A. GJEVORI, B. ZIBERI: INFLUENCE OF DOPING WITH CARBON AND NITROGEN ON THE PHOTOACTIVITY ...
581–585
INFLUENCE OF DOPING WITH CARBON AND NITROGEN ON
THE PHOTOACTIVITY OF TiO
2
THIN FILMS OBTAINED
WITH MePIIID
VPLIV DOPIRANJA Z OGLJIKOM IN DU[IKOM NA
FOTOAKTIVNOST TANKIH FILMOV TiO
2
IZDELANIH Z MePIIID
Altin Gjevori
1*
, Bashkim Ziberi
2
1
Physics Engineering Department, Polytechnic University of Tirana, Mother Theresa Square 4, 1000, Tirana, Albania
2
Physics Department, University of Tetova, Ilinden Street nn, 1200, Tetova, Republic of North Macedonia
Prejem rokopisa – received: 2023-04-29; sprejem za objavo – accepted for publication: 2023-09-17
doi:10.17222/mit.2023.867
Titanium dioxide is well known as a photoactive material activated under ultraviolet irradiation. It is either employed as a
photocatalyst or it exhibits superhydrophilic behavior after obtaining reduced surface energy under illumination used for
self-cleaning or anti-fogging surfaces. For increasing the reactivity of thin films under solar illumination, a reduced band gap is
desired. Doping with transition metals or nitrogen has been reported in the literature. However, the incorporation of nitrogen
into a growing film is a much more complex process that is presently not completely understood. TiO2 thin layers were pro-
duced by metal plasma immersion ion implantation and deposition at room temperature at a pulse voltage of 0–5 kV and a duty
cycle of 9 % for an apparently amorphous layer. An auxiliary rf plasma source was employed to increase the growth rate at low
gas flow ratios. By adjusting the geometry between the incident ion beam, sputter target and substrate independently of the pri-
mary ion energy and species, a controlled deposition of samples was possible. Conventional ion implantation was employed to
implant either carbon or nitrogen ions below the surface for bandgap engineering. The resulting thin films were subsequently in-
vestigated for optical properties, stoichiometry, structural properties, surface topography and photoactivity.
Keywords: TiO2, MePIIID, SRIM, photoactivity
Titanov dioksid (TiO2) je dobro znani fotoaktivni material, ki se aktivira pod vplivom sevanja ultravijoli~ne svetlobe. Zato se
lahko uporablja kot fotokatalizator. Pod vplivom osvetljevanja se obna{a kot super hidrofilni (dobro omo~ljiv) material, tako
zmanj{a povr{insko energijo in je lahko zato uporaben kot samo ~istilna povr{ina in povr{ina, ki se ne rosi. Pove~anje
reaktivnosti tankih filmov pod vplivom sevanja son~ne svetlobe zahteva pove~anja {irine prepovedanega pasu za elektrone. O
dopiranju TiO2 s prehodnimi kovinami ali du{ikom so poro~ali `e {tevilni avtorji v znanstveni literaturi. Vendar pa je vgradnja
du{ikovih atomov med rastjo oziroma tvorbo tankega filma zelo kompleksen proces, ki {e ni popolnoma raziskan. Avtorja v
pri~ujo~em ~lanku opisujeta izdelavo tankih filmov TiO2 s plazemsko imerzijsko (potopno) ionsko implantacijo kovin
(MePIIID; angl.: Metal Plasma Immersion Ion Implantation and Deposition) pri sobni temperaturi, napetostnih pulzih od 0 kV
do5kVinz9%izk oristkom za izdelavo navidezno amorfnega filma. Uporabila sta {e pomo`ni izvor radiofrekven~ne plazme
za pove~anje hitrosti rasti tankih plasti pri majhnih pretokih plina. Z namestitvijo geometrije med vpadnim kotom curka ionov,
tar~e za napr{evanje in podlage so lahko neodvisno od primarne energije ionov in vrste kontrolirali rast vzorca tankega filma.
