preparation and photocatalytic activity of carbon- modified titania

Preparation and photocatalytic activity of carbon-

modified titania

Herstellung und photokatalytische Aktivität von

Kohlenstoff-modifiziertem Titandioxid

Den Naturwissenschaftlichen Fakultätender Friedrich-Alexander-Universität Erlangen-Nürnberg

zurErlangung des Doktorgrades

vorgelegt vonPrzemysᐠaw Zၐbekaus Krak㳀w (Polen)

Als Dissertation genehmigt von den Naturwissenschaftlichen Fakultäten der Universität Erlangen-Nürnberg

Tag der mündlichen Prüfung: 07. 05. 2010

Vorsitzender der Promotionskommission: Prof. Dr. Eberhard Bänsch

Erstberichterstatter: Prof. Dr. Horst Kisch

Zweitberichterstatter: Prof. Dr. Franziska Gröhn

1

Acknowledgements

I wish to thank Prof. Dr. Horst Kisch for the supervision of this work and many invaluable discussions. I am particularly grateful for his advice and his generous support of my work.

Parts of this work would not have been possible without the help of several people.

Dr. Dariusz Mitoraj, Dr. Radim Beránek, and Dr. Francesco Parrino are acknowledgedfor many valuable discussions. I thank also Christina Wronna for elemental analyses, Susanne Hoffman for XRD measurements, Siegfried Smolny for surface area measurements, and Helga Hildebrand for XPS measurements. Manfred Weller, Peter Igel and their co-workers from the “Werkstatt” are acknowledged for the assistance with technical problems. Various kinds of wood shavings were generously supplied by Holzwelt Murau and Das Holzmuseum, Murau (Austria). I am also indebted to Dr. Matthias Moll for his manifold help, Uwe Reißer for his assistance with electronic equipment, Ronny Wiefel for glass work, and to Dr. Jörg Sutter for computer assistance. I would like to thank Dr. Andreas Scheurer for helping in the translation of Chapter 10 of this thesis.

Many thanks to all my colleagues for contributing to the very good atmosphere in the group – Dariusz, Francesco, Radim, Joachim, and Sina.

I am very grateful to my parents for their lifelong love and encouragement, and I dedicate this work to them and also to Stefania, without her this work would not have been possible.

2

Die vorliegende Arbeit entstand in der Zeit von September 2006 bis März 2010 am

Department Chemie und Pharmazie der Universität Erlangen-Nürnberg unter

Anleitung von Herrn Prof. Dr. Horst Kisch.

Pracᆐ tᆐ dedykujᆐ rodzicom w 30-tၐ rocznicᆐ ᖰlubu, kt㳀ry odbyᐠ siᆐ 5 kwietnia 1980 roku.

3

Content

Acknowledgements .......................................................................................... 1Content ............................................................................................................. 3Symbols and Abbreviations............................................................................. 51. Fundamentals of semiconductor photocatalysis ...................................... 72. Previous work.......................................................................................... 11

2.1 “Carbon” modification of titania..................................................................112.1.1 Carbidic species ............................................................................................122.1.2 Carbonates ...................................................................................................122.1.3 Amorphous carbon ........................................................................................15

2.1.3.1 Aromatic hydrocarbons, soot and coke-like species..............................................162.1.4 Unusual carbon species .................................................................................172.1.5 Colour centers ...............................................................................................172.1.6 Theoretical calculations on the nature of carbon species ................................17

2.2 Goals of the work ........................................................................................19

3. General experimental part...................................................................... 203.1 Techniques ..................................................................................................203.2 Preparation of TiO2 .....................................................................................213.3 Photodegradation/Photomineralization of 4-CP and other pollutants ...........22

4. Graphite and graphite oxides as modifiers ............................................ 244.1 Introduction.................................................................................................244.2 Experimental ...............................................................................................25

4.2.1 Graphite oxidation methods...........................................................................254.2.2 Preparation of modified titania and photocatalytic tests .................................26

4.3 Results and discussion .................................................................................284.3.1 Oxidation of graphite.....................................................................................284.3.2 Titania modification by graphite and graphite oxides.....................................32

5. Organic acids as modifiers...................................................................... 345.1 Introduction.................................................................................................345.2 Experimental ...............................................................................................35

5.2.1 Preparation of TiO2 .......................................................................................355.2.2 Preparation of titania modified with benzoic acid ..........................................355.2.3 Preparation of titania modified with phthalic acid..........................................355.2.4 Preparation of titania modified with trimellitic acid .......................................365.2.5 Preparation of titania modified with shikimic acid .........................................365.2.6 Photodegradation of 4-CP .............................................................................38

5.3 Results and discussion .................................................................................385.3.1 Benzoic acid..................................................................................................385.3.2 Phthalic acid..................................................................................................385.3.3 Trimellitic acid..............................................................................................40

4

5.3.4 Shikimic acid ................................................................................................42

6. Modification of titania with toluene ........................................................436.1 Introduction ................................................................................................ 436.2 Experimental............................................................................................... 43

6.2.1 Preparation of TiO2 .......................................................................................436.2.2 Preparation of TiO2 modified with toluene ....................................................436.2.3 Photomineralization of 4-CP .........................................................................44

6.3 Results and discussion ................................................................................ 44

7. Commercial carbon-modified titania......................................................457.1 Introduction ................................................................................................ 457.2 Experimental............................................................................................... 467.3 Results and discussion ................................................................................ 487.4 Conclusions ................................................................................................ 57

8. Polyol derived carbon-modified titania...................................................598.1 Introduction ................................................................................................ 598.2 Experimental............................................................................................... 608.3 Results and discussion ................................................................................ 65

8.3.1 Preparation and characterization of photocatalysts.........................................658.3.2 Photomineralization of 4-CP .........................................................................688.3.3 On the nature of carbon sensitizer..................................................................698.3.4 Alkaline stability at room temperature...........................................................73

8.4 Conclusions ................................................................................................ 74

9. Wood shavings as renewable modifiers ..................................................759.1 Introduction ................................................................................................ 759.2 Experimental............................................................................................... 759.3 Results and Discussion................................................................................ 77

9.3.1 Preparation and characterization of photocatalysts.........................................779.3.2 Photomineralization of 4-CP .........................................................................81

9.4 Conclusions ................................................................................................ 83

10. Summary ...............................................................................................8410. Zusammenfassung ................................................................................8711. References .............................................................................................90

Symbols and Abbreviations

5


A electron acceptor ATR attenuated total reflection spectroscopyBET Brunauer-Emmett-TellerCB conduction bandD electron donor DFT density functional theoryDRS diffuse reflectance spectroscopyEC B conduction band edge

nEF* quasi-Fermi level of electrons

pEF* quasi-Fermi level of holesEbg bandgap energyEG ethylene glycolEPR electron paramagnetic resonanceEVB valence band edgee– electrone r

– reactive electronF(R∞) Kubelka-Munk function

Φp quantum yield of the product formationh+ holeh r

+ reactive hole

hν energy of lightIa intensity of lightIR infrared spectroscopyIFET interfacial electron transfer

λ wavelengthLED light-emitting diodMOCVD metal-organic chemical vapour depositionNHE normal hydrogen electrodePE pentaerythritolR∞ diffuse reflectance of the sample relative to the reflectance of a standardSEM scanning electron microscopy


6

TBAH tetrabutylammonium hydroxideTEM transmission electron spectroscopyTiO2-C carbon-modified titaniaTOC total organic carbonVB valence bandVis visible lightXPS X-ray photoelectron spectroscopyXRD X-ray diffractometry

1. Fundamentals of semiconductor photocatalysis

7


At the beginning of this dissertation the definitions of photocatalysis,

photocatalyst, and photosensitizer will be given. The properties of TiO2 will be also

briefly discussed. Finally, the mechanism of semiconductor photocatalysis including

kinetic aspects will be presented.

Photocatalysis is a change in the rate of chemical reactions or their generation

under the action of light in the presence of substrates – called photocatalysts – that

absorb light quanta and are involved in the chemical transformations of the reaction

participants. And a photocatalyst is a substance that is able to produce, by absorption

of light quanta, chemical transformations of the reaction participants. Photosensitizer

is an agent that absorbs light and subsequently initiates a photochemical or

photophysical alteration in the system, the agent being not consumed therewith. In

case of chemical alteration, the photosensitizer is usually identical to

a photocatalyst.∗[1, 2]

Titanium dioxide, as a typical semiconductor photocatalyst, exhibits two energy

bands: the highest occupied energy band (filled with electrons, so-called valence

band), and the lowest unoccupied energy band (so-called conduction band). These

energy bands determine the optical properties of a semiconductor. It is noteworthy,

that the conduction band edge states have predominantly Ti 3d character, while the

valence band edge states have the O 2p character. The energy difference between the

lower edge of the conduction band and the higher edge of the valence band is defined

as bandgap energy (3.2 eV for TiO2). The generation of electron-hole pairs in the band

structure of a semiconductor occurs if it is irradiated by light with energy equal or

greater than the bandgap energy. If a semiconductor (here: TiO2) is modified with

carbon as in this dissertation, the charge carriers can be generated by light with energy

smaller than the bandgap energy (Ebg < 3.2 eV, 㮰 ≥ 390 nm), due to the presence of so-

called surface states arising from carbon modification.

∗ The definitions of photocatalysis, photocatalysts, and photosensitizer come from the Glossary of Terms in Photocatalysis and Radiation Catalysis, IUPAC Congress/General Assembly, July 2001


8

The primary processes occurring at a semiconductor particle in heterogeneous

photocatalysis are presented below in Figure 1.1.

Figure 1.1: Simplified representation of the key processes of a photocatalytic reaction at a semiconductor particle. For details see the text. Adopted from ref.[3]

Process 1 illustrates the generation of an electron-hole pair upon photon

absorption. Thereafter, an electron located in the conduction band and a hole located in

the valence band can either undergo primary recombination (process 2), or be trapped

at reactive surface sites (process 3). As reported for TiO2, electrons are trapped in the

form of TiIII centers and holes as surface-bound hydroxyl radicals {≡TiIVOH•}+.[4]

Trapping of holes requires 10-100 ns, while this process is faster for electrons and

proceeds in some hundreds of picoseconds.[4] The trapped charge carriers can again

either recombine (secondary recombination, process 4), or undergo an interfacial

electron transfer, whereby the electron reduces an electron acceptor species A to

a primary reduction product A– • (process 5), and the hole oxidizes an electron donor

species D to a primary oxidation product D+• (process 6). To avoid primary (9) and

secondary (i.e., oxidation of A– • by a reactive hole or reduction of D+• by a reactive

electron) back-electron transfer, A– • and D+• must then undergo a rapid conversion to

the final reduction (Ared) and oxidation (Dox) products (processes 7 and 8). In a typical

photocatalytic oxidation of organic water pollutants on TiO2 the reacting holes are

scavenged either directly by the pollutant or by adsorbed hydroxyl ions, what results in

the formation of hydroxyl radicals, which can further oxidize the pollutant due to their


9

high oxidizing power. The photogenerated electrons reduce molecular oxygen to

a superoxide radical, which can further react to produce hydroxyl radicals via

following reactions:[4-8]

O2 + eCB– → O2

•– (1.1)

O2•– + H+ → HO2

• (1.2)

HO2• + HO2

• → H2O2 + O2 (1.3)

O2•– + HO2

• → O2 + HO2– (1.4)

HO2– + H+ → H2O2 (1.5)

H2O2 + O2•– → •OH + OH– (1.6)

H2O2 + eCB– → •OH + OH– + O2 (1.7)

As produced hydroxyl radicals can again be involved in the mineralization of the

pollutants, like 4-chlorophenol (4-CP) used in this dissertation. This compound

represents a class of organic substances, which contaminate natural environment as

a result of human activities such as water disinfection, uncontrolled applications of

pesticides and herbicides, etc. Halogenated hydrocarbons provide serious problems

due to their biological accumulation. More dangerous halogenated substances are

pentachlorophenols (PCP), polychlorinated biphenyls (PCB) or polychlorinated

dibenzofurans (PCDF). 4-CP was chosen for experiments in this work because its

degradation can be easily followed by UV-Vis spectroscopy or TOC measurements

(See also Chapter 3.3). The general equation of 4-CP mineralization in the presence of

TiO2 can be written as follows:

2 p-ClC6H4OH + 13 O2 → 12 CO2 + 2 H+ + 2 Cl– + 4 H2O (1.8)

According to the primary processes presented in Figure 1.1 it is recalled that the

rate of a photochemical reaction is defined by the quantum yield of the product

formation Φp and absorbed light intensity Ia (Eq. 1.9).

apIrate Φ= (1.9)


10

The quantum yield can then be expressed by the efficiencies of light induced

charged separation of the reactive electron-hole pairs − er−, hr

+ (ηcs), IFET processes

(ηIFET) and the final product formation (ηp) (Eq. 2.0).

pIFETcsp ηηη=Φ (2.0)

2. Previous work

11

2. Previous work

2.1 “Carbon” modification of titania

Titanium dioxide has received great attention both in fundamental and applied

photocatalysis due to its low cost, non-toxicity, and stability against photocorrosion.[4,

9-13] Unfortunately, it can utilize only the very small UV part (about 3%) of solar light

arriving at the earth surface. However, also the visible part (㮰 > 400 nm) may induce

photocatalysis if titania is modified by transition[14-26] or main group elements. Out of

the latter, especially nitrogen[27-37] and carbon were reported to give the most active

photocatalysts. In this dissertation carbon modification of titania, often referred to as

“carbon-doping” (abbreviated later as “C-doping” or “C-doped” for carbon-doped

materials) will be presented. Up to date, various modification procedures employing

diverse carbon compounds have been reported in the literature. In most cases they

exist of thermal treatment of titania or titania precursors with an organic carbon

compound at 250-600°C in the presence of air. Typical carbon precursors employed

are alcohols[38-40], sugars[41-44], adamantane[45], tetrabutylammonium hydroxide[46],

n-hexane[47], 2,4-pentanedione[48], oxalic acid[49], the alkoxide groups of titanium

alcoholates[50-54], and even carbonic ink.[55] Another method is oxidative annealing of

titanium carbide at about 600°C.[56-60] All these so called “C-doped” titania materials

exhibit a weak absorption shoulder between 400 and 800 nm, the intensity of which

increases with increasing carbon content.[46, 54] Most of them are active in visible light

photo-oxidations of various organic pollutants. In addition to 4-chlorophenol[41, 61]

employed in this work, also isopropanol,[56] gaseous benzene,[39] and nitrogen

oxides[40, 45, 50] were photo-oxidized by visible light irradiation of “C-doped” titania.

However, in all cases the nature of the carbon species in these TiO2-C powders,

responsible for Vis light photooxidation of various pollutants, could not be unravelled.

Proposed species from carbide[38, 45, 56-60, 62], carbonates[38, 39, 49-53, 60], and aromatic

hydrocarbons[41, 43, 47, 49, 51, 54] to oxygen vacancies[63, 64], were suggested on the basis of

C1s XPS binding energies. XPS seems to be one of the most common method used by

many scientists to identify relevant carbon species. Thus, further details about

proposed carbon species on the basis of this spectroscopy are summarized in Chapter

2. Previous work

12

7.1, and ref.[65], and extended in Chapters 2.1.1, 2.1.2, and 2.1.3. Also other

experimental techniques i.a. EPR (See Chapter 2.1.2), SEM and TEM (See Chapter

2.1.4) employed for determining the nature of carbon species will be mentioned in the

further part of this introduction.

2.1.1 Carbidic species

The work of Irie et al.[56] presents “C-doped” anatase powders prepared by

annealing of commercial TiC under flow of O2 at 600°C. C1s XPS measurements

revealed a peak at 281.8 eV assigned to Ti-C bond and two peaks at 285 and 276 eV

corresponded to kind of elemental carbon. The reference TiO2 did not show a peak at

281.8 eV. The modified material exhibited activity in the decomposition of

isopropanol under Vis light (400-530 nm).

In the work of Gu et al.[60] the micro-mesoporous “C-doped” TiO2 was prepared by

a special low-temperature procedure, wherein at the final step the powders were

calcined at 120°C for 48 h. The C1s XPS revealed of three peaks at 281.9, 284.8 and

288.6 eV. The first one was assigned to carbidic species. Two latter ones were

suggested as an adventitious carbon and carbonates, respectively. This “C-doped”

TiO2 showed high activity in the decomposition of methylene blue (㮰 > 420 nm), as

compared with reference P25.

Also Choi et al.[59] prepared “C-doped” anatase materials by oxidative annealing of

TiC powders at 350°C. From XRD and C1s binding energy of 281.9 eV it was

concluded that carbon is incorporated into titania lattice in the place of oxygen. The

decomposition of methylene blue was observed under Vis light irradiation (420-500

nm).

2.1.2 Carbonates

One of the first papers presenting evidence for carbonates was published by

Sakthivel et al.[46] C1s binding energies of 285.6, 287.5 and 288.5 eV were found for

the sample as obtained by hydrolysis of titanium tetrachloride with

tetrabutylammonium hydroxide as a carbon source, followed by calcination at 400°C

2. Previous work

13

for 1 h (TiO2-C1b). The first value arises from adventitious elemental carbon; the latter

two values suggest the presence of carbonate species. Contrary to this, in an undoped

sample no carbon peaks were detected. The IR spectrum of TiO2-C1b exhibited low-

intense peaks at 1738, 1096, and 798 cm−1 characteristic for carbonate ion, what

supports the previous assumption. TiO2-C1b induced complete mineralization of 4-CP

after 3 h irradiation in Vis light (㮰 ≥ 455 nm). Also oxidation of gaseous acetaldehyde,

benzene, and carbon monoxide occurs. Carbon modification generates surface states –

localized intrabandgap energy levels. Their approximate energy was deduced from the

wavelength dependence of OH-radical formation. For details see ref.[46] On the basis

of this experiment it was conluded that the surface states are localized close to the

valence band edge. It is noteworthy, that such mechanistical investigations are not

commonly performed.

In subsequent work of Sakthivel et al.[31] it was noted that carbonates are not

responsible for visible light activity since they were also present in unmodified titania

prepared from titanium tetrachloride and sodium carbonate. This material was not

photoactive upon irradiation at 㮰 ≥ 455 nm.

Next work of Li et al.[39] reported about “C-doped” titania prepared by

temperature-programmed carbonization of anatase titania in a flow of argon saturated

by cyclohexane as an organic modifier in temperature range of 450-500°C. C1s

binding energies were formed at 284.6 and 288.2 eV. The former arises from an

adventitious elemental carbon and was present also in pure titania. The latter one was

ascribed to carbonate species according to results of Sakthivel et al.[46]

Tseng et al.[50, 51] prepared visible light responsive nano-TiO2 by hydrolysis of

tetrabutyl orthotitanate, followed by calcination at high temperatures for 5 h. C1s

binding energies peaks were in the range of 282-292 eV and ascribed to C-C, C-O and

O=C-O species. For the samples calcined at 400°C and higher temperatures coke-like

species as playing the role of sensitizer were postulated. XRD revealed about mixed

anatase and brookite phase for sample prepared at 500°C and complete rutile phase for

other one calcined at 600°C. The best material prepared at 200°C exhibited ca. 80% of

2. Previous work

14

NOx removal under UV, blue, green and red light emitted by diods (LED). Contrary to

this, commercial TiO2 (UV100) possessed very little visible light reactivity.

