, the ellipsometry
data (cf. Figure 5) in fact revealed saturation coverage
slightly below the theoretical limit, which may explain the
somewhat-higher TFET adsorption currently observed.
In contrast to these basically promising results, the films
for sample 3a exhibit a significant increase in the fluorine
content with increasing incubation time, thereby giving,
once more, evidence for the poor monolayer formation of
this molecule. Presumably, the 3a molecules reorganize
with increasing incubation time in such a way that even
more adsorption places for TFET are generated than in
the initial phase of film formation. A corresponding mole-
cular rearrangement can also be observed in the IRRA
spectra, in particular, for the region of the CH stretching
vibrations (vide supra). Because the post-adsorption time
for TFET was kept constant at 24 h in all cases, this
scenario is more likely than desorption caused by TFET.
on gold are consistent with literature values. For the layers
of 3a on gold, the higher contact angle of 90° can be
attributed to the hydrophobic influence of the hexyl side
chain. The stoichiometric ratios C 1s/S 2p of samples 5a and
5b on gold determined by XPS match very well with
calculated data when a layer model is applied, thus yielding
evidence for well-ordered layers with a well-defined orienta-
tion and alignment of the molecules. In contrast, the
samples of 3a on gold are better-described by a statistic
model with an irregular orientation of the molecules on the
surface.
In the IRRA spectra of sample 5a on gold, the bands
clearly indicate that the molecules are oriented close to the
gold surface normal. The same bands are found in the
spectra of sample 5b on gold, and, in combination with
the layer thicknesses determined by ellipsometry, evi-
dence is given that this molecule is bound to the gold
surface with only one S atom and that it is oriented
orthogonally. In the CH-region of the spectra of sample
3a on gold, the hexyl chain is clearly identified by four
distinct aliphatic CH stretching vibrations. With increas-
ing incubation times, these bands are diminishing, as a
consequence of molecular reorientation with immersion
of the hexyl chains into the phenothiazine layer and/or
desorption of the 3a molecules.
Post-adsorption studies with trifluoroethanthiol (TFET)
show that the layers of sample 5a form almost-perfect
SAMs on gold (see Chart 1), where only 6.9% of the fluo-
rine amount is post-adsorbed. The films of 5a and 3a
insert fluorine amounts of 18.5% and even 48.2%. Hence,
compound 3a does not form close-packed films on gold,
as a consequence of the disturbing hexyl side chain. The
higher quality of the films of compound 5b on gold is
supposedly a consequence of the presence of two thiol
groups, which increases the adsorption probability, de-
spite the bulkiness of the molecule. This is a very inter-
esting observation, since, a priori, it might have been
expected that the molecule adsorbs with both thiol groups
on the gold surface, thereby bridging between two differ-
ent adsorption sites. The fact that this is not the case sheds
light on the importance of the Au-S-C angle for SAM
structure.30 In addition, with the free thiol groups on top
Conclusions
Sonogashira coupling of alkynyl (oligo)phenothiazines
with 1-(S-acetylthio)-4-iodo benzene furnishes phenylethy-
nyl phenothiazinyl thioacetates, which is a class of redox-
active, fluorescent π-systems with acetyl-protected “alliga-
tor clips” for chemisorption to conductive gold surfaces.
These thioacetates show reversible one-electron oxidations
at relatively low oxidation potentials. The reversible oxida-
tion behavior is maintained after in situ deprotection,
leading to chemisorption on the electrode surface. After
multisweep experiments, no traces of desorbed molecules
canbedetectedintheelectrolyte, indicating the formation of
stable adsorbates. Furthermore, the oligomeric phenothia-
zinyl representatives reveal considerable greenish-blue lum-
inescence with substantial fluorescence quantum yields and
large Stokes shifts. Adsorption kinetics and structures of the
formed monolayers of three 4-mercapto phenylalkynyl
phenothiazinyl molecules on gold were investigated using
several techniques. Ellipsometry shows that all molecules
are forming films on gold within 300 min without multilayer
formation. The final coverage of samples 5a and 5b on gold
are in very good agreement with the calculated layer
thicknesses, assuming an orthogonal orientation of the
molecules on the surface. The sample film of compound
˚
3a reveals a deviation of almost 5 A from the calculated
values. The water contact angles of the layers of 5a and 5b
(30) Rong, H. T.; Frey, S.; Yang, Y. J.; Zharnikov, M.; Buck, M.;
Wuhn, M.; Woll, C.; Helmchen, G. Langmuir 2001, 17, 1582–1593.

Article
Chem. Mater., Vol. 22, No. 1, 2010 63
of the monolayer, SAMs of 5b promise to be ideal
candidates for the formation of intermetal junctions
and, thus, molecular diode fabrication.
can be very useful, for example, for coadsorption of the
moieties with a second, nonconductive molecule, which
shall serve as insulating matrix. Further studies directed
toward such more-complex oligophenothiazine SAMs on
gold and functionalized redox manipulable surfaces are
currently underway.
In summary, our study reveals that phenothiazines with
thioacetate functionalities are well-suited for SAM for-
mation on gold electrodes and surfaces and, thus, are
excellent candidate systems for the buildup of complex
redox switchable ultrathin films for molecular electronics
applications. Besides the high degree of order achieved,
which warrants well-defined electronic and optical prop-
erties of the structures formed, one interesting aspect of
the systems studied is related to the possibility of trigger-
ing the chemisorption process by the addition of ammo-
nium hydroxide. This chemical trigger enables better
control of film formation and adsorption kinetics, which
Acknowledgment. This project was supported by the DFG
(SFB 486) and the Fonds der Chemischen Industrie. S.S. and
M.H. acknowledge gratefully partial support by the DFG
under Hi 693/2-2.
Supporting Information Available: Analysis of the structure of
SAMs formed from 3a, 5a, and 5b by IRRAS. (PDF) This
information is available free of charge via the Internet at http://
pubs.acs.org/.