In`eniring {irine prepovedanega pasu za elektrone sta avtorja izvajala s konvencionalno ionsko implantacijo za implantacijo
ogljikovih ali du{ikovih ionov pod povr{ino. Sledile so raziskave na izdelanih vzorcih tankih plasti (filmih) oziroma dolo~itev
njihovih opti~nih lastnosti, stehiometrije, strukturnih lastnosti, povr{inske topografije in fotoaktivnosti.
Klju~ne besede: TiO2, MePiiiD, SRIM, fotoaktivnost
1 INTRODUCTION
Titanium dioxide is well known as a photoactive ma-
terial activated under ultraviolet (UV) irradiation,
1,2
and
employed either as a photocatalyst or exhibiting
superhydrophilic behavior after a reduction of the sur-
face energy under illumination needed for self-cleaning
or anti-fogging surfaces.
3
As an alternative to powders,
TiO
2
can be produced as thin films using chemical
sol-gel processes, operating in air, or physical vapor de-
position (PVD) processes where either high temperature
or increased ion energy is necessary to obtain a
photoactive phase.
4
An even coating of membranes is
possible using a more complex process.
5
There, tempera-
ture sensitive substrates require low temperatures, ideally
near room temperature where an amorphous photoactive
phase can be formed under certain conditions.
6,7
Regardless of the phase composition, the band gap is
always more than 3 eV (the exact value depends on the
polymorph), thus necessitating UV-A radiation for an ac-
tivation. For increasing the reactivity of thin films under
solar illumination, a reduced band gap is desired. Doping
with transition metals or with nitrogen has been reported
in the literature.
8,9
The latter dopant is – theoretically –
readily accessible during PVD processes. However, the
incorporation of nitrogen into a growing film, in contrast
to ion implantation into TiO
2
thin films or selective oxi-
dation of TiN, is a much more complex process, which is
presently not completely understood.
10
Alternatively,
TiO
2
nanotubes with a reduced band gap, visi-
Materiali in tehnologije / Materials and technology 57 (2023) 6, 581–585 581
UDK 661.882:539.21 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 57(6)581(2023)
*Corresponding author's e-mail:
a.gjevori@fimif.edu.al (Altin Gjevori)

ble-light-active TiO
2
nanorods N-doped with a hydro-
thermal treatment or hydrazine doping of brookite
nanorods at 200 °C for 18 h have been proposed re-
cently.
11–13
Despite a large amount of published work,
even the photoactivity of pure TiO
2
thin films is still a
matter of controversy.
Therefore, an experiment was set up, which provided
an opportunity to reduce the bandgap of TiO
2
thin films,
increasing the surface energy and leading to improved
photoactive properties. TiO
2
thin films were produced
with metal plasma immersion ion implantation and depo-
sition (MePIIID). Then, we employed mixtures of ener-
gies/species, 65 keV C + 75 keV N, 30 keV C + 35 keV
Nand35keVN+75keVN,toincrease the local dopant
concentration and adjust the depth of the penetration
range of the light. The purpose was to investigate
whether synergy effects exist for a combination of car-
bon and nitrogen dopants. It was found that ion implan-
tation at concentrations of less than 10 x/% can lead to
an increased photoactivity under UV illumination. Re-
ducing the bandgap with carbon and/or nitrogen implan-
tation before changing from a semiconductor to a
semimetal is challenging. This was known for direct
co-deposition during PVD by vacuum arc, and now it
was shown to be also true for the post-implantation after
MePIIID.
2 EXPERIMENTAL PART
TiO
2
thin films are produced with MePIIID. A d.c.
cathodic arc with a pure titanium (99.99 %) cathode,
running at 100 A was employed to generate Ti ions. An
oxygen supply of 45 sccm to 70 sccm was used to estab-
lish a working pressure between 0.03 Pa and 0.3 Pa.