Kang et al.[38] prepared “C-doped” material by ball miling of pristine TiO2 with

ethanol as a carbon source, followed by heating at 200 or 400°C. C1s binding energies

of 282 and 288.6 eV were assigned to carbidic, and carbonate species, respectively.

The product heated at 400°C does not exhibit Ti-C bonds and its photocatalytic

activity in decomposition of NOx is lower (ca. 12%, 㮰 ≥ 510 nm) than that of the

sample obtained at 200°C (ca. 25%, 㮰 ≥ 510 nm). The authors emphasize the

contribution of C-O (carbonate species) and Ti-C groups for visible light activity.

They further report that heating TiO2 with various carbon compounds usually leads to

relatively thermal stable C-O bonds. Contrary to this, oxidative annealing of TiC

materials does not generate C-O bonds.

Chen et al.[66] prepared his photocatalysts by hydrolysis of titanium tetra-n-butylate

with tetrabutylammonium hydroxide as a carbon source, followed by calcination at

high temperatures. C1s binding energies were found at 282.5, 286.5, and 288.4 eV.

The first peak was assigned to “active carbon” originating from residual TBAH, and

the other two suggest the presence of C-O and C=O groups, probably carbonate

species. A peak around 281 eV arising from a Ti-C bond was not detected. The authors

speculated that the doped carbon forms a layer on the surface of TiO2 nanoparticles

with a mixture of active carbon and carbonate species. The photocatalytic activity was

evaluated by measuring the decomposition rate of methylene blue under Vis light

irradiation (㮰 ≥ 420 nm). Whereas 26% of methylene blue decomposed in the presence

of pure TiO2, even 70% was noticed for modified sample.

One of the latest works of Reyes-Garcia et al.[49] reported about a brown “C-TiO2”

anatase material, whose preparation can be described as follows: titanium

isopropoxide suspended in a solution of ethylenediamine was mixed with an aqueous

solution of oxalic acid (organic modifier), followed by amine removal, and finally

calcination at 400-500°C. 13C-NMR spectroscopy revealed the presence of carbonate,

bicarbonate or even polycarbonate species. Also carbon species corresponding to sp2

hybridization like in graphitic materials were detected. EPR spectroscopy revealed

2. Previous work

15

paramagnetic signals (at g = 1.93 and 1.99) arising from point defects such as Ti3+

(= an electron trapped at an oxygen vacancy) and oxygen anion radicals (at g = 2.02).

C1s XPS peaks of the sample calcined at 400°C appeared at 285.9 and 288.9 eV and

were both assigned to carbonate species. There were no peaks around 281 eV, what

indicates the absence of carbidic species. Unfortunately, no photocatalytic tests in

presence of either dyes or organic pollutants were performed.

Carbonate species were also proposed in the work of Lee et al.[48] as evidenced by

a C1s XPS peak of 288.2 eV. The “C-doped” anatase material was prepared by

hydrolysis of titanium isopropoxide with 2,4-pentanedione as an organic precursor,

followed by calcination at 200°C. EPR spectroscopy revealed a paramagnetic signal at

g = 1.997 ascribed to trapped electrons. The intensity of this signal was enhanced after

starting of irradiation, and with increasing the wavelengths of light up to 550 nm. The

photocatalysts induced ca. 60% of phenol mineralization, whereas reference P25 did

not exhibit any activity.

2.1.3 Amorphous carbon∗

The pyrolysis of many organic compounds leads to formation of different products

called amorphous carbon species. On the basis of crystallographic data it has been

proven that amorphous carbon consists of microcrystallites having a defect graphite

lattice. If the pyrolysis occurs at ca. 700K (427°C), carbon in the form of soot is

formed. Increasing the temperature to ca. 1100K (827°C) leads to formation of coke.

Thus, both soot and coke as microcrystallites of graphite contain in their structure

(polycyclic) aromatic hydrocarbons, which can be extracted from coke-industry

wastes.[67] Aromatic hydrocarbons, soot, and coke-like species or generally

carbonaceous species were also proposed to be present in carbon-modified titania.

∗ Information about amorphous carbon was taken from “Podstawy Chemii Nieorganicznej 2”, Wydawnictwo Naukowe PWN, Warszawa 2006, Adam Bielaᑀski

2. Previous work

16

2.1.3.1 Aromatic hydrocarbons, soot and coke-like species

For the first time Lettman et al.[54] proposed that coke-like species are responsible

for the Vis light activity of anatase materials. They were prepared by a sol-gel process

using different titania alkoxide precursors followed by calcination at 250°C. The coke-

like species were evidenced on the basis of EPR signals at g = 2.005[68, 69], and IR data

(See also Chapter 4.1). The best material was obtained from titanium isobutoxide. It

induced ca. 50% of 4-CP mineralization after 100 min irradiation

(㮰 ≥ 400 nm) and the strongest absorption in the visible. Alternatively, commercial

TiO2 – Hombikat UV 100 was impregnated with various alcohols becoming as active

as the materials prepared by sol-gel methods.

Kuo et al.[70] prepared carbon-modified anatase by metal-organic chemical vapour

deposition (MOCVD) using titanium isopropoxide as a precursor. The optimum

preparation temperature was found as 500°C in an oxygen-free atmosphere. On the

basis of C1s XPS and Raman spectroscopy coke-like species were proposed as

responsible for Vis light absorption and photoactivity. Carbon-modified samples

exhibited a relative high photocatalytic activity in NOx oxidation under visible light

illumination.

Chou et al.[51] postulated coke-like species as a sensitizer for his carbon-modified

material on the basis of XPS spectroscopy (for details see Chapter 2.1.2).

Khan et al.[41] prepared carbon-modified anatase by hydrolysis of titanium

tetrachloride in the presence of tetrabutylammonium hydroxide followed by

calcination at 400°C or in the presence of glucose and sodium hydroxide (calcined at

500°C). On the basis of differences in absorption spectra and colours of samples

prepared from TBAH or glucose it was concluded that coke-like products were

produced in the former sample. For powders obtained from glucose it was suggested

that carbon substitutes the lattice oxygen atoms. The modified samples exhibited

ca. 60% 4-CP degradation under Vis light (㮰 ≥ 380 nm).

2. Previous work

17

2.1.4 Unusual carbon species

Wang et al.[43] reported on “C-doped” TiO2 hollow spheres prepared after rapid

combustion of carbonaceous polysaccharide microspheres at high temperature (for

details see Chapter 9.1). The hollow spheres were identicated on the basis of scanning

electron microscopy (SEM) and transmission electron microscopy (TEM). In other

work, Wu et al.[53] revealed of “C-doped” nanospheres prepared via chemical vapor

deposition from titanium tetrabutoxide as a carbon source. Argon served as the carrier

gas. As prepared materials exhibited much higher photocurrents, than a reference P25

sample under Vis light irradiation. Additionally, quantum calculations indicated that

insertion of carbon into the TiO2 lattice both as anion or cation form is possible.

2.1.5 Colour centers

Contrary to the generally made assumption that the carbon species are the origin of

visible light photocatalysis, Serpone et al.[63, 64] reported that formation of lattice

defects associated with oxygen vacancies (here: colour centers), generates light

absorption and photocatalytic activity. Difference diffuse reflectance spectra of various

“C-doped” samples before and after heat treatment exhibited strong similarities. Thus,

it was concluded, that the origin of visible light absorption bands is independent on the

preparation method of these “C-doped” materials. The origin of visible light activity of

“C-doped” titania was ascribed to electronic transitions involving colour centers (so

called F-centers arising from the reduction of TiO2 after some form of heat treatment).

In our opinion, the strong similarities of DRS spectra for samples obtained from

different organic precusors seem to be unsufficient to conclude that in all modified

materials the same type of lattice defects is involved. However, the presence of such

defects cannot be excluded, but their participation in the photocatalysis of “C-doped”

titania remains questionable.

2.1.6 Theoretical calculations on the nature of carbon species

Density functional theory was also employed for investigating the carbon species

in carbon-modified titania.[71] The authors analyzed the effect of both substitutional

and interstitial carbon introduced into the anatase and rutile lattice. The stability of

2. Previous work

18

carbon species as a function of oxygen partial pressure was also investigated.

Additionally, the electronic structure of each material after the addition of either

substitutional or interstitial type of carbon was calculated. For both anatase and rutile

TiO2 at low partial pressures of oxygen, it was found that carbon preferentially

substitutes oxygen atoms. At high partial pressures of oxygen, both interstitial and

substitutional (whereas carbon substitutes titanium atoms) carbon species were

preferred (See Figure 2.1). For both anatase and rutile phase, the modification with

carbon results in a series of localized occupied states within the bandgap. Their

location depends on the kind of carbon position in the crystal lattice.[72]

Figure 2.1: Partial geometry of the models for carbon atoms substituting oxygen (CS-O) (a), or titanium (CS-Ti) (b), for an interstitial C atom (CI) (c), and for an interstitial C atom nearby an oxygen vacancy (CI + VO) (d) in the anatase supercell. The yellow spheres represent O atoms, the small brown spheres represent Ti atoms, and the black spheres represent the carbon impurity. Adopted from ref.[71]

2. Previous work

19

2.2 Goals of the work

According to the previous knowledge as presented in Chapter 2.1 the major goals

of this dissertation are summarized as follows:

• Improving the photocatalytic activity of carbon-modified titania by

preparation of new photocatalysts based on low-valent carbon compounds

e.g. graphite oxides, organic acids, and toluene.

• Unravelling the origin of visible light activity and nature of the “carbon

species” (commercially available carbon-modified titania - VLPcom).

• Preparing of new photocatalysts based on polyols as organic modifiers and

investigating their chemical stability.

• Using renewable organic modifiers.

3. General Experimental Part

20

3. General experimental part

3.1 Techniques

Diffuse reflectance spectra were obtained relative to BaSO4 with

a Shimadzu UV-Vis recording spectrophotometer equipped with a diffuse reflectance

accessory. Samples were prepared by mechanical mixing of 2 g of BaSO4 with 25 mg

of the corresponding powder sample. Reproducibility of Ebg was better than ± 0.05 eV.

Quasi-Fermi levels of electrons (nEF*) were measured according to the

literature[73]; 50 mg of catalyst and 10 mg of methylviologen dichloride were

suspended in 50 mL of 0.1 M KNO3 in a 100 mL two-necked flask. A platinum flag

and Ag/AgCl served as working and reference electrodes and a pH meter for following

the proton concentration. Concentrated HNO3 and NaOH (0.1, 0.01, and 0.001 M)

were used to adjust the pH value. The suspension was magnetically stirred and purged

with nitrogen gas throughout the experiment. Before starting the measurement the pH

value of the suspension was adjusted to pH 2. Irradiation was performed with the full

light of an Osram XBO 150 W xenon arc lamp. Stable photovoltages were recordable

about 30 min after changing the pH value. The pH0 values obtained from the inflection

points were converted to the Fermi potential at pH 7 by the equation nEF* (pH 7) =

−0.445 V + 0.059 V (pH0 − 7).[73] Reproducibility of nEF* was better than ± 0.05 V.

Reported values are the average of three measurements.

Binding energies were measured by X-ray photoelectron spectroscopy (XPS) with

a PHI 5600 XPS instrument employing pressed powder pellets contacted by silver

lacquer with an aluminium foil. Argon ion sputtering was performed using a Penning

source, Specs PS IQP 10/63 (p = 10−8 Torr; voltage = 3.5 kV) and the sputtering rate

was estimated by calibration with a SiO2 standard of known thickness. All XPS spectra

were referenced to the C1s peak of adventitious hydrocarbon contamination located at

284.8 eV. Fitting of the XPS data was accomplished using XPSPEAK41 software.

A Shirley-type background subtraction was used.

X-Ray diffractometry (XRD) measurements were performed with a Philips X´Pert

PW 3040/60 diffractometer.


21

Emission and excitation spectra were recorded by means of spectrofluorometer

Jasco FP- 6200.

UV-Vis spectra were recorded on Varian, Cary 50, UV-Visible Spectrophotometer.

Elemental analyses (EuroVector, CHNSO, E.A. 3000 equipped with a GC

detector) were conducted by dynamic spontaneous combustion.

Surface areas were determined by the BET (Brunauer-Emmet-Teller) method

using a Gemini 2360 V5.00 instrument.

3.2 Preparation of TiO2

Room temperature preparation: A solution of 0.25 M TiOSO4 (Alfa Aesar) was

prepared by stirring the required quantity of TiOSO4 in 1250 mL of doubly distilled

water at room temperature until a clear solution was obtained. Thereafter, the solution

was alkalized with 1 M NaOH until a pH value of 6.0 was reached. The formed gel

was kept under stirring at room temperature for overnight aging. Finally, titanium

hydroxide was filtered off, washed with water, and dried under air at 70°C. To obtain

titania, titanium hydroxide was calcined in a rotating flask for 1 h at 400°C. This self-

prepared titania was used for modification reactions presented in Chapter 4, 5, and 7.

Preparation at 80°C: A solution of 0.25 M TiOSO4 (Alfa Aesar) was prepared by

stirring the required quantity of TiOSO4 in 1250 mL of doubly distilled water at room

temperature until a clear solution was obtained. Thereafter, it was heated to 80°C and

a solution of 1.25 M NaOH was slowly added until a pH value of 5.5 was reached. The

formed gel was kept at 80°C for 1 h under stirring followed by overnight aging at

room temperature. Finally, titanium hydroxide was filtered off, washed with water,

and dried under air at 70°C. Subsequent calcination was performed in a rotating flask

for 1 h at 400°C (for TiO2 used in Chapter 6) or at 250°C (for TiO2 used in Chapter 8).

Titanium hydroxide prepared as described above was used for corresponding

modification reactions in Chapter 6 and 8.

The experimental set-up for the standard modification procedure is presented

below in Fig. 3.1.


22

3.3 Photodegradation/Photomineralization of 4-CP and other

pollutants

Standard photocatalytic experiment: 20 mL of a photocatalyst suspension in

aqueous 4-CP (2.5∙10−4 M) were sonicated for 15 min in a solidex glass cuvette before

irradiating with visible light (㮰 ≥ 455 nm, Osram XBO 150 W xenon arc lamp) for 3 h.

The concentration of the photocatalyst is given in each chapter. 2 mL of the

suspension were withdrawn with a syringe both before irradiating and after 60, 120,

and 180 min of irradiation time. Thereafter, the samples were filtered through

a micropore filter (pore size 0.22 µm, Rotilabo Roth) and degradation rate of 4-CP was

measured by recording the UV-Vis spectra of each sample. The decrease of the

maximum absorbance of 4-CP at 225 nm was observed in Chapter 4 and 5. To

determine the mineralization rate of 4-CP the TOC value∗ was estimated using

Shimadzu Total Carbon Analyzer TOC-500/5050 with a NDIR optical system

detector. For TOC measurements performed in Chapters 6, 7, 8, and 9, 2 mL of the

suspension were withdrawn with a syringe both before irradiating and after 90, 180

min of irradiation time.

In Chapter 4 and 7 diethylphthalate (DEP, 2.5∙10−4 M) was used as an alternative

pollutant. The photocatalyst suspension was prepared analogously as described above.

In Chapter 4 the decrease of the maximum absorbance of DEP was observed at 230

nm.[74, 75] In Chapter 7 the TOC value was estimated.

Alternatively, in Chapter 7 also dichloroacetic acid (DCE, 2.5∙10−3 M) was used as

an organic pollutant. Its mineralization rate was determined by TOC measurements.

For DEP and DEC degradation experiments the photocatalyst suspensions were

irradiated with visible light of 㮰 ≥ 455 nm for 3 h.

The experimental set-up for the standard photocatalytic experiment is presented

below in Fig. 3.2.

∗ Total organic carbon (mg∙L−1) was calculated according to the following equation:TOC (Total Organic Carbon) = TC (Total Carbon) − IC (Inorganic Carbon), whereas TC and IC are measured values.


23

B

A

C

Figure 3.1: Experimental set up for the standard modification procedure. A – rotator, B – oven, C – reactor. Adopted from ref.[76]

E

D

ABC

F

Figure 3.2: Experimental set up for the standard photocatalytic experiment. A – power supply, B – xenon-arc lamp with water cooling, C – IR water filter, D – cut-off filter, E – glass cuvette, F – magnetic stirrer. Adopted from ref.[76]

4. Graphite and graphite oxides as modifiers

24


4.1 Introduction

In the first paper concerning carbon-modified titania it was proposed that highly

condensed, coke-like species are responsible for the Vis light activity of materials

prepared through a sol-gel process by using different titanium alkoxides.[54] Highly

conjugated, probably polyaromatic compounds were proposed on the basis of EPR

signals.[68, 69] Also high C/H ratios of app. 200 assigned to the carbonaceous species

and complete absence of C-H bands in IR spectrum supported the previous suggestion.

In a subsequent work carbon-modified powders were prepared by hydrolysis of

titanium tetrachloride with tetrabutylammonium hydroxide followed by calcination at

high temperatures.[46] C1s binding energies were found at 285.6, 287.5 and 288.5 eV.

The first value arises from adventitious elemental carbon; the latter two values suggest

the presence of carbonate species. However, carbonates are not responsible for

photocatalytic activity of carbon-modified materials[31] (See also Chapter 2.1.2).

Searching for low-valent carbon compounds of coke-like structures (vide supra) we

have decided to use graphite/graphite oxides as precursors (abbreviated as G/GO).

Graphite oxide was first prepared more than 150 years ago by Brodie et al.[77] An

excellent study on the formation and structure of graphite oxide was reported by

Boehm et al.[78, 79] The authors concluded that graphite is first oxidized to the graphite

salt Cn+A− ∙ 2HA like C24

+HSO4− ∙ 2H2SO4 or C24

+NO3− ∙ 2HNO3. Hydrolysis of

graphite salts finalized formation of graphite oxide. Due to the binding of oxygen

atoms to ring carbon atoms during oxidation the six-membered carbon rings lose their

aromatic character (Eq. 4.1).

C24+A− + H2O C24-OH + HA (4.1)

Acidic hydroxyl groups and small amounts of carboxyl groups located at the edge

of carbon layers are produced.

A few structures of graphite oxide were proposed in the last century.[80-87] The

model of Scholz et al.[86] (See Figure 4.1) exhibits a quinoide carbon network

consisting of two regions (of trans-linked cyclohexane chairs and ribbons of flat


25

carbon hexagons connected by C=C double bounds) and functional groups such as

tertiary OH, 1,3-ethers, ketones, quinones, carboxylates, and alcohols.