Converting the nominal Ti ion current from the cathode
of about 10 A at an average charge state of 2.1 into a par-
ticle flux and comparing this value with the oxygen flow,
a O/Ti ratio of roughly 2.5:1 to 3.5:1 was established in
the gas phase (assuming a homogeneous distribution
within the vacuum chamber, which is definitely not true
for the Ti flux). Silicon (100) coupons were used as the
substrate material, mounted at a fixed distance of 39 cm
from the cathode. No additional heating system was in-
stalled, thus the temperature of the substrate always
started at room temperature (RT), rising to 50 °C or less
at the end of the experiment. The nominal kinetic ener-
gies of 10–50 eV for Ti ions emanating from the d.c. arc
were enhanced by negative high voltage pulses between
1 and 5 kV with a length of 30 μs at a repetition rate of
1–3 kHz (a duty cycle of 3–9 %). Furthermore, a
40.68 MHz rf plasma source was selectively employed at
a fixed power of 150 W to create a background oxygen
plasma. Without arc, an oxygen plasma density of
6×10
9
cm
–3
at an electron temperature of 1 eV was mea-
sured by a Langmuir probe. The utilized geometry with
the substrates facing towards the arc is depicted in Fig-
ure 1. A total deposition time of 5 min was used for all
samples.
The film morphology was investigated with scanning
electron microscopy (SEM), while the surface roughness
was determined by atomic force microscopy (AFM).
X-ray diffraction (XRD) measurements were performed
in  /2 -geometry. Elastic recoil detection analysis
(ERDA) measurements were carried out using 200 MeV
197
Au
15+
ions at a 19° incident angle and a detector placed
at a scattering angle of 37° for a typical fluence of
3×10
11
ions. Using the integrated count rates, O/Ti ra-
tios with a relative accuracy of 0.25 x/% were derived.
Raman spectra were measured using a LabRam HR Evo-
lution Raman spectrometer with a spectral resolution of
0.7 cm
–1
at 473 nm laser excitation (a grid with
1800 L/mm). The entrance slit of the spectrometer was
150 μm, the pin-hole of the confocal setup was 300 μm
in diameter. The film thickness was derived from second-
ary ion mass spectrometry (SIMS) in a time-of-flight
setup, stopping right at the interface and determining the
sputter rate and absolute film thickness with mechanical
profilometry.
The surface energy was derived from contact angle
measurements (CAM) with a data analysis according to
the Owens–Wendt method.
14
Contact angle measure-
ments were performed on TiO
2
samples using the Krüss
contact angle measuring system G2, and the results were
obtained from the DSA II software (drop shape analysis
II). In the dynamic CAM mode, two different liquids
were used, i.e., deionized water and ethylene glycol, em-
ploying a manual drop position on the surface with 2–4
drops (because of high wettability of the sample surfaces
after UV irradiation). From each liquid with a start vol-
ume of 3 μL and end volume of 4 μL, 10 successive
snapshots were taken for each drop.
Photoactivity was examined by exposing the samples
to UV-A light, generated from an actinic tube in a spec-
A. GJEVORI, B. ZIBERI: INFLUENCE OF DOPING WITH CARBON AND NITROGEN ON THE PHOTOACTIVITY ...
582 Materiali in tehnologije / Materials and technology 57 (2023) 6, 581–585
Figure 1: Schematic of the MePIIID deposition system with an auxil-
iary plasma source

tral range of 300–460 nm and with a maximum of
365 nm, at an intensity of 1 mW/cm
2
, for up to 3 h.
Then the samples were doped with carbon and nitro-
gen, using a conventional ion implanter, IMC-200. The
aim was to reduce the band gap of the UV active TiO
2
thin films as well as to investigate the effect of implanta-
tion on non-photoactive TiO
2
thin films. The following
implantation parameters were used: energies for C
+
ions
– 30 keV and 65 keV, energies for N
+
ions – 35 keV and
75 keV, the average implantation time – 35 min, and the
total fluency–2×10
16
cm
–2
.