Figure 4.1: Model of graphite oxide structure adopted from Scholz et al.[86]

Our goal was to utilize the carboxylate groups of GO for esterification with surface

hydroxyl groups of titanium dioxide (Eq. 4.2):

[TiO2]-OH + HOOC-[GO] [TiO2]OOC-[GO] + H2O (4.2)

4.2 Experimental

4.2.1 Graphite oxidation methods

Brodie et al.: 1 g of graphite powder (Chempur, Feinchemikalien und

Forschungsbedarf GmbH, <100 㯀m) was added under stirring into a 100-mL beaker

containing 20 mL of fuming nitric acid. Thereafter 8 g of KClO3 were given to the

suspension. After 3 h the acid suspension was transferred into ca. 500 mL of distilled

water to terminate the reaction. After filtration the solid residue was washed with

methanol until the pH value of the filtrate reached pH 5 or more affording a dark

brown powder.[82]

Staudenmaier et al.: Graphite powder was oxidized by fuming nitric acid, 97%

H2SO4, and KClO3. A 300 mL flask containing 97% H2SO4 (50 mL) and 25 mL of

fuming nitric acid was cooled down to 5°C in an ice bath. Thereafter, 1 g of graphite

powder was slowly added to the reaction flask. During three days 2 g of KClO3 were

added eight times (per day). Finally, the acid suspension was transferred into ca. 1 L of

distilled water to terminate the reaction and immediately filtrated. The residue was


26

washed with methanol until the pH value of the filtrate reached pH 5 or more affording

the black powder.[82]

Hummers et al.: The graphite oxide was prepared by stirring 1 g of graphite

powder and 0.5 g of NaNO3 in 23 mL of 97% H2SO4. The suspension was cooled

down to 0°C in an ice-bath. After adding 3 g of KMnO4 the colour of the suspension

changed to green. After removal of the ice-bath the temperature of the reaction mixture

increased to ca. 40°C. The reaction mixture was left at this temperature for 30 min.

Subsequently, 46 mL of distilled water were slowly added causing a violet fume,

increase of temperature to ca. 80°C, and colour change to brown. After 15 min the

suspension was diluted to approximately 100 mL of warm distilled water and treated

with ca. 12 mL of 3% H2O2 to reduce the residual MnO4− and MnO2 to colourless and

soluble MnSO4. Upon treatment with the peroxide, the suspension turned to bright

yellow. The warm suspension was filtrated to avoid precipitation of the slightly

soluble salt of mellitic acid formed in a side reaction. The yellowish-brown powder

was washed three times with 140 mL of warm water. Thereafter, the washing

procedure was repeated with distilled water, 3 M HCl, 3 M NaOH, and finally again

with distilled water affording a black powder.[88]

Szabo et al.: 1 g of graphite and 8.5 g of KClO3 were mixed in a round bottom

flask. Next, 6 mL of fuming HNO3 were slowly dropped. The obtained dark green

slurry was left overnight for aging at ambient temperature. The loss of HNO3 due to

evaporation was compensated by adding another portion of acid (4 mL). The slurry

was then heated to 60°C for 8 h. The reaction was terminated by transferring the pasty

mixture into 100 mL of distilled water. The suspension was washed 5 times with 20

mL of 3 M HCl and 7 times with 100 mL of distilled water to remove acidic and saline

impurities. The residue was separated by filtration and finally dried at 60°C for 1 h

affording bright brown powder.[86]

4.2.2 Preparation of modified titania and photocatalytic tests

Modification of TiO2 by graphite and graphite oxides: Modified powders were

prepared by grinding 1 g of TiO2 (prepared as in Chapter 3.2) with graphite or graphite


27

oxides followed by calcination in a rotating flask for 1 h at various temperatures.

Alternatively, air was blown into the rotating flask.

Alternatively, the graphite oxides were sonicated for 15 min in 20 mL of distilled

water. Next, 3 M HCl was added to the suspension, until the pH value of ca. 2.3 was

reached. Thereafter, the suspension was heated at ca. 70-80°C for 1 h under stirring.

Part of such prepared brown powder was further calcined at 400°C for 1 h in a rotating

flask affording grey TiO2-GOsol-400∗.

Table 4.1: List of samples prepared with G/GO as modifiers.

Sample Modifier wt % of modifier

Calcination temperature

[°C]

Colour of

sampleTiO2-G-400 G 5 400 greyTiO2-G-500 G 5 500 greyTiO2-G-600 G 5 600 greyTiO2-GO-200 GO 20 200 grey

TiO2-GO-300 GO 5 300 slight grey


TiO2-GOair-350 GO 5 350 grey


TiO2-GOsol -400 GO 5 400 greyTiO2-GO-500 GO 5 500 whiteTiO2-GO-600 GO 5 600 white

Photodegradation of 4-CP and DEP: Photocatalytic tests in presence of 4-CP or

DEP were performed according to the standard procedure presented in Chapter 3.3.

The concentration of catalysts was equal to 0.5 g∙L−1.

∗ Explanation of samples nomenclature: e.g. TiO2-GOsol-400 means that titania was modified with graphite oxides (GO) in water suspension, followed by calcination at 400°C.


28

4.3 Results and discussion

4.3.1 Oxidation of graphite

A few methods were employed to oxidize graphite[82, 86, 88] (See also Chapter

4.2.1). According to the Brodie method, fuming HNO3 and KClO3 were used as

oxidation agents.

4000 3500 3000 2500 2000 1500 1000 500

47

48

49

50

51

52

53

Tran

smitt

ance

/ %

Wavenumber / cm-1

Figure 4.2: IR spectrum of graphite oxide prepared according to Brodie et al.

The IR spectrum of the oxidation product (Figure 4.2) exhibits peaks at 3420 and

1624 cm−1 (theor. 3400 and 1620 cm−1) arising probably from water located between

GO layers and KBr. The peaks at 1562 and 1462 cm−1 can be ascribed to carboxylate

ions (theor. 1550-1610 and 1300-1400 cm−1). The peak at 1382 cm−1 arises probably

from anti-stretching vibration of carboxylates. The peak at 1624 cm−1 may arise from

the deformation vibration of water (theor. 1620 cm−1). The absorption of chinon

systems, conjugated carbon-carbon double bound and microcrystalline carbon may be

ascribed to the peak at 1600 cm−1.[79] The peak at 1093 cm−1 comes probably from the

stretching vibration of C-OH group (theor. 1070 cm−1) and the peak at 800 cm−1 from

tertiary alcohols, peroxides or hydroperoxides (theor. 820-920 cm−1). The intense peak

at 2925 cm−1 corroborates well with the theoretical value of 2900 cm−1 characteristic


29

for C-H vibrations. The peak at 2373 cm−1 is probably adsorbed CO2 (theor. 2350

cm−1). The theoretical IR values given above are from ref.[79] Although, elemental

analysis for as prepared GO of C 63.85, O 33.78, and H 1.20% is in good agreement

with the theoretical values of C 62.0, O 36.2, and H 1.82%, the oxidation of graphite to

graphite oxide was not successful because of the absence of the peak at 1720 cm−1

characteristic for a carbonyl group. Thus, the other oxidation method developed by

Staudenmaier et al.[82] was also performed. Unfortunately, no changes in the spectrum

of oxidation product in comparison with pure graphite were detected (Figure 4.3).

Even the colour of the powder remained black, in opposite to the dark brown powder

obtained by Brodie method. The elemental analysis of C 73.8, H 1.1% did not agree

with the literature values of C 57.1, H 1.7%.[79]

4000 3500 3000 2500 2000 1500 1000 50045

50

55

60

65

70

75

Tran

smitt

ance

/ %

Wavenumber / cm-1

Figure 4.3: IR spectrum of graphite oxide prepared according to Staudenmaier et al.

The third method applied in this dissertation was first proposed by Hummers et

al.[88] Unfortunately, except the peaks 3441, 2931, 1645 and 1106 cm−1, whose origin

was already explained in Figure 4.2 (vide supra), there was no maximum at 1720 cm−1

characteristic for carbonyl group (Figure 4.4). Also the colour of the powder did not

change. The elemental analysis of C 50.0 and H 0.9% did not fit the literature values

of C 60.9 and H 1.6%.


30

4000 3500 3000 2500 2000 1500 1000 500

68

69

70

71

72

73

74

75

Tran

smitt

ance

/ %

Wavenumber / cm-1

Figure 4.4: IR spectrum of graphite oxide prepared according to Hummers et al.

Finally, the carbonyl band of 1720 cm−1 was found in the spectrum of GO prepared

by Szabo et al.[86] (Figure 4.5). Also peaks at 1625, 1387 and 1078 cm−1 (explained by

description of Figure 4.2) were present. The bright brown colour of the powder and IR

maximum at 1720 cm−1 prove successful oxidation of graphite to graphite oxides. It is

noteworthy that intensity of GO carbonyl peak is relative low because of keto-enole

tautomerism.[79]

4000 3500 3000 2500 2000 1500 1000 500

26

28

30

32

34

36

38

Tran

smitt

ance

/ %

Wavenumber / cm-1

~ 1720 cm−1

Figure 4.5: IR spectrum of graphite oxide prepared according to Szabo et al.


31

The oxidation of graphite according to Szabo was repeated three times. With each

next cycle the increase of carbonyl band at 1720 cm−1 was observed (Figure 4.6). Also

the change of the colour from black characteristic for graphite to light brown

characteristic for GO was noticed.

4000 3500 3000 2500 2000 1500 1000 50025

30

35

40

45

50

55

60

65

~ 1720 cm−1

d

c

b

Tran

smitt

ance

/ %

Wavenumber / cm-1

a

Figure 4.6: Changes in IR spectra of graphite after oxidation according to Szabo et al.: graphite (a), GO after 1st (b), 2nd (c), and 3rd (c) oxidation cycle.

To sum up, only technique of Szabo et al. resulted in the presence of the IR

diagnostic peak at 1720 cm−1 (Figure 4.6). A few details considering the oxidation

condition seem to be crucial for the effectivity of this technique in comparison with

other ones. First of all, the graphite and KClO3 were ground before addition of fuming

HNO3. Moreover, the reaction mixture was heated up to 60°C and the whole process

was repeated three times (See Table 4.2).


32

Table 4.2: Comparison of all oxidation methods.

Oxidation method Oxidation agents

Time of reaction Temperature

BrodieFuming HNO3, KClO3

~ 3h RT

Staudenmaier

Fuming HNO3, KClO3, conc. H2SO4

3 daysCooling

down to the 5°C in an ice

bath

Hummers

conc. H2SO4,

3% H2O2, NaNO3, KMnO4

~ 2h

Without cooling.

Temperature was

increased up to 80°C

SzaboFuming HNO3, KClO3

1.5 day Heating at 60°C for 8h

4.3.2 Titania modification by graphite and graphite oxides

Titanium dioxide was modified by graphite/graphite oxides by grinding of these

two components, followed by calcination at various temperatures (See Chapter 4.2.2).

An esterification between surface hydroxyl groups of titania and carboxylate groups of

GO was assumed (See Eq. 4.2). The reference self-prepared TiO2 did not induce any

degradation of 4-CP.

Unfortunately, both TiO2-G and TiO2-GO materials did not exhibit any

degradation of 4-CP under visible light irradiation (㮰 ≥ 455 nm). Only the material

obtained from acidic suspension of graphite oxide and titania (See Chapter 4.2.2)

resulted in 20% of 4-CP degradation. However, after calcination at 400°C, the activity

completely ceased. Both TiO2-GO-200 and TiO2-GO-350 showed only 10% of 4-CP

degradation. To check if the TiO2-GO samples can induce degradation of other


33

pollutants than 4-CP, a photocatalytic test in presence of diethyl phthalate (DEP)[74, 75]

was performed (See Chapter 4.2.2). Unfortunately, no degradation was observed.

Moreover, the 4-CP suspensions of TiO2-G and TiO2-GO prepared for degradation

experiments were not homogeneous. Particles of probably coke or unreacted GO were

present in the suspensions. This suggests that the carbon species responsible for

photocatalytic activity were not chemically attached to titania. Probably the amount of

carboxylates at the edges of GO layers was not sufficient for the anticipated

esterification with titania hydroxyl groups (See Eq. 4.2).

Employing other modification methods may result in TiO2-G/TiO2-GO powders

active in visible light oxidations of organic pollutants. First of all, the stability of

modified materials needs to be improved. Secondly, titanium hydroxide instead of

titanium dioxide could be used because of the higher density of surface hydroxyl

groups.

5. Organic acids as modifiers

34


5.1 Introduction

Modification of titanium dioxide with graphite and graphite oxides did not result in

materials active in Vis light photocatalysis. Thus, we decided to use simple aromatic

systems like benzoic, phthalic or trimellitic acid as modifiers to render titania active.

We supposed that using such organic molecules facilitates understanding the origin of

visible light activity of carbon-modified titania. For comparison also the non-aromatic

shikimic acid with substituted hydroxyl groups was investigated. The powders were

prepared by grinding those compounds with titanium dioxide followed by calcination

at high temperatures. The Vis light activity was examined by degradation of

4-chlorophenol (4-CP) (according to Chapter 3.3). For materials prepared from

phthalic acid and trimellitic acid, the optimum preparation conditions were

determined. An esterification between surface hydroxyl groups of titania and

carboxylic groups of organic acids was assumed (See Eq. 4.2).

Figure 5.1: Structural formula of organic modifiers: benzoic acid (A), phthalic acid (B), trimellitic acid (C), and shikimic acid (D).


35

5.2 Experimental

5.2.1 Preparation of TiO2

Titanium dioxide was prepared according to the preparation method described in

Chapter 3.2.

5.2.2 Preparation of titania modified with benzoic acid

Standard modification procedure: Modified powders were prepared by grinding

0.5 g of TiO2 with benzoic acid or other corresponding acid followed by calcination

under air in a rotating flask for 1 h at 200-600°C. The samples modified with benzoic

acid are named as TiO2-BA (See Table 5.1).

5.2.3 Preparation of titania modified with phthalic acid

The same procedure as described in Chapter 5.2.2 was employed. The modified

samples are named as TiO2-PA∗ (See Table 5.1). Alternatively, air or argon were blown

through the reactor.

In a modified version, the preparation was performed in water suspension as

follows: Phthalic acid was ground with TiO2. The prepared solid mixture was

suspended in 20 mL of doubly distilled water, sonicated for 15 min, heated at 100°C

for 1 h under reflux, and left overnight under stirring. Thereafter, the solid was washed

three times with doubly distilled water by centrifugation and dried at 100°C for 1 h,

followed by grinding. Finally, the powder was calcined at high temperatures for 1 h in

a rotating flask affording creamy TiO2-PA-H2O-300 and colourless TiO2-PA-H2O-500

(See Table 5.1).

Alternatively, the preparation was performed in acetone suspension as follows:

Phthalic acid was ground with TiO2. As prepared solid mixture was suspended in 20

mL of acetone and left overnight under stirring. Thereafter, acetone was removed by

∗ Explanation of samples nomenclature: e.g. TiO2-PA-H2O-300 means that titania was modified with phthalic acid (PA) in water suspension, followed by calcination at 300°C.


36

centrifugation. The powder was ground and calcined at 500°C for 1 h in a rotating

flask affording grey TiO2-PAacet-500 (See Table 5.1).

5.2.4 Preparation of titania modified with trimellitic acid

The same standard modification procedure as described in Chapter 5.2.2 resulted

in formation of TiO2-TA materials (See Table 5.1).

Alternatively, the preparation in water suspension was performed according to

Chapter 5.2.3 with one difference, that the water was removed by evaporation under

vacuum at 100°C for 3 h. The powder was calcined at 400°C for 1 h in a rotating flask

affording slight yellow TiO2-TA-H2O-400 (See Table 5.1).

Preparation of titania modified with trimellitic acid was performed in acetone

suspension as described in Chapter 5.2.3. After calcination at 400°C slightly yellow

TiO2-TAacet-400 was obtained.

5.2.5 Preparation of titania modified with shikimic acid

Modification with shikimic acid was carried out according to the modification of

titania with phthalic acid in water suspension (See Chapter 5.2.3). Calcination at

300°C afforded slightly yellow TiO2-SA-H2O-300.

Table 5.1: List of samples prepared from organic acids and titania. ∗

Sample ModifierCalc. temp. [°C]

Colour of

sample

TiO2-BA-200 benzoic acid

200 creamy

TiO2-PA-200 phthalic acid

200 white


300 slightly yellow

∗ For each sample 20 wt% of an organic modifier was used.


37

TiO2-PA-H2O-300

phthalic acid

300 creamy


400 slightly brown


450 slightly brown


500 greyish

TiO2-PAair-500 phthalic acid

500 slightly yellow

TiO2-PAAr-500 phthalic acid

500 dark grey

TiO2-PA-H2O-500

phthalic acid

500 white

TiO2-PAacet-500 phthalic acid 500 grey


550 grey


600 white

TiO2-TA-200 trimellitic acid

200 creamy


300 slight yellow


400 slight brown

TiO2-TA-H2O-400

trimellitic acid

400 slight yellow

TiO2-TAacet-400 trimellitic acid

400 slight yellow


500 slight yellow


600 creamy

TiO2-SA-H2O-400

shikimic acid

400 slight yellow


38

5.2.6 Photodegradation of 4-CP

Photocatalytic tests in presence of 4-CP were performed according to the standard

method presented in Chapter 3.3. For titania modified by trimellitic acid the maximum

absorbance of 4-CP was observed at 280 nm. The concentration of catalysts was equal

to 2.0 g∙L−1.


5.3.1 Benzoic acid

Benzoic acid is known as an organic pollutant for testing the photoactivity of UV

irradiated TiO2 (P25).[89] It was also used to study the adsorption on a TiO2 surface.[90,

91] In this dissertation benzoic acid was employed as an organic modifier. Simple

procedure with the calcination of a solid mixture of benzoic acid and TiO2 afforded

creamy TiO2-BA-200, which photocatalyzed only 7% of 4-CP degradation (㮰 ≥ 455

nm).

5.3.2 Phthalic acid

To our knowledge phthalic acid was not used previously for the modification of

TiO2. However, in work of Sun et al.[92] phthalic acid was mixed with zinc oxide

affording at 120°C anhydrous zinc phthalate. Benzophenone and anthraquinone were

obtained after pyrolysis of zinc phthalate at 400-450°C in argon atmosphere. The

presence of aromatic systems after pyrolysis of zinc phthalate is relevant to

forthcoming parts of this dissertation, wherein aromatic hydrocarbons are proposed as

relevant species (See Chapter 7). Modification was performed by grinding phthalic

acid with titania. The calcination at 500°C resulted in a photocatalyst (TiO2-PA-500)

of the highest photoactivity (33%) in degradation of 4-CP under Vis light irradiation

(Figure 5.2).