3 RESULTS AND DISCUSSIONS
The films produced with this method were slightly
substoichiometric titania films with an O/Ti ratio of
1.90–1.95, as determined by the ERDA. The film thick-
ness was always close to 300 nm, indicating a growth
rate of about 1 nm/s and no strong influence of the vary-
ing ion bombardment on the growth rate. No crystalline
anatase or rutile phases could be detected at RT for the
pulse voltage range from 0 kV to 5 kV, being independ-
ent of the gas flow or plasma activation, and indicating
either an amorphous structure or nanocrystallites with a
size of less than 5 nm. Similar results were obtained for
the film morphology when investigating thin-film
cross-sections with SEM. No contrast was visible for a
film deposited at room temperature witha5k Vpulse
voltage. AFM results confirmed SEM results with regard
to the morphology; the measurements showed that the
surface was very smooth, i.e., the root mean square (rms)
roughness was 0.22 nm. Detailed results for the
structural properties, surface topography and influence
of the deposition conditions on the growth rate of the
TiO
2
thin films produced with this method are given in
previous works.
10,15,16
From the samples produced, we selected four pairs,
within which one sample was promising to be
photoactive, while the other was not. The sample param-
eters and photoactivity characteristics are depicted in Ta-
ble 1.
Then, the above listed amorphous samples were
doped with carbon and nitrogen, using the IMC-200 ion
implanter. The aim was to reduce the band gap of the UV
active TiO
2
thin films as well as to investigate the effect
of implantation on non-photoactive TiO
2
thin films.
The concentration profiles of carbon and nitrogen,
calculated with the stopping and range of ions in matter
(SRIM) analysis, are shown in Figure 2.
After implantation the samples were illuminated with
UV light for three hours and the surface energy was de-
rived from contact angle measurements. The results of
contact angle measurements are presented in the Fig-
ure 3. We observe an increase in the surface energy of
A. GJEVORI, B. ZIBERI: INFLUENCE OF DOPING WITH CARBON AND NITROGEN ON THE PHOTOACTIVITY ...
Materiali in tehnologije / Materials and technology 57 (2023) 6, 581–585 583
Table 1: Selected sample list from the photoactivity results obtained with contact angle measurements
No. of
pairs
Sample ID HV Pulse length Repetition rate RF plasma
Distance from
axis
Photoactivity (surface
energy, mN/m)
1
9020 0 kV 30 μs 3 kHz 150 W/1 % 20 cm 42.3 (No)
9021 5 kV 30 μs 3 kHz 150 W/1 % 20 cm 82.9 (Yes)
2
9022 3 kV 30 μs 3 kHz 150 W/1 % 20 cm 100 (Yes)
9024 5 kV 30 μs 3 kHz 150 W/1 % 10 cm 40.6 (No)
3
9025 3 kV 30 μs 3 kHz 150 W/1 % 10 cm 95.2 (Yes)
9027 0 kV 30 μs 3 kHz 150 W/1 % 10 cm 40.3 (No)
4
9029 3 kV 30 μs 3 kHz 150 W/1 % 0 cm 73.6 (Yes)
9030 1 kV 30 μs 3 kHz 150 W/1 % 0 cm 47.5 (No)
Substrate holder diameter  = 7.8 cm (9020–9031)
Figure 3: Surface energy of selected samples before and after implan-
tation with/without UV-illumination
Figure 2: Concentration profiles of carbon and nitrogen

some samples after implantation, but for other samples
no change or decrease in the surface energy can be seen
on this graph.
For clarifying these findings, Raman measurements
were performed with a 473 nm laser excitation. Two
spectra, one with a laser intensity of 2 mW on a sample
and one with a laser intensity of 8 mW on a sample were
measured at different positions. The acquisition time was
20 s for each measurement. In order to filter out random
spikes, each measurement was done twice.
As seen in Figure 4, the spectra obtained here for the
films after implantation prepared with MePIIID are very
similar to that of the nanocrystalline rutile layer on Ti,
shown as a reference. The low intensity of the signal
must be a combination of a small grain size and low
crystalline quality.
18
The next experiment consisted of doping samples
with carbon and nitrogen using ion implanter IMC-200.
We employed mixtures of energies/species, 65 keV C +
7 5k e VN ,3 0k e VC+3 5k e VN ,a n d3 5k e VN+
75 keV N, to increase the local dopant concentration and
adjust the depth of penetration range of the light. The
purpose was to investigate whether synergy effects exist
for a combination of carbon and nitrogen dopants. We
had combinations of fluencies up to7×10
16
cm
–2
, corre-
sponding to local concentrations of up to 10 at.%. Fig-
ure 5 presents calculated depth profiles for selected sam-
ples.