39

200 300 400 500 6000

20

40

60

80

100

% d

eg. 4

-CP

calcination temperature / °C

Figure 5.2: Dependence of 4-CP degradation on calcination temperature of the TiO2-PA series.

To improve photocatalytic activity phthalic acid was mixed with TiO2 in water

suspension (See Chapter 5.2.3). Unfortunately, this procedure resulted in only 6% of

4-CP degradation found for white TiO2-PA-H2O-500 (See Figure 5.3 e). However,

a significant value of 28% was noticed for creamy TiO2-PA-H2O-300 (See Figure 5.3

d). A mutual influence of reaction solvent was also investigated. Unfortunately, using

acetone to prepare the suspension of phthalic acid and titania did not result in any

improvement of activity (See Figure 5.3 f).

Calcination conditions should also influence the properties of the samples. For

TiO2-PAair-500 calcined in a flow of air, a degradation value of 23% was found. For

TiO2-PAAr-500 calcined in a flow of argon, the degradation value of only 3% was

observed (See Figure 5.3 b, c). In summary, the best preparation procedure was just

grinding followed by calcination at 500°C. The resulting TiO2-PA-500 photocatalyzed

33% of 4-CP degradation (See Figure 5.3 a). Contrary to this, self-prepared TiO2 did

not show any activity in visible light.


40

0 60 120 1800

20

40

60

80

100

4-C

P/4

-CP

0/ %

time / min

abcdef

Figure 5.3: Photodegradation of 4-CP in presence of TiO2-PA-500 (a), TiO2-PAair-500 (b), TiO2-PAAr-500 (c), TiO2-PA-H2O-300 (d), TiO2-PA-H2O-500 (e), and TiO2-PAacet-500 (f).

5.3.3 Trimellitic acid

To date, trimellitic acid was employed as an organic pollutant for titania catalyzed

photoactivity tests.[93] We have decided to use trimellitic acid as an organic modifier of

TiO2. The increase of the amount of carboxylic functional groups may facilitate the

esterification with titania hydroxyl groups (See Eq. 4.2). The powders were prepared

by calcination of the solid mixture of an organic acid and TiO2. The calcination at

400°C resulted in the most photoactive TiO2-TA-400 (33% of 4-CP degradation) (See

Figure 5.4).


41

200 300 400 500 6000

20

40

60

80

100

% d

eg. 4

-CP

calcination temperature / °C

Figure 5.4: Dependence of 4-CP degradation on calcination temperature of the TiO2-TA series.

Water or acetone suspensions (See Experimental 5.2.4) gave materials of

comparable activities of 33 and 23% (See Figure 5.5 b, c). Therefore, no improvement

of photocatalytic activity was observed in comparison to the solid grinding procedure.

In summary, the most effective preparation was grinding followed by calcination at

400°C. The resulting TiO2-TA-400 photocatalyzed 33% of 4-CP degradation (See

Figure 5.5 a). Contrary to this, pristine TiO2 did not show any activity in visible light.


42

0 60 120 1800

20

40

60

80

100

4-C

P/4

-CP

0/ %

time / min

abc

Figure 5.5: Photodegradation of 4-CP in presence of TiO2-TA-400 (a), TiO2-TA-H2O-400 (b), and TiO2-TAacet-400 (c).

5.3.4 Shikimic acid

Shikimic acid was also used as an organic modifier, due to the presence of three

hydroxyl groups capable of esterification with Ti-[OH] groups (See Eq. 4.2). The

resulting TiO2-SA-H2O-300 exhibited 24% of 4-CP degradation, whereas reference

TiO2 did not induced any activity.

In summary, modification of titania with organic acids like phthalic, trimellitic,

and shikimic acid afforded materials inducing only ca. 30% activity in visible light.

6. Modification of titania with toluene

43


6.1 Introduction

As it was already mentioned in Chapters 2.1.3.1 and 4.1 coke-like species formed

during calcination were proposed as responsible for Vis light photosensitization.[54] It

is known that such carbonaceous polymers (coke) are produced through condensation

reactions of gaseous benzene and toluene over zeolites.[94] Typical temperatures are in

the range of 400-500°C.[95] We therefore anticipated that calcination of titania at

400°C in the presence of toluene may produce similar carbon species. Thus, we have

used also toluene as an organic modifier.

6.2 Experimental

6.2.1 Preparation of TiO2

Titanium hydroxide and titanium dioxide were prepared according to preparation

method described in Chapter 3.2.

6.2.2 Preparation of TiO2 modified with toluene

Onto 0.5 g of titanium hydroxide or titanium dioxide a portion of 0.5 mL of

toluene p.a. (Carl Roth) was slowly dropped. Thereafter, the suspension was sonicated

for 15 min, followed by calcination at 400°C for 1 h in a rotating flask. To remove all

organic and inorganic impurities the calcined powders were suspended in 100 mL of

doubly distilled water and heated at 90°C for 1 h with stirring, followed by three times

centrifuging with 50 mL of doubly distilled water. After drying at 100°C for 1 h the

brown powders TiO2-TOL-1a and TiO2-TOL-1b were obtained (See Table 6.1) ∗.

∗ Explanation of samples nomenclature: e.g. TiO2-TOL-1a means that titania was modified with toluene (TOL), followed by calcination at high temperature.


44

6.2.3 Photomineralization of 4-CP

Photocatalytic tests in presence of 4-CP were performed according to the standard

method presented in Chapter 3.3. The concentration of catalysts was equal to 2.0 g∙L−1.


Samples were prepared by dropping onto titanium dioxide/titanium hydroxide the

suitable amount of toluene, followed by calcination at 400°C for 1 h. The resulting

slight brown TiO2-TOL-1a induced 24% of mineralization. Contrary to this, reference

TiO2 did not show any activity in visible light.

To check a mutual influence of titania hydroxyl groups titanium hydroxide instead

of TiO2 was used in the modification with toluene. As can be seen from Table 6.1 the

material TiO2-TOL-1b prepared from titanium hydroxide exhibits a much higher

activity (42% of mineralization) as compared to the titania based material – TiO2-

TOL-1a (24% of mineralization). Another parameter, which has an influence on the

mineralization rate, is the pH value of the 4-CP suspensions. The pH values of various

photocatalyst suspension before starting the irradiation have to be the same to get

comparable results.[96]

Table 6.1: Photocatalytic activity of samples prepared from TiO2 and Ti(OH)4.

pHSample Starting

material Colour0h 3h

Mineralization %

TiO2-TOL-1a TiO2 slight brown

8.2 7.8 24

TiO2-TOL-1b Ti(OH)4 brown 8.5 7.1 42

From the results presented above; one has to be concluded that modification of

titania or titanium hydroxide by liquide toluene does not effort photocatalysts of high

activity in 4-CP mineralization.

7. Commercial carbon-modified titania

45

7. Commercial carbon-modified titania∗

7.1 Introduction

Recently, we have reported on a carbon-modified titania (TiO2-C) prepared from

TiCl4 and tetrabutylammonium hydroxide.[46, 97, 98] Thereafter a technical process was

developed rendering this visible light photocatalyst commercially available (VLPcom).

It consists of calcining titania in the presence of an organic compound as a carbon

source.[98] It was generally proposed that the presence of some carbon species in titania

is responsible for the visible light activity. However, the chemical nature of the

“carbon dopant” is still a matter of discussion. The C1s binding energy, as easily

obtained by XPS, was taken as diagnostic tool for the type of carbon present. From

corresponding values of 284.8-285.7 eV[42, 46, 56, 57, 61, 70, 97, 99] the presence of elemental

carbon and graphitic or coke-like carbon was proposed.[54, 70] It is noted that the

binding energies of carbidic carbon of 281.8-284.3 eV[45, 56-59, 99] and aromatic ring

carbon atoms of 284.3-284.7[90, 100, 101] fall in similar range. Also surface carbonates

were proposed as relevant species 286.5-289.4 eV[39, 46, 66, 70, 97, 102], but it was shown

that they are not responsible for visible light activity.[31] Binding energies of 288.6 and

288.9 eV were thought to arise from structural fragments like Ti-O-C[42] and

Ti-OCO.[103] Density functional theory calculations suggest that substitutional (of

lattice oxide) and interstitial carbon atoms are present.[71] Contrary to the common

opinion it was proposed that not carbon species itself but oxygen vacancies, generated

only in the presence of a carbon source, are responsible for visible light activity.[63, 64]

It is noted that the relevant species may be different, depending on the nature of the

carbon source.

In anatase powders prepared from different titanium alkoxides as a carbon source

a symmetric paramagnetic signal was observed at g = 2.005 by EPR spectroscopy,

assigned to aromatic coke like species.[54] Similar results were obtained for the

∗ Most of the work presented in this chapter has been published as:Zabek, P.; Eberl, J.; Kisch, H., Photochem. Photobiol. Sci., 8, 264 (2009).


46

commercial VLPcom product revealing that signal intensity increases with carbon

content.[97] Although the intensity increased upon Vis irradiation, it could not be

concluded that the corresponding radical is involved in the photocatalysis process

since the concentration of radicals was about five to six orders of magnitude lower

than the total carbon content.[97] Contrary to this, it was proposed that in carbon-

modified titania prepared from gaseous cyclohexane this paramagnetic signal arises

from an electron trapped at an oxygen vacancy.[39]

To get further information on the basic question whether a carbon species or just

oxygen vacancies are responsible for the visible light activity, we have performed

some simple chemical experiments with the commercial VLPcom material.

7.2 Experimental

Preparation of TiO2: Titanium dioxide was prepared according to the preparation

method described in Chapter 3.2.

Extraction of the photosensitizer: 5 g of C-modified titania [VLPcom, a research

product from Kronos Internat., Inc., prepared from titania modification with

pentaerythritol (C 0.36, H 0.40%), specific surface area of 150 m2∙g−1] were suspended

in 50 mL of doubly distilled water and brought to pH ~ 12.5 with 0.5 M NaOH.

Thereafter the suspension was heated for one hour at 90°C and aged over night under

vigorous stirring. A brown extract (Sensorg) and pale brown powder were obtained

after centrifugation. The solid was extracted three times using the same procedure in

order to remove all organic and inorganic impurities affording an almost colourless

powder (VLPcom, res, C 0.44, H 0.33%), which was dried one hour at 100°C. The brown

solid obtained after evaporation of Sensorg to dryness afforded the following elemental

analysis of C 7.86, H 2.10, O 47.75%.

Re-assembling reaction: 50 mL of the brown extract were acidified to pH ~ 3 with

1 M HCl and after adding 1 g of VLPcom, res the pH value was adjusted to 5.5 by

addition of 1 M HCl. Thereafter the suspension was heated for one hour at 90°C and

stirred overnight at ambient temperature. The brownish powder obtained after

centrifugation was dried at 100°C and calcined for one hour at 200°C in presence of


47

air in a rotating flask affording VLPcom, reas (C 0.65, H 0.34%). The procedure was

repeated as described above, but 1 g of self-prepared TiO2 was used instead of VLPcom,

res affording slightly yellow TiO2-C.

Preparation of P25-C: As described above for the re-assembling reaction but

VLPcom, res was replaced by the commercial titania powder P25 (Degussa) affording

P25-C material with elemental analysis of C 0.35, H 0.15%.

Photomineralization of 4-CP: Photocatalytic tests in presence of 4-CP were

performed according to the standard photocatalytic experiment presented in Chapter

3.3. The concentration of catalysts was equal to 2.0 g∙L−1. Whereas VLPcom and

VLPcom, reas suspended in 4-CP induced a solution of pH ~ 5.8, VLPres gave rise to pH

~ 10 and therefore a few drops of 1M HCl was added to obtain pH ~ 5.8. Only this pH

adjustment allows a reliable comparison since the degradation rates are pH

dependent.[96]

Photomineralization of DEP and DCE: Photocatalytic tests in the presence of DEP

and DCE were performed according to the standard photocatalytic experiment

presented in Chapter 3.3. The concentration of catalysts was equal to 2.0 g∙L−1.

Photosensitized degradation of 4-CP: 19 mL of Sensorg were added to a 5∙10−3 M

4-CP solution and brought to pH 5.8 by addition of 1 M HCl. Thereafter irradiation

was performed with visible light (㮰 ≥ 455 nm). TOC values were measured as

described in Chapter 3.3.

Photostability test: A suspension of 2.0 g∙L−1 of beige VLPcom in 190 mL of

2.5∙10−4 M aqueous 4-CP was sonicated in a 200 mL immersion lamp apparatus for 15

min. After measuring the TOC value, the suspension was irradiated for 3 h with the

filtered light (1 M solution of sodium nitrite, λ ≥ 400 nm) of a tungsten halogen lamp.

Subsequently the TOC was again measured and new 4-chlorophenol solution was

added to the concentration (See Figure 7.7). This procedure was repeated five times

resulting in overall six degradation runs. Thereafter the remaining yellowish powder

(C 2.70, H 0.54%) was extracted with NaOH as described above affording the whitish

VLPcom res,long term (C 1.84, H 0.57%). When this material was employed in the standard


48

photodegradation of 4-CP (See Chapter 3.3) it induced only 20% of TOC decrease

after 7.5 h of irradiation with the light of 㮰 ≥ 420 nm.


When an alkaline suspension of the beige VLPcom powder was heated at 90°C,

a brown extract (SENSorg) and an almost white residue (VLPcom, res) were obtained after

centrifugation. This suggests that the visible light activity of VLPcom is due to the

presence of a molecular sensitizer. If this is the case, it should be possible to re-

assemble the original VLPcom by treating VLPcom, res with the extract SENSorg. This was

achieved by adjusting the pH of the corresponding suspension to pH 5.5, followed by

heating at 90°C and removal of water. Subsequent calcination at 200°C afforded the

slightly brownish powder VLPcom, reas. As indicated by XRD analysis VLPcom,

VLPcom, reas, VLPcom, res, and TiO2 all consist of anatase (See Figure 7.1). The peak at

44.7° observed for VLPcom, reas arises from aluminum added to the powder to fill up the

sample holder. Crystallite sizes of 10, 13, 12, and 6 nm are obtained for VLPcom,

VLPcom, reas, VLPcom, res and TiO2, respectively, through application of the Scherrer

Equation to the peak at 25.3°.


49

10 20 30 40 50 60 70 80

2θ / deg

a

b

c

de

f

Figure 7.1: XRD diffraction patterns of VLPcom a), VLPcom, reas b), VLPcom, res c), self-prepared TiO2

d), anatase e), and aluminum f). Vertical solid and dashed lines represent reference signals of anatase (ASTM 84-1286) and aluminum (ASTM 4-0787), respectively.

Diffuse reflectance spectra of VLPcom and VLPcom, reas exhibit a weak but

significant absorption in the visible, which is absent in the case of VLPcom, res (See

Figure 7.2 c). Assuming that all three materials are indirect semiconductors like

anatase, optical bandgaps of 3.22, 3.21, and 3.27 eV are obtained for VLPcom, VLPcom,

reas, and VLPcom, res, respectively, from a plot of transformed Kubelka-Munk function

vs. the energy of the absorbed light.[104, 105]


50

350 400 450 5000.00

0.01

0.02

0.03

0.04

F(R

∞)/

a.u

.

λ / nm

a

bc

Figure 7.2: Diffuse reflectance spectra of VLPcom a), VLPcom, reas b), and VLPcom, res c).

Although the similar diffuse reflectance spectra of VLPcom and VLPcom, reas

suggested that the re-assembling reaction between VLPcom, res and extract SENSorg has

reformed VLPcom, XPS experiments, photoelectrochemical measurements, and

photocatalytic activity tests were performed to prove or disprove this assumption.

XPS analysis of VLPcom, reas and VLPcom revealed the presence of three peaks in the

range of C1s binding energies (See Figure 7.3 A and B). Curve fitting analysis affords

almost identical values of 284.8, 286.5, 288.4 eV and 284.8, 286.3, 288.8 eV for

VLPcom, reas and VLPcom, respectively. The only significant difference is the higher

intensity of the 286.5 eV peak in VLPcom, reas. This excellent agreement further

supports the success of the re-assembling reaction. Peaks at 284.8 eV are assumed to

arise predominantly from adventitious carbon, although aromatic and graphitic carbon

have similar binding energies in the range of 284.3-284.7 eV.[90, 99-101] Peaks at the

higher binding energies may originate from carbonate or carboxylate groups. Presence

of the former seems unlikely since in the IR spectrum no corresponding absorption can

be observed at 1720-1740 cm−1; it is known that carbonate does not induce visible

light activity.[31, 54] More likely is that these peaks arise from two types of

carboxylates. From the fact that benzoate bridging two titanium centers exhibits

a binding energy of 288.6 eV[90], the peaks 288.4 and 288.6 eV can be assigned to an


51

arylcarboxylate group. The remaining peaks at 286.5 and 286.3 eV are tentatively

assigned to a bidentately-bound arylcarboxylate, which should have a slightly lower

C1s binding energy. A symmetrical surface binding of the arylcarboxylate is also

supported by the O1s binding energies. The two peaks at 530.0 and 531.5 eV can be

assigned to oxygen atoms in titania and in the coordinated carboxylate, respectively.

Corresponding values of 529.9 and 531.3[106]/531.5 eV[90] were reported for benzoic

acid coordination.


52

292 290 288 286 284 282 2800

1x103

2x103

3x103

4x103

Inte

nsity

/ a.

u.

Binding energy / eV

ab

c

A

292 290 288 286 284 282 2800

1x103

2x103

3x103

4x103

Inte

nsity

/ a.u

. a

bc

Binding energy / eV

B

Figure 7.3: XPS C1s spectra of VLPcom, reas (A: 284.8 (a), 286.5 (b), 288.4 (c) eV) and VLPcom (B: 284.8 (a), 286.3 (b), 288.8 (c) eV).

Although both diffuse reflectance and XPS data further corroborate the similarity

of VLPcom and VLPcom, reas, it is at least of equal importance comparing also their

photoelectrochemical properties. This was performed by measuring the quasi-Fermi

level of electrons (nEF*) for all three powders through recording the photovoltage of

a powder suspension as a function of pH value (See Figure 7.4).[73] Again VLPcom and


53

VLPcom, reas exhibited the same result, which was in this case one unique nEF* value of

−0.50 V (NHE), whereas −0.56 V was obtained for VLPcom, res at pH 7. The latter value

suggests that the powder remaining after extraction consists of unmodified anatase, in

accordance with the XRD data, whereas the anodically shifted value is significant for

carbon-doped titania.[46] Assuming that the energy difference between Fermi level and

conduction band edge is negligible for these n-type powders, the valence band edge

potentials of VLPcom and VLPcom, reas were obtained as 2.72 and 2.71 V, respectively,

by adding the bandgap energy to the quasi-Fermi energy.

2 3 4 5 6 7 8 9

-0.2

0.0

0.2

0.4

0.6

VpH

/ V v

s. A

g/A

gCl

pH

a

bc

Figure 7.4: Variation of photovoltage with pH value for the suspension of various photocatalystsin the presence of methylviologen dichloride: VLPcom a), VLPcom, reas b), and VLPcom, res

c). Dashed lines indicate pH values of inflection points from which the quasi-Fermi level can be calculated.36

The results presented above suggest that the alkali extract SENSorg contains

carboxylic groups, which during the re-assembling reaction undergo esterification with

titania surface hydroxyl groups. This is corroborated by IR analysis of the brown

residue obtained from SENSorg after the removal of water from the basic solution (pH

12) (See Figure 7.5). The sample exhibited intense peaks at 1580 and 1420 cm−1,

assignable to asymmetric and symmetric stretching vibrations of an arylcarboxylate

group. Corresponding values of e.g. free sodium benzoate are 1552/1414 cm−1 (KBr).

Unfortunately no corresponding peaks could be obtained in the IR spectra of VLPcom


54

and VLPcom, reas, both measured conventionally or by the ATR technique. This failing

suggests that the sensitizer is present as thin surface layer of too low concentration to

be detected by IR spectroscopy. The fact that the significant carbon peaks at 286.3 eV

and 288.8 eV in the XPS spectrum completely disappear after sputtering with argon

for 3 min, what is sufficient to remove a 4 nm thin surface layer, corroborates this

explanation.

4000 3500 3000 2500 2000 1500 1000 5000

10

20

30

40

50

60

70

Tran

smitt

ance

/ %

Wavenumber / cm-1

Figure 7.5: IR spectrum (KBr) of solid residue of SENSorg.

The excellent agreement between the physical properties of VLPcom and VLPcom,

reas as reported above, suggests that also the photocatalytic activities may be very

similar. In fact, visible light photodegradation experiments with 4-CP revealed that

after 3 h irradiation time VLPcom and VLPcom, reas induced mineralizations of 70% and

77%, respectively. Contrary to this, VLPcom, res exhibited only 18% mineralization (See

Figure 7.6). To confirm that VLPcom contains an organic sensitizer, this material was

calcined at 400°C for 3 h affording white VLPcom, calc containing 0.09% C. It induced

only 13% mineralization. This fact points to almost complete oxidation of the organic

sensitizer.


55

0 60 120 180

20

40

60

80

100

[TO

C]/[

TOC

] 0/ %

time / min

ab

cd

Figure 7.6: Photomineralization of 4-CP with visible light (㮰 ≥ 455 nm) in the presence of VLPcom

(a), VLPcom, reas (b), VLPcom, res (c), and SENSorg (d). Suspensions (a) and (b) exhibited pH ~ 5.8, whereas suspension (c) and solution (d) needed addition of acid to obtain this pH value.

It is noted that the photodegradation rate strongly depends on the pH value at

which the re-assembling reaction was conducted. At 3 h of irradiation time the VLPcom,

reas prepared at pH ~ 12 afforded only 13% mineralization, whereas 44% were obtained

at pH ~ 7. It is also noteworthy that suspensions of VLPcom and VLPcom, reas in 4-CP

exhibit the same pH value of 5.8 suggesting a similar surface structure.

To investigate if VLPcom can induce mineralization of other organic pollutants, also

diethylphthalate (DEP) and dichloroacetic acid (DCE) were investigated.

Unfortunately, VLPcom did not exhibit any activity in mineralization of DEP under

standard irradiation conditions. Only 10% mineralization was noticed for DCE.

When the re-assembling reaction was performed with the titania powder P25, the

resulting material P25-C under standard irradiation conditions induced 30%

mineralization after 3 h as also observed for unmodified P25.

Additionally, self-prepared TiO2 was modified with Sensorg according to the re-

assembled reaction (See Chapter 7.2) affording slightly yellow TiO2-C inducing

ca. 50% activity in Vis light mineralization of 4-CP. Contrary to this, unmodified


56

titania did not show any activity. This result proves that the extracted Sensorg contains

indeed carbon species rendering TiO2 and VLPcom, res active.

To address the problem of sensitizer photostability, an eighteen hour

photomineralization of 4-CP in the presence of VLPcom was performed in an

immersion lamp apparatus. After every third hour the irradiation was interrupted for

the TOC measurement followed by addition of a new portion of 4-CP solution and

subsequent TOC determination (See Figure 7.7). Whereas after the second 4-CP

addition the reaction rate was decreased by only 20%, it amounted 60% for the third to

fifth addition, and 70% for the sixth addition. During the experiment the colour of

VLPcom changed from beige to slightly yellow suggesting accumulation of coloured

reaction intermediates, as also indicated by its increased of carbon content (See

Chapter 7.2). Both competitive light absorption by the intermediates and

photodegradation of the sensitizer may rationalize the observed rate decrease.

However, a considerable amount of the sensitizer has survived as indicated by the

following experiment. Alkaline treatment of the VLPcom photocatalyst obtained after

the final sixth degradation step afforded a whitish powder which induced only 15%

mineralization after 3 h irradiation time (㮰 ≥ 420 nm).

0 3 6 9 12 15 18

6

8

10

12

14

16

18

20

22

time / min

[TO

C]/

mg

⋅L−1

Figure 7.7: Decrease of TOC value upon addition of 4-CP after every third hour (dashed lines) during Vis irradiation of VLPcom (㮰 ≥ 400 nm). See text.


57

This suggests that the brownish extract contained the sensitizer as also evidenced by

the congruence of its UV-Vis spectrum with that of SENSorg. It exhibits absorption

shoulders at 460 and 260 nm (See Figure 7.8).

To exclude a mutual homogeneous photosensitized degradation a blank experiment

was carried out employing only the extract SENSorg.

200 400 600 800 10000.0

0.5

1.0

1.5

2.0

Abs

orba

nce

λ / nm

Figure 7.8: Absorption spectrum of the as-obtained extract solution SENSorg diluted by 1:8.

It did not induce any photomineralization of 4-CP (See Figure 7.6 d). To check the

possibility if oxygen defects may also induce visible light activity, VLPcom, res was

heated in vacuo for 1 h at 150°C. Under the standard experimental conditions this

material enabled within 3 h a 4-CP mineralization of 10%.

7.4 Conclusions

The results presented above indicate that the commercial visible light photocatalyst

VLPcom contains an organic sensitizer which can be extracted with alkaline water. Both

the extract and the remaining solid do not exhibit significant activity in the

photomineralization of 4-chlorophenol. However, when these two components are re-

assembled, the activity, quasi-Fermi level, diffuse reflectance spectrum, and XPS data

are almost identical with those of original VLPcom. Thus, the visible light activity of


58

VLPcom and probably of many previously reported so-called "C-doped” titania

powders is not due the presence of lattice carbon atoms or oxygen defects as generally

reported. However, this finding does not exclude the presence of defects in VLPcom but

there is no experimental evidence that they are involved in the visible light induced

charge separation at the sensitizer-modified anatase surface.

8. Polyol derived carbon-modified titania

59

8. Polyol derived carbon-modified titania∗

8.1 Introduction

It is noted that all the proposals for carbon species (See Chapter 2.1 and 7.1) being

responsible for Vis light photooxidations of various pollutants were based on rather

few experimental evidence; mainly on physical data like ambiguous carbon 1s binding

energies. Contrary to that, we thought that chemical data like stability against acids

and bases may afford more valuable information.

This was confirmed in Chapter 7 by finding that the beige carbon-modified

commercial titania visible light photocatalyst VLPcom contains structurally unknown

aromatic hydrocarbon compounds responsible for Vis light activity.[98] Surprisingly,

treating VLPcom with NaOH afforded a dark brown soluble organic photosensitizer

(Sensorg). The almost colourless residue VLPcom,res induced only 18% mineralization of

4-chlorophenol (4-CP), in comparison with 70% as observed for VLPcom. After re-

assembling Sensorg with VLPcom,res, the resulting powder VLPcom,reas had the same

quasi-Fermi level of −0.50 V (vs NHE) and the same Vis light activity in 4-CP

mineralization like pristine VLPcom. Contrary to this, the much less active residue

VLPcom,res exhibited a quasi-Fermi level of −0.56 V. From elemental analysis and the

presence of intense peaks at 1580 and 1420 cm-1 in the IR spectrum it was concluded

that SENSorg contains organic carboxylate groups. Standard analysis by

chromatography and NMR spectroscopy revealed that SENSex is a multi-component

mixture and all purification attempts failed[65].

However, the results presented above clearly indicate that the Vis light activity of

carbon-modified VLPcom is due to the presence of covalently attached organic

photosensitizers generated during the modification reaction conducted at 250-500°C.

The first step of the corresponding mechanism should be complexation of

pentaerythritol by under-coordinated titania surface atoms. Corresponding

∗ Most of the work presented in this chapter has been published as:Zabek, P.; Kisch, H. J. Coord. Chem., accepted.


60

coordination complexes between titania and various polyols are well known in the

literature.[107, 108] Subsequent aerial oxidation of the polyol is expected to produce

a mixture of carboxylic acids. Esterification of the latter with titania hydroxyl groups

eventually affords VLPcom. From this one expects that using ethylene glycol, an

alcohol simpler than the employed tetra-alcohol but still capable of forming stable

titania chelate complexes, may lead to a more selective modification reaction. This

may generate only one unique photosensitizer enabling structural characterization. In

the following we report on the preparation of the corresponding modified titania

powders, on their photoelectrochemical characterization, photocatalytic activity, and

on their stability against alkali and acid treatments. For the matter of comparison also

pentaerythritol was employed as a modifier. Ethylene glycol was often used as

a structural directing/modifying agent to obtain titania crystals of various

morphology.[109-115] Its action is based on the selective formation of labile titanium-

ethylene glycol coordination complexes. Only two papers dealing with polyols were

reported in the literature. In the first one, the modified sample was prepared by stirring

a water suspension of commercial TiO2 and glycerol at room temperature followed by

calcination at 300°C.[97] According to the second paper ethylene glycol and citric acid

were employed together with titanium isopropoxide in a special sol-gel method.[116]

8.2 Experimental

Preparation of titanium hydroxide and titanium dioxide: Titanium hydroxide and

titanium dioxide were prepared according to the preparation method described in

Chapter 3.2. White TiO2calc has a specific area of 68 m2g−1 and anatase structure as

indicated by XRD analysis.

Preparation of UVLP-EG and TiO2-EG:

UVLP-EG: 0.12 mL of ethylene glycol p.a. (Acros Organics) were diluted with

4.88 mL of doubly distilled water. Thereafter the solution was added to 1 g of the

commercial TiO2 powder UVLP 7500 (Kronos Titan GmbH; abbreviated in the

following as UVLP), followed by 30 min of sonication. After stirring for 24 h the

suspension was kept for 3 h at 120°C to evaporate water. The dried powder was


61

ground and then calcined in a rotating flask for 1 h at 250°C. To remove inorganic and

organic impurities the powder was washed three times by centrifuging for 15 min in

ca. 50 mL of doubly distilled water. Final drying at 100°C for 1 h afforded beige

UVLP-EG having a specific surface area of 175 m2∙g−1 and anatase structure as

indicated by XRD analysis.

TiO2-EG: 1 g of self-prepared titanium hydroxide was used instead of UVLP as

described above affording slightly yellow TiO2-EG having a specific surface area of

133 m2g−1 and anatase structure as indicated by XRD analysis.

Preparation of UVLPcalc-EG: As described above but UVLP was calcined before

use for 1 h at 250°C affording UVLPcalc (C 0.02, H 0.39%) having a specific surface

area of 112 m2∙g−1. After modification as described above, the creamy anatase powder

UVLPcalc-EG (C 0.67, H 0.38%) was obtained.

Preparation of UVLP-PE: 0.11 g of pentaerythritol (98%, Acros Organics) were

diluted in 5 mL of doubly distilled water. Thereafter the solution was added to 1 g of

UVLP followed by 30 min of sonication. After stirring for 24 h the suspension was

kept for 3 h at 120°C to evaporate water. The dried powder was ground and then

calcined in a rotating flask for 1 h at 250°C. To remove inorganic and organic

impurities the powder was washed as described above. Subsequent drying at 100°C for

1 h afforded brownish UVLP-PE (C 2.24, H 0.92%) having anatase structure.


62

Table 8.1: Elemental analyses, amounts of mineralization, and photoelectrochemical properties.

Catalyst % C % H % Mineralisation Ebg [eV] nEF* [V] EVB [V]

TiO2calc 0.07 0.56 20 3.06∗ − 0.59 2.47

TiO2-EG 0.76 0.96 68 3.04* − 0.58 2.46

TiO2-EGres 0.67 0.85 37 3.18 − 0.52 2.66

UVLP 0.04 1.14 60 3.22 − 0.50 2.72

UVLPcalc 0.02 0.39 21 3.21 − 0.53 2.68

UVLP-EG 1.05 0.76 63 3.19 − 0.52 2.67

UVLP-EGres 0.42 0.76 75 3.21 − 0.55 2.66

UVLP-EGreas 0.41 0.56 60 3.21 − 0.55 2.66

UVLP-PE 2.24 0.92 70 2.97* − 0.46 2.51

UVLP-PEres 0.49 0.90 83 3.16 − 0.50 2.66

Attempted alkaline desorption of photosensitizer: 5 g of beige UVLP-EG were

suspended in 50 mL of doubly distilled water and brought to pH 12 with 0.5 M NaOH.

Thereafter the suspension was heated for one hour at 90°C and kept overnight at RT

under vigorous stirring. A brown supernatant (Orgdes) and a beige powder were

obtained after centrifugation. The solid was extracted three times using the same

procedure and finally washed three times by centrifuging with doubly distilled water in

order to remove organic and inorganic impurities affording a beige residue UVLP-

EGres, which was dried one hour at 100°C. The experiment was repeated as described

above, employing only 0.2 g instead of 5 g of UVLP-EG.

As described above, but 0.5 g of UVLP-PE was used. UVLP-PE was extracted

only once. After washing and drying the beige residue UVLP-PEres was obtained.

As described above, but 0.3 g of TiO2-EG were used. TiO2-EG was extracted only

once. After washing and drying the slightly yellow residue TiO2-EGres was obtained.

∗ Apparent bandgap; due to the presence of a broad shoulder the extrapolation of the linear absorption part may contain a large error.


63

Attempted acidic desorption of photosensitizer: 0.5 g of UVLP-EG were

suspended in 50 mL of doubly distilled water and brought to pH ~ 2 with 1 M HNO3.

Thereafter the suspension was heated for one hour at 90°C and kept overnight at RT

under vigorous stirring. A colourless supernatant and beige powder were obtained after

centrifugation. The solid was washed three times with doubly distilled water and dried

for 1 h at 100°C affording UVLP-EGres,acid (C 0.90, H 0.60%). Repetition was

performed using 0.2 g instead of 0.5 g of UVLP-EG.

Attempted re-assembling reaction: 50 mL of the brown extract obtained from the

alkaline desorption experiment were acidified to pH ~ 3 with 1 M HCl. After adding of

1 g of UVLP-EGres the pH value of the suspension reached ~ 5.5. Therefore it was not

necessary to adjust again the pH by addition of 1 M HCl as reported recently in an

analogous experiment.[65] Thereafter the suspension was heated for one hour at 90°C

and stirred overnight at ambient temperature. The powder obtained after centrifugation

was dried at 100°C for 1 h, calcined in a rotating flask for 1 h at 250°C, washed as

described above, and dried again at 100°C for 1 h affording beige UVLP-EGreas.

Attempted Lewis acid catalyzed Ti-O-C cleavage:[117] 0.1 g of beige UVLP-EG

were suspended in 5 mL of CH2Cl2 p.a. (Fisher Scientific) and cooled in an ice bath.

Thereafter 0.3 mL of BBr3 (Acros Organics) were dropped to the reaction suspension

affording an intense yellow solid. After addition of BBr3 the reaction mixture was

heated 1 h at 80°C. Thereafter 10 mL of dichloromethane were added and the

suspension was kept overnight under stirring at ambient temperature. The rest of

solvent was removed by keeping for 1 h at 70°C. Thereafter the residue was washed

two times with dichloromethane and two times with doubly distilled water followed by

1 h drying at 70°C. The as-obtained material had a similar beige colour like UVLP-EG

before the BBr3 treatment. The reaction was repeated with an excess of boron

tribromide to exclude that only surface OH groups may have been reacting. In this

experiment 0.2 g of UVLP-EG in 20 mL of CH2Cl2 p.a. and 2 mL of BBr3 were used.

The obtained residue was washed two times with dichloromethane affording a yellow

supernatant, two times with methanol/water, and finally once with an ethanol/water

mixture affording in all cases colourless washings. After 1 h drying at 70°C a beige

material was obtained (C 0.98, H 1.34%).


64



3.3. All photocatalysts 4-CP suspensions in this chapter needed addition of 1 M HCl to

obtain a pH value of 3.5-4.0, the optimum condition for the degradation process.[96]

The concentration of catalysts was equal to 2.0 g∙L−1. Reproducibility of TOC values

was in the range of ±10%.

Photostability test: A suspension of 2.0 g∙L−1 of beige UVLP-EG in 190 mL of

2.5∙10−4 M aqueous 4-CP was sonicated for 15 min in a 200 mL immersion lamp

apparatus. After measuring the TOC value, the suspension was irradiated for 3 h with

the filtered light (1 M solution of sodium nitrite, λ ≥ 400 nm) of a tungsten halogen

lamp. Subsequently the TOC was again measured and new 4-CP solution was added.

This procedure was repeated five times resulting in overall six degradation runs.

Thereafter the slightly yellow powder was washed as described above affording

UVLP-EGres,washed (C 2.18, H 0.55%). When this material was employed in the

standard photomineralization of 4-CP, it induced within 3 h a decrease of the TOC

value by 48%.

Alkaline stability at room temperature: 1 g of TiO2-C1b (C 0.17, H 0.34%)[46],

VLPcom (C 0.36, H 0.40%)[65], UVLP-EG (C 1.05, H 0.76%), TiO2-EG (C 1.29, H

0.87%), and 0.5 g of UVLP-PE (C 2.24, H 0.92%) were suspended in 50 ml of doubly

distilled water and kept under stirring for 30 min. Thereafter the pH value of each

suspension was brought to ~ pH 13 by addition of 1 M NaOH and stirred overnight at

RT. The powders were washed three times with doubly distilled water by centrifuging

and then dried at 80 °C for 1 h.


65


8.3.1 Preparation and characterization of photocatalysts

Slightly yellow TiO2-EG, beige UVLP-EG, and brownish UVLP-PE were obtained

by impregnating titanium hydroxide and the commercial titania powder UVLP with

aqueous solutions of ethylene glycol and pentaerythritol, respectively, followed by

removal of water and subsequent calcination in an open rotating glass flask at 250°C.

Increasing the latter temperature to 350°C and 400°C decreased the visible light

(㮰 ≥ 455 nm) photomineralization rate of 4-chlorophenol (4-CP). Elemental analysis

revealed carbon contents of 0.76% for TiO2-EG, 1.05% for UVLP-EG, and 2.24% for

UVLP-PE (See Table 8.1 in Chapter 8.2). As indicated by XRD analysis of these

powders and their derivatives (vide infra) they all are present in the anatase phase.

Crystallite sizes of 15, 7, 13, 15, 17, 21, and 15 nm were obtained for TiO2calc, TiO2-

EG, UVLPcalc, UVLP-EG, UVLP-EGres, UVLP-EGreas, and UVLP-PE, respectively,

through application of the Scherrer Equation to the diffraction peak at 25.3°. For the

matter of comparison UVLP was calcined at 250°C in the absence of any precursor

affording UVLPcalc containing 0.02% carbon according to elemental analysis.

XPS spectra of UVLPcalc, UVLP-EG, and UVLP-PE were measured in the range of

C1s binding energies. Curve fitting analysis reveals best results when the broad signal

of UVLPcalc was decomposed into four peaks located at 284.8, 285.6, 286.8, and 289.0

eV. After argon ion sputtering for three minutes, only a very weak signal is remaining

at 286.0 eV. Unfortunately the same result was obtained for the modified sample

UVLP-EG. Only in the case of UVLP-PE the intensity of the peak at 285.6 eV

observed after sputtering is a little higher. In view of these results no firm conclusion

on the nature of the carbon species is possible. It is noted that due to the relative large

errors of up to 0.4 eV occurring in measurements of such binding energies[118], many

assignments made in the literature for various carbon species in titania appear as

questionable.

The diffuse reflectance spectra of TiO2-EG, UVLP-EG and UVLP-PE exhibit

a significant absorption in the visible (Figure 8.1 b, d, and e, respectively), which is


66

absent in the case of TiO2calc and UVLPcalc (Figure 8.1 a, c). The stronger Vis light

absorption of UVLP-PE (Figure 8.1 e) as compared to UVLP-EG (Figure 8.1 d) is in

accord with its much higher carbon content. Both UVLP-EGres and UVLP-EGreas

exhibit very weak almost negligible absorption in the visible. TiO2-EGres and UVLP-

PEres show less intense absorption shoulders than respective outgoing materials.

400 500 600 700 8000.00

0.01

0.02

0.03

0.04

edc

ba

F(R

∞)/

a.u

.

λ / nm

Figure 8.1: Diffuse reflectance spectra of TiO2calc∗ (a), TiO2-EG (b), UVLPcalc (c), UVLP-EG (d),

and UVLP-PE (e).

∗ 100 mg of TiO2 calc were taken to record the spectrum (See Chapter 3.1)


67

1 2 3 4 50.0

0.5

1.0

1.5

2.0

edc

b

a

(F(R

∞)E

)1/2 / a

.u.

E / eV

Figure 8.2: Plot of transformed Kubelka-Munk function versus the energy of the light absorbed ofTiO2calc (a), TiO2-EG (b), UVLPcalc (c), UVLP-EG (d), and UVLP-PE (e).

Assuming that TiO2calc, TiO2-EG, UVLPcalc, UVLP-EG, and UVLP-PE, are

indirect semiconductors like anatase, optical bandgaps of 3.06, 3.04, 3.21, 3.19, and

2.97 eV (Figure 8.2), respectively, were obtained for these materials from a plot of

transformed Kubelka-Munk function vs. the energy of the absorbed light.[104, 105] Thus,

carbon-modification does not significantly change the bandgap as also recently

observed for commercially available VLPcom[65] (See also Chapter 7.3).

Measurement of the quasi-Fermi potentials by recording the photovoltage of the

semiconductor suspension as function of pH value[73] afforded −0.58, −0.52, and −0.46

V for TiO2-EG, UVLP-EG, and UVLP-PE at pH 7, respectively. The value of UVLP-

PE is more positive by 0.04 V as compared to VLPcom suggesting the presence of

different sensitizers in the two photocatalysts. TiO2-EGres, UVLP-EGres, and UVLP-

PEres gave rise to a quasi-Fermi level of −0.52, −0.55 and −0.50 V, respectively.


68

2 3 4 5 6 7 8 9-0.4

-0.2

0.0

0.2

0.4

0.6

cbV p

H/ V

vs.

Ag/

AgC

l

pH

a

Figure 8.3: Variation of photovoltage with pH value for a suspension of UVLP-EG (a), UVLP-EGres (b), and UVLP-PE (c) in the presence of methylviologen dichloride. Solid (UVLP-EG), dashed (UVLP-EGres), and dotted lines (UVLP-PE) indicate the pH value of inflection points from which the quasi-Fermi level can be calculated.[73]

8.3.2 Photomineralization of 4-CP

The photocatalytic activity of the novel powders was tested in the mineralization

of 4-CP at pH 3.5-4.0, the optimum pH value.[96] Unless otherwise noted, all activities

refer to the TOC (Total Organic Content) value measured after irradiating for 3 h at

㮰 ≥ 455 nm. It is further noted, that due to the high catalyst concentration of

2 g∙L−1 the amount of light absorbed is expected to be about the same in all

experiments and therefore the reaction rates become comparable. The surprising

activity of 60% mineralization measured for the as received UVLP (C 0.04%) is due to

formation of a charge-transfer (CT) complex as already observed with other titania

samples like P25.[119-121] Since the modified powders were prepared by calcination of

UVLP in the presence of the carbon precursor, UVLP was analogously treated at

250°C in the absence of any additive in order to obtain the appropriate reference

material UVLPcalc. The fact that the latter contained only 0.02% C but induced 21% of

mineralization (Figure 8.4 a) is rationalized again by CT complex formation.

Similarly, TiO2calc (C 0.07%) exhibits an activity of 20%. Much higher values of 68%,

63%, and 70% were observed for TiO2-EG, UVLP-EG, and UVLP-PE, respectively


69

(Figure 8.4 b, d). The Table 8.1 in Chapter 8.2 summarizes chemical and physical

properties of the various photocatalysts.

0 90 1800

20

40

60

80

100

dcb

a[T

OC

]/[TO

C] 0

/ %

time / min

Figure 8.4: Photomineralization of 4-CP with visible light (㮰 ≥ 455 nm) in the presence of UVLPcalc (a), UVLP-EG (b), UVLP-EGres (c), and UVLP-PE (d).

8.3.3 On the nature of carbon sensitizer

As summarized in the introductory part (See Chapter 8.1) we recently found that

commercial VLPcom containes an alkali soluble organic photosensitizer that can be

easily desorbed.[65] To investigate if similar carbon species are present in the self-

prepared TiO2-EG, UVLP-EG, and UVLP-PE, the powders were subjected to the same

and to two new desorption procedures.

After boiling TiO2-EG at pH 12 a pale brown solution and slightly yellow residue

TiO2-EGres were obtained. The residue exhibited only 37% 4-CP mineralization as

compared to 68% found for pristine TiO2-EG. When UVLP-EG was analogously

treated, again a brown solution (Orgdes) and a beige residue (UVLP-EGres) were

obtained. Although UVLP-EGres contained only 0.42% C as compared to 1.05% C

found for UVLP-EG, its mineralization activity of 75% was even a little higher than

that of the starting material (63%, Figure 8.4 b). Performing the re-assembling reaction

gave rise to UVLP-EGreas having about the same activity as UVLP-EGres. In accord

with this surprising result is the finding that both powders exhibited the same quasi-


70

Fermi level of −0.55 V and the same carbon content of 0.42%. Thus, different from

VLPcom (C 0.36%)[65], the photosensitizer cannot be desorbed from the titania surface

through an alkaline treatment. But it could be removed by treating UVLP-EG in the

presence of air at 400°C affording UVLP-EG400 (C 0.11%) which displayed no visible

light absorption and induced only 9% mineralization, probably due to CT

complexation of 4-CP (vide supra). The brown colour of the desorbed species Orgdes

therefore is not related to the photosensitizer responsible for the visible light activity of

UVLP-EG but to an impurity, the removal of which leads to a more efficient

photocatalyst. Accordingly, different from VLPcom, where the Fermi level of the

residue is shifted slightly cathodically by 0.06 V[65], it stays constant in the case of

UVLP-EG. In agreement with the conclusion that the sensitizer could not be desorbed

is also the difference in emission spectra of the two extracts. Whereas the brown

solution of Sensorg as obtained from VLPcom exhibited two maxima at 380 and 400 nm

and shoulders at 428 and 488 nm (Figure 8.5 b), brown Orgdes extracted from UVLP-

EG showed only one maximum at 428 nm and a shoulder at 488 nm (Figure 8.5 c).

The fact that the maximum of Sensorg at 380 nm is absent in the spectrum of Orgdes,

suggests that this signal originates from the photosensitizer.

200 300 400 500 600 7000

10

20

30

40

c

b

a

Inte

nsity

/ a.u

.

λ / nm

Figure 8.5: Emission spectra of H2O employed (a), Sensorg from VLPcom at pH = 12 (b), and Orgdes

from UVLP-EG at pH = 12 (c). 㮰exc = 254 nm. Dilution factor; 1 : 1.


71

Surprisingly, in contrast to VLPcom also the activity of the self-prepared

pentaerythritol derived photocatalyst UVLP-PE did not change after a corresponding

alkaline desorption treatment. This may be caused by differing reaction conditions and

slightly different titania starting materials. As mentioned above the density of surface

hydroxyl groups should significantly influence the structure and reactivity of the

initially generated titanium-polyol surface complexes. It seems unlikely that this

surface property is identical for the two titania powders. Furthermore the amount of

aerial oxygen available for generation of the carboxylic acid strongly depends on

reactor geometry and other technical details. It is very unlikely that they are the same

in the commercial and home-made reactor.

The inertness of UVLP-EG and UVLP-PE against alkali treatments suggests that

the organic sensitizer is not bound to the titania surface through an Ti-O-C(O)Ar group

as proposed for VLPcom.[65] An alternative binding could occur via a Ti-OAr bond,

more susceptible to an electrophilic cleavage. Therefore UVLP-EG was refluxed in

acidic solution of pH 2. However, no significant decrease of activity was observable

for the obtained solid residue containing 0.90% C. Even a vigorous treatment with the

strong Lewis acid BBr3[117] was not successful. The remaining beige residue (C 0.98%)

had about the same activity as the starting material.

These results suggest that the mutual Ti-OAr fragment is of unusual inertness or

that the organic sensitizer forms a Ti-C bond stable under the applied desorption

conditions. Eventually, it may be insoluble under alkaline or acidic conditions and not

chemically bound to the titania surface. This seems rather unlikely since the removal

of surface OH groups by precalcining UVLP at 250°C before the treatment with

ethylene glycol decreased the amount of mineralization to 27%, indicating that these

groups are involved in the modification reaction.

To address the problem of sensitizer photostability, an eighteen hour

photomineralization of 4-CP in the presence of UVLP-EG was performed in an

immersion lamp apparatus (Figure 8.6). After every third hour the irradiation was

interrupted for the TOC measurements followed by addition of a new portion of 4-CP

solution and subsequent irradiation and TOC determination. Whereas in the first cycle


72

64% of the pollutant was mineralized, only 28% were found after the second 4-CP

addition. This decrease continued and the oxidation completely ceased after the fifth

addition suggesting photocorrosion of the photocatalyst. Most likely this is due to

accumulation of coloured intermediates preventing light absorption by the

photocatalyst as suggested by a colour change from beige to yellow. Accordingly,

when the residue after the sixth degradation cycle was washed with water,

mineralization increased to 48% (under standard irradiation conditions).

0 3 6 9 12 15 1802468

10121416182022

[TO

C]/

mg

⋅L−1

time / min

Figure 8.6: Decrease of TOC value upon addition of 4-CP after every third hour (dashed lines) during Vis light irradiation of UVLP- EG (㮰 ≥ 400 nm).


73

8.3.4 Alkaline stability at room temperature

Table 8.2: List of the samples treated with 1M NaOH at room temperature.

Mineralization %∗1 Carbon content %before after before afterSample

alkaline treatmentTiO2-C1b∗2 72 48 0.17 0.15VLPcom 47 43 0.36 0.24UVLP-PE 70 66 2.24 0.72UVLP-EG 63 72 1.05 0.61TiO2-EG 68 60 1.29 1.02

To control if the various carbon-modified materials are stable under alkaline

conditions the photocatalysts were treated at pH 13 at RT (See Chapter 8.2). After this

treatment the mineralization of VLPcom (C 0.36%) of 47% was slightly decreased to

43% for creamy VLPcom, alk (C 0.24%). Almost no decrease (~ 4%) was noticed for

brownish UVLP-PE (C 2.24%), which after treatment in NaOH afforded slightly

brownish UVLP-PEalk (C 0.72%). For slightly yellow TiO2-EG (C 1.29%) and beige

UVLP-EG (C 1.05%) only ~ 8% decrease and ~ 9% increase of mineralization were

observed, respectively. This slightly increase of activity of UVLP-EG corroborates the

previous suggestion that organic impurities can be removed through the alkaline

treatment both at RT and 90°C. The slightly yellow TiO2-EGalk (C 1.02%) and creamy

UVLP-EGalk (C 0.61%) were obtained. Contrary to those results, a significant decrease

of ~ 24% mineralization was noticed for yellowish TiO2-C1b (C 0.17%), which as

only one was not prepared by surface modification. Alkaline treatment afforded

creamy TiO2-C1balk (C 0.15%).

∗1 All the suspensions needed addition of acid to obtain the optimum pH for mineralization of 4-CP in the range of 3.5-4.0.∗2 Prepared by hydrolysis of titanium tetrachloride with tetrabutylammonium hydroxide.[46]


74

8.4 Conclusions

Slightly yellow TiO2-EG, beige UVLP-EG, and brownish UVLP-PE carbon-

modified anatase are prepared by calcining titanium hydroxide or the commercial

titania powder UVLP at 250°C in the presence of ethylene glycol or pentaerythritol.

These surface-modified titania materials photocatalyze the visible light mineralization

of 4-chlorophenol. Removal of surface hydroxyl groups from titania before the

modification process results in a photocatalyst of much lower photocatalytic activity,

suggesting that these groups are involved in the reaction with the polyol. Different

from the pentaerythritol derived commercial photocatalyst VLPcom, self-prepared

UVLP-PE and ethylene glycol derived UVLP-EG contain a sensitizer which cannot be

desorbed from the titania surface either under severe alkaline or acidic reaction

conditions. Whereas no significant difference is observable between the diffuse

reflectance spectra, quasi-Fermi levels of −0.58, −0.52 and −0.46 V are observed for

TiO2-EG, UVLP-EG, and UVLP-PE, respectively. However, when UVLP-EG is

treated at 400°C in the presence of air, the visible light absorption disappears and

activity strongly decreases, pointing to oxidation of a carbon sensitizer. Surprisingly,

TiO2-EG as prepared from titanium hydroxide loses a significant part of its activity

after alkaline treatment. This emphasizes an important role of the nature of the titania

precursor on the stability of carbon-modified titania photocatalysts. Additionally, the

modified powders were treated under severe alkaline conditions at room temperature.

Except TiO2-C1b prepared not from TiO2, but from TiCl4, all investigated powders

were stable under these conditions. An increase of activity of 9% was noticed for

UVLP-EG suggesting removing of organic impurities during the alkaline treatment.

9. Wood shavings as renewable modifiers

75


9.1 Introduction

Wood shavings described in this chapter consist of ca. 50% of cellulose,

a polysaccharide with 㬠(1,4)-linked D-glucose units. Up to date, only a few papers

reported about glucose applied as an organic titania modifier. Khan et al.[41] obtained

modified titania by hydrolysis of TiCl4 in the presence of glucose, followed by 150 h

aging at RT and 5 min calcination at 500°C. The resulting samples exhibited ca. 60%

4-CP degradation under Vis light (㮰 ≥ 380 nm). Ren et al.[42] calcined the reaction

mixture of amorphous titania and glucose at 160°C for 12 h. In 2009 Wang et al.[43]

treated the mixture of (NH4)2TiF6 and glucose at 160°C affording carbonaceous

polysaccharide microspheres transformed after rapid combustion at 500°C to “carbon-

doped” TiO2 hollow spheres. The visible light activity for the degradation of gaseous

toluene (㮰 ≥ 425 nm) of ca. 35% was reported for the best sample. Lin et al.[44]

impregnated titania with aqueous solutions of saccharose followed by calcination at

high temperatures in a steam of N2. The most structurally similar carbon source to our

wood shavings was employed by Nagaoka et al.[122] The cellulose/TiO2 microspheres

composite were made by using an aqueous solution of cellulose xantate and sodium

polyacrylate, wherein P25 was dispersed. Thereafter, cellulose/TiO2 was calcined at

600°C for 5h affording carbon/TiO2 microsphere composites. Unfortunately, the

photocatalytic decomposition of acetaldehyde was performed under UV irradiation in

the range of 300-400 nm. Up to date, there were no simple methods proposed for

modification of titania using wood shavings as modifier.

9.2 Experimental∗

Preparation of titania modified with wood shavings from larch: Wood shavings

(C 47.12, H 5.95%) were ground by planetary ball milling for 15 min (with

a frequency of 20.0 Hz). Thereafter, 50, 106, or 212 mg of wood shavings were milled

∗ Detailed information on the optimalization is presented in Table 9.1


76

again with 1 g of commercial TiO2 – UVLP7500 (Kronos Titan GmbH; abbreviated in

the following as UVLP) for 15 min, followed by calcination in a rotating flask at

250°C for 1 h. After finding the optimum mass of larch as 106 mg the calcination was

performed also at 350 and 400°C. To remove inorganic and organic impurities the

powders were washed three times by centrifuging for 15 min in ca. 50 mL of doubly

distilled water. Final drying at 70°C for 1 h afforded pale brown materials (abbreviated

in the following as UVLP-L).

Alternatively, 106 mg of wood shavings were milled with 1 g of UVLP for 15 min.

Thereafter, such prepared mixture was suspended in 20 mL of doubly distilled water.

After stirring for 24 h, the suspension was kept for ca. 20 min at 110°C under vacuo

with stirring, followed by calcination of the obtained powder in a rotating flask at

250°C for 1 h. To remove inorganic and organic impurities the powder was washed

three times by centrifuging for 15 min in ca. 50 mL of doubly distilled water. Final

drying at 70°C for 1 h afforded pale brown UVLP-Lsol.

UVLP-LAr was prepared by blowing argon during the calcination at 250°C for 1h

in a rotating flask. The milling procedure was the same as described above.

Attempted alkaline desorption of photosensitizer: The alkaline desorption was

performed according to the procedure presented in Chapter 8.2. 0.5 g of UVLP-L were

extracted only once affording a creamy residue UVLP-Lres (C 1.12, H 0.59%).

Alkaline stability at room temperature: The alkaline stability test was performed

according to the procedure presented in Chapter 8.2. The pale brown UVLP-Lalk

(C 1.63, H 0.72%) was obtained.



3.3. All photocatalysts 4-CP suspensions in this chapter needed addition of 1 M HCl to

obtain a pH value of 3.5-4.0, the optimum condition for the degradation process.[96]

The concentration of catalysts was equal to 2.0 g∙L−1. Reproducibility of TOC values

was in the range of ±10%.


77

9.3 Results and Discussion

9.3.1 Preparation and characterization of photocatalysts

Pale brown UVLP-L powders were obtained by milling of UVLP with wood

shavings. The milled mixture of titania and wood shaving was calcined at 250°C (350

or 400°C). XRD analysis of these powders revealed that they consist of the anatase

phase. Crystallite sizes of 13 and 17 nm were obtained for UVLPcalc and UVLP-L,

respectively, through application of the Scherrer Equation to the peak at 25.3°. As

already presented in Chapter 8.3.1 UVLP was calcined at 250°C in the absence of any

precursor affording reference UVLPcalc containing 0.02% carbon according to

elemental analysis.

The diffuse reflectance spectrum of UVLP-L exhibits a weak absorption in the

visible (up to ca. 550 nm) (Figure 9.1 b), which is absent in the case of UVLPcalc

(Figure 9.1 a). Very weak almost neglected absorption is observed in the spectra of

UVLP-L-350 and UVLP-L-400 (Figure 9.1 c, d), calcined at 350 and 400°C,

respectively. Elemental analysis revealed carbon contents of 2.49% for UVLP-L,

1.60% for UVLP-L-350, and 1.06% for UVLP-L-400.

400 500 600 700 8000.000

0.005

0.010

0.015

0.020

F(R

∞)/

a.u

.

λ / nm

a

b

cd

Figure 9.1: Diffuse reflectance spectra of UVLPcalc (solid line) (a), UVLP-L (dashed line) (b), UVLP-L-350 (dot line) (c), and UVLP-L-400 (dash dot line) (d).


78

The absorption becomes more intense with increasing the mass of larch, resulting

in an increase of the carbon content. The strongest absorption (up to ca. 700 nm) is

present for the sample obtained by doping with 212 mg of larch – UVLP-L212 (Figure

9.2 d) with carbon content of 5.91%. Contrary to this UVLP-L (C 2.49%) and UVLP-

L50 (C 1.35%) show weak absorption shoulder, up to ca. 550 nm (Figure 9.2 b, c).

400 500 600 700 8000.000

0.005

0.010

0.015

0.020

dc

b

a

F(R

∞)/

a.u

.

λ / nm

Figure 9.2: Diffuse reflectance spectra of UVLPcalc (solid line) (a), UVLP-L (dashed line) (b), UVLP-L50 (dot line) (c), and UVLP-L212 (dash dot line) (d).

The strongest absorption up to ca. 700 nm was observed for brown UVLP-LAr

(C 5.61%) prepared by blowing argon during calcination at 250°C (Figure 9.3 d).

Contrary to this, UVLP-L (C 2.49%) prepared by milling and UVLP-Lsol (C 2.64%)

obtained from solution exhibit weaker absorption, up to ca. 550 nm (Figure 9.3 b, c).

Only the white material UVLP-L400 (C 0.08%) obtained after again calcination of

UVLP-L at 400°C for 3 h does not show any absorption in visible (Figure 9.3 e), the

same as pristine UVLPcalc (C 0.02%) (Figure 9.3 a).


79

400 500 600 700 8000.000

0.005

0.010

0.015

0.020

ed

cb

a

F(R

∞)/

a.u

.

λ / nm

Figure 9.3: Diffuse reflectance spectra of UVLPcalc (solid line) (a), UVLP-L (dashed line) (b), UVLP-Lsol (dot line) (c), UVLP-LAr (dash dot line) (d), and UVLP-L400 (dash dot dot line) (e).

Refluxing of UVLP-L (C 2.49%) at pH 12 at 90°C afforded the residue UVLP-Lres

(C 1.12%), which exhibited an almost identical absorption in visible as compared to

the starting material (Figure 9.4 a, b). This correlates well with the similar activities of

these powders of 34-38% and 27%, respectively. The significant decrease of carbon

content may be explained by removal of organic impurities during desorption

procedure, which do not absorb in visible. Surprisingly, UVLP-Lalk (C 1.63%), as

obtained by stirring of UVLP-L at pH 13 at RT exhibits a strong absorption shoulder,

more intense than that of the starting material (Figure 9.4 c).


80

400 500 600 700 8000.000

0.005

0.010

0.015

0.020

F(R

∞)/

a.u

.

λ / nm

a, bc

Figure 9.4: Diffuse reflectance spectra of UVLP-L (a), UVLP-Lres (b), and UVLP-Lalk (c).

1 2 3 4 50

1

2

b

(F(R

∞)E

)1/2 / a

.u.

E / eV

a

Figure 9.5: Plot of transformed Kubelka-Munk function versus the energy of the light absorbed of UVLPcalc (a), and UVLP-L (b).

Assuming that UVLPcalc, UVLP-L, and UVLP-Lres are indirect semiconductors like

anatase, optical bandgaps of 3.21, 3.20 (Figure 9.5, vide supra), and 3.21 eV,

respectively, were obtained for these materials from a plot of transformed Kubelka-

Munk function vs. the energy of the absorbed light.[104, 105] Thus, modification with

larch does not significantly change the bandgap as also recently observed for


81

commercially available VLPcom[65] (See Chapter 7), UVLP-EG and UVLP-PE (See

Chapter 8).

Measurement of the quasi-Fermi potentials of UVLPcalc and UVLP-L by recording

the photovoltage of the semiconductor suspension as function of pH value[73] afforded

−0.53 and −0.52 V (NHE) at pH 7 , respectively (Figure 9.6 a, b). UVLP-Lres gave rise

to a value of −0.53 V. The latter two results are comparable with the values of −0.52 V

and −0.55 V noticed for UVLP-EG and UVLP-EGres, respectively.

2 3 4 5 6 7 8 9

-0.4

-0.2

0.0

0.2

0.4

0.6

V pH

/ V v

s. A

g/A

gCl

pH

a

b

Figure 9.6: Variation of photovoltage with pH value for a suspension of UVLPcalc (a), and UVLP-L (b) in the presence of methylviologen dichloride. Solid (UVLPcalc) and dashed (UVLP-L) indicate the pH value of inflection points from which the quasi-Fermi level can be calculated.[73]

9.3.2 Photomineralization of 4-CP∗

The photocatalytic activity of the novel powders was tested in the mineralization

of 4-chlorophenol (4-CP) at pH 3.5-4.0, the optimum pH value[96], as already found in

Chapter 8.3.2. Unfortunately, the powders suspensions were not completely

homogeneous. Dark brown, almost black particles, probably of coke or not completely

∗ Detailed information to this chapter concerning elemental analysis and Vis light activities is presented in Table 9.1. Sample nomenclature was defined as follows: UVLP-L-XXX means, that sample was prepared with 106 mg of larch and calcined at XXX°C, given from the table. The


82

burnt larch were visible in the 4-CP suspensions after sonication. Whereas the

reference UVLPcalc induced only 21% of mineralization (probably because of CT

complex formation, as explained in Chapter 8.3.2), values of 37, 34-38, and 37% were

observed for UVLP-L50, UVLP-L, and UVLP-L212, respectively. 50, 106 and 212

mg of larch were used for modification with pristine UVLP at 250°C, respectively

(vide supra). The optimum mass of modifier (larch) was found as 106 mg. Calcination

of UVLP with 106 mg of larch at 350 and 400°C, afforded pale brown UVLP-L-350

and UVLP-L-400 with 27 and 29% of mineralization, respectively. Samples prepared

from water suspensions exhibited 35% of mineralization. Performing the calcination

under argon did not improve photocatalytic activity. The brown UVLP-LAr induced

only 27% of mineralization.

In the previous chapter we have reported about carbon-modified titania prepared

by calcining UVLP at 250°C in the presence of ethylene glycol and pentaerythritol. It

was found that UVLP-PE and UVLP-EG contain an organic sensitizer which could not

be desorbed from the titania surface either under severe alkaline or acidic reaction

conditions. After boiling pale brown UVLP-L at pH 12 (according to the procedure

presented in Chapter 8.2) a brown solution (ORGdes) and a creamy residue (UVLP-

Lres) were obtained. Although UVLP-Lres contained only 1.12% C as compared to

2.49% found for the starting UVLP-L, its activity of 27% was only a little lower in

comparison with 34-38% noticed for the starting material.

When UVLP-L was treated at pH 13 at RT the resulting pale brown UVLP-Lalk

induced 32% of mineralization. Thus, the carbon species responsible for visible light

activity cannot be desorbed from the surface of larch modified titania through the

alkaline treatment either at room or 90°C temperature.

When UVLP-L was calcined at 400°C for 3 h in the presence of air (analogously

as ready materials from Chapters 7 and 8), the activity completely decreased and

visible light absorption disappeared (Figure 9.3 e), resulting in white UVLP-L400. This

optimum conditions (106 mg of larch, and 250°C as calcination temp.) are not given in the sample name (UVLP-L).


83

suggests oxidation of Vis light active carbon species, in accordance with the decrease

of carbon content from 2.49 to 0.08%.

Table 9.1: Colour, elemental analysis and 4-CP mineralization.

Sample Colour Elemental analysis%

Mineralization %

UVLP-L50 Pale brown C 1.35, H 0.00, S 0.00 37

UVLP-L Pale brown C 2.49, H 0.49, S 0.17 34-38

UVLP-L-350 Pale brown C 1.60, H 0.39, S 0.16 27

UVLP-L-400 Pale brown C 1.06, H 0.17, S 0.12 29

UVLP-Lsol Pale brown C 2.64, H 0.56, S 0.17 35

UVLP-LAr Brown C 5.61, H 1.28, S 0.00 27

UVLP-L212 Pale brown C 5.91, H 1.25, S 0.13

37

UVLP-L400 White C 0.08, H 0.20, S 0.19

No mineralization

9.4 Conclusions

The modification with larch as a natural modifier afforded powders, which

photocatalyze the visible light mineralization of 4-CP to about 35%. The suspensions

of modified samples are not homogeneous since black particles of coke or not

completely burnt larch can be observed. Thus, preparation of modified samples needs

to be improved. Similar to UVLP-EG and UVLP-PE (See Chapter 8), UVLP-L

contains an organic sensitizer which cannot be desorbed under alkaline reaction

conditions both at RT and at 90°C. However, when UVLP-L is treated at 400°C for

3 h, the visible light absorption disappears, suggesting oxidation of relevant organic

carbon species.

10. Summary

84

10. Summary

In this dissertation titania (both self-prepared and commercial) was modified with

different organic precursors to obtain highly efficient photocatalysts and to unravel the

nature of the “carbon” species responsible for visible light activity. Based on previous

knowledge, the “carbon” species may be a condensed aromatic system of high thermal

stability. Therefore first a modification with graphite/graphite oxides (G/GO) was

performed. Unfortunately, this modification did not afford active in Vis light (㮰 ≥ 455

nm) materials (Chapter 4). Subsequently organic acids were employed. This led to

powders with a little better activity as compared to graphite/graphite oxides modified

titania. Phthalic acid and trimellitic acid afforded photocatalysts inducing ca. 30% of

4-chlorophenol (4-CP) degradation. Contrary to this, self-prepared TiO2 was inactive

(Chapter 5). Finally, a little higher activity of ca. 40% 4-CP mineralization was

observed for titania modified with toluene. Therefore, further investigations with this

material should be continued (Chapter 6).

To understand the origin of visible light activity of carbon-modified titania the

commercial product VLPcom prepared by pentaerythritol modification was

investigated. The results indicated that VLPcom contains an organic sensitizer which

can be desorbed with alkaline water at 90°C. The remaining colourless solid residue

and a brown organic extract did not exhibit significant activity in the

photomineralization of 4-CP. However, after re-assembling of these two components,

the activity of the original VLPcom was regenerated. Additionally, modifying self-

prepared TiO2 with the organic extract afforded a photocatalyst inducing ca. 50%

mineralization. Thus, it was concluded that the visible light activity of VLPcom and

probably of other reported so-called “C-doped” titania does not originate from lattice

defects (e.g. F-centers) or substitutional carbon atoms, as generally proposed, but from

an aromatic hydrocarbon sensitizer. Unfortunately, all attempts failed to isolate and

solve the structure of the various compounds present in the organic extract. As

a characteristic property the extract exhibited at room temperature moderate emission

peaks at 380 and 400 nm upon excitation at 㮰 = 254 nm (Chapter 7).

10. Summary

85

In Chapter 8 titanium hydroxide and commercial TiO2 - UVLP were modified with

aqueous solutions of ethylene glycol. Additionally, UVLP was modified also with

pentaerythritol to have a material comparable with commercial VLPcom. All these

powders induced a relative high mineralization of 4-CP of ca. 60-70%. Different from

VLPcom, self-prepared pentaerythritol derived UVLP-PE and ethylene glycol derived

UVLP-EG contain a sensitizer which could not be desorbed from the titania surface

either under severe alkaline or acidic conditions. Although, again in this case a brown

extract was obtained upon alkaline treatment, the solid residue had an improved

activity and re-assembling did not improve its activity. This failure of the re-

assembling reaction became understandable when it was observed that in the emission

spectrum the characteristic peak at 380 nm was absent. TiO2-EG as prepared from self-

prepared titanium hydroxide loses a significant part of its activity after alkaline

treatment.

Finally renewable organic modifiers – wood shavings – were also employed to

obtain Vis light active titania. The resulting powders induced ca. 35% mineralization

of 4-CP. Unfortunately, black particles of coke or not completely burnt larch

components were found in the 4-CP suspensions. Similarly to results from Chapter 8

the larch modified material contains an organic sensitizer which could not be desorbed

under alkaline conditions (Chapter 9).

In summary; the following general conclusions can be drawn:

- Graphite oxides, organic acids, polyols, toluene, and wood shavings upon

heating to 200-500°C in the presence of anatase produced powders active in

visible light mineralization of 4-chlorophenol. The quasi-Fermi level of these

materials is in the range of −0.6 to −0.5 V.

- This modification is successful only when the density of titania surface OH

groups is high enough. The presence of an organic sensitizer is suggested by

the observation that prolonged heating at 400°C destroys the visible light

activity.

- Depending on the details of the modification process, the sensitizer may be

desorbed from the titania surface by an alkaline treatment.

10. Summary

86

- These results clearly prove that the visible light activity of most of the so-

called “carbon-doped” or “carbon-modified” titania photocatalysts originates

from a covalently attached organic sensitizer, and not from lattice defects or

carbidic species as proposed in the literature.

10. Zusammenfassung

87

10. Zusammenfassung

Vor Beginn dieser Dissertation gab es in der Literatur widersprechende

Vorschläge, welche Kohlenstoffspezies für die photokatalytische Aktivität sogenannter

„Kohlenstoff-dotierter“ Titandioxide mit sichtbarem Licht verantwortlich sei. In der

vorliegenden Arbeit wurden deshalb verschiedene Titandioxidpulver mit einigen

anorganischen und organischen Kohlenstoffverbindungen modifiziert. Da einer der

Vorschläge auf kondensierte Aromaten als photochemisch aktive Kohlenstoffspezies

hinwiesen, wurden zunächst Graphit/Graphitoxide mit Titandioxid bei 200-600°C

umgesetzt. Die erhaltenen Pulver besaßen allerdings in der Mineralisierung von 4-

Chlorophenol (4-CP) nur eine geringe photokatalytische Aktivität (Kapitel 4). Diese

verbesserte sich auf etwa 30%, wenn Phthalsäure oder Trimellitsäure für die

Modifizierung verwendet wurden. Unmodifiziertes Titandioxid war unter den

Standardbelichtungsbedingungen (㮰 ≥ 455 nm) inaktiv. Eine Steigerung auf 40%

Mineralisierung wurde für Toluol-modifiziertes Titandioxid beobachtet.

Um grundlegende Information über den Ursprung der photokatalytischen

Aktivität mit sichtbarem Licht zu erhalten, wurde das kommerzielle Produkt VLPcom,

hergestellt mit Pentaerythrit als Modifizierungagens, untersucht. Behandlung mit

Natronlauge bei 90°C ergab einen weißen Rückstand und ein braunes Extrakt. Keines

der beiden besaß eine signifikante Aktivität im Abbau von 4-CP. Wurde dagegen der

Rückstand mit dem Extrakt vereinigt und auf 200°C erhitzt, wird das ursprüngliche

VLPcom regeneriert. Der gleiche aktive Photokatalysator wird auch erhalten, wenn

selbsthergestelltes Titandioxid mit dem Extrakt behandelt wird. Diese Ergebnisse

beweisen, dass die photokatalytische Aktivität von VLPcom und vermutlich der meisten

sogenannten „Kohlenstoff-dotierten“ Titandioxide auf die Anwesenheit eines

organischen Sensibilisators zurückzuführen ist. Die in der Literatur gemachten

Vorschläge, dass Gitterdefekte oder karbidischer bzw. atomarer Kohlenstoff für die

photokatalytische Aktivität essentiell ist, muß daher als nicht zu treffend betrachtet

werden. Intensive Untersuchungen zur Strukturaufklärung des braunen Extrakts

blieben leider erfolgslos. Als eine charakteristische Eigenschaft besitzt das Extrakt im

10. Zusammenfassung

88

Raumtemperatur – ein Emissionsspektrum (㮰exc = 254 nm) mit zwei Maxima bei 380

und 400 nm.

Im Kapitel 8 wird über Modifizierung von Titanhydroxid und kommerziellem

TiO2 – UVLP durch Ethylenglykol und Pentaerythrit berichtet. Die daraus erhaltenen

Photokatalysatoren induzieren die Mineralisierung von 4-CP zu 60-70%. Im

Unterschied zu kommerziellem VLPcom erhält man bei der Alkaliextraktion zwar ein

braunes Extrakt, allerdings führt dessen Umsetzung mit dem festen Rückstand oder

mit selbsthergestelltem Titandioxid nicht zu einer Regeneration der photokatalytischen

Aktivität. Im Einklang damit fehlt im Emissionsspektrum dieses Extrakts die

charakteristische Bande bei 380 nm.

Im letzten Kapitel (Kapitel 9) werden Sägespäne aus Lärchenholz als nachwachsende

Modifikatoren verwendet. Auch die daraus resultierenden Photokatalysatoren

ermöglichen eine Mineralisierung (35%) von 4-CP durch sichtbares Licht. Allerdings

sind sie nicht homogen, da sie unvollständig verbrannte Kohlenstoffpartikel enthalten.

Wie bei den in Gegenwart von Ethylenglykol oder Pentaerythrit erhaltenen

Photokatalysatoren wird von Natronlauge auch hier kein Photosensibilisator

desorbiert.

Die wichtigsten Ergebnisse lassen sich wie folgt zusammenfassen:

- Modifizierung von Titandioxid mit Graphitoxiden, organischen Säuren, Polyolen,

Toluol und Sägespänen bei 200-500°C ergibt Photokatalysatoren für die mit

sichtbarem Licht ablaufende Mineralisierung von 4-CP. Das quasi-Ferminiveau dieser

Halbleiter liegt bei −0.6 bis −0.5 V.

- Diese Modifizierung ist nur erfolgreich, wenn genügend freie OH Gruppen auf der

Titandioxidoberfläche vorhanden sind. Die Anwesenheit eines organischen

Sensibilisators in diesen modifizierten Titandioxiden wird durch die Zerstörung der

photokatalytischen Aktivität bestätigt, wenn die Pulver längere Zeit auf 400°C erhitzt

werden.

- In Abhängigkeit der Reaktionsbedingungen erhält man Photokatalysatoren die gegen

Alkalilauge stabil sind oder eine Extraktion des Sensibilisators ermöglichen.

10. Zusammenfassung

89

- Diese Ergebnisse beweisen, dass die photokatalytische Aktivität der meisten

sogenannten „Kohlenstoff-dotierten“ oder „Kohlenstoff-modifizierten“ Titandioxide

mit sichtbarem Licht auf einen kovalent gebundenen, organischen Sensibilisator

zurückzuführen ist, und nicht auf Gitterdefekte oder atomare Kohlenstoffspezies.

11. References

90

11. References

[1] S. E. Braslavsky, K. N. Houk, Pure Appl. Chem., 60, 1055 (1988).[2] J. W. Verhoeven, Pure Appl. Chem., 68, 2223 (1996).[3] R. Beranek, PhD, From Photocatalysis to Optoelectronic Switches: Studies of Visible

Light Active Photoelectrodes Based on Surface-Modified Titanium Dioxide, Friedrich-Alexander-Universität Erlangen-Nürnberg, 2007.

[4] M. R. Hoffmann, S. T. Martin, W. Choi, D. W. Bahnemann, Chem. Rev., 95, 69 (1995).

[5] R. B. Cundall, R. Rudham, M. S. Salim, J. Chem. Soc., Faraday Trans. 1, 72, 1642 (1976).

[6] D. T. Sawyer, M. J. Gibian, Tetrahedron, 35, 1471 (1979).[7] K. Okamoto, Y. Yamamoto, H. Tanaka, M. Tanaka, A. Itaya, Bull. Chem. Soc. Jpn.,

58, 2015 (1985).[8] H. Gerischer, A. Heller, J. Phys. Chem., 95, 5261 (1991).[9] A. Fujishima, T. N. Rao, D. A. Tryk, J. Photochem. Photobiol. C, 1, 1 (2000).[10] D. A. Tryk, A. Fujishima, K. Honda, Electrochim. Acta, 45, 2363 (2000).[11] O. Carp, C. L. Huisman, A. Reller, Prog. Solid State Chem., 32, 33 (2004).[12] A. L. Linsebigler, G. Lu, J. T. Yates, Jr., Chem. Rev., 95, 735 (1995).[13] T. L. Thompson, J. T. Yates, Jr., Top. Catal., 35, 197 (2005).[14] L. Palmisano, V. Augugliaro, A. Sclafani, M. Schiavello, J. Phys. Chem., 92, 6710

(1988).[15] A. Di Paola, E. Garcia-Lopez, S. Ikeda, G. Marci, B. Ohtani, L. Palmisano,

Catal. Today, 75, 87 (2002).[16] L. Palmisano, M. Schiavello, A. Sclafani, C. Martin, I. Martin, V. Rives, Catal. Lett.,

24, 303 (1994).[17] N. Serpone, D. Lawless, J. Disdier, J.-M. Herrmann, Langmuir, 10, 643 (1994).[18] J. Soria, J. C. Conesa, V. Augugliaro, L. Palmisano, M. Schiavello, A. Sclafani,

J. Phys. Chem., 95, 274 (1991).[19] W. Choi, A. Termin, M. R. Hoffmann, J. Phys. Chem., 98, 13669 (1994).[20] S. T. Martin, C. L. Morrison, M. R. Hoffmann, J. Phys. Chem., 98, 13695 (1994).[21] M. I. Litter, J. A. Navio, J. Photochem. Photobiol., A, 98, 171 (1996).[22] M. Anpo, Pure Appl. Chem., 72, 1787 (2000).[23] M. Anpo, M. Takeuchi, J. Catal., 216, 505 (2003).[24] R. Janisch, P. Gopal, N. A. Spaldin, J. Phys.: Condens. Matter, 17, R657 (2005).[25] Y. Wang, Y. Hao, H. Cheng, H. Ma, B. Xu, W. Li, S. Cai, J. Mater. Sci., 34, 2773

(1999).[26] Y. Wang, H. Cheng, Y. Hao, J. Ma, W. Li, S. Cai, Thin Solid Films, 349, 120 (1999).[27] T. Hirai, I. Tari, J. Yamaura, Bull. Chem. Soc. Jpn., 51, 3057 (1978).[28] S. Sato, Chem. Phys. Lett., 123, 126 (1986).[29] R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki, Y. Taga, Science, 293, 269 (2001).[30] S. Sakthivel, H. Kisch, ChemPhysChem, 4, 487 (2003).[31] S. Sakthivel, M. Janczarek, H. Kisch, J. Phys. Chem. B, 108, 19384 (2004).[32] T. Lindgren, J. M. Mwabora, E. Avendano, J. Jonsson, A. Hoel, C.-G. Granqvist,

S.-E. Lindquist, J. Phys. Chem. B, 107, 5709 (2003).[33] H. Irie, Y. Watanabe, K. Hashimoto, J. Phys. Chem. B, 107, 5483 (2003).[34] C. Burda, Y. Lou, X. Chen, A. C. S. Samia, J. Stout, J. L. Gole, Nano Lett., 3, 1049

(2003).[35] M. Sathish, B. Viswanathan, R. P. Viswanath, C. S. Gopinath, Chem. Mater., 17, 6349

(2005).[36] D. Mitoraj, H. Kisch, Angew. Chem., Int. Ed., 47, 9975 (2008).[37] D. Mitoraj, H. Kisch, Chem. Eur. J., 16, 261 (2010).[38] I.-C. Kang, Q. Zhang, S. Yin, T. Sato, F. Saito, Appl. Catal., B, 80, 81 (2008).[39] Y. Li, D.-S. Hwang, N. H. Lee, S.-J. Kim, Chem. Phys. Lett., 404, 25 (2005).

11. References

91

[40] S. Y. Treschev, P.-W. Chou, Y.-H. Tseng, J.-B. Wang, E. V. Perevedentseva, C.-L. Cheng, Appl. Catal., B, 79, 8 (2008).

[41] C. Xu, R. Killmeyer, M. L. Gray, S. U. M. Khan, Appl. Catal., B, 64, 312 (2006).[42] W. Ren, Z. Ai, F. Jia, L. Zhang, X. Fan, Z. Zou, Appl. Catal., B, 69, 138 (2007).[43] H. Wang, Z. Wu, Y. Liu, J. Phys. Chem. C, 113, 13317 (2009).[44] L. Lin, W. Lin, Y. X. Zhu, B. Y. Zhao, Y. C. Xie, Y. He, Y. F. Zhu, J. Mol. Catal.

A: Chem., 236, 46 (2005).[45] S. Yin, M. Komatsu, Q. Zhang, F. Saito, T. Sato, J. Mater. Sci., 42, 2399 (2007).[46] S. Sakthivel, H. Kisch, Angew. Chem., Int. Ed., 42, 4908 (2003).[47] M. Janus, B. Tryba, M. Inagaki, A. W. Morawski, Appl. Catal., B, 52, 61 (2004).[48] S. Lee, C. Y. Yun, M. S. Hahn, J. Lee, J. Yi, Korean J. Chem. Eng., 25, 892 (2008).[49] E. A. Reyes-Garcia, Y. Sun, K. R. Reyes-Gil, D. Raftery, Solid State Nucl. Magn.

Reson., 35, 74 (2009).[50] Y.-H. Tseng, C.-S. Kuo, C.-H. Huang, Y.-Y. Li, P.-W. Chou, C.-L. Cheng,

M.-S. Wong, Nanotechnology, 17, 2490 (2006).[51] P. W. Chou, S. Treschev, P. H. Chung, C. L. Cheng, Y. H. Tseng, Y. J. Chen,

M. S. Wong, Appl. Phys. Lett., 89, 131919/1 (2006).[52] G. Yu, Z. Chen, Z. Zhang, P. Zhang, Z. Jiang, Catal. Today, 90, 305 (2004).[53] G. Wu, T. Nishikawa, B. Ohtani, A. Chen, Chem. Mater., 19, 4530 (2007).[54] C. Lettmann, K. Hildenbrand, H. Kisch, W. Macyk, W. F. Maier, Appl. Catal., B,

32, 215 (2001).[55] Y. Xie, X. Zhao, Y. Chen, Q. Zhao, Q. Yuan, J. Solid State Chem., 180, 3576 (2007).[56] H. Irie, Y. Watanabe, K. Hashimoto, Chem. Lett., 32, 772 (2003).[57] H. Liu, A. Imanishi, Y. Nakato, J. Phys. Chem. C, 111, 8603 (2007).[58] H. Irie, S. Washizuka, K. Hashimoto, Thin Solid Films, 510, 21 (2006).[59] Y. Choi, T. Umebayashi, M. Yoshikawa, J. Mater. Sci., 39, 1837 (2004).[60] D. Gu, Y. Lu, B.-c. Yang, Y.-d. Hu, Chem. Commun. (Cambridge, U. K.), 2453

(2008).[61] Y. Cheng, H. Sun, W. Jin, N. Xu, Chem. Eng. J. (Amsterdam, Neth.), 128, 127 (2007).[62] M.-S. Wong, S.-W. Hsu, K. K. Rao, C. P. Kumar, J. Mol. Catal. A: Chem., 279, 20

(2008).[63] N. Serpone, J. Phys. Chem. B, 110, 24287 (2006).[64] V. N. Kuznetsov, N. Serpone, J. Phys. Chem. B, 110, 25203 (2006).[65] P. Zabek, J. Eberl, H. Kisch, Photochem. Photobiol. Sci., 8, 264 (2009).[66] D. Chen, Z. Jiang, J. Geng, Q. Wang, D. Yang, Ind. Eng. Chem. Res., 46, 2741 (2007).[67] N. V. Zhuravleva, Koks Khim., 35 (2007).[68] J. P. Lange, A. Gutsze, H. G. Karge, J. Catal., 114, 136 (1988).[69] H. G. Karge, J. P. Lange, A. Gutsze, M. Laniecki, J. Catal., 114, 144 (1988).[70] C.-S. Kuo, Y.-H. Tseng, C.-H. Huang, Y.-Y. Li, J. Mol. Catal. A: Chem., 270, 93

(2007).[71] C. Di Valentin, G. Pacchioni, A. Selloni, Chem. Mater., 17, 6656 (2005).[72] T. L. Thompson, J. T. Yates, Jr., Chem. Rev., 106, 4428 (2006).[73] A. M. Roy, G. C. De, N. Sasmal, S. S. Bhattacharyya, Int. J. Hydrogen Energy, 20,

627 (1995).[74] M. Muneer, J. Theurich, D. Bahnemann, J. Photochem. Photobiol., A, 143, 213

(2001).[75] G. Mailhot, M. Sarakha, B. Lavedrine, J. Caceres, S. Malato, Chemosphere, 49, 525

(2002).[76] D. Mitoraj, PhD, Origin of Visible Light Activity in Urea Modified Titanium Dioxide,

Friedrich-Alexander-Universität Erlangen-Nürnberg, 2009.[77] B. Brodie, Ann. Chim. Phys., 45, 351 (1855).[78] H. P. Boehm, M. Eckel, W. Scholz, Z. Anorg. Allg. Chem., 353, 236 (1967).[79] W. Scholz, H. P. Boehm, Z. Anorg. Allg. Chem., 369, 327 (1969).[80] U. Hofmann, R. Holst, Ber. Dtsch. Chem. Ges. B, 72B, 754 (1939).[81] M. Mermoux, Y. Chabre, A. Rousseau, Carbon, 29, 469 (1991).[82] T. Nakajima, Y. Matsuo, Carbon, 32, 469 (1994).[83] F. Cataldo, Fullerenes, Nanotubes, Carbon Nanostruct., 11, 1 (2003).

11. References

92

[84] C. Hontoria-Lucas, A. J. Lopez-Peinado, J. d. D. Loepz-Gonzalez, M. L. Rojas-Cervantes, R. M. Martin-Aranda, Carbon, 33, 1585 (1995).

[85] A. Lerf, H. He, T. Riedl, M. Forster, J. Klinowski, Solid State Ionics, 101-103, 857 (1997).

[86] T. Szabo, O. Berkesi, P. Forgo, K. Josepovits, Y. Sanakis, D. Petridis, I. Dekany, Chem. Mater., 18, 2740 (2006).

[87] G. Ruess, Monatsh. Chem., 76, 381 (1946).[88] W. S. Hummers, Jr., R. E. Offeman, J. Am. Chem. Soc., 80, 1339 (1958).[89] D. Gumy, S. A. Giraldo, J. Rengifo, C. Pulgarin, Appl. Catal., B, 78, 19 (2008).[90] J. Schnadt, J. N. O'Shea, L. Patthey, J. Schiessling, J. Krempasky, M. Shi,

N. Martensson, P. A. Bruhwiler, Surf. Sci., 544, 74 (2003).[91] Q. Guo, I. Cocks, E. M. Williams, Surf. Sci., 393, 1 (1997).[92] J. Sun, L. Yuan, K. Zhang, D. Wang, Thermochim. Acta, 343, 105 (2000).[93] A. Assabane, Y. A. Ichou, H. Tahiri, C. Guillard, J.-M. Hermann, Appl. Catal., B,

24, 71 (2000).[94] J. R. Anderson, Q. N. Dong, Y. F. Chang, R. J. Western, J. Catal., 127, 113 (1991).[95] B. An, S. Kaida, T. Miyake, T. Tani, I. Kashimoto, H. Kominami, J. Jpn. Pet. Inst.,

50, 283 (2007).[96] J. Theurich, M. Lindner, D. W. Bahnemann, Langmuir, 12, 6368 (1996).[97] E. A. Konstantinova, A. I. Kokorin, S. Sakthivel, H. Kisch, K. Lips, Chimia, 61, 810

(2007).[98] J. Orth-Gerber, H. Kisch, (Kronos International Inc., Germany). US 7 615, 512 B2,

2005.[99] K. Nagaveni, G. Sivalingam, M. S. Hegde, G. Madras, Appl. Catal., B, 48, 83 (2004).[100] J. Riga, J. J. Pireaux, R. Caudano, J. J. Verbist, Phys. Scr., 16, 346 (1977).[101] R. Larsson, B. Folkesson, Chemica Scripta, 9, 148 (1976).[102] X. Yang, C. Cao, K. Hohn, L. Erickson, R. Maghirang, D. Hamal, K. Klabunde,

J. Catal., 252, 296 (2007).[103] X. Zhang, M. Zhou, L. Lei, Carbon, 44, 325 (2005).[104] J. Tauc, R. Grigorovici, A. Vancu, J. Phys. Soc. Jpn., Suppl., 21, 123 (1966).[105] B. Karvaly, I. Hevesi, Z. Naturforsch., A, 26, 245 (1971).[106] R. Rahal, S. Daniele, L. G. Hubert-Pfalzgraf, V. Guyot-Ferreol, J.-F. Tranchant,

Eur. J. Inorg. Chem., 980 (2008).[107] T. Lana-Villarreal, A. Rodes, J. M. Perez, R. Gomez, J. Am. Chem. Soc., 127, 12601

(2005).[108] Y. Liu, J. I. Dadap, D. Zimdars, K. B. Eisenthal, J. Phys. Chem. B, 103, 2480 (1999).[109] M. Pal, J. G. Serrano, P. Santiago, U. Pal, J. Phys. Chem. C, 111, 96 (2007).[110] X. Jiang, T. Herricks, Y. Xia, Adv. Mater. (Weinheim, Ger.), 15, 1205 (2003).[111] D. Wang, R. Yu, Y. Chen, N. Kumada, N. Kinomura, M. Takano, Solid State Ionics,

172, 101 (2004).[112] H. K. Yu, T. H. Eun, G.-R. Yi, S.-M. Yang, J. Colloid Interface Sci., 316, 175 (2007).[113] X. Jiang, Y. Wang, T. Herricks, Y. Xia, J. Mater. Chem., 14, 695 (2004).[114] Q. Li, B. Liu, Y. Li, R. Liu, X. Li, D. Li, S. Yu, D. Liu, P. Wang, B. Li, B. Zou,

T. Cui, G. Zou, J. Alloys Compd., 471, 477 (2009).[115] H. Thoms, M. Epple, M. Fröba, J. Wong, A. Reller, J. Mater. Chem., 8, 1447 (1998).[116] Q. Xiao, J. Zhang, C. Xiao, Z. Si, X. Tan, Sol. Energy, 82, 706 (2008).[117] F. L. Benton, T. E. Dillon, J. Am. Chem. Soc., 64, 1128 (1942).[118] E. A. Hoffmann, T. Koertvelyesi, E. Wilusz, L. S. Korugic-Karasz, F. E. Karasz,

Z. A. Fekete, THEOCHEM, 725, 5 (2005).[119] S. Kim, W. Choi, J. Phys. Chem. B, 109, 5143 (2005).[120] A. G. Agrios, K. A. Gray, E. Weitz, Langmuir, 19, 1402 (2003).[121] A. G. Agrios, K. A. Gray, E. Weitz, Langmuir, 20, 5911 (2004).[122] S. Nagaoka, Y. Hamasaki, S.-i. Ishihara, M. Nagata, K. Iio, C. Nagasawa, H. Ihara,

J. Mol. Catal. A: Chem., 177, 255 (2002).

LebenslaufPersönliches:

Name: Przemysᐠaw Zၐbek

Geburtsdatum: 16. September 1982

Geburtsort: Krak㳀w

Eltern: Barbara Zၐbek, geb. Sobczak

Witold Zၐbek

Staatsangehörigkeit: polnisch

Familienstand: ledig

Studium:

September 2006 Beginn der Promotion an der Friedrich-Alexander Universität Erlangen-Nürnberg, Department Chemie und Pharmazie unter Anleitung von Herrn Prof. Dr. Horst Kisch

Oktober 2003 - Juni 2006 Studium der Pädagogik der Chemie an der Jagiellonen Universität in Krakau

Oktober 2001 - Juni 2006 Chemiestudium an der Jagiellonen Universität in Krakau mit dem Abschluss Magister

Schulbildung:

September 1997 - Juni 2001 Allgemeinbildende Oberschule in Krakau (abgeschlossen mit Abitur)

September 1989 - Juni 1997 Grundschule in Krakau

preparation and photocatalytic activity of carbon- modified titania

Documents