The effect of the ion implantation fluency on the
photoactivity is depicted in Figure 6 below. No system-
atic differences were found before illumination. From
the data obtained for the surface energies after illumina-
tion, a strong, positive influence of the ion fluency on the
photoactivity is observed. The time between the end of
the illumination and the contact angle measurements was
always less than 1 h; no influence of the small changes in
this delay time on the results was observed. For the sam-
ples implanted with combinations of fluencies up to
7×10
16
cm
–2
, the surface energies of around 100 mN/m,
indicating photoactivity, were obtained.
A. GJEVORI, B. ZIBERI: INFLUENCE OF DOPING WITH CARBON AND NITROGEN ON THE PHOTOACTIVITY ...
584 Materiali in tehnologije / Materials and technology 57 (2023) 6, 581–585
Figure 4: Raman spectra of selected samples after implantation, together with the reference spectra of a rutile thin film and anatase powder (APS
32 nm, Alfa-Aesar).
17
The prominent peak at 520 cm
–1
arises from the Si substrate.
Figure 6: Variation of surface energy with ion fluency before and af-
ter illumination
Figure 5: Concentration profiles of carbon and nitrogen obtained with
SRIM

4 CONCLUSIONS
Using the PVD processes – metal plasma immersion
ion implantation and deposition and conventional ion im-
plantation – the formation of photoactive TiO
2
thin films
is reported.
It can be stated that ion implantation at concentra-
tions lower than 10 at.% can lead to an increased
photoactivity under UV illumination. Reducing a band
gap with carbon and/or nitrogen implantation before
changing a semiconductor into a semimetal is challeng-
ing. This was known about direct co-deposition during
PVD by a vacuum arc, and now it is shown to be also
true for post-implantation after metal plasma immersion
ion implantation and deposition. Furthermore, the doping
process increases the surface energy, which leads to im-
proved photoactive properties. The doping influence on
the surface energy is interesting, promising and not often
reported in the literature. However, further work is nec-
essary to elucidate its mechanisms and establish whether
this is a doping or a damage effect.
Acknowledgment
This study was supported by the DAAD Grant, Re-
search Stays for University Academics and Scientists,
(57378441), personal ref. no. 91656534. We would like
to thank S. Mandl, the supervisor, who also assisted us
with Raman and IMC-200 Implanter measurements, and
Darina Manova for valuable discussions during the eval-
uation of the results.
5 REFERENCES
1
W. Hofmeister, E. Tillmanns, W. H. Bauer, Refinement of barium
tetratitanate, BaTi4O9, and hexabarium 17-titanate, Ba6Ti17O40, Acta
Crystallogr. C, 40 (1984), 1510–1512, doi:10.1107/
S0108270184008544
2
A. Fujishima, K. Honda, Electrochemical photolysis of water at a
semiconductor electrode, Nature, 238 (1972), 37–38, doi:10.1038/
238037a0
3
R. Wang, K. Hashimoto, A. Fujishima, M. Chikuni, E. Kojima, A.
Kitamura, M. Shimohigoshi, T. Watanabe, Light-induced amphi-
philic surfaces, Nature, 388 (1997), 431–432, doi:10.1038/41233
4
P. Löbl, M. Huppertz, D. Mergel, Nucleation and growth in TiO2
films prepared by sputtering and evaporation, Thin Solid Films, 251
(1994), 72–79, doi:10.1016/0040-6090(94)90843-5
5
K. Fischer, M. Kühnert, R. Gläser, A. Schulze, Photocatalytic degra-
dation and toxicity evaluation of diclofenac by nanotubular
titaniumdioxide–PES membrane in a static and continuous setup,
RSC Adv., 5 (2015), 16340–16348, doi:10.1039/C4RA16219F
6
J. M. Schneider, S. Rohde, W. D. Sproul, A. Matthews, Recent devel-
opments in plasma assisted physical vapour deposition, J. Phys. D.
Appl. Phys., 33 (2000), R173–R186, doi:10.1088/0022-3727/33/
18/201
7
D. Manova, A. Gjevori, F. Haberkorn, J. Lutz, S. Dimitrov, J. W.
Gerlach, E. Valcheva, S. Mändl, Formation of hydrophilic and
photocatalytically active TiO2 thin films by plasma based ion im-
plantation and deposition, Phys. Status Solidi A, 206 (2009) 71–77,
doi:10.1002/pssa.200723600
8
T. Morikawa, R. Asahi, T. Ohwaki, K. Aoki, Y. Taga, Band-gap nar-
rowing of titanium dioxide by nitrogen doping, Jpn. J. Appl. Phys.,
40 (2001), L561–L563, doi:10.1143/JJAP.40.L561
9
D. H. Lee, Y. S. Cho, W. I. Yi, T. S. Kim, J. K. Lee, H. Jin Jung,
Metalorganic chemical vapor deposition of TiO2:N anatase thin film
on Si substrate, Appl. Phys. Lett., 66 (1995), 815–816, doi:10.1063/
1.113430
10
I. Asenova, D. Manova, S. Mändl, Incorporation of nitrogen into
TiO2 thin films during PVD processes, J. Phys. Conf. Ser., 559
(2014), 012008, doi:10.1088/1742-6596/559/1/012008
11
B. Choudhury, S. Bayan, A. Choudhury, P. Chakraborty, Narrowing
of band gap and effective charge carrier separation in oxygen defi-
cient TiO2 nanotubes with improved visible light photocatalytic ac-
tivity, J. Colloid Interface Sci., 465 (2016), 1–10, doi:10.1016/
j.jcis.2015.11.050
12
S. Abu Bakar, G. Byzynski, C. Ribeiro, Synergistic effect on the
photocatalytic activity of N-doped TiO2 nanorods synthesised by
novel route with exposed (110) facet, J. Alloys Compd., 666 (2016),
38–49, doi:10.1016/j.jallcom.2016.01.112
13
J. Pan, S. P. Jiang, Synthesis of nitrogen doped faceted titanium-
dioxide in pure brookite phase with enhanced visible light
photoactivity, J. Colloid Interface Sci., 469 (2016), 25–30,
doi:10.1016/j.jcis.2016.02.013
14
D. K. Owens, R. C. Wendt, Estimation of the surface free energy of
polymers, J. Appl. Polym. Sci., 13 (1969) 8, 1741–1747,
doi:10.1002/app.1969.070130815
15
A. Gjevori, K. Nonnenmacher, B. Ziberi, D. Hirch, J. W. Gerlach, D.
Manova, S. Mändl, Investigation of nucleation and phase formation
of photocatalytically active TiO2 films by MePBIID, Nucl. Instrum.
Method. Phys. Res. B, 267 (2009), 1658–1661, doi:10.1016/
j.nimb.2009.02.047
16
A. Gjevori, J. W. Gerlach, D. Manova, W. Assmann, E. Valcheva, S.
Mändl, Influence of auxiliary plasma source and ion bombardment
on growth of TiO2 thin films, Surface & Coatings Technology,
(2011), S232–S234, doi:10.1016/j.surfcoat.2011.02.046
17
G. Thorwarth, S. Ma¨ndl, B. Rauschenbach, Plasma immersion ion
implantation using titanium and oxygen ions, Surf. Coat. Technol.,
128–129 (2000), 116–120, doi:10.1016/S0257-8972(00)00586-7
18
S. Ma¨ndl, G. Thorwarth, M. Schreck, B. Stritzker, B. Rauschenbach,
Raman study of titanium oxide layers produced with plasma immer-
sion ion implantation, Surf. Coat. Technol., 125 (2000) 1, 84–88,
doi:10.1016/S0257-8972(99)00559-9
A. GJEVORI, B. ZIBERI: INFLUENCE OF DOPING WITH CARBON AND NITROGEN ON THE PHOTOACTIVITY ...
Materiali in tehnologije / Materials and technology 57 (2023) 6, 581–585